University of Pennsylvania

ScholarlyCommons
Publicly Accessible Penn Dissertations

2017

Flavor Added: The Sciences Of Flavor And The

Industrialization Of Taste In America
Nadia Berenstein
University of Pennsylvania, [email protected]

Follow this and additional works at: https://repository.upenn.edu/edissertations

Part of the Food Science Commons, and the History Commons

Recommended Citation
Berenstein, Nadia, "Flavor Added: The Sciences Of Flavor And The Industrialization Of Taste In America" (2017). Publicly Accessible
Penn Dissertations. 2715.
https://repository.upenn.edu/edissertations/2715

This paper is posted at ScholarlyCommons. https://repository.upenn.edu/edissertations/2715

For more information, please contact [email protected].
Flavor Added: The Sciences Of Flavor And The Industrialization Of Taste
In America
Abstract
In the mid-nineteenth century, flavor additives - volatile organic chemicals with desirable aromatic qualities -
began to be used to flavor sugary confections, carbonated beverages, and other mass-marketed delights. By the
mid-twentieth century, added flavors had become ubiquitous in processed, packaged foods; a sophisticated,
technoscientific, and globe-spanning industry had emerged that specialized in their production. Drawing on
history of science and technology, business history, and cultural history, "Flavor Added" investigates the
history of synthetic flavor additives, the systems of scientific and technical knowledge that emerged to create
these substances, and their social and cultural consequences. Focusing primarily on the United States, "Flavor
Added" traces the origins and development of both flavor chemistry and sensory science, illuminating their
entangled roots in private industry, regulatory laboratories, USDA research experiment stations, the US
military, and academic institutions. Several chapters take on the technologies and tools of flavor creation,
including the taste panel, the flavor profile, and the combined gas chromatograph-mass spectrometer. This
dissertation also documents the professional history of flavorists, the highly specialized scientific craft-workers
who develop and design flavor additives.

Degree Type
Dissertation

Degree Name
Doctor of Philosophy (PhD)

Graduate Group
History and Sociology of Science

First Advisor
John Tresch

Keywords
Flavor Additives, Flavor Chemistry, Flavor industry, Food Industrialization, Food Technology, Sensory
Science

Subject Categories
Food Science | History

This dissertation is available at ScholarlyCommons: https://repository.upenn.edu/edissertations/2715

FLAVOR ADDED: THE SCIENCES OF FLAVOR AND THE INDUSTRIALIZATION OF TASTE IN

AMERICA

Nadia Berenstein

A DISSERTATION

in

History and Sociology of Science

Presented to the Faculties of the University of Pennsylvania

in

Partial Fulfillment of the Requirements for the

Degree of Doctor of Philosophy

2018

Supervisor of Dissertation

John Tresch

Associate Professor, History and Sociology of Science, University of Pennsylvania

Graduate Group Chairperson

Beth Linker

Associate Professor, History and Sociology of Science, University of Pennsylvania

Dissertation Committee

John Tresch, Associate Professor, History and Sociology of Science, University of Pennsylania

Adelheid Voskuhl, Associate Professor, History and Sociology of Science, University of

Pennsylvania

Steven Shapin, Franklin L. Ford Research Professor in the History of Science, Harvard University
FLAVOR ADDED: THE SCIENCES OF FLAVOR AND THE INDUSTRIALIZATION OF TASTE IN

AMERICA

COPYRIGHT

2018

NADIA BERENSTEIN

This work is licensed under the

Creative Commons Attribution-
NonCommercial-ShareAlike 3.0
License

To view a copy of this license, visit

https://creativecommons.org/licenses/by-nc-sa/3.0/us/
 ACKNOWLEDGMENTS

While I was in the thick of writing this thing, I’ll admit I was pretty miserable.

How does anyone do this? I was desperate for clues. I developed the habit of closely

reading the acknowledgements sections of other dissertations, hoping that somehow in

the gratitude-chains and the naming-names, the alchemical formula, the one weird trick

for a done dissertation, would rise shimmering and present itself to me.

So my first acknowledgment here is to you, reader, and to those among you who

approach the thankless task of reading a turgid, tumultuous dissertation through the

rockiest of roads, the acknowledgements section, and especially to graduate students or

other thinkers, writers, artists who are puzzling over not only a particular set of questions

relating to flavor and its role in our sensual lives, but also, perhaps, struggling with how

to write about this, or anything, at all.

Ultimately, I don’t really know how it happened, getting to the end. The whole

thing was a mess, a disaster, pages and pages of unmoored maunderings. “There’s no

train of thought here,” I kept saying to myself, scrolling through the damage on Microsoft

Word. I had worked so long, done so much research, so many people had trusted me with

their stories, I cared so much about “my” subject — but all you could see was the sweat-

stains and the tear-stains and the ugly residues of exertion, none of the love. It was a

wreck, and so was I. It was August, or maybe it was December, or maybe it was March.

What did seasons matter? Each day was indistinguishable from the one previous or the

one succeeding, a dismal blur of pixels; nothing felt like progress.

iii
 The secret to writing is not writing. Or maybe the secret to writing is writing

every day. Or maybe the secret to writing is... there’s no secret. The only thing is to

follow what interests you. One day, after many many many revisions, the stories started

to come together, slough off their excesses and excrescences, and began to explain

themselves to me.

All genuine scholarship is fundamentally, irreducibly collaborative, a dialogue

with historical sources, with other scholars (whether credentialed or not), with colleagues,

with friends. But this kind of production also arises from contingent conditions, life

circumstances — don’t forget, ignore, or evade yourself when writing. Idea-seeds

cultivated in this dissertation were sown in casual conversations with strangers in the

supermarket; during random TV-watching binges during anxiety meltdowns (food

advertising is wonderfully suggestive); walking around with a tiny dog at the end of a

leash; asleep at night, in bed. Life spat, sputtered, spattered, always bursting with flavor,

and I was there to be splashed in its spray. It took me a while to realize that this was also

what I should be registering, recording, and responding to. The conditions of writing are

usually strenuously isolating, and systems of academic reputation-making generally

reward the “individual talent,” the fantasy of the dissertation-producing brain in the vat.

But don’t forget this, readers/toilers in need of uplift, comfort, consolation! Move away

from the laptop, open the window, talk to other people, talk to me, let yourself wander. I

don’t know whether this will seem obvious to almost everyone, but it was something that

I kept forgetting.
iv
 This project began with curiosity and coincidence. In my second year of graduate

school, I was in Ruth Schwartz Cowan’s history of technology seminar, and I had to

come up with an idea for a paper. I lit upon a phenomenon I had recently observed. Why

do concord grapes, the deep purply grapes you can find more easily at east coast farmer’s

markets than at grocery stores, taste so much like grape jolly ranchers, fake grape? If one

considered grape flavor as a technology, a technology that emerged at a specific historical

moment and under particular material and social conditions, could it perhaps illuminate

this apparent slippage between the uber-fake and the ultra-authentic?

I wondered if I could find an answer to that question at Monell Chemical Senses

Center, a low-slung brick building on Market Street, fronted by an extremely compelling

work of public art — “Face Fragment,” a 1975 sculpture by Arlene Love that featured

giant gilded lips and a giant gilded schnoz. Every day on my way to campus, I would

walk right past this gleaming mouth and nose, and mull over that unfamiliar, alluring,

term: “chemical senses.”

It turned out that Monell — an active research institution — housed, within its

warren of laboratories, the library of the Society of Flavor Chemists, a collection of such

strangeness, richness, and untapped splendor that it inspired this whole crazy journey.

Dani Reed, at Monell, was instrumental in making this resource available to me, and

encouraging me to drink deeply. This project also would never have been completed

without the generous, gracious support and encouragement of Alfred Goossens, of the

Society of Flavor Chemists. Many of the readers of this dissertation will not know Alfred,

so allow me to sketch a portrait for you. He is unfailingly elegant or even courtly – he

v
always wears the most gorgeous suits — and speaks English wish a soft Dutch accent. A

retired flavor chemist with a long and distinguished career at Naardens, Alfred was

unstintingly generous with his knowledge, his expertise, his time. He was always ready to

answer my questions, or introduce me to fascinating people in the flavor world. I owe

him so much.

This work was supported by Chemical Heritage Foundation, where I spent a

marvelous and productive year as a Beckman Fellow. My fellow dissertation fellows,

Tim Johnson, Edward Driggers, Britt Dahlberg, and Dan Liu, were incredible colleagues

and friends. Dan, in particular, deserves credit for helping me uncrumple some of my

much-crumpled ideas, detangle the skeins of thought, and also for sharing the bounty of

his incredible culinary skills. The staff of scholars, archivists, and librarians at CHF are

nonpareil, and I’d like to thank in particular: Andy Mangavite, Carin Berkowitz, and

Ronald Brashear. I also am incredibly grateful to the archival staff at the Hagley Museum

and Library (a great source for all kinds of food technology and food additive-related

treasures, and much more); the Smithsonian’s invaluable trade literature collection, at the

National Museum of American History; the somewhat surly but extremely

knowledgeable librarians at the National Archives in College Park, keepers of the

bureaucratic detritus of governance that holds so many of our nation’s stories; the

fabulous librarians at UC Davis’s special collections library, who helped me grind

through the unique and irreplaceable material in the A.W. Noling Beverage Literature

Collection, during one intense summer week of research; and the staff at the MIT

vi
Institute Special Collections, who guided me through the Arthur D. Little files, and

pointed me to the Emily Wick papers.

I had the good fortune to fall in with a congenial multidisciplinary bunch of “early

career” scholars, who were, like me, investigating questions about the technosciences of

the senses, sensory knowledge, and pleasure — the things that sometimes get bundled

together under the term “sciences of subjectivity.” Jake Lahne, who combines an actual

food science background with philosophical aplomb; he and his partner Kieran

Hutchinson are also marvelous people to enjoy food, spirits, and board games with. Ella

Butler, canny anthropological observer. Sarah Tracy, blazingly brilliant. Ana Ulloa,

whose fearlessness I hope to emulate. Christy Spackman, whose provocative questions

often led me to realize the parts of my work that had grown stale. I also want to thank Ai

Hisano, whose work on synthetic color in foods parallels my own story, and whose

inspired insights into the ambivalent appeal of the synthetically enhanced influenced my

own.

Sometimes during this, I would think, in despair. “The life of the mind! The life

of the mind is no life at all!” My advisor John Tresch, with his patience, enthusiasm, and

ecumenical intellect, reminded me — through his example and through his sage advice

— that thinking, writing, doing, can be lively and electric and rigorous, at once. My other

committee members, Heidi Voskuhl and Stephen Shapin, also provided intellectual and

scholarly support and nurturance for this project.

vii
 I’m indebted to many other scholars, who shaped this work by offering productive

critiques, encouragement, or other forms of intellectual nourishment: Ruth Schwartz

Cowan, Lissa Roberts, Dan Raff, Roger Horowitz. Robbie Aronowitz, Regina Blaszczyk,

Rachel Laudan, and Harold McGee.

Finishing (or even conceiving of) this dissertation would not have happened

without my fellow grad students at Penn. My cohort, Eram Alam and Rosanna Dent,

possibly the two kindest people I’ve ever met. Their compassion illuminates their fiercely

brilliant and important scholarship. (And Jen Goldsack, who left our cohort after year two

to accomplish exciting things in real life.) L. Ruth Rand, sassy empath, aviatrix, brilliant

human. Deanna Day, with her intellectual courage, scrupulous honesty, and sparkling wit.

Peter Collopy, whose sense of justice was a guiding light for me. Samantha Muka, who

somehow combines no nonsense with maximum fun. Marissa Mika, thoughtful and

generous. Elaine LaFay, a good spirit. Mary X. Mitchell, I learned so much from her

boldness. Whitney Laemmli, tough as nails. Kate Dorscht, always asking amazing

questions.

And there are even more people to thank, in the human network that buoyed me

through this: Ann Heppermann, who ran around in circles with me for fun, and her

partner Jason Cady, a talented composer; Mollie Goldstrom, for all the seaweed talk;

Anne Guthrie, who also made it through a dissertation and lived to tell the tale; Amanda

Gill, gentle spirit, an old and true friend. Elena Stover, Sami Stover, and Jessica Lazar

Bates, my oldest friends, tremendous spirits. Emma Boast, Catherine Piccoli, Peter Kim,

and the wonderful staff at the Museum of Food and Drink. Paul Adams, who really thinks
viii
I can write. Nicky Twilley, a brilliant writer, whose enthusiasm for food and flavor

helped me reignite some of the embers here. Bojack Horseman also really came through

for me during a tough time.

Bela Shayevich. Bela deserves her own paragraph just for the sheer quantity of

my bellyaching that she had to sit through. Bela is a freethinker, an artist, and I have

learned so much from her, including the virtue (necessity?) of feeling what you are

writing.

Joanna Radin, my north star and kindred spirit. Joanna was on her way out of

Penn as I was shimmying in, and I consider it one of the luckiest breaks in my life for our

paths to have crossed.

My partner in life, for better and for worse, Robbie Lee. Robbie never stopped

believing in me, even when I out ringing cowbells, gathering rosebuds, and calling it

quits. Thank you, Robbie, for being a friend.

And my family, the rooms of the heart where I was first fed. My tia Mabel. My

grandmother, Haydee DeJean Garcia. And of course, my mom and my dad, Elsa and

Carlos Berenstein, without whom none of this would have been possible.

ix
 ABSTRACT

FLAVOR ADDED: THE SCIENCES OF FLAVOR AND THE INDUSTRIALIZATION OF TASTE IN

AMERICA

NADIA BERENSTEIN

JOHN TRESCH

In the mid-nineteenth century, flavor additives — volatile organic chemicals with

desirable aromatic qualities — began to be used to flavor sugary confections, carbonated

beverages, and other mass-marketed delights. By the mid-twentieth century, added

flavors had become ubiquitous in processed, packaged foods; a sophisticated,

technoscientific, and globe-spanning industry had emerged that specialized in their

production. Drawing on history of science and technology, business history, and cultural

history, “Flavor Added” investigates the history of synthetic flavor additives, the systems

of scientific and technical knowledge that emerged to create these substances, and their

social and cultural consequences. Focusing primarily on the United States, “Flavor

Added” traces the origins and development of both flavor chemistry and sensory science,

illuminating their entangled roots in private industry, regulatory laboratories, USDA

research experiment stations, the US military, and academic institutions. Several chapters

take on the technologies and tools of flavor creation, including the taste panel, the flavor

profile, and the combined gas chromatograph-mass spectrometer. This dissertation also

documents the professional history of flavorists, the highly specialized scientific craft-

workers who develop and design flavor additives.

x
 Table of Contents
Acknowledgements iii

Abstract x

List of Illustrations xii

Introduction: Welcome to a New World of Flavor xiii

Chapter 1: Flavor by Formula 1

Chapter 2: A Flavor You Can’t Forget 77

Chapter 3: Assembling the Human Instrument 146

Chapter 4: Fresh, Easy, New 212

Chapter 5: Designing Flavors for Mass Consumption 277

Chapter 6: The Sniffing Machine 332

Chapter 7: The Creative Flavorist at Work 384

Bibliography 457

xi
 LIST OF ILLUSTRATIONS

Kletzinsky’s Table of Artificial Fruit Essences………………………………………….37

The Late Alois von Isakovics…………………………………………………………….47

“All the Flavor of the Vineyards in this Bottle”………………………………………….79

Tasting Booth at the Quartermaster Food Acceptance Research Branch………………..188

Floor Plan of Taste Panel Room………………………………………………………….198

“The Flavor is LOCKED-IN”…………………………………………………………….257

Flavor Profile of Carbonated Beverage and Seasoned Summer Squash…………………291

Flavor Profile Response Sheet……………………………………………………………292

The Role of the Creative Flavorist………………………………………………………..440

xii
 INTRODUCTION:
Welcome to a New World of Flavor

“There are still luxury foods, like caviar and guinea hen,” remarks a food

manufacturer in Lucy Kavaler’s 1963 book, The Artificial World Around Us, “but there

are no longer any luxury flavors.”1

Kavaler’s book, a work of popular science written for young adults, describes the

contours of a postwar world remade by synthetic chemistry. “We are living in an age of

chemicals,” she announces in the first chapter (“What Men Are Doing to Things”), an age

of both better things and better living, where the necessities of daily life, as well as its

pleasures and comforts, are made possible by the chemical laboratory.

Kavaler’s unnamed food manufacturer celebrates one of these signal

achievements: the synthetic production of the complex chemical mixtures that replicated

flavor — the essential experience of a food — and made that experience, and its

associated pleasures, widely available. Although his proclamation was a bit of an

overstatement —plenty of flavors cannot be effectively duplicated — the kernel of his

meaning remains. Even if some products remained the rarefied tidbits of the elite, the

essential experiences of sensory pleasure were available, and at a price that anyone could

afford. The unnamed food manufacturer paints a picture of a postwar abundance in which

1
Lucy Kavaler, The Artificial World Around Us, (New York: John Day Company, 1963):
61.
xiii
class divisions persist, although drawn along different lines. When everyone (allegedly)

has enough, the distinction between haves and the have-nots transforms into a division

between those who can feast on the real and those who must satisfy themselves with the

synthetic.

Why is the history of synthetic flavors worth telling? Understanding the advent of

synthetic flavors, the workers who devise them, and the technosocial networks in which

they were produced, used, valued, and consumed not only enhances our understanding of

the history of industrial food and the chemical industry in the United States, but also

elucidates the role of the senses and of sensory knowledge in technical and chemical

practice and in consumer culture. Further, flavor additives have shaped cultural, social,

and legal notions about what is natural and what is not, and playing a leading role in

debates about the perils or promises of science and technology in modernity.

Flavor, as those who sought to study it scientifically inevitably discovered, was a

scientific object whose contours and modes of research were underdetermined and

changing. Flavor research occurs in two distinct but inextricably intertwined fields —

chemistry and sensory science—a mixed discipline that draws on physiology,

psychology, among other fields. The subject has implications for our understanding of the

relationship between the human sciences and "hard" sciences, as well as the position of

the senses in scientific knowledge more generally.

The individuals who designed and developed flavor additives at flavor

companies— flavorists and their occupational forebears—are not the only people who

xiv
worked with flavors who appear in this history. A constellation of chemists, food

scientists, home economists, psychologists, sensory scientists, and engineers make up the

professional network of flavor science. Flavor science takes place not only in corporate

laboratories and industrial factories, but also in USDA regional research centers,

regulatory testing laboratories, military research installations, and academic labs. The

multidisciplinarity of this network can be attributed to the complexity of flavor itself.

Integral to, and woven within, these narratives about professionals and scientific experts

are the sensate bodies of consumers, whose experiences, behaviors, and appetites were

studied, measured, and refashioned, in a world of consumer goods whose sensory

qualities were increasingly designed in accordance with the technoscientific practices

described here.

I take synthetic flavors on their own terms as new things that also had to make a

case for themselves by demonstrating their utility and value. Synthetic flavorings were

not “invented” to satisfy some pre-existing need, nor can they be explained simply as

substitutes or imitations of other, scarcer or more expensive, things.2 Flavor additives

emerged from a world whose material, economic, social, and cultural substrate was

rapidly being reorganized by science, technology, and global market forces, and they

found a place in a mass consumer economy whose contours, dynamics, technics, and

meanings were in the process of formation.

This work can be considered historical ontology of added flavor, surveying the

attempts of a heterogeneous group of scientific and technical workers to define its

2
An apposite comparison could be drawn with the early history of plastics.
xv
material and phenomenological boundaries, to develop a standard set of terms, tools, and

methods to investigate its constituents, causes, and effects, and to apply technoscientific

knowledge toward the production, reproduction, and control of flavor.

But what should be included in a study science of flavor, and what should be

excluded? How should the lines be drawn? What, exactly, are flavors?

Flavors Are Chemicals

Many substances used throughout history, and around the world, can be described

as ‘flavor additives’: spices and essential oils, sugar, salt. These materials were

considered to have medicinal as well as gustatory virtues, and could serve other

functions, such as food preservation.3 The preparation of, and trade in, flavoring

substances (and related perfumery materials) required special technologies, skills, and

expertise, and involved practices which could be considered alchemical or proto-

chemical: distillation, fermentation, extraction, as well as sensory and chemical methods

to determine purity or detect adulteration.4 Indeed, corporate histories of companies in the

flavor and fragrance industry frequently begin with a nod toward the spice trade, gilding

their ultra-modern chemical business with the radiance of more heroic ages.

3
Paul Freedman, Out of the East: Spices and the Medieval Imagination, (New Haven:
Yale UP, 2008); Wolfgang Schivelbusch, Tastes of Paradise: A Social History of Spices,
Stimulants, and Intoxicants, trans. David Jacobson, (New York: Vintage, 1993).
4
The alchemical heritage is more extensively plumbed in the history of perfumes and
perfumery than in spices and seasonings. See, for instance, Mandy Aftel, Essence and
Alchemy: A History of Perfume, (New York: North Point Press, 2001). For commentary
regarding why the modern perfume industry chooses to invoke a pre-Enlightenment
heritage, see Maksym Klymentiev, “Creating Spices for the Mind: The Origins of
Modern Western Perfumery,” Senses & Society 9.2 (2014): 212-31.
xvi
 Throughout this work, I use the term “synthetic flavor” rather than the more

familiar (to us) “artificial flavor.” “Artificial flavor” is, in ordinary speech, largely

construed in contradistinction with “natural flavor.” Both of these terms have specific—

though largely mysterious to everyday consumers—regulatory definitions in the U.S.

Federal Code that determine whether and how they are listed on food labels. But both so-

called natural and artificial flavors are artefactual — deliberately designed, produced, and

added, for specific purposes, to manufactured foods.

My use of the term “synthetic flavor” also deliberately signals a historiographical

rupture from stories that run smoothly from the antique spice trade to the modern

manufacture of exquisitely engineered flavoring compounds. Specifically, I situate the

roots of my narrative within a particular context in the history of chemistry — the

emergence of organic chemistry and the development of a chemical industry based on

chemical synthesis in the mid-nineteenth century. These associated events are both bound

up with industrialization, which provided the carboniferous raw material for synthetic

chemical processes, the commercial rationale for increasing technical and scientific

control over those processes, and the social, technological, and economic bonds

connecting the chemical laboratory to the factory and the consumer economy.5

5
The history of modern chemistry wrestles with two narrative arcs: chemists' quest for a
distinctive chemical theory, and the continuing importance of artisanal and empirical
processes of investigation in chemical research. The role of sensory knowledge in this
drama is an unresolved question in the history of chemistry. See for instance, Catherine
M. Jackson, “Synthetical Experiments and Alkaloid Analogues: Liebig, Hoffman, and the
Origins of Organic Synthesis,” Historical Studies of the Natural Sciences 44.4 (2014):
319-63; Mary Jo Nye, From Chemical Philosophy to Theoretical Chemistry: Dynamics of
Matter and Dynamics of Disciplines, 1800-1950, (Berkeley: University of California
xvii
 As the material world came to be regarded as a thing whose basic composition,

contours, and qualities could be manipulated by chemical processes and techniques,

flavor was only one part of the world that began to be considered in chemical terms.6 Yet

flavor presented unique challenges to those who wished to understand it, analytically, as

a chemical phenomenon. The chemical constituents of flavor in foods are scarce, fleeting,

labile, and promiscuous — present in minute quantities, highly volatile, generally

unstable, and tending to intermingle and react with other compounds in food or

packaging. For much of the period covered by this dissertation, analytic knowledge of the

chemical constituents of flavors in foods was hard-won and uncertain. But the path from

chemical to flavor did not exclusively (or even primarily) run through chemical analysis.

Press, 1993); Bernadette Bensaude-Vincent and Isabelle Stengers, A History of

Chemistry, trans. Deborah van Dam, (Cambridge: Harvard UP, 1996); and especially
Lissa Roberts, "Death of the Sensuous Chemist: The 'New' Chemistry and the
Transformation of Sensuous Technology," Studies in the History and Philosophy of
Science 26, no 4 (1995): 503-529. The disciplinary process of subordinating the
discerning body to the quantifying procedures of experimental machines that Roberts
finds to be characteristic of chemistry after Lavoisier may be complicated by modes of
organic chemistry, including flavor chemistry, which uses the senses to constitute its
subject.
6
Historians and sociologists of science have explored the emergence of chemical
synthesis in the nineteenth century, uncovering the contested circumstances under which
atoms and molecules, once theoretical objects, gained material reality, and documenting
how this new disciplinary paradigm was brought into alignment with the systems and
processes of industrial production. In particular, scholars have described the ways in
which ideas, ideologies, and social relations are embedded in the material cultures and
experimental practices of chemistry. See Christoph Meinel, “Molecules and Croquet
Balls,” in Soraya de Chadarevian and Nick Hopwood, eds, Models: The Third Dimension
of Science, (Stanford: Stanford UP, 2004): 242-275; Ursula Klein, “Technoscience Avant
la Lettre,” Perspectives on Science 13.2 (2005): 226-66; Klein, Experiments, Models,
Paper Tools: Cultures of Organic Chemistry in the Nineteenth Century, (Chicago:
UChicago, 2003); Alan J. Rocke, Image and Reality: Kekule, Kopp, and the Scientific
Imagination, (Chicago: UChicago, 2010); and David M. Knight and Helge Kragh, eds.
The Making of the Chemist: A Social History of Chemistry in Europe, 1789-1914,
(Cambridge: Cambridge UP, 1998.)
xviii
A chemical could also become a “flavor chemical” as a result of its sensory qualities,

independent of any experimental confirmation of its presence in foods.

Recognizing the extreme technical difficulty of producing chemical knowledge

about the molecular constituents of flavor, and of applying physicochemical methods to

the task of controlling, stabilizing, and standardizing the flavor of foods, is necessary

context for understanding the development of flavor chemistry as a commercial

enterprise and a scientific field. Chemical knowledge about flavor and flavor additives

themselves are both laboriously manufactured. In this way, this project draws deeply

from a venerable and vital spring in history of science and technology that has shown

scientific knowledge, including the standard objects of chemistry, to be the outcome of

coordinated social and technological processes, rather than prior natural givens.7

Flavor is not just an object of chemical inquiry, but also a product of chemical

industry.8 Classic studies of the chemical industry have focused on large, vertically

7
In particular, David Singerman’s work on how sugar became sucrose. David Roth
Singerman, “Inventing Purity in the Atlantic Sugar World, 1860-1930,”(PhD diss,
Massachusetts Institute of Technology, 2014).
8
To my knowledge, there have been no extended scholarly studies on the history of the
flavor manufacturing, and, as a result, the industry retains a reputation for extreme
“secrecy” that it has (deliberately) not done much to dispel. Much of the published
secondary literature on the subject is internalist, produced by workers in the industry or
published under the banner of corporate histories or trade organizations (which is not to
impugn its validity or reliability, as some of these studies have proven to be extremely
valuable resources, but simply to note the implied absence of a critical distance toward
their subject, and their often limited scope), or journalism, which has, generally,
investigated and “exposed” the ways of the flavor industry in the context of broader
critiques of industrial foods. Among internalist histories: Wayne E. Dorland and James
A. Rogers, Jr. monograph remains a useful source for company histories, historical
production processes and equipment, glossaries of materials, and related organizations.
See: Dorland and Rogers, The Fragrance and Flavor Industry, (Mendham, NJ: Wayne E.
xix
integrated corporations, such as DuPont in the United States.9 Historians and sociologists

of science have examined the organization and administration of research and

development programs in large companies, the paths by which scientific discoveries were

translated into commercial products, as well as the linkages between industry, academy,

and state.10 Following Alfred Chandler’s framework, which linked the establishment of

industrial research and development to structures of managerial capitalism, business

scholars have scrutinized the means by which the chemical industry achieved scale and

scope in its operations.11

Dorland Co., 1977). The centennial volume published by the Flavor and Extract
Manufacturer’s Association (FEMA) is fantastically illustrated with photographs and
documents and stuffed with historical anecdotes about the political, economic, and
technological incidents that shaped the industry (I remain hopeful that I will one day be
permitted to view their archival records): Flavor and Extract Manufacturers Association,
FEMA 100: A Century of Great Taste, (Washington, D.C.: FEMA, 2009). Useful
historical studies can be found in trade journals relevant to the flavor and fragrance
industries. See, for instance: “50 Years of Service by Van Dyk & Co.,” American
Perfumer and Essential Oil Review 63 (June 1954): 453-5; Paul Z. Bedoukian, “The
Perfumery Aromatics Industry in the United States, Parts I-III” American Perfumer and
Aromatics (November 1957): 33-6, (December 1957): 31-5, (January 1958): 43-50;
Gabriel Sink, “A Tribute to the Oldest American Flavor and Fragrance House,” Perfumer
and Flavorist 17 (January/February 1992): 37-9. For journalistic accounts: Raffi
Khatchadourian, “The Taste Makers,” The New Yorker (November 23, 2009): 129-35;
Eric Schlosser, Fast Food Nation: The Dark Side of the All-American Meal, (New York:
Perennial, 2002); and Michael Moss, Salt, Sugar, Fat: How the Food Giants Hooked Us,
(New York: Random House, 2013).
9
David A. Hounshell and John Kenly Smith, Science and Corporate Strategy: DuPont
R&D, 1902-1980, (Cambridge: Cambridge UP, 1988).
10
The scholarship on the emergence, evolution, and structure of industrial R&D is vast.
Notable titles include: Pap Ndiaye, Nylon and Bombs: DuPont and the March of Modern
America, trans. Elborg Forster, (Baltimore: JHU Press, 2007); David A. Hounshell, "The
Evolution of Industrial research in the United States," In R. Rosenbloom and W. Spencer
(Eds.), Engines of innovation: U.S. industrial research at the end of an era, (Boston:
Harvard Business School Press, 1996).
11
Alfred D. Chandler, Scale and Scope: The Dynamics of Industrial Capitalism,
(Cambridge: Belknap, 1990) and Shaping the Industrial Century: The Remarkable Story
xx
 This work focuses instead on the small-scale manufacture of chemical specialties

designed to preserve and achieve desired sensory qualities at mass-scale, with significant

impact on the much larger industries that relied on its products—food and beverages, as

well as pharmaceuticals, tobacco products, and animal feeds.12 While most scholars have

focused on the growth of the American chemical industry after the First World War, this

study begins in the nineteenth century, tracing the roots of the highly specialized

American flavor and fragrance industry to a heterogeneous network of chemical

producers and brokers: pharmacists, distillers, confectioners, essential oil importers, and

other “practical chemists” who made or sold these goods as part of more diversified

businesses. 13 Further, while much secondary literature on the flavor industry fails to

of the Evolution of the Modern Chemical and Pharmaceutical Industries, (Cambridge:

Harvard UP, 2005).
12
On this point, this project owes a debt to Philip Scranton’s meticulous, substantial, and
lucid work on American methods of custom- and batch-production in the late nineteenth
and early twentieth centuries, and the role of these manufacturing strategies in producing
a variegated world of goods for mass consumer markets. Philip Scranton, Endless
Novelty: Specialty Production and American Industrialization, 1865-1925, (Princeton:
Princeton UP: 1997).
13
Aside from histories of pharmacy and monographs about large chemical companies,
scholarship about small- and mid-scale is almost all antique; this field cries out for
renewed scholarly inquiry. The classic account of the American chemical industry is
William Haynes’ six-volume history, published between the late 1940s and 1950s — a
massive undertaking sustained by funding from the Chemical Foundation, Dow, and
Monsanto. For a sense, see, in particular: William Haynes, American Chemical Industry:
A History, Volume I: Background and Beginnings (Toronto, NY, London: D. Van
Nostrand, 1954). For a useful overview of the history of pharmacy, see Gregory Higby
and Elaine C. Stroud, eds, American Pharmacy (1852-2002): A Collection of Historical
Essays, (Madison: American Institute of the History of Pharmacy, 2005). John Kenly
Smith proposes several historiographical frameworks for studies of chemical modes of
production (rather than histories of particular products or particular firms) in “The
Evolution of the Chemical Industry: A Technological Perspective,” in Seymour
Mauskopf, ed., Chemical Sciences in the Modern World (Philadelphia: University of
Pennsylvania Press, 1993). A recent, solid monograph that considers the role played by
the First World War in reorganizing the political and economic conditions of the
xxi
recognize its scope or significance before the Second World War, “Flavor Added”

documents the diversity of chemical methods of production outside of well-organized,

vertically integrated industries, while also documenting the perseverance of craft methods

within the chemical industry.14 The peculiar chemical, social, and regulatory

requirements of flavor additives make the flavor industry a prime candidate to enhance

our understanding of how chemical knowledge translates into manufacturing processes

and consumer products.

Flavors Are Multisensory Perceptual Phenomena

In his story “Tlön, Uqbar, Orbis Tertius,” Jorge Luis Borges describes the syntax

of the languages spoken in Tlön, an imaginary planet whose customs, history, and

geography are elaborated in a series of obscure and pirated encyclopedias. In the

Northern hemisphere of the planet, nouns do not exist in their own right, but are instead

devised by “an accumulation of adjectives.” He explains:

There are objects composed of two terms, one of visual and another of
auditory character: the color of the rising sun and the faraway cry of a
bird. There are objects of many terms: the sun and water on a swimmer's

American chemical industry is: Kathryn Steen, The American Synthetic Chemicals
Industry: War and Politics, 1910-1930, (Chapel Hill, UNC Press: 2014).
14
For instance, Constance Classen, David Howes, and Anthony Synnott have written:
“Artificial flavours were invented in the late nineteenth century, but didn’t become
prevalent until the 1960s.” At the very least, I hope this dissertation puts that
misconception to rest. Classen, Howes, and Synnott, “Artificial Flavors,” in The Taste
Culture Reader: Experiencing Food and Drink, ed. Carolyn Korsmeyer, (Oxford and
New York: Berg, 2005 [2007]): 337.
xxii
 chest, the vague tremulous rose color we see with our eyes closed, the
sensation of being carried along by a river and also by sleep.15

Clusters of sensations, bodily states, affective feelings, and other mental

phenomena — the very stuff of subjectivity — are, in Tlön, materialized as nouns, the

denotative objects of language. “There are famous poems,” Borges writes, “made up of

one enormous word.”

This study concerns a series of attempts by scientific and technical workers to

describe, determine, and comprehend flavor, an object of many terms, in the absence of a

grammar that could cast it as one comprehensive, enormous, denotative word. During the

hundred years, give or take, covered by this dissertation, flavor came to be associated

with specific molecules and chemical processes, but also simultaneously understood as an

experiential response — an embodied reaction which could only partially or imperfectly

be described in terms of the chemical presences that were the apparent occasion for

sensations. By the 1930s, the researchers, regulators, and flavor makers who were

concerned with the determination, measurement, control, and production of flavor

recognized it as a multisensory phenomenon — involving not only the chemical senses

(smell and taste, broadly constituted), but also somatosensations within the oral cavity

(such as mouthfeel), visual factors, and auditory components. These researchers were

also increasingly aware of the influence of the personal and physiological circumstances

15
Jorge Luis Borges, “Tlön, Uqbar, Orbis Tertius,” trans. James E. Irby, in Donald A.
Yates and Irby, eds. Labyrinths: Selected Stories and Other Writings, (New York: New
Directions, 1964): 9.
xxiii
of the eater. The state of the body, the atmospheric and environmental conditions of the

room, social influences, prior experiences, all seemed to affect not just stated preferences,

but the perceptual experience of flavor itself.

Historians of scientific objectivity, most prominently Lorraine Daston and Peter

Galison, have described the epistemic virtue of objectivity not as a quality achieved by

certain forms of knowledge, but as a set of practices, or perhaps as a style, assumed by

the investigator.16 In taking as its subject scientific atlases — visual representations of

natural specimens and other external, more or less observable phenomena — Daston and

Galison’s Objectivity, which has gained landmark status, may misrepresent this specific

scenario as prototypical of the production of objectivity in scientific knowledge-making

more generally. In parallel to the scientific atlas-makers, a group of scientific workers

known as experimental psychologists and psychophysicists labored to measure, represent,

and determine psychological phenomena, subjective states of mind, and the structures of

consciousness. 17 Notable recent works of scholarship have examined how scientists,

technicians, engineers, and designers grapple with the practical and epistemological

questions raised by sensory objects, and the concurrent formation of "sciences of

subjectivity" to organize the study of these phenomena.18 Rather than the exception to the

16
Lorraine Daston and Peter Galison, Objectivity, (Cambridge: Zone Books, 2007).
17
See, for instance, Christopher Green’s critique of Objectivity along these lines in:
Christopher D. Green, “Scientific Objectivity and E.B. Titchener’s Experimental
Psychology,” Isis 101 (2010): 697-721.
18
See in particular, Steven Shapin, "Sciences of Subjectivity," Social Studies of Science
42, no 2. (2012): 170-184. Also: Bruno Latour, "How to Talk About the Body? The
Normative Dimension of Science Studies," Body & Society 10, no. 2-3 (2004): 205-229;
Rebecca Lemov, Database of Dreams: The Lost Quest to Catalog Humanity, (New
Haven: Yale UP, 2015); Sophia Roosth, "Screaming Yeast: Sonocytology, Cytoplasmic
xxiv
rule, these “world-making” disciplines, to use Steven Shapin’s phrase, should be

considered equally central to the project of understanding the peculiar technoscientific

constitution of modernity.

My project contributes to this literature by documenting the practices, values, and

experiences of the technicians and scientists who worked to create a science of flavor.19

In particular, I examine the technologies and instruments that these workers used to

produce scientific and standard knowledge about flavor from the data of human

Milieus, and Cellular Subjectivities," Critical Inquiry 35, no. 4 (2009): 332-350; Jonathan
Sterne, The Audible Past: Cultural Origins of Sound Reproduction, (Durham: Duke
University Press, 2003); Emily Thompson, The Soundscape of Modernity: Architectural
Acoustics and the Culture of Listening in America, (Cambridge: MIT Press, 2004).
19
Alternately, anthropologists have explored flavor and the production and constitution
of sensing bodies and subjective knowledge in the contexts of Western industrial food
systems. See, for instance, Ana Ulloa’s ethnographic studies of flavor research
laboratories, flavor industry laboratories, and “molecular gastronomy” kitchens, which
scrutinizes the modes of practice and ways of knowing particular to flavor-work; Ella
Butler’s ongoing anthropological studies of sensory science and sensory evaluation; Jake
Lahne’s anthropologically informed inquiries into the methods of sensory evaluation and
the construction of sensory knowledge; David Howes’ ethnographic investigations into
the development of sensory techniques; and Amy Trubek’s study of the social, cultural,
and political production of the set of sensible qualities known as terroir. Ana Maria Ulloa,
Josep Roca, and Heloise Vilaseca, “From Sensory Capacities to Sensible Skills:
Experimenting with El Celler de Can Roca,” Gastronomica: The Journal of Critical
Food Studies 17.2 (2017): 26-38 and forthcoming PhD dissertation; Ella Butler,
forthcoming PhD dissertation; Jake Lahne, “Sensory Science, the Food Industry, and the
Objectification of Taste,” Anthropology of Food 2016, and “Tasting in Context:
Consumer Sensory Perception of Vermont Artisan Cheese,” (PhD dissertation, University
of Vermont, 2014); David Howes, “The Science of Sensory Evaluation: An Ethnographic
Critique,” in The Social Life of Materials: Studies in Materials and Society, ed. Adam
Drazin and Susanne Küchler, (London: Bloomsbury, 2015): 81-97; and Amy Trubek, The
Taste of Place: A Cultural Journey into Terroir, (Oakland: University of California Press,
2008). See also the work of Annemarie Mol on sensual forms of knowledge production,
especially: "Tasting Food: Tasting Between the Laboratory and the Clinic," in A
Companion to the Anthropology of the Body and Embodiment, ed. Frances E. Mascia-
Lees, (Wiley-Blackwell, 2011): 467-80, "I Eat an Apple. On Theorizing Subjectivities,"
Subjectivity 1 (2008): 28-37, and "Good Taste: The embodied normativity of the
consumer-citizen," Journal of Cultural Economy 2, no 3 (November 2009): 269-284.
xxv
experiential responses and the chemical components of foods. If any general statement

can be made about them, it is this: these technologies can never be disaggregated from

human bodies, and always require human substrate, whether they are the technologies of

laboratory taste panels or the physicochemical machines introduced into the flavor

laboratory during the "instrumental revolution" in chemistry.20

"Flavor Added" draws from the history of the senses, especially work that attends

to the material, cultural, and social dimensions of sensory experience. 21 Scholars such as

Melanie Kiechle, Mark Jenner, Mark Smith, David Howes, Carolyn Korsmeyer, and

Alain Corbin have shown that our sensory worlds are not only personal, but are

20
Christy Spackman’s ongoing work on the use of analytic and other technologies in the
production of scientific knowledge about sensation demonstrates the potentials of this
field of inquiry for STS scholars. Christy Spackman, “Transforming Taste: The
Twentieth-Century Aesthetic Remaking of Water,” (PhD diss., New York University,
2015). Ingemar Pettersson’s dissertation work at the University of Uppsala is another
recent project to examine the science of sensory analysis. For work on the "instrumental
revolution," more generally, see Peter J.T. Morris, ed. From Classical to Modern
Chemistry: The Instrumental Revolution, (Cambridge: Royal Society of Chemistry,
2000); Frederick Holmes and T.H. Levere, eds., Instruments and Experimentation in the
History of Chemistry, (Cambridge: MIT Press, 2000); Carsten Reinhardt, Shifting and
Rearranging: Physical Methods and the Transformation of Modern Chemistry,
(Sagamore Beach, Mass.: Science History Publications/USA, 2006).
21
The history of the senses is doggedly interdisciplinary, combining methodologies from
history, anthropology, philosophy, and the social sciences. See, for instance, Melanie
Kiechle, “’The Air We Breathe’: Nineteenth-Century Americans and the Search for Fresh
Air,” (PhD diss., Rutgers, 2012); Peter Charles Hoffer, Sensory Worlds in Early America,
(Baltimore: JHU Press, 2003); David Howes, Sensual Relations: Engaging the Senses in
Culture and Social Theory, (Ann Arbor: University of Michigan Press, 2003); Carolyn
Korsmeyer, Making Sense of Taste: Food and Philosophy, (Ithaca: Cornell UP, 2002);
Alain Corbin, The Foul and the Fragrant: Odor and the French Social Imagination,
trans. Miriam L. Kochan, Roy Porter, and Christopher Prendergast, (Cambridge: Harvard
UP, 1986); Adam Mack, "'Speaking of Tomatoes': Supermarkets, the Senses, and Sexual
Fantasy in Modern America," Journal of Social History 43 (Summer 2010): 815- 842;
and Mark Smith, How Race is Made: Slavery, Segregation, and the Senses (Chapel Hill:
UNC Press, 2006).
xxvi
fundamentally historical. The sensible things we attend to, and the meanings we make of

them, are shaped by social, cultural, economic forces. The “authoritative” and apparently

incontrovertibly personal knowledge of our senses, in other words, is always already

informed by authorities and forms of knowledge other than our own experience.

Following the example of historians of the senses, I have tried to be careful not to

lapse into the assumption that there is anything prior or “natural” about the flavor of

things, or anything trans-historical about how flavor is perceived and understood.

Although the label of “imitation” haunted (and still haunts) many foods and beverages

containing “artificial flavors,” to reflexively assume that there was an “original” to which

the imitation aspired mistakes both the dynamic historical circumstances in which

synthetic flavorings emerged, as well as the consequences of their availability for the

development of food, drinks, and other sensible goods. Thus, in pursuing this research, I

have tried to avoid questions such as, “What is the flavor of an apple?” and, “Did this

flavoring successfully replicate it?”

Considering flavor as a historical construct means asking, instead, questions such

as: What were the forms and conditions in which apple flavor was available to eaters

(including, but not limited to, apples themselves)? What were the material, social,

cultural, and scientific contexts within which apple flavor was consumed, considered,

discussed, and valued? How have sensory expectations around apples changed, and what

forces may have contributed to these changes? In other words, there is no pre-existing,

immutable, or trans-historical "apple flavor." the phenomenon of apple flavor comes into

being, in all its specificity, only upon the meeting of a certain historical body and a
xxvii
certain comestible, under social circumstances where the sensation produced is, both

intimately and intersubjectively, recognized as “apple.” Likewise, in linking particular

chemicals with recognizable, named sensory effects, I have tried to avoid foreclosing the

recognition of other possible sensory experiences resultant from exposure to these

compounds, and have also tried to avoid the presumption that a certain chemical found in

a food represented or produced a definite, universally agreed-upon sensation.

Flavors Are Food Technologies

Gaston Bachelard, ruminating on the conditions of scientific knowledge, observed

“there are no simple phenomena; every phenomenon is a fabric of relations,” produced by

and embedded within practices, machines, ideologies. While proper scientific objects

must be wrest from the context of ordinary sensual life and reconstituted by laboratory

labor, their return to the world lays bare the special conditions of the scientific mode of

knowing. “Application is complication.”22

This history of flavor science is fundamentally about applied knowledge. It thus

intersects with, and illuminates, the development of an American system of food

production that resulted in an abundance of cheap calories.23 The sciences of flavor were

22
Gaston Bachelard, The New Scientific Spirit, [1934], 147-8.
23
Some of the most compelling work in this area has come from historians of technology
and of business, who situate consumers, producers, and resources in dynamic
technosocial systems of food production and distribution. For studies that consider the
design of sensory qualities of food in industrial food systems, see, especially, Gabriella
M. Petrick, “The Arbiters of Taste: Producers, Consumers, and the Industrialization of
Taste in America, 1900-1960,” (PhD diss., University of Delaware, 2006); and Ai
Hisano, “Eye Appeal is Buy Appeal: Business Creates the Color of Foods, 1870-1970,”
(PhD. Diss, University of Delaware, 2016). Shane Hamilton, Suzanne Friedberg, Anna
Zeide, Anne Vileisis, Paul Josephson, and Amy Bentley, among others, take as case
xxviii
ultimately applied sciences, intended to produce not abstract knowledge, but actual things

— commercial products — that yielded intended (perceptual) effects. As such, workers in

this field often found themselves grappling with the complications of application. The

scenarios in which they cultivated the sciences of flavor — industry, progressive

government agencies, military research — framed problems of flavor in the same way as

other problems of production: as technoscientific problems, with technoscientific

solutions. In these contexts, certain types of solutions were pursued in preference to

others. For instance, although some agricultural research was devoted to developing

“better tasting” varieties of fruits, vegetables, and meats, most agricultural science was

studies specific technologies of production (refrigeration, freezing, canning) or food

products (the tomato, the fish stick, baby food), illuminating the web of technosocial
relations that shape systems, goods, and consumers in twentieth century economies. See:
Shane Hamilton, “Cold Capitalism: The Political Ecology of Frozen Concentrated
Orange Juice,” Agricultural History 77, no. 4 (Autumn 2003): 557–581 and “The
Economies and Conveniences of Modern-Day Living: Frozen Foods and Mass
Marketing, 1945-1965,” The Business History Review 77, no. 1 (Spring 2003): 33–60;
Susanne Freidberg, Fresh: A Perishable History, (Cambridge: Harvard UP, 2009); Anna
Zeide, “In Cans We Trust: Food, Consumers, and Expertise in Twentieth-Century
America,” (PhD Diss., University of Wisconsin-Madison, 2014); Ann Vileisis, "Are
Tomatoes Natural?" in Martin Reuss, Stephen H. Cutliffe, eds., The Illusory Boundary:
Environment and Technology in History, UVA Press, 2010; Paul Josephson, "The
Ocean's Hot Dog: The Development of the Fish Stick," Technology and Culture 49.1
(January 2008): 41-61; and Amy Bentley, Inventing Baby Food: Taste, Health, and the
Industrialization of the American Diet, (Oakland: University of California Press, 2014).
For more general accounts of the food system and food production informed by history of
technology, see: Rachel Laudan, Cuisine and Empire: Cooking in World History,
(Oakland: University of California Press, 2013); Warren Belasco and Roger Horowitz,
eds. Food Chains: From Farmyard to Shopping Cart, (Philadelphia: UPenn Press, 2009);
Nancy F. Koehn, “Henry Heinz and Brand Creation in the Late Nineteenth Century:
Making Markets for Processed Food,” The Business History Review 73, no. 3 (Autumn
1999): 349-393; and Warren Belasco and Philip Scranton, eds. Food Nations: Selling
Taste in Consumer Societies, (New York: Routledge, 2002).
xxix
oriented toward increasing yields and efficiency.24 Solutions to the flavor deficiencies of

industrial food were generally sought at the level of manufacturing and distribution: in

the development and deployment of new chemical additives, or improvements to

processing, packaging, quality control, and transportation technologies.

Many historians and other commentators have correctly drawn attention to the

negative consequences of this abundance: the inequitable distribution of its rewards, its

sham choices and false promises, its detrimental effects on the health and well-being of

certain populations, its effects on environments, economies, and traditional ways of life.25

These narratives are elaborations upon what Harvey Levenstein has called “the paradox

of plenty”: the political, social, and cultural anxieties about food consumption that

accompanied the proliferation of food calories.26 The argument made in many of these

accounts is that the industrial food system achieves its apparent cheapness and abundance

at great cost, to human health and lives, as well as planetary well-being.

24
See, for instance: William Boyd, "Making Meat: Science, Technology, and American
Poultry Production," Technology and Culture 42. 4 (October 2001): 631-664; Deborah
Fitzgerald, “Deskilling Farmers: Hybrid Corn and Farmers’ Work,” Technology and
Culture 34.2 (April 1993): 324-43.
25
For a recent bibliography of food history that emphasizes social and cultural history,
see Marion Nestle and W. Alex McIntosh, “Writing the Food Studies Movement,” Food,
Culture and Society: An International Journal of Multidisciplinary Research 13, no. 2
(2010): 159–179. A comparable resource for works on the anthropology of food can be
found at Sidney W. Mintz and Christine M. Du Bois, “The Anthropology of Food and
Eating,” Annual Review of Anthropology 31 (2002): 99–119; Warren J. Belasco and
Philip Scranton, eds., Food Nations: Selling Taste in Consumer Societies, Hagley
Perspectives on Business and Culture (New York: Routledge, 2002), provides a solid,
international set of examples about how cultural and social food habits are susceptible
and resistant to change.
26
Harvey Levenstein, Paradox of Plenty: A Social History of Eating in Modern America,
(New York: Oxford UP, 1993).
xxx
 Food studies scholars and critics, documenting the industrialization of the food

system, have often ignored or dismissed the sensual aspects of these changing

technologies of food production. Many commentators have taken the position that flavor

did not matter to manufacturers, nor to the scientific workers (such as nutritionists and

food technologists) who, with their expert labor, supported the industrialization of the

food system; or, alternately, that flavor played second fiddle to other concerns, such as

nutrition, safety, and profit.27 In response to these claims, I argue the following. The

evidence that flavor was a primary concern of food manufacturers and the industrializing

food system is plentiful. So why has its role been overlooked? Discussions and

technological interventions aimed at shaping, controlling, and improving the sensory

qualities of food in the food industry and its technosciences often do not coincide with

lay notions of how flavor ought to be talked about. Although its outcomes may not be

congruent with prevailing ideas about “good flavor,” when one looks for the evidence of

how flavor mattered to the food manufacturers and the industrializing food system, one

finds it plentifully in evidence.

In this dissertation, I consider flavor additives as technologies — as deliberately

designed artifacts that operate within the context of a broader food system. This system

27
For some commentators, this has resulted in an extreme skepticism that approaches a
disavowal of food science and food technology as such. Take for instance, this statement
from Michael Pollan in a recent interview with Lucky Peach’s Rachel Khong: “I
sometimes find myself wondering whether we can post or imagine a food science that is
actually improving food in the way that cooking for most of its history succeeded in
doing…. We’ve had food science and food technology now for a hundred and fifty years,
and so far, not so good. So far we haven’t done anything useful. But we understand a lot
more, and we should be able to improve on things, not just make money and entertain
people.” Rachel Khong Interviews Michael Pollan, “The End of the World as We Know
it,” Lucky Peach, (September 10, 2014).
xxxi
includes, most immediately, the other ingredients that constitute the food, wrapping and

packaging materials, the machines and methods that produce the food and make it

available to consumers for a fixed price, and, more distantly, the cultural, social, and

environmental context within which the food is consumed. The precise form that flavor

additives took, as well as the purposes that they were expected to serve, vary over the

course of the century or so discussed in this dissertation, reflecting changes to the

methods, institutional arrangements, and technosocial networks of flavor science, as well

as changes to the market for food and other consumer goods. At different points in this

dissertation, flavor additives are technologies that can efficiently convert commodities

into consumer products, confer uniqueness or distinction to branded goods, enforce

standard uniformity on items made from variable raw materials, minimize unpleasant or

unpalatable sensations, enhance and extend pleasurable and desirable sensations, and

deliver precisely calibrated sensory experiences.28

Crucially, flavor is a technology that becomes effective only by acting directly on

the body and mind of the consumer. But precisely how the body is believed to be

susceptible to flavor, the terms under which flavor’s effect on the body is theorized and

28
A relevant body of literature here is the history of advertising and consumer culture,
particularly accounts that investigate the social and psychological sciences that inform
advertising, marketing, merchandising, and other consumer-oriented business practices.
See, for instance: Lawrence R. Samuel, Freud on Madison Avenue: Motivational
Research and Subliminal Advertising in America, (Philadelphia: University of
Pennsylvania Press, 2011); William Leach, Land of Desire: Merchants, Power, and the
Rise of a New American Culture, (New York: Pantheon, 1993); Roland Marchand,
Advertising the American Dream: Making Way for Modernity, 1920-1940, (Berkeley:
University of California Press, 1985); Susan Strasser, Satisfaction Guaranteed: The
Making of the American Mass Market, (New York: Pantheon, 1989); Pamela Walker
Laird, Advertising Progress: American Business and the Rise of Consumer Marketing
(Baltimore: JHU press, 1998).
xxxii
measured, varies over the course of this history, as does the imagined relationship

between flavor sensations and resultant psychic and physiological phenomena, such as

perception, appetite, affect, and behavior.29 Of course, these changing understandings of

the sensible body — its appetites, responses, and needs — reflect changing concerns,

ideologies, and interests that deeply inform the designs and purposes of flavor

technologies.30

Novel preservation technologies, such as flash freezing; new materials, such as

polyethylene plastics, which lined bags and containers; improvements to the cold chain

that kept foods chilled from factory to supermarket — in addition to performing other

functions in the food system, these should also be considered technologies of flavor.

Their consequences for the sensory qualities of food shaped the ultimate forms that these

29
Sarah Tracy’s ongoing research into MSG and the taste modality known as umami has
been an intellectual inspiration to this scholar ever since I heard one of her papers, about
umami and the democratization of deliciousness, at a Hagley Library conference on the
history of the senses all the way back in 2013. Her exemplary work draws on scholarship
in science and technology studies to draw connections between the intimate self and
social phenomena, sensual possibilities and biopolitical contexts. Sarah E. Tracy,
“Delicious: A History of Monosodium Glutamate and Umami, the Fifth Taste Sensation,”
(PhD diss., University of Toronto, 2016). See also: Joel Dickau, “Inventing Texture:
Edible Science and the Management of Familiarity, 1963-1975,” Global Food History 3
(2017): 1-23.
30
Recent scholarship has used the history of diet and dietetics as a means to sound the
resonating strings that connect histories of the body, histories of medicine and science,
and political and social history. See for instance: E. Melanie DuPuis, Dangerous
Digestion: The Politics of American Dietary Advice, (Oakland: University of California
Press, 2015); Helen Zoe Veit, Modern Food, Moral Food: Self-Control, Science, and the
Rise of Modern American Eating in the Early Twentieth Century, (Chapel Hill: UNC
Press, 2013); Jessica Mudry, "The Mindful Measurement of Food: Quantification, the
Food Pyramid and Discourses of Taste," Material Culture Review 70 (Fall 2009), 12-22;
David Schleifer, "The Perfect Solution: How Trans Fats Became the Health Replacement
for Saturated Fats," Technology and Culture 53, no 1 (January 2012): 94-119; and Chin
Jou, Controlling Consumption: The Origins of Modern American Ideas about Food,
Eating, and Fat, 1886-1930, PhD Dissertation, Princeton, 2009.
xxxiii
technologies took in the world. Taking the long view, one can discern a distinct tendency

in the development of food technologies over the course of the twentieth century: towards

production, packaging, and distribution methods that preserve foods not only from

spoilage, but also from any chemical changes that could alter the sensory qualities of

food. In other words, the users of these technologies aspired toward maximizing their

control over the sensible matter of food, between the site of manufacture and the ready

mouth of the consumer.

Although this dissertation is primarily concerned with added flavors, the changing

material and technological conditions of food production and distribution provide

necessary context for understanding the role that flavor additives played in this system.

The contemporary flavorists’ work differs from that of her or his predecessor of fifty

years ago, not only because of the expanded palette of flavoring materials and the

growing share of knowledge about the chemistry of flavor, but also because different

production methods and packaging materials require different performances and

properties from flavoring, while also affording distinct sensory possibilities.31

As is the case with other technologies, the uses and meanings of technologies of

flavor were never exclusively determined by their creators. I follow the model of social

historians of technology, who have emphasized the manifold ways in which artifacts are

31
Gary Cross and Robert Proctor’s study of packaging, which persuasively connects the
changing forms and functionalities of containers to an intensifying attention on the
sensual possibilities for the thing contained, is particularly recommended to readers
interested in the largely overlooked (but crucial) history of packages and containers. Gary
S. Cross and Robert N. Proctor, Packaged Pleasures: How Technology and Marketing
Revolutionized Desire, (Chicago: UChicago Press, 2014).
xxxiv
shaped by the technosocial worlds in which they circulate, and the users who determine

their meaning and value, and, in this case, incorporate them into their bodies, habits, and

social lives.32

Flavors are Deliberately Designed Artifacts

If there are any protagonists in this dissertation, they are the skilled workers who,

in the late 1940s, begin to call themselves “flavorists.”

When this story begins in the last half of the nineteenth century, these workers

may have been called ‘practical chemists’ or ‘manufacturing chemists.’ They may have

been trained in a pharmacy, or worked in one of several new branches of food

manufacturing: producing flavoring syrups for bottled carbonated beverages or soda

fountains; flavor extracts for candies, confectionery, or other sweet things; essences for

liquors and spirits; or household extracts for home kitchens. They may have been

employed in the essential oil trade, in spice milling, or in the nascent synthetic perfume

industry. For much of the period covered by this dissertation, they lacked a single job title

or occupational identity. The professional titles most commonly used by those who make

flavors today — ‘flavorist’ and ‘flavor chemist’ — only entered the vernacular after the

Second World War.

32
For instance, Donald MacKenzie and Judy Wajcman, eds., The Social Shaping of
Technology: How the Refrigerator Got its Hum, (Philadelphia: Open U Press, 1985);
Wiebe E. Bijker, Thomas Parke Hughes, and Trevor Pinch, The Social Construction of
Technological Systems: New Directions in the Sociology and History of Technology,
(Cambridge: MIT Press, 1987); Jeffrey Meikle, American Plastic: A Cultural History,
(New Brunswick: Rutgers University Press, 1995); and Regina Lee Blaszczyk, The Color
Revolution, (Cambridge: MIT Press, 2012).
xxxv
 Historians of science and technology have produced numerous important studies

of the work-lives of professionals in technical and scientific industries.33 The story I tell

here of the professionalization of flavorists takes two bodies of scholarship — by

historians of chemistry, and by business historians on the origins and organization of the

chemical industry — as a foundation and backdrop. Rather than the academic chemist

negotiating the disciplinary boundary with physics while investigating the structure of

matter, or the industrial chemist tasked with generating new basic knowledge in the

33
Landmark studies of technicians and the application of technical knowledge in diverse
contexts include: Blaszczyk 2012; Nathan Ensmenger, The Computer Boys Take Over:
Computers, Programmers, and the Politics of Technical Expertise, (Cambridge: MIT
Press, 2010); Deborah Fitzgerald, Every Farm a Factory: The Industrial Ideal in
American Agriculture, (New Haven: Yale UP, 2003); Margarete Sandelowski, Devices
and Desires: Gender, Technology, and American Nursing, (Chapel Hill: University of
North Carolina Press, 2000); Steven Shapin, The Scientific Life: A Moral History of a
Late Modern Vocation, (Chicago: University of Chicago Press, 2010); and Walter
Vincenti, What Engineers Know and How They Know It: Analytical Studies From
Aeronautical History, (Baltimore: Johns Hopkins University Press, 1990). Natasha Dow
Schüll provides an anthropological investigation of the design process of gambling
machines, revealing the expert forms of knowledge, interventions, and adjustments in the
dyadic relation of susceptible bodies and machine affordances, necessary to produce
subjective states of deep play: Natasha Dow Schüll, Addiction by Design: Machine
Gambling in Las Vegas, (Princeton and Oxford: Princeton UP, 2012). My analysis is also
indebted to the sociology of technical work, in particular that of Stephen Barley and
Julian Orr: Stephen R. Barley, "Technicians in the Workplace: Ethnographic Evidence
for Bringing Work Into Organizational Studies," Administrative Science Quarterly 41,
no. 3 (1996); Julian E. Orr, Talking About Machines: An Ethnography of a Modern Job,
(Ithaca, NY: ILR Press, 1996); Stephen R. Barley and Julian E. Orr, eds. Between Craft
and Science: Technical Work in U.S. Settings, (Ithaca: IRL Press, 1997). See also: Steven
Shapin, "The Invisible Technician," American Scientist 77, no. 6 (1989): 554-563; Adele
Clarke and Joan Fujimura, eds., The Right Tools for the Job: At Work in Twentieth-
Century Life Sciences, (Princeton: Princeton UP, 1992); Pamela H. Smith The Body of the
Artisan: Art and Experience in the Scientific Revolution, (Chicago: University of Chicago
Press, 2004); Lissa Roberts, Simon Schaffer, and Peter Dear, eds. The Mindful Hand:
Inquiry and Invention from the Late Renaissance to Early Industrialization, (Chicago:
University of Chicago Press, 2008).

xxxvi
corporate research and development laboratory, I offer a different kind of narrative about

chemical work and careers in chemistry. The skilled workers who made chemicals into

flavors often did not have extensive academic training in chemistry, but they gained deep

practical experience with chemicosensory aspects of phenomena. They gained expertise

through direct experience with sensible materials in the laboratory, but their success also

depended on a thorough knowledge of contexts and applications outside of the laboratory

— how these substances were liable to change, react, and perform under varied

conditions of industrial production and consumption.

In other words, this is a story of chemistry as scientific craftwork, sensual

practice, embodied skill, and constant improvisation.

Flavors are Vectors of Cultural and Social Meaning

Over the years I’ve spent researching and writing this dissertation, I’ve

encountered one question more than any other when I mention the subject of my work,

especially to people outside the academic bubble. It often goes something like this: “I bet

you eat a lot less processed food, since you began working on this?” It’s often not really a

question, actually, but more like a kind of assertion — or a request for confirmation, like

sticking out your arm and waiting for a handshake.

In order to get at what I think this frequent question means, both for the subject of

my work and for me personally as the author of it, allow me to tell you a little bit about

myself, the conditions of my life as an eater, my particular relationship to foods and their

flavors. I was — luckily, I think — raised in a household that took the pleasure of food

xxxvii
very seriously. My mother will gladly recount, in fine detail, a meal that she ate in the

1970s at L’Auberge D’Ill, in the Alsatian town of Illhaeusern, or expound upon the

virtues of tramezzini, vitello tonatto, or the Argentine tart known as pastafrola. My father

was more ecumenical in his tastes, but no less intense in his enthusiasms. He could not

restrain his glee at encountering a good Reuben sandwich, a plate of alfajores, a crispy

apple. He liked to tell us that when he was a child, his grandmother would make two

platters of latkes on the holidays: one for the rest of the family, and one for him.

My mother is an accomplished cook, but she is also a research scientist who toiled

long hours in the laboratories of the National Institutes of Health. My father, like many

men of his generation, was rather helpless in the kitchen. The person who most often

prepared meals at home during my childhood was my grandmother, Haydee Garcia; I

called her nona. She passed away well over a decade ago, but is still intensely missed. I

feel confident in saying that no one who had ever tasted the silken flan that she produced

in a bundt cake pan, and sluiced with translucent caramel, will ever forget it. Friends of

the family, when traveling through Spain, knew to bring back for her tiny glass vials

containing a half-dozen or so scarlet saffron pistils, which she would accept with delight,

and incorporate one at a time into pilafs of fulvous rice studded with tiny green peas. She

boiled cans of Borden’s sweetened condensed milk, for hours, in a big pot on the stove.

When cooled, the cans yielded dulce de leche. I often ate this caramelized goo, spread on

saltines, for breakfast.

As an allegedly adult person, living between Brooklyn and Philadelphia, I’ve

spent many hours at farmer’s markets, weensy gourmandish stores, ethnic groceries, and
xxxviii
mega-supermarkets, looking, trying, talking, buying; in the kitchen, my own or that of

friends, knife or wooden spoon in one hand, a glass of wine in the other, flames in the

background; at the counter, table, or to-go window of hubs of sophisticated noshing “at

all price points,” to use the parlance of our times. I’m a snob, but like all proper snobs,

my list of favorites is always in flux. I like to think I’m open to anything. But yet I find it

really, really difficult—impossible, actually—to eat from a bag of Doritos, to drink a

Coca-Cola, to take my dinner from the freezer and heat it in the microwave. I could not

tell you the exact location of any McDonald’s in New York City. Just the other day,

around the corner from the rented apartment I’ve lived in for nearly a year, on a block I

walk down nearly every day on my way to the park with my dog, I noticed, for the first

time, a Domino’s Pizza, its tenancy in this neighborhood far longer than my own.

In his disquisition on the cultural economies of taste, the uber-French sociologist

Pierre Bourdieu explains that the social logics and practices by which the foods that

people choose to eat, the things that they relish and enjoy, are both constitutive of their

identities and also reflective of their social positions.34 “Tell me what you eat, and I’ll tell

you who you are,” hummed Brillat-Savarin.

The question of how this project has affected my eating habits carries with it two

presumptions: first, that synthetic flavors — and the science and technology of food and

flavor, more generally — only affect “industrial” and processed foods, and that there are

other, “better,” foods out there that are innocent of these interventions. Second, that the

34
Pierre Bourdeiu, Distinction: A Social Critique of the Judgment of Taste, trans. Richard
Nice, (Cambridge: Harvard UP, 1984).
xxxix
more information one possesses about the chemicals that are put into our foods, the more

repulsed and disgusted one is certain to be. Both of these assumptions carve the world

into two distinct parts: the realm of good food with real flavors, and the realm of bad

food, made of chemicals.

As a historian of technology, I’m obligated to point out that all food production

involves technology. Assuming a pre-existing distinction between "real food" and

everything else — a distinction which, today, generally pits the produce aisle ("natural,"

"authentic") against packaged and processed foods ("artificial," "fake") — erases the

manifold ways in which all food is "artificial," mixed up as it is with human knowledge

and labor. When we grow plants for food, we transform them and we transform the

environment they grow in. We also don't just grow food: We cook it, preserve it, ferment

it, subject it to processes that also transform it chemically and nutritionally, that change

its flavor and aroma, endow it with durability, social purpose, and cultural meaning. At

what point do things stop being "real food"? When they are harvested by machines?

When we alter their genomes using biotechnology rather than selective breeding? When

somebody else makes them? When they are made in a factory? Food isn't just edible stuff

out there in the world. It is, and has always been, embedded in human cultures; it

constitutes culture.

As noted earlier, many recent accounts of industrial food have focused on loss,

specifically on the loss of flavor in “real” food—or, in parallel, on the proliferation of

“bad” industrial flavor. This flavor story has a clear moral dimension, distinguishing the

good and the real from the bad and the fake. Currently, it has achieved the status of
xl
conventional wisdom — that we chose—or had thrust upon us—beautiful, insipid apples

over gnarled and speckled fruits of depth and complexity.

There is something inevitably wistful and elegiac about these arguments—a

lament that the rich sensory world, in all its fetid pungency, has been replaced by

scentless, climate-controlled spaces filled with piped-in music; smooth, bland armpits;

Kraft singles sheathed in cellophane; vanilla ice cream containing neither vanilla, nor

cream. This is, I think, a form of declensionist narrative—that standard plot that has been

so potently critiqued by environmental historians. It represents the present world in terms

of its losses, tatters, and absences, rather than its fluxes, flows, and dynamic relations,

emergent forms, new potentials.

These narratives make the serious mistake of taking our current, exceptionally

high valuation of and appetite for intense, distinctive flavor as trans-historical, and even

as biologically natural to human bodies (and to “real” foods). Even as the cultural

relativism of tastes are acknowledged by food scholars in all disciplines, the flavor of

food is almost always regarded as a quality of paramount importance. Was there a time

when flavor, the sensory qualities of food, was not comprehended as part of the

phenomenological world of food and drink? Perhaps not — but what, precisely, flavor

was, ontologically speaking, what its relation was to the substance, material, and value of

foods, the effects it had upon bodies — all these things have changed tremendously, even

in the cultural West, even in the relatively short span of what we call modernity.35 Just as

35
For a comparative overview surveying the changing relation of the flavor of foods to
the bodies of eaters, see: Steven Shapin, “Changing Tastes: How Things Tasted in the
xli
the meaning, power, and ontology of flavor has changed historically, so have the cultural

calculations of its value, as well as the instruments and other means by which the value of

food is calculated. 36 I might even venture to argue that the current high-foodie-culture

valuation of flavor, the valor assigned to unique, distinct, or intense flavors, and the

expectation that “heirloom” varieties of produce also deliver on promises of intensive

flavors, is a result of the flavor-culture produced by industrial food, rather than a rejection

of it.

Another version of this story, the one told most vividly, perhaps, by journalist

Michael Moss in his recent book, Salt, Sugar, Fat, describes flavor technologies that have

been refined to a degree of effectiveness such that we are more or less biologically unable

to resist them. This familiar narrative is one where scientific knowledge has been

leveraged to steamroller the authentic desires and needs of consumers, turning the body

against its own best interests in service of the interests of powerful, multinational

corporations. These narratives are, in essence, a species of technological determinism, but

they are not inventions out of whole cloth. Certainly, the technologies and sciences of

Early Modern Period and How They Taste Now,” Hans Rausing Lecture 2011, Uppsala
University. For an illuminating case study that examines how the relation between the
qualities of food, chemical compounds, scientific knowledge, body, and state was
constituted in Revolutionary France, see E.C. Spary, Feeding France: New Sciences of
Food, 1760-1815, (Cambridge: Cambridge UP, 2014). An informative discussion of the
heated debates over the virtues of culinary “refinement” can be found in Rebecca L.
Spang, The Invention of the Restaurant: Paris and Modern Gastronomic Culture,
(Cambridge: Harvard UP, 2000).
36
For instance, historian Andrew Haley has argued that American diners in the late
nineteenth and early twentieth centuries paid little attention to how food tasted,
evaluating its appeal instead in terms of social contexts, class cues, and gendered
practices of eating, as well as prevailing physiological concerns about ‘digestibility.’ See:
Andrew P. Haley, "The Nation Before Taste: The Challenges of American Culinary
History," The Public Historian 34, no 2 (Spring 2012): 53-78.
xlii
flavor discussed in this account derive their meaning, motive, and value from a particular

social, cultural, and economic context: consumer capitalism. As readers of this

dissertation shall see, flavor additives—and the sciences that construed them—were

explicitly designed to influence consumer behaviors. But to give these technologies

ultimate power not only misrepresents the complexity of social relations around food for

the purpose of designating villains; it also undermines an examination of the other

substantial, structural realities that constrain our choices and limit our possibilities.

In many of these accounts, information is the answer; the individual citizen must

arm herself with knowledge, which will guide her to the foods that will recognize her

goodness (and reward her good choices) by endowing her with historically appropriate

virtues.37 Slimness underscored by strength; a clarity of thought and action reflected in

unblemished, glowing skin; unanxious, energetic vigor; “wellness.” In order to achieve

this, the knowledge of good and evil flavor must be accompanied by an individual

retraining of the will. We must learn to correct our desires, to build an appetite for only

the virtuous things, to be disgusted by the false enticements of industrial foods. Is it a

coincidence that these virtuous foods (constantly changing) are often also the chosen

tidbits of economic elites?

This is not the place to launch into a full-scale critique of these declensionist,

determinist, elitist narratives, nor is this dissertation meant as a defense of the flavor

37
Xaq Frohlich offers a sustained and fascinating critique of what he calls the
“informational turn” in food labeling in his recent work. Xaq Frohlich, “The
Informational Turn in Food Politics: The US FDA’s Nutrition Label as Information
Infrastructure,” Social Studies of Science 2016: 1-27.
xliii
industry, or of the industrialized food system as it is currently constituted. But what I will

say here is this: It seems to me that the interest in a system that provides greater food

equity, that takes less of a toll on the planet, and that does a better job of sustaining

bodies, lives, and communities, is poorly served by critiques that draw hard and fast

distinctions between the virtuous (“local,” authentic, healthy) “real” and the wicked

(industrial, chemical, unhealthy) “fake.”

If we accept that increasing the availability of food that sustains us and provides

pleasure is a desirable social goal and a common good, then achieving this goal will

demand not only changes in individual behavior, but also social investments, economic

and agricultural policies, and educational programs. It perhaps demands even more

profound social and cultural reorganizations and redistributions. It will mean addressing

questions of labor, including uncompensated domestic labor—after all, one of the

implications of "eat less processed food" is "do more of your own food processing." It

will mean drawing on food science and food technologies.

Of course, this doesn't mean a blind acceptance of every "innovation.” An

historical orientation means being well aware that nothing comes from nowhere. We

must follow the multiple, variable, and heterogeneous routes by which things came to be,

delineating the networks of dependencies that shape materials, human bodies and bodies

of knowledge, and life chances. It means trying very, very hard to acknowledge social

and individual costs and consequences, including those that may be less visible,

discounted, or submerged. It also means recognizing that nothing is inevitable, and that

no system is immutable.
xliv
Chapter Summary and Overview

The seven chapters of this dissertation are grouped into three sections, each of

which has a distinct chronological and historiographic orientation. The first section

considers the social, political, and commercial conditions that shaped the market for, and

meanings of, synthetic flavors in the U.S. before the Second World War. Much of the

published secondary literature about the flavor industry begins its story in the postwar

period. By tracing the roots of flavor additives, and the companies and individuals who

manufactured them, to the late nineteenth century, I shed light on how these molecules

became so ubiquitous within the food system and how they contributed to the shape of

industrial systems of food production.

The opening chapter tracks the growth of synthetic flavor manufacturing in the

U.S. from the mid-nineteenth century through the end of the 1930s. By attending to the

contexts in which synthetic flavors were made and used, as well as the networks within

which knowledge about flavoring materials circulated, this chapter traces the increasing

specialization of flavor companies and of the workers they employed.

The second chapter introduces one of the recurring themes of this project: the

contested and evolving meaning of “natural.” I take as my central example the case of

NuGrape soda, whose “genuine” grape flavoring made by a Brooklyn flavor company

was alleged to be “imitation” by regulators. Assessing the conflicting positions of food

officials, flavor manufacturers, and consumers, I untangle the multiple, competing

definitions of “natural” that prevailed in the period between the passage of the 1906 Pure

Food & Drug Act and its 1938 revision and expansion.
xlv
 Part Two of “Flavor Added” concerns a crucial era for the history of

synthetic flavors: the Second World War and the 1950s, a period marked by the

accelerating industrialization of the food system and the growing centrality of processed

food in the American diet. The three chapters in this section consider flavor additives as

deliberately designed technological artifacts, whose inclusion in systems of industrial

food production was mediated by emergent, increasingly professionalized

technoscientific practices of flavor research.

The industrial food system required more than just cheap flavorings. It required a

science of flavor, one that could credibly investigate questions related to the sensory

qualities of food, and develop and implement technical programs for controlling,

standardizing, and improving flavor in manufactured foods. This required not only

identifying the chemical components of foods, but also measuring experiential effects on

sensible bodies. This measurement of sensation is the subject of the third chapter, which

locates the origins of sensory science in attempts to objectively determine flavor qualities

by using panels of human tasters, efforts which began in the 1930s but crystalized during

the Second World War at the U.S. Army Quartermaster Food Acceptance Laboratory.

The expanding variety of packaged “convenience foods” in postwar America

provides the context for the fourth chapter, which examines the relationship between

flavor companies and food manufacturers in the research, development, and production

of new types of food products. I detail the formation of advanced research and

development operations within the flavor industry, and show how new flavor

technologies created by postwar flavor companies was an essential part of their business
xlvi
strategy. The fifth chapter takes a close look at the ideologies, values, and concerns that

informed processes of flavor design by investigating the history of one of the most widely

used tools of flavor evaluation: the flavor profile, a method developed by chemists at a

Cambridge contract research and consulting company. I argue that the flavor profile was

not a neutral technique, and that it profoundly shaped the sensory qualities of postwar

foods in ways that reflected the needs of large food companies producing highly

processed comestibles.

The final section of “Flavor Added” shifts focus to the practices, work-

lives, and epistemic virtues shared by the newly professionalized experts who worked

with flavor in postwar America, during a period when chemistry was transformed by

what scholars have called the “instrumental revolution.” The introduction of powerful

analytic technologies such as gas chromatography and mass spectrometry reshaped the

chemical laboratory, refashioned the identities of analytic chemists, and redefined

industries including petrochemicals, pharmaceuticals, and polymers. This section

interrogates the consequences of instrumental research for flavor science.

In chapter six, I trace the gradual development of the instrumental assemblage of

basic flavor research in the USDA and the academy, beginning with the first gas

chromatography units in the early 1950s and ending in the early 1970s when conjoined

capillary column gas chromatography-mass spectrometry had become standard in the

field. I attend to the specific techniques, technical modifications, and embodied practices

that distinguish flavor researchers from other users of these machines, and consider the

xlvii
special problems of correlating “objective” information about chemical identity with

“subjective” information about its perceptual meaning.

My final chapter considers how the expanding body of basic research about flavor

was applied to the design and development of synthetic flavors at flavor companies. A

rising cohort of creative flavorists, most of them hired after the war, redefined social,

material, and professional norms in their field, and managed an increasingly complex set

of knowledge practices related to chemicals, regulations, and commercial conditions. I

track these changes by following the training regimes, professional virtues, and career

ideals of members of the Society of Flavor Chemists during its first twenty years,

between 1954 and 1974.

xlviii
Chapter 1

Flavor by Formula: Making, Using, and Consuming

Synthetic Flavors Before the Second World War

August Hofmann sucked on a pear drop, and wondered. Hofmann, director of the

Royal College of Chemistry, was a member of the jury for the Great Exhibition of the

Works of Industry of All Nations, the sprawling Victorian fair that, in 1851, assembled

the world’s accumulated technological marvels, priceless gems, mass-manufactured

novelties, and assorted bric-a-brac within a dazzling glass and iron enclosure upon a hill

in Hyde Park, London.38

Among the ferrovitreous arcades of the 1851 Exhibition, the pear drop was, in its

way, as much a marvel as any of the other industrial products on display at the Crystal

Palace. The barley-sugar lozenge had the fruity aroma of a Jargonelle pear, a variety

well-known in England, Hofmann’s adopted homeland. But its resemblance to the pear

was arrived at not by way of the ripened fruit, but from a chemical compound,

synthesized from one of the byproducts of industrial alcohol distillation. As Hofmann

wrote to Justus Liebig, his erstwhile professor, “pear oil,” the substance used to flavor the

candies, was nothing more than amyl acetate, a compound whose odor they both knew

well, diluted in several volumes of neutral alcohol.39 And this was only one of the many

38
Jeffrey Auerbach, The Great Exhibition of 1851: A Nation on Display, (New Haven:
Yale UP, 1999); Peter Sloterdijk, “The Crystal Palace,” Public 37 (2008): 11-16.
39
August Hofmann, “Chemistry Applied to Arts and Manufactures: Application of
Organic Chemistry to Perfumery, from a Letter written by Dr. Hofmann to Prof. Liebig,”
1
“artificial essences” showcased at the exhibition. Perfumers, druggists, and makers of

fine chemicals from Britain, France, and Germany displayed fragrant vials of substances

that evoked the odors of apple, pineapple, and other fruits, as well as “artificial” oil of

bitter almond and of wintergreen. All of these things captivated the scientific interest and

fancy of many observers, who may well have already consumed these synthetic

compounds in candies, liquors, and other beverages.

Hofmann was not able to identify the chemical compounds that comprised all the

artificial fruit essences he sampled. However, he recognized that most of them were

members of a group of organic chemicals then called “compound ethers.”40 As he wrote

to Liebig: “The remarkable fruity odor of many of these ethers had not been overlooked

by chemists.” Indeed, what chemist had not noticed the “insupportable odor of rotten

apples” that “filled the laboratory” when preparing valerianic acid?41 Even if smell and

taste had lost their primary evidentiary status in the quantitative chemistry that prevailed

after Lavoisier, the balance, thermometer, and other instruments had not mitigated the

stinkiness of the chemical laboratory.42

The Chemical Gazette 10 (March 1, 1852): 98-99. Originally appeared in the Ann. der
Chem. und Pharm. vol 71.
40
“Compound ethers” belong to the class of chemicals that are now referred to as esters:
organic compounds comprising an oxygen atom bonded to an alcohol radical and an acid
radical. Compound esters with a fatty acid radical are generally described as having a
fruity smell.
41
Hoffmann 1852: 98.
42
Lissa Roberts, “Death of the Sensuous Chemist: The ‘New’ Chemistry and the
Transformation of Chemical Technology,” Studies in the History and Philosophy of
Science 26.4 (1995): 503-29.
2
 Nonetheless, it would not be the academic chemists, insisted Hofmann, who

would turn their sensory observations about chemical compounds into commercial

products, in order to peddle a dilute solution of amyl valerianate as “apple oil” to

confectioners, or suggest ethyl butyrate as a way of adding a redeeming pineapple flavor

to “bad rum,” or amyl acetate to give a sugar-drop a kiss of pear. “It was reserved to

practical men to make the selection and ascertain the proportions in which certain of

these compounds resembled in so great a degree the aroma of particular fruits that we

almost feel ourselves led to the idea, that these very compounds are the cause of the odor

of the fruits in question.”43 Hofmann went on to speculate that chemical analysis of fruits

may one day prove this to be the case, and that these synthetic compounds might indeed

be identical to those that gave ripe fruits their distinct flavors.

Like the elm trees enclosed within the Crystal Palace pavilion’s glazed interior, or

like the structural architecture of the building itself—designed by gardener Joseph Paxton

after the branching venation along the underleaf of the Victoria regia, the colossal

Amazonian waterlily— the lozenges’ chemical evocation of Jargonelle pear wrenched

nature into novel material contexts and juxtapositions, recalling familiar experiences but

offering entirely new sensations.

How did chemicals become flavors, and how did flavor become a chemical

phenomenon?

43
Hofmann 1852: 99.
3
 The answer to these questions lies with the “practical men,” who had recognized

and seized upon the commercial potential of these newly available materials. Before

researchers began analyzing fruits and flowers to determine the chemical causes of their

flavors and aromas, a diverse group of skilled workers, trained in chemical methods if not

schooled in chemical theory, were capitalizing upon these similarities, refining the

sensible qualities of chemically produced flavorings, and building a market for their

production and use. Within a year of Hofmann’s visit to the Crystal Palace, synthetic fruit

ethers were commercially available in the United States, generally imported from

England, France, or Germany. Soon after, these chemicals began to be produced

domestically in the United States, and found increasingly widespread applications in

manufactured sweets, beverages, liquors, and household extracts. By the end of the First

World War, synthetic flavorings were commonplace. They made the child’s red candy

‘cherry’ or ‘strawberry;’ added the savors of ‘peach’ and ‘vanilla’ to the lady’s afternoon

dish of ice-cream; put the ‘bourbon’ in the laborer’s nightly draught.

When this chapter opens, in the second half of the nineteenth century, synthetic

flavor additives are generally one product among many manufactured by pharmacists and

others with skill as “practical chemists” who make a range of chemical goods. When this

chapter ends, after the First World War, flavor additives have become specialty products

made by workers with unique training and skills, including specialized knowledge in the

manipulation of the sensory qualities of chemical materials. This chapter traces the

emergence and growth of a specialized industry producing synthetic flavors in the United

States, and the concurrent appearance of individuals who fashioned themselves as experts

4
in the creation of flavorings, by examining three dimensions of this transformation: to

materials, methods, and manufacturers.

This chapter unfolds in three parts. I begin by considering the materiality of

synthetic flavors — the chemical compounds that constituted flavorings, their origins,

and the material contexts in which they circulated and mingled. Two substances in

particular are crucial to understanding the expanding market for synthetic flavors in the

nineteenth century: sugar and alcohol. Synthetic flavorings were intimately bound up

with the same industrial processes that made these substances into standardized

commodities, and played an instrumental role in their conversion back into desirable,

sensually attractive consumer goods.

Next, I examine the limited number of chemical compounds used in flavorings,

and consider how they were made to reproduce a cornucopia of fruit flavors. What types

of knowledge and what kinds of skills were required to make synthetic flavors out of

chemicals, and how did this knowledge circulate, accumulate, and change? The formula

is the central figure in this part of the story. On the one hand, if flavors are chemicals,

then anyone with access to those chemicals and some basic chemical training should be

able to make them by following rote formulas. But if the chemical aspect of flavor is

perceived not as a grounds for replication and imitation, but as an opportunity for

innovation, distinction, and discovery — a field for the rapid production of novelty,

bringing food and flavor under the cultural logic of fashion — then an expert with a

different set of skills and resources is needed. The flavor formula — whether public or

5
proprietary, whether conclusive or a starting point — is bound up with the identity and

professional prestige of the individuals who made synthetic flavors.

From here, I turn to the relation between these makers and the companies they

worked for and transacted business with. The earliest makers and users of synthetic

flavorings included pharmacists, distillers, perfumers and essential oil dealers, and

manufacturers of confectionery and syrups, as well as makers of household extracts for

domestic consumers. The groups involved in the production of synthetic flavorings often

had special access to raw materials, special skills in practical chemistry, or some

combination of the two. The claims these early manufacturers made about the virtues of

their products were most often about their chemical purity and freedom from

adulteration, rather than the uniqueness of their qualities or the skill in their blending.

This started to change around the turn of the twentieth century, when specialized

flavor and fragrance companies began to offer proprietary formulations, produced by

skilled workers who blended different flavoring compounds into a finished product, as

well as technical assistance directly to manufacturers. By the end of the First World War,

a growing industry in synthetic flavorings had taken root in the United States, separate

and distinct from its precursors in pharmacy and distillery. As a specialized chemical

industry, the flavor business was generally oriented toward making products that served

the needs of other manufacturers — namely, food and beverage producers — rather than

products intended for ordinary, household consumers.

6
 In this final section, I focus on the story of one exemplary synthetic flavor and

fragrance company, Synfleur, and the career of its founder and chief chemist, Alois von

Isakovics. Beginning in the 1890s as a manufacturer and mail-order dealer of proprietary

medicines, perfumes, and other small retail goods, with the advent of the twentieth

century, the company shifted to producing synthetic aromatic raw materials for

manufacturers. Synfleur staked its place in the market by offering not only a wide variety

of quality chemical compounds and specialties, but also expert advice and products

customized for individual manufacturers. Meanwhile, Isakovics tirelessly educated

manufacturers, students, and other segments of the chemically-interested public about the

science underlying the production of the synthetic aromatic chemicals, aligning this

category of products with those of other progressive industries, and articulating an

argument for the superiority of synthetics over “natural” materials. Embedded in

Isakovics’ chemical writings is a theory of flavor design, one which interrogates the

material conditions which produce naturalistic sensory effects.

I conclude with a brief account of new chemical materials and flavor companies

after the first world war, examining how these companies portrayed themselves as part of

a scientific industry.

I. From Gross Materials to Ethereal Delights

Shortly after the 1851 Exhibition, compound ethers — generally using the name

“fruit essences” — began to appear in the United States, where they were imported,

produced domestically, and sold by druggists, manufacturing chemists, and dealers in

7
essential oils and perfumery.44 From the outset, synthetic flavorings rapidly found their

way into a variety of sweet confections and refreshments, such as confectionery, jellies,

sauces, pastries, syrups, carbonated beverages, and other sweet and sweetened things.

While commercial food manufacturers could purchase flavorings in wholesale quantities,

households could purchase one- or two-ounce retail bottles of flavoring extracts, for use

in baking and cooking.45

In addition to sugar drops, bon-bons, and soda fountain syrups, the compound

ethers also flavored less innocent pleasures: liquors and spirits. The same synthetic

chemicals used in sweets found a ready place in the production of alcoholic beverages,

where they gave neutral spirits the semblance of rum, whiskey, cordials, brandies, or just

about any other liquor imaginable, and “improved” lackluster swill by imparting the

qualities of age and refinement. Amyl acetate, for instance, the substance that added pear

flavor to sugar lozenges, was also recommended for use in “old rye, Bourbon, and

44
“Compound ethers” belong to the class of chemicals that are now referred to as esters,
organic compounds comprising an oxygen atom bonded to an alcohol radical and an acid
radical. Compound ethers with a fatty acid radical were known to have a fruity smell.
45
See, for instance, Centennial Cookbook: J.W. Colton’s Choice Cooking Recipes,
Preparation, and Calendar for 1876-1877, [pamphlet], (Westfield, MA: J.W. Colton Co.,
1876), courtesy Alfred Goossens. This booklet contains recipes for crullers, sponge cake,
and other foods using Colton’s Select Flavors addressed to housewives, endorsements
from politicians and medical professionals, and testimonials from confectioners, hotel
operators, and other commercial food producers. It also contains advertisements for
Colton’s patent medicine formulations, including Nervine tonic. Typical of flavoring
extracts produced at the time, there is little distinction between household and
commercial markets for these products, and, as goods, they are classified with proprietary
medicines, soaps, and toilet articles.
8
Roanoke whiskey” as “its soft, mellow odor” imparted “to any kind of liquor the fine,

soft mellowness of age.”46

In the case of both sweet things and booze, the synthetically produced fruit ethers

were used in conjunction with many other flavoring and coloring materials, both

botanically derived and chemically created. Vanilla bean extract, for instance, became

commercially available around the same time that the fruit essences began to circulate;

vanilla flavorings, often produced by combining genuine beans with synthetic

compounds including vanillin and coumarin, gained rapid popularity as the nineteenth

century drew to a close.47 Taken as a whole, what these new technologies of flavor made

possible was the continual, efficient production of variety, allowing manufacturers large

and small to offer goods that conformed to the fluctuations of consumer desires rather

than simply reflecting natural cycles of availability.48

The meaning of synthetic flavors was also entwined with their distinctly

unappealing, chemical origins. “Some of the most esteemed modern scents are made by

46
Pierre Lacour, The Manufacture of Liquors, Wines, and Cordials without the Aid of
Distillation, (New York: Dick & Fitzgerald, 1868) [Originally published by subscription,
New York: R. Craighead, 1853]: 57.
47
Nadia Berenstein, “Making a Global Sensation: Vanilla Flavor, Synthetic Chemistry,
and the Meanings of Purity,” History of Science 54.4 (December 2016): 399-424.
48
This type of relationship is described in detail by Philip Scranton in his study of
specialty production in American manufacturing during this period. Scranton argues that
custom and batch production played an unheralded role in the expansion of mass
production and the formation of a consumer economy, allowing for flexible, rapid
response to fluctuating market demands. Like many of the specialty and custom
manufacturers Scranton describes, flavor manufacturers clustered in particular urban
regions, employed a specialized labor force, and tended toward a competitive strategy
that emphasized novelty and quality over price reductions. Philip Scranton, Endless
Novelty: Specialty Production and American Industrialization, 1865-1925, (Princeton
UP: 1997).
9
chemical means, from materials which are generally considered anything but pleasant,”

marveled one account of synthetic perfumes and flavorings displayed at the 1853 New

York Exhibition, articulating a sentiment that would often be repeated in popular

scientific literature.49 Although it underscored the rude origins of synthetics, the

sentiment was not entirely negative. It is as though the triumph of chemistry over nature

was magnified by the reclamation of pleasurable substances from repulsive materials. As

\ a book of practical chemical formulas published in 1860 informed readers, "The

majority of the fruit extracts which are manufactured for sale are artificial…. Some of

them — I will not say what ones — are made from the drippings of horse stables, and

most delicious to the taste!”50

As far as I can tell, horse excrement was not a component of any known

flavorings.51 However, most of the synthetic fruit ethers were made from another noxious

substance: ‘fusel oil.’ This foul-smelling, sickening liquid was a mixture of compounds,

mainly amyl alcohol and other higher alcohols — ie, those with more carbons than ethyl

alcohol’s two — separated from ethyl alcohol and other desirable substances through

distillation. In other words, it was a waste product. “It will strike the reader as not

unworthy of remark,” instructed one popular chemistry textbook published in the 1850s,

49
Charles Rush Goodrich, ed. Science and Mechanism: Illustrated by Examples in the
New York Exhibition 1853-1854 (New York: Putnam & Company, 1854): 242.
50
A.W. Chase, Dr. Chase’s Recipes, or, Information for Everybody: An Invaluable
Collection of About Six Hundred Practical Recipes… 8th ed. (Ann Arbor, MI: A.W.
Chase, 1860): 177.
51
I believe Chase’s reference is not to a flavoring, but to the alleged origins of the
perfume (and medicine) known as “eau de millefleurs,” which was, by many accounts,
made from cow urine or dung, and dates to the reign of Louis XV, if not before. The
emphasis on the abject origins of these substances of pleasure was familiar rhetoric for
both perfumes and flavors.
10
that the same substance that “because of its offensive smell and taste is carefully removed

by the rectifier from the ardent spirits he distils, should, under the hands of the chemist,

become possessed of the most agreeable and coveted fragrance!”52 (‘Pineapple essence’

— ethyl butyrate — was the exception; rather than being synthesized from fusel oil, it

was “obtained by fermenting a mixture of sugar, sour milk, a little old cheese and some

chalk,” according to contemporary sources.)53

The process of converting fusel oil into fruit essences was a chemical procedure,

requiring other harsh and unpalatable substances — such as potash and ‘oil of vitriol’

(sulfuric acid) — as well as substantial quantities of neutral spirit (ethyl alcohol). But to

many contemporary commentators, the creation of these products seemed to require more

than chemical skill. One British account of the operations of a London chemical

manufacturer portrays Mr. Routledge, the firm’s extract maker, as a fine artist:

“With sundry bottles of ethereal compounds before him, ranged like the
colours in a painter's palette, he adds ounces of one, drops of another, and
mere hints of others, until he ultimately finds that he has made the essence
required. We might as well ask the artist how he mixes his russets and
purple grays, as ask Mr. Routledge how he makes artificial ribstone
pippins and raspberry out of ethers whose origin is to be sought for in
stinking cheese and the foulest fusel oil."54

52
James F.W. Johnston, The Chemistry of Common Life, Vol. 2, 8th ed., (New York: D.
Appleton, 1856): 202.
53
W. Bastick, “Artificial Essence of Pineapple,” in G.W. Septimus Piesse, The Art of
Perfumery…2nd American edition, (Philadelphia: Lindsay & Blakiston, 1867): 384-5.
54
“A Visit to Messrs. Davy and Macmurdo’s Chemical Works at Bermondsey and Upper
Thames Street,” in Dr. G.L.M Strauss et al. England’s Workshops, (London:
Groombridge & Sons, 1864): 158-9.
11
 The skill required to make these flavorings was craft-based and artisanal, the

same sort of tacit knowledge about materials that guided a painter’s use of paint. At once

“the products of the chemist’s science and the manufacturer’s art,” gross materials were

transformed into ethereal substances that could then impart the flavor of “strawberry,

pineapple, apricot, quince, raspberry, green gage [plum], mulberry, black currant, &c.” to

“syrups, jellies, blanc mange, cordials” and other confections.55

Making Commodities into consumer goods

Synthetic flavorings bound together two substances transformed by

industrialization: sugar and alcohol. Sugar and alcohol, eminently versatile materials,

have played multiple and changing roles in cultural and social life: as components of

luxury goods, medicines, and preservatives, as well as agents of sweetness and

intoxication. The meanings and uses of these substances, and the sensations and pleasures

associated with them, shape and are shaped by historical, economic, and technological

forces.56 During the nineteenth century, both of these substances became mass-produced

commodities, and were made homogenous, pure, and standard by new chemical

55
“Art vs. Nature,” The American Journal of Pharmacy and the Sciences, (April 1852):
184.
56
Sidney Mintz, Sweetness and Power: The Place of Sugar in Modern History, (New
York: Viking, 1985); Wendy Woloson, Refined Tastes: Sugar, Confectionery, and
Consumers in Nineteenth-Century America (Baltimore: Johns Hopkins University Press,
2002); Iain Gately, Drink: A Cultural History of Alcohol, (New York: Gotham, 2008).
12
techniques and industrial technologies.57 But these commodities were not yet consumer

goods. Their standard and uniform properties — a chemical purity that could come to

seem like an absence of qualities — had to become specific and ‘impure’ in order to gain

maximum value in the consumer market.58 Drinkers craved whiskey, not ethanol; bon-

bons were delectable, not sucrose.

Synthetic flavorings, and the sensations they produced, played a crucial role in

transforming these standard commodities into desirable consumer goods, into objects of

fashion, pleasure, and value. At the same time, the standard chemical composition of

industrially produced alcohol and sugar made these materials the ideal media for the

conveyance of deliberate, designed flavorful experiences. But, for the status and

reputation of synthetic flavorings, getting mixed up with these two commodities had

different outcomes. The dubious reputation of alcoholic beverages during a time of

growing temperance sentiment, as well as commercial and political divisions within the

spirits industry, cast suspicion on the quality, safety, and honesty of added flavorings. In

this context, synthetic flavors came to be seen as inherently fraudulent. And while

57
For an illuminating account of the scientific, social, and technological labor necessary
to make sucrose the standard of sugar, see David Roth Singerman, “Inventing Purity in
the Atlantic Sugar World,” (PhD diss., MIT, 2014).
58
The meanings, and relative values, of purity and its opposite would come to be a
subject of contention during the debates around the 1906 Pure Food and Drugs Act. For
chemists, a pure substance is one that is chemically homogenous; outside of the
laboratory, purity indicates wholesomeness and soundness, and carries with it moral,
physical, and even spiritual virtues. Although regulators enforced the meaning of purity
consistent with public understanding, as chemists they had to concede that for many
products – such as maple sugar, and whiskey -- the consumer could be said to pay a
premium for the impurities. See, for instance, James H. Shepard, “Like Substances,” Pure
Products 3.11 (November 1907): 507-13. Shepard was a chemist with the South Dakota
Pure Food Commission.
13
synthetic flavors did carry an ambivalent reputation among manufacturers and consumers

of sugary treats and refreshments, their use also facilitated the production of an

expanding and dazzling array of sweet substances, and made a new kinds of consumer

experiences — the experience of limitless variety, of immediate pleasure — imaginable,

possible, and accessible to an emerging mass market.

‘Any kind of liquor that you want in five minutes’ notice’ 59

Industrial alcohol distillers and rectifiers played an early, crucial role in the

emergence of a trade in synthetic flavorings. The connections between industrial

distillation and flavor manufacturing are manifold, on both the supply and the demand

sides. Distillers and rectifiers were users of flavoring extracts, and also supplied flavoring

manufacturers with raw materials (fusel oil, esters, and purified ethanol, the universal

menstruum for flavoring extracts).

Alcohol is a naturally occurring byproduct of fermentation. As yeasts and other

microorganisms colonize a fruit juice or grain mash, they break its sugars down into

molecules of ethanol, meanwhile perpetrating chemical transformations that yield higher

alcohols as well as other compounds, both fragrant and obnoxious. Spirits such as

whiskey and brandy are produced by further distillations of this initial fermentation. The

key tool of distillation is the still, where the fermentation liquid is heated, and

differentials in boiling point are used to separate ethanol from water and other chemical

59
From the Testimony of James M. Veazey to the U.S. House of Representatives,
Committee on the Judiciary, Report on the Whisky Trust Investigation, 52nd Congress,
2d. Session, (March 1, 1893): 10.
14
components — including toxic methanol and acetone, fusel oil, and a group of substances

known as ‘congeners,’ which contributed flavor, richness, and body.60

Until around 1830, the copper pot still was the standard technology of distillation.

Producing spirits in this way was a batch process, and required skilled, attentive labor and

plenty of fuel. The end result was not purified ethanol, but ethanol mixed with selectively

limited quantities of fusel oil and congeners. Further steps, including aging in wood casks

and a series of subsequent distillations known as rectification, were generally required to

develop desirable flavors and diminish harsh and uninviting ones.61 A skilled distiller was

a respected artisan who could bring out the treasured qualities of a spirit through careful

management of the process of production. The quality of the spirit was often greatly

influenced by the quality (and cost) of the raw material used in distillation, as well as the

time spent aging, and it was difficult to maintain standard properties from batch to batch.

Although the highest quality spirits continue to be produced using pot distillation, the

60
Harold McGee, On Food and Cooking: The Science and Lore of the Kitchen, revised
edition, (New York: Scribners, 2004): 713-8, 758-71. Congeners included compounds
such as esters, terpenes, and phenolics, which added characteristic and valued flavors,
richness, and body. While most spirits are largely composed of ethanol, different
congeners account for the distinct taste of whiskey and bourbon, or rum and rye.
61
Aging (especially in wood) improves the flavor of whiskey in various ways. Fusel oil
and congeners oxidize, developing into molecules with more prized sensory qualities.
Compounds from the wood barrel itself also leach into the alcohol, reacting with
chemicals in the whiskey and undergoing other desirable chemical changes. Some of the
oxidative changes that occur to fusel oil during the process of aging result in the same
ester compounds that chemists synthesized.
15
technical and material requirements of this technology limit the scale and speed of

alcohol production.62

Distillation can also be performed with a fractionating column, which separates

substances of different boiling points at condensing plates arranged within an elongated

cylinder. Several attempts to render this principle into a practicable continuous

distillation device preceded the successful design patented in 1830 by Aeneas Coffey, a

retired Irish exciseman. Coffey’s two-column, steam-heated ‘patent still’ was extremely

efficient, could be operated continuously, and did not require the close monitoring of an

expert distiller. It reliably produced a concentrated spirit containing between 86 and 96

percent ethanol. This efficiently achieved purity meant that variations were kept to a

minimum. Coffey’s patent still, and similar devices that followed, made it possible to

produce a spirit that approached the status of a homogenous commodity.63 Once alcohol

became a standard commodity, it could readily assume industrial applications and

purposes. It also made the other, secondary, compounds — such as the fraction of fusel

oil — available as raw materials for other chemical processes.64

62
McGee 2004: 761-3; R.B. Weir, “Distilling and Alcohol, 1870-1939,” Agricultural
History Review 32.1 (1984): 50.
63
Weir 1984: 50.
64
Fusel oil had once been considered largely a waste product of distilling. One of the
earliest uses for fusel oil was the production of synthetic flavorings, and the importance
of the material grew as it began to be used in an increasing number of chemical
processes, including the manufacture of celluloid, pyroxylin varnishes, photographic
films, and alkaloids. Before the First World War, Russia was a major exporter of fusel oil
to the United States. The Russian revolution and the expansion of prohibition caused
industrial chemists to search for alternate raw materials to replace the tightening supply
of this material. See R. Schupphaus, “On the Alcohols of Fusel Oil,” Journal of the
American Chemical Society 14.3 (1892): 45-60; Benjamin T. Brooks, Dillon F. Smith,
16
 But even as the continuous still created new efficiencies, markets, and

opportunities for alcohol distillers, commodity alcohol was of little value to consumers.

Simply put, the same processes that made alcohol standard, also stripped it of the

compounds that produced its flavor and other sensible qualities. In order to become a

desirable beverage, to compete with products with known market value, manufacturers

had to add substances that contributed color, flavor, and aroma.

In the United States, this was performed by a group of licensed professionals

known as rectifiers, who blended neutral spirits with synthetic and botanical flavorings,

or with “straight” (ie, distilled and aged) liquor, to produce branded products that were

then sold to wholesalers.65 These “blended” spirits also tended to be safer for consumers,

as their mode of production meant they generally contained lower quantities of fusel oil.66

Blending allowed for the large-scale, efficient production of liquors and spirits; cost

savings could then be passed on to consumers. But it also divided the industry, pitting the

interests of producers of “straight” goods from those who made and sold “blended”

liquors. It also divided lawmakers and consumers, many of whom did not consider the

rectified and flavored product to be “genuine” liquor, but imitations, of lesser quality and

with diminished medicinal effectiveness.

and Harry Essex, “The Manufacture of Amyl Acetate and Similar Solvents from
Petroleum Pentane,” Journal of Industrial and Engineering Chemistry 10.7 (July 1918):
511-15.
65
Jack High and Clayton A. Coppin, “Wiley and the Whiskey Industry: Strategic
Behavior and the Passage of the Pure Food Act,” Business History Review 62.2 (Summer
1988): 286-309; Werner Troesken, “Exclusive Dealing and the Whiskey Trust,” Journal
of Economic History 58.3 (1998): 755-78.
66
High and Coppin 1988: 291.
17
 Because synthetic flavorings were necessary to the production of blended

whiskies, in the 1890s, the products were caught right in the middle of one of the earliest

scandals of monopoly capitalism, and drawn into the midst of Congressional hearings on

the Whiskey Trust. The Whiskey Trust, or, as it was officially known, the Distilling and

Cattle Feeding Company, was an organization of distillers who produced neutral spirits

for the manufacture of rectified whiskey, not “straight whiskey.” The industrialization of

alcohol manufacturing had lead to overproduction and falling prices, exacerbated by

rising imports of potato- and grain-based spirits from Europe.67 The Whiskey Trust

formed in the 1880s in response to these perilous economic conditions. By limiting

production across their network of distillers, they kept prices from plunging below

sustainable levels. By the time of the 1893 Judiciary Committee investigation, they

dominated the market — producing more than 95 percent of all the spirits legally

manufactured in the United States.68 This market dominance allowed the Trust to develop

a system of rebates to compel wholesalers and merchants to buy exclusively from them;

the effect was to further drive competitors out of the market and exert a monopolistic

control over prices.

The main question before the House Judiciary Committee was whether the

Whisky Trust was engaging in anti-competitive practices.69 Inextricable from this

investigation of commercial practices was an inquiry into the substance of the product

67
High and Coppin 1988; Troesken 1998; Williams Haynes, American Chemical
Industry: A History, Vol I: Background and Beginnings, (Toronto: D. Van Nostrand,
1954): 320-1.
68
Troesken 1998: 760.
69
U.S. Congress. House. Committee on the Judiciary. Report on the Whisky Trust
Investigation, 52nd Cong., 2nd sess., 1893, H. Rep. 2601.
18
they were manufacturing, whether there was something suspect or against public interest

inherent in the very nature of rectified whisky. Indeed, many in Congress wondered

whether it could rightfully be called whisky at all.

The Judiciary Committee hearings kicked off with a bombshell whistleblower as a

witness, James Veazey. Born in 1854, in Hamilton County, Ohio, Veazey had worked as

a traveling liquor salesman since 1878, peddling whiskies, brandies, gins, and other

spirits for a half dozen companies in Ohio, Kentucky, and Illinois, before a health crisis

precipitated his retirement from the road. This included three years working for

Alexander Fries & Brothers, chemists, of Cincinnati, where, he became privy to "what is

known as the 'secrets of the liquor trade.'" He assures the Judiciary Committee: "I

became acquainted with its entire manipulation."70

Over two days of testimony, Veazey let the members of the Committee in on the

"secrets of the liquor trade," showing them exactly how a dealer could produce "any kind

of liquor that you want" with "five minutes' notice." The transcripts record a man

unspooling an easy, confiding patter:

"Say an order comes in for any class of goods, say Jamaica rum; Jamaica
rum essence is put into [spirits] and it is colored with burnt sugar and the
name branded upon it as the law requires it shall be stamped, and away it
goes. Say another order comes in for gin, and the spirits is filled out of the
same tub, flavored with gin essence, colored with sugar, sirup, or glucose,

70
Testimony of James M. Veazey, Saturday, February 4, 1893, Report on the Whisky
Trust Investigation: 1-16.
19
 and away that goes. Yes, sir; anything you want, and it is generally in use,
and represents to-day one-half of the liquor business of this country."71

Veazey answered the Congressmen's questions, providing documentation at

times, but drawing dramatic authority as a witness by invoking his personal experience.

For instance, when asked if the flavoring essences are poisonous, he replies: "I am not a

chemist, but I have been warned when in the employ of these people not to take the crude

material in my mouth."72

On his second day of testimony, Veazey added some show to his big tell. He

brought in two demijohns of spirits, as well as "a number of bottles containing essential

oils, essences, etc.," and stirred up a full bar's worth of libations for the Judiciary

Committee.73 Beginning with neutral spirits, he added a drop of Jamaica rum essence,

some coloring, some simple syrup, and passed out tumblerfuls for the members to

sample. "Does it smell like rum and taste like it?" he asked. I picture the tippling

congressmen nodding in affirmation, all except the most teetotal of the bunch, who

perhaps deigns only to stick his long and disapproving nose into his tumbler to take a

long and disapproving sniff. Veazey then demonstrated the effect of another additive

("bead oil") that altered a watered-down rum so that it ran thicker, with the viscosity of

full-strength liquor. He mixed up rye whiskey, then "aged" it with other essences, prune

71
Ibid: 14.
72
Ibid.: 7.
73
For a similar performance of the fraudulence and allure of ready-made liquors, see: Eli
Johnson, Drinks from Drugs, or the Magic Box: A Startling Exposure of the Tricks of the
Liquor Traffic, (Chicago: Revolution Temperance Publishing House, 1881).
20
juice, and raisin oil, to imitate successively older bottlings — three-year, five-year, and

even "velvet" whisky, aged thirty years in oak casks.74

Throughout his testimony, Veazey underscored that the ultimate dupe is the

consumer. "The average man... is unable to protect himself, not understanding these

imitations... at the time of purchase... falsely represented to him."75

But what, really, makes the imitation so deplorable? Consider that the

persuasiveness of Veazey's demonstration depended on the undetectability of the

imitation, on the high quality of the flavoring. If whisky, rum, cognac made from alcohol

and flavoring essences were bad imitations, then they would be less of a problem; frauds

could be sniffed out, unscrupulous agents and manufacturers driven out of the market if

substantially inferior to the real thing.

From the perspective of the chemists who manufactured flavoring essences, their

products were directly related if not chemically identical to the compounds that gave

"straight" liquors their flavors. Entered into the Congressional Record of this

investigation is the complete text of a Manual for Compounders, published by Fries &

Brothers — a handbook for users of their flavoring essences. "All natural old liquors

(straight goods) contain certain odorous compound ethers arising from fermentative

processes and slow oxidations," instructed the manual. But these sluggish processes can

74
Testimony of James M. Veazey, Saturday, February 4, 1893, Report on the Whisky
Trust Investigation: 14.
75
Veazey’s testimony was itself a fraudulent act, connected with a naked short selling
stock scam. See: Nadia Berenstein, “Who’s Afraid of the Whisky Trust,” Flavor Added
Blog, entry posted October 10, 2015, http://nadiaberenstein.com/blog/2015/10/30/whos-
afraid-of-the-whisky-trust (accessed August 15, 2016).
21
be abbreviated by chemical reactions, producing ethers that are "the synthetical

reproduction of those manufactured in nature's laboratory." Moreover, chemists who

manufacturing flavoring essences often began with a raw material sourced from alcohol

distillation — fusel oil, those higher alcohols, removed during distillation and otherwise a

waste product. The question was whether the transformation of an undesirable waste

material to a pleasant and valuable one would be effected by the oxidative effects of time,

or the directed and deliberate efforts of the manufacturing chemist.

In other words, if the way that whisky changes as it ages in the barrel can be

comprehended as a chemical process, then why not reproduce that process more

efficiently, and thus more cost-effectively? Was this not one of the imperatives toward

improvement that drives innovation? Yet this argument failed to be persuasive to many of

the Congressional inquisitors and witnesses, who seemed to accept that there was

something inherently inferior about whiskey produced this way.

The Congressional inquiry had little effect — it was unclear whether it possessed

the legal authority to break up the corporation — though the Trust itself filed for

bankruptcy in 1895, and subsequently reorganized in a less market-dominant form.

However, the legitimacy and value of blended spirits, made with synthetic flavorings,

continued to be in doubt, and would inform debates into the Pure Food and Drugs Act.76

As Harvey Wiley, Chief Chemist of the Bureau of Chemistry and one of the law’s

proponents, explained in testimony, artificial flavorings were “chemically the same as

those which are produced by the natural methods of aging in whiskey,” but “there is

76
High and Coppin 1998.
22
something lacking… While you can imitate nature, you cannot substitute the artificial for

natural products without impairing the quality of the product.” This was an “almost

indescribable” distinction that exceeded the powers of chemistry to define. “The stomach

and the system are very expert wine tasters and whiskey experts, and they will detect the

difference… which the chemical laboratory fails to distinguish.”77

The imitation represented a diminution of quality, a difference in “effect,” that

only the sensing body, not the skilled chemist, could register. There appeared to be a

connection between the (allegedly) illicit profits of the Whisky Trust, and the specious

flavor of ready-made whisky — both seemed unearned, dubious, untethered from solid

virtues and values. This low reputation would continue to bedevil both manufacturers of

synthetic flavors and the products they manufactured.

Sweetness and Variety

The industrialization of sugar production made that substance, once a refined

luxury, a “prolific necessity.”78 As domestic sugar production increased, and with sugar

cane cultivation in the American territory of Hawaii, American consumption of refined

sugar surged, particularly in the last quarter of the nineteenth century.79

Much of Americans’ increased sugar consumption can be accounted for in

manufactured foods, which packaged sweetness in a growing range of forms. New steam-

77
Testimony of Harvey Wiley, February 28, 1900, Senate Committee on Manufactures,
Adulteration of Food Products, 56th Cong., 1st sess., 1900, S. Rep. 516: 56.
78
Woloson 2002: 3.
79
Woloson 2002: 5-6.
23
powered machines made it possible to efficiently turn refined sugar into cheap candies:

wafers, lozenges, cream centers, bon-bons, kisses, gum drops, and more.80 The value of

candy manufactured in the U.S. grew from $3 million in 1850, to more than $60 million

at the century’s end.81 Variety was the soul of the candy business; constant novelty was

an imperative. One business expert, in 1915, estimated that the average wholesale

confectioner offered between fifteen and six hundred different kinds of candy, with some

listing more than a thousand.82

Soda pop was another increasingly popular product that owed much of its appeal

to sweetness. By the beginning of the twentieth century, parched Americans could

quench their thirst at one of approximately a hundred thousand soda fountains, from

ornate marble-and-mirrors fountains in ice cream parlors and upscale drugstores,

department stores, ice cream parlors, to humbler fountains in train stations, five-and-ten

cent stores, and sidewalk stands.83

Soda fountains competed for trade through the encyclopedic range of flavors they

made available, with new offerings creatively named to latch onto the latest trend. (For

instance, at the height of the bicycle fad in the 1880s, fountains offered the “pedal

pusher,” “sprocket foam,” and “cycla-phate.”)84 Large soda fountains might have more

80
Woloson 2002: 35-9; Samira Kawash, Candy: A Century of Panic and Pleasure, (New
York: Faber and Faber, 2013): 27-43.
81
Kawash 2013: 29.
82
Kawash 2013: 40.
83
Anne Cooper Funderburg, Sundae Best: A History of Soda Fountains, (Popular Press,
2002): 101.
84
Funderberg 2002: 45.
24
than a hundred flavors on their menus, ready to prepare at a customer’s request.85 An

article in Scientific American, in the summer of 1899, explained the economic role of the

new and unique flavor in the soda fountain trade. A soda dispenser’s “knowledge of

syrups, waters, and chemicals enables him to mix different ingredients together which

will produce a flavor peculiar to itself.” This dreamed-up flavor, available nowhere else,

“may have no other virtue. But if it is properly named and skillfully advertised, it may

have a ‘run’ or a season that will pay big profits.” The soda fountain operator did not

expect to profit from this novelty forever. “He is satisfied if it will take for a few weeks

or months.” Of the countless new flavors introduced every year, fewer than one percent

ever had any lasting success, according to the writer. But this cycle of novelty was a

driver of sales as much as the reliable familiar flavors.86

Joining the soda fountain was a business in bottled carbonated beverages, which

began to expand rapidly when the price of sugar dropped after the Civil War. Bottlers’

flavors became (and remain) a specialized branch of the flavoring industry, as these

products have unique technical requirements dictated by their method of production and

distribution. The dominant economic model for the manufacture of brand-name

carbonated beverages consists of the distribution of flavorings to regional, independent

bottling plants, which manufacture and bottle beverages under contract.87 This places a

high premium on batch consistency, flavor stability, and price control, properties that

85
Funderburg 2002: 44.
86
G.E.W., “Some Soda Water Fountain Statistics,” Scientific American 81.7 (August 12,
1899): 99.
87
See Bartow J. Elmore, Citizen Coke: The Making of Coca-Cola Capitalism, (New
York: Norton, 2015) for an illuminating account of some of the business factors that
produced this model.
25
synthetic flavorings could deliver much more readily than fruit-based syrups or botanical

extracts.

Synthetic flavorings (and colors) did not simply make it possible to manufacture

candy and soda pop at low cost for wide distribution. They made possible an experience

of dazzling variety and choice that had been previously unimaginable, at least among less

elite eaters. Synthetic flavors were not limited by seasonal and geographical patterns of

cultivation that governed the fruits of the vine and the orchard; a synthetic pineapple did

not have to be grown in Hawaii, a synthetic strawberry could be sampled in dead white

winter. Soda fountains and confectioners did also use “true fruit” flavors, concentrated

juices and syrups deriving their flavor only from fruit and sweeteners, but these “natural”

flavors had liabilities. Fruit juices were difficult to concentrate and preserve from

fermentation without developing an undesirable, ‘cooked’ taste.88 Further, they tended to

cloud or leave undesirable sediment in bottled beverages.89 Synthetic flavors presented

none of these challenges.

Then there is the question of intensity. “‘There is mighty little genuine fruit

extract in the sirups and flavors of commerce,’” remarked the chemist of a flavor

manufacturing house, quoted in a syndicated article from 1881, while “pushing aside

glass jars, strainers, and retorts, so as to make a clear space for some of his books and

88
Andrew Smith, Drinking History: Fifteen Turning Points in the Making of American
Beverages, (New York: Columbia UP, 2013): 143. Smith observes the most common
way of preserving fruit juice was by turning it into alcohol — fermenting it. He notes that
the Shakers and the Oneida Community both developed techniques for concentrating and
preserving unfermented fruit juice, which may be attributed to their prohibition on
alcohol.
89
Funderburg 2002: 46.
26
formulas. ‘Natural flavors are both weak and costly.’”90 The weakness of nature is

contrasted with the power of the synthetic, its efficiency in delivering flavor sensations.

But even though synthetic flavors offered great advantages, and were widely

commercially available by the 1870s, they were not universally used. The choice to use

synthetic flavors appears to have depended on the reputation of the manufacturer or

merchant, and the class of customer served. This is documented in an 1873 report on

flavoring additives prepared by Henry K. Oliver, a medical doctor, for the

Massachussetts Board of Health.91 His investigation began as an attempt to determine

whether artificial essences were harmful, and in what quantities they could be safely

consumed — questions to which he ultimately could not provide definite answers,

although he warned against “habitual indulgence.” Oliver interviewed confectioners and

makers of fruit jellies, visited druggists and apothecaries who operated soda fountains,

and wrote to flavoring manufacturers and liquor dealers, inquiring into their use of these

products. His report offers a picture of the market for artificial fruit essences at the time.

In order to find foods made with synthetic substances, Oliver had to do a bit of

slumming. Tracking down jellies made with synthetic flavors necessitated a visit to a

district of “second-class grocers,” where Oliver found deep-hued “currant” and other

fruit-flavored jellies selling for 20 or 25 cents, less than half the price of the presumably

90
“How Flavoring Extracts are Made,” Iron County [MO] Register, July 21, 1881.
[Reprinted from New York Sun]
91
Henry K. Oliver, M.D. “Report on the Character of Substances Used for Flavoring
Articles of Food and Drink,” Annual Report of the State Board of Health of
Massachusetts, Vol. 4, (January 1873): 145-74.
27
genuine jellies sold by more prestigious grocers.92 Boston confectioners “of excellent

repute” did not use artificial essences, and thus were able to offer only a limited number

of flavors as a result.93 Meanwhile, a manufacturer of popular candies “sold principally in

the street and in places of public resort, railroad stations, etc.” used artificial flavors

exclusively.94 As for alcoholic beverages, he found that most of the spurious liquors were

sold not in Boston, but by low-class dealers in small towns along the city’s margins.95The

more elite the clientele, and the more well-heeled the district, the less likely Oliver was to

encounter merchants that admitted to using artificial flavorings.

But it cannot be presumed that artificial essences were mainly consumed by the

down-and-out. Oliver summarized the advantages of artificial fruit flavors for

manufacturers. “The list of flavors could be greatly enlarged; perishable and rare fruits

could be cheaply imitated in flavor by substances unchangeable and always at hand, and

most persons would fail to detect the imposition.”96 There were practical reasons for

using synthetics. Oliver mentions “S,” a candy manufacturer of “good reputation,” who

nonetheless used some artificial flavors in his products. “Desires to have a good list of

flavors,” Oliver noted, “and finds it difficult to use fruit-juices in any but soft candy, on

account of their watery element.” In other words, S. used artificial flavors not only to

expand his range of flavors, but because they were materially more compatible with his

production processes for hard candy, as they were more concentrated and in alcoholic,

92
Ibid: 165. Oliver testified that the cheaper jellies were mostly bland apply jelly,
doctored with flavorings and colors.
93
Ibid: 160.
94
Ibid: 161.
95
Ibid: 169.
96
Ibid: 148.
28
rather than aqueous, solution.97 The rumored “opinion” of “some chemists” that “the odor

and flavors of flowers and fruits are really due to the presence of these ethers” had

probably also “greatly encouraged their employment.”98

Oliver also discovered that artificial flavorings themselves varied in quality and

price, and manufacturers had options when it came to procuring or even making their

own flavorings. One manufacturer of popular candies claimed to make his own essences,

“from the best materials. They cost him nearly twice as much as those which he formerly

bought…. Thinks the cheap essences are bad, but has a very different opinion of those

made by himself.”99 Another manufacturer of artificially flavored jellies claimed to pay

“the highest price” for artificial fruit essences from a company in New York; his

customers, he said, could not tell them from the real thing. Oliver himself agreed with

this after sampling the currant jelly: “the taste decidedly resembled the currant flavor, so

that it would generally pass for the genuine article.”100 His report repeats the claim of one

New York imitation fruit essence manufacturer: “when properly made,” he wrote, the

artificial essences “are often preferred to pure fruit.”101

Nonetheless, the diminished reputation of synthetic flavors, their low-class

cultural associations, the possibility of fraudulence and unwholesomeness that hovered

around them, meant that even well-made synthetics bore the stigma of their chemical

97
Ibid: 160.
98
Ibid: 148.
99
Ibid: 161.
100
Ibid.: 165.
101
Ibid.: 163.
29
origins. Under these circumstances, it might be better not to disclose the use of

synthetics, and instead, allow the flavor to speak for itself.

What did it mean for these products to be “properly made”? How did information

about making flavors circulate? How did manufacturers attempt to improve the sensory

qualities or use-value of their products, or distinguish themselves from competitors? The

flavor formula, a chemical tool that could be published and shared or kept secret and

obscure, sheds light on how synthetic flavor makers in the late nineteenth and early

twentieth century made their products and built their trade.

FLAVOR BY FORMULA
The artificial flavoring extracts are frequently known as ‘Fruit Ethers,’
and sometimes ‘Fruit Oils.’ Many of the ethereal ingredients of these
extracts have received in the trade special, significant names. For
example, amyl acetate is known as ‘Pear Oil,’ amyl valerianate as ‘Apple
Oil,’ butyric ether as ‘Pineapple Oil’ and ‘Rum Ether,’ oenanthic ether as
‘Oil of Wine’ and ‘Grape Oil,’ and sometimes as ‘Cognac Oil,’ although
various mixtures are also frequently sold under the latter designation. –
A. Emil Hiss, The Standard Manual of Soda and Other Beverages, 1901102

Until the beginning of the twentieth century, the synthetic chemical compounds

used in flavoring additives were more or less limited to the small set of compound ethers

listed above, as well as a group of synthetic materials that claimed identity with naturally

occurring chemicals: vanillin (artificial vanilla), coumarin (artificial tonka bean),

102
A. Emil Hiss, The Standard Manual of Soda and Other Beverages: A Treatise
Especially Adapted to the Requirements of Druggists and Confectioners, 10th ed.
(Chicago: G.P Engelhardt & Co., 1901 [1897]): 36.
30
benzaldehyde (artificial oil of bitter almonds), and methyl salicylate (wintergreen oil).103

Yet the variety of flavoring extracts available was diverse and dazzling. Emil Hiss, in the

manual quoted above, provides more than twenty densely printed pages of extract and

essence formulations, including five different formulas for banana essence, two for

blackberry, two for gooseberry, four for nectarine, and five for peach.104

How were a relatively small number of chemicals made to stand in for an

expanding array of distinct flavorings? An examination of late-nineteenth and early-

twentieth-century flavor formulas, and of the conditions and contexts of their circulation

and dissemination, illuminates the changes to the practices, markets, and social networks

involved in the production of synthetic flavorings.

Flavor and the Pharmaceutical Formulary

A nineteenth-century confectioner in search of one of the new synthetic flavors

would most likely find them at the local druggists’ establishment. An 1855 advertisement

from Samuel Simes, whose retail drug store and chemical manufacturing business was

housed in a large, four-story building that took up the Northwest corner lot on Chestnut

103
In the final years of the nineteenth century, makers of flavorings also began to use
some newer synthetic materials primarily used in perfumery: citral, often derived from
lemongrass, which was used in orange and lemon flavorings; ionone, a violet-scented
ketone synthesized from citral, sometimes used in raspberry; and linalyl formate, whose
odor resembling bergamot, can be found in some formulas for peach, apricot, and other
stone fruit.
104
Hiss 1901: 36-59.
31
and Twelfth Streets in Philadelphia,105 boasted that the fruit essences he manufactured

“expressly for confectioners” gave candies and other sweets “the rich and luscious flavors

of the different fruits more decidedly than the fruits themselves.”106 He offered

Pineapple, Strawberry, Raspberry, and Jargonelle Pear, as well as Vanilla, Orange,

Blackberry, “and all other kinds.”

For much of the nineteenth century, druggists were the primary distributors, if not

also the major manufacturers, of artificial flavoring essences, as well as other flavoring

products in the United States. Botanical extracts and essential oils had long had a place in

pharmacopeias, where they were included both for their purported therapeutic virtues as

well as for their ability to make difficult-to-swallow medicines more palatable. Pharmacy

trade journals and textbooks were early and important sources for formulas for synthetic

flavorings.107

Synthetic flavors fit nicely into the expanding portfolio of pharmaceutical

products and practices. In the mid-nineteenth century, pharmacy was in the process of

establishing itself as a modern professional discipline, one distinct from but in service to

medicine, whose practitioners received scientific — and particularly chemical —

education and training. Many druggists, such as Samuel Simes mentioned above, were

105
"Samuel Simes, Operative & Dispensing Chemist," engraving mounted on paper,
October 1856, Poulson Scrapbooks, vol. 8, Library Company of Philadelphia.
[(8)2526.F.10]
106
“Fruit Essences…” advertisement, North American & US Gazette [Philadelphia],
(February 20, 1855): n.p.
107
See, for instance, [M.] Fehling, “Artificial Fruit Essences,” American Journal of
Pharmacy, March 1853): 155; Edward Parrish, An Introduction to Practical Pharmacy,
2nd ed. (Philadelphia: Blanchard & Lea, 1859).
32
not only retailers, but also manufacturers and wholesalers, who produced their own

pharmaceutical preparations in a dedicated laboratory space, using chemical processes

such as distillation, extraction, and so on. The “manufacturing pharmacist” generally had

the chemical know-how to understand flavor formulas and processes, the specialized

glassware and other tools to produce his own ‘compound ethers’ and to assess the purity

and contents of commercially available essential oils and extracts, as well as the access to

raw materials necessary for the production of flavorings.108 As early as the 1850s,

wholesale druggists’ supply houses and chemical supply catalogs began listing

“compound ethers” among offerings, sometimes with a descriptive commercial term

(e.g., ‘apple oil’) alongside the standard chemical name (amyl valerianate).109

Druggists were also users of flavoring materials. In the battle for professional

standing waged between doctors and pharmacists, the pharmacists’ ability to offer a more

palatable preparation provided an advantage.110 The US Pharmacopeia and the National

108
Gregory J. Higby, “American Pharmacy’s First Great Transformation: Practice, 1852-
1902,” in Higby and Elaine C. Stroud, eds. American Pharmacy (1852-2002): A
Collection of Historical Essays, (Madison: American Institute for the History of
Pharmacy, 2005): 1-4; John Parascandola, “The Pharmaceutical Sciences in America,
1852-1902,” in ibid., 19-23.
109
See, for instance, “Descriptive Catalogue of Chemical Apparatus, Chemicals and Pure
Reagents, Manufactured, Imported and Sold by Edward N. Kent, Practical Chemist, No.
116 John Street, Near Pearl,” (New York: Van Norden & Amerman, 1854); and “List of
Chemical Preparations and Pure Reagents Imported and For Sale by JF Luhme & Co.,
556 Broadway,” (New York: MW Siebert, 1856). Both catalogs list various compound
ethers, under chemical names supplemented by commercial names indicating the fruit the
chemical suggests or imitates.
110
John S. Haller, Jr., “With a Spoonful of Sugar: The Art of Prescription Writing in the
Late 19th and Early 20th Century,” Pharmacy in History 26.4 (1984): 171-8. See also the
advertisement for John Wyeth & Brother, Philadelphia Manufacturing Chemists, in the
1889 Meyer Brothers Catalog, which states that their products are regularly prescribed
and preferred by a majority of doctors in the US and Great Britain, “on account of their
33
Formulary, the standard professional texts for drug formulation, both included formulas

and instructions for preparing flavoring extracts. By the 1860s, these texts also included

formulas for synthetic fruit flavors.

The professionalizing pharmacist was a follower of formulas, a producer of

standard products that elicited standard effects on the human body (and sensorium). As

flavor manufacturing became a specialized chemical industry, it would move to

distinguish itself from this formula-bound model.

The Rise and Fall of Kletzinsky’s Table of Artificial Fruit

Essences

Perhaps because of the professionalization of pharmacy, and the concomitant

standardization of its practices and procedures, druggists’ trade journals hosted the

earliest American appearances of what would be the most influential and widely

circulated set of flavor formulas: Kletzinsky’s table of artificial fruit essences. Vincenz

Kletzinsky (1826-1882) — sometimes spelled Kletzinski — was an Austrian chemist

known for his work in ‘animal chemistry.’ That is, he studied the chemical reactions

underlying the physiological processes of life: digestion, metabolism, health and disease,

the ways that drugs worked upon the body.

superior quality, strength, elegance, and agreeable flavor.” [218] Meyer Brothers Drug
Co., “Annual Catalogue and Prices Current,” St. Louis, MO, August 1889: 218.
Smithsonian Libraries Trade Literature Collection.
34
 Kletzinsky's Table of Formulas for "Artificial Fruit Essences" was first released

into the world in 1865, when it appeared in his report of the latest pure and applied

chemical research.111 It began its circulation when it appeared in the pages of Dingler's

Polytechnisches Journal, a widely-read German technical journal, the following year.112

The table made its print debut in the United States in April 1867, in the Druggists’

Circular and Chemical Gazette, and the following month, in the American Journal of

Pharmacy.113

For at least fifty years, Kletzinsky's table and its associated formulas percolated

through the written record: first in trade journals and professional reference books for

pharmacists, confectioners, ice cream makers, and those in the beverage or soda fountain

trade; later in miscellanies and formula books for amateurs. The formulas are included in

two of the earliest American monographs on the subject of manufacturing and using

flavoring extracts: Charles Herman Sulz’s 1888 Compendium of Flavorings,114 and

111
V. Kletzinsky, Mittheilungen aus dem Gebiete der reinen und angewandten
Chemie, (Vienna: Selbstverlag des Verfassers, 1865): 45
112
“Ueber die sogenannten Fruchtessenzen,” Dingler’s Polytechnische Journal 180
(1866): 77.
113
M. Kletzinski [sic], “On Fruit Essences,” Druggists’ Circular and Chemical Gazette
(April 1, 1867): 82; “On Fruit Essences,” American Journal of Pharmacy (May 1867):
238. Both of these early reprints contain an error, in that the column for “oil of persicot”
(ie, essential oil of bitter almond, or benzaldehyde) is empty. These seem to be
transcribing an error from the reprint of these formulas in the London Pharmacy Journal;
as the original table in Dingler’s Polytechnic contains quantities in this column.
114
This text was a selection and abridgment of a much larger volume published the same
year, A Treatise on Beverages, or The Complete Practical Bottler. (Sulz described
himself as a “technical and analytical chemist” with experience as a “practical bottler.”)
While A Treatise on Beverages was a comprehensive manual on nearly every aspect of
producing bottled carbonated beverages, Compendium on Flavorings was intended to be
work of broader utility, intended for all users of flavorings, with some recognition of the
different needs these products had to fulfill in different contexts. For instance, Sulz drew
35
Joseph Harrop’s 1891 Monograph on Flavoring Extracts with Essences, Syrups, and

Colorings.115

By following Kletzinsky’s table, the flavor-maker could summon the aromatic

specters of fifteen distinct fruity flavors: pineapple, melon, strawberry, raspberry,

gooseberry, grape, apple, orange, pear, lemon, cherry and black cherry, plum, apricot,

and peach.

a distinction between “extracts, essences, and tinctures made for the druggist,
confectioner, and carbonator.” While concentrated flavorings best served the purposes of
the druggist and confectioner, the beverage bottler had other requirements: flavors that
would “yield clear and bright syrups,” that wouldn’t separate or become turbid on the
shelf, and that were water-soluble.
115
Joseph Harrop, Monograph on Flavoring Extracts with Essences, Syrups, and
Colorings. Also Formulas for the Preparation with Appendix. Intended for the Use of
Druggists. (Columbus, OH: Harrop & Co, 1891.)
36
 Fig 1. Kletzinsky’s table of artificial fruit essences in one of
its first appearances in the United States. M. [sic] Kletzinski,
“On Fruit Essences,” American Journal of Pharmacy (May
1867): 238.

Kletzinsky outlined a basic set of chemical materials that would be used in the

production of synthetic flavors. These included a range of ethers and amyl ethers, a

couple of essential oils, a pair of aldehydes (including benzaldehyde and acetyl aldehyde,

which was listed as “aldehyd” after Liebig’s usage), a handful of organic acids, and other

constituents including chloroform, nitrous ether, glycerin, and, especially, alcohol. These

compounds could readily be purchased from druggists’ wholesalers and chemical supply

houses, as well as from many essential oil dealers. Following the model of some earlier

flavor formulas, Kletzinsky’s table specified ratios rather than fixed quantities: the

proportional quantities of one or two esters dissolved in 100 parts of alcohol. Expressing

the formula as a ratio of chemicals rather than as measured quantities suggests that users

could scale production up or down as needed.

By presenting each compound in a range of different flavor applications, the

sensory meaning of each of these ethereal chemicals was ultimately not fixed to one

particular fruit; it could vary depending on concentration, as well as chemical and local

contexts. Consider the case of amyl acetate, the essence of Jargonelle pear, often sold as

‘pear oil.’ In Kletzinsky’s table, it also plays a role in strawberry, raspberry, and orange

flavorings. In the United States, this chemical was also frequently sold as ‘banana oil,’

named for its apparent evocation of the odor of that fruit, and was used as a component of

varnishes in addition to its role in flavorings. (The candy-banana smell of isoamyl acetate
37
remains familiar to us.) Indeed, reprints of Kletzinsky’s formulas in American

publications often included an additional formula for banana essence (usually a

combination of amyl acetate and ethyl butyrate) indicating the popularity of this

flavoring.116

The text accopanying Kletzinsky’s table was spare — one scant paragraph. It

underscored the importance of using only chemically pure substances, including pure

alcohol. It also explained that the glycerine was included in nearly all of the formulas

because it “appears to blend the different odors, and to harmonize them.”117 Glycerine is

a simple sugar alcohol, a viscous liquid derived from fatty substances such as palm oil,

valued for its efficacy as a solvent. It had multiple applications in the nineteenth century,

including in pharmacy, surgery, and the preparation of scientific specimens.118 (It remains

important in flavor production to this day.) Kletzinsky’s articulation of the idea that

“blending” and “harmonization” were virtues to which artificial flavors should aspire

would remain important, as we shall see. The production of synthetic flavors exhibiting

“blendedness” and “harmony” — a condition in which the individual chemicals

contributing distinct sensory qualities to a substance were not detectable to the senses, but

were submerged into and contributing to a single, irreducible perceptual experience —

116
Nadia Berenstein, “The History of Banana Flavoring,” Lucky Peach (August 2016):
http://luckypeach.com/the-history-of-banana-flavoring/
117
Kletsinki 1867. The version of the table published in Dingler’s uses a phrase from
perfumery, describing glycerine as causing the “individual flavor and odor notes” to
blend into “a single sensory chord.”
118
Wm. Abbots Smith, On Glycerine, and Its Uses in Medicine, Surgery, & Pharmacy.
Being Principally an Abstract of M. Demarquay's Treatise, 'De La Glycerine,' &c.
(London: H.K. Lewis, 1863).
38
would also come to trouble efforts to create and enforce a definition of these flavors that

distinguished them from the strictly “natural.”

Kletzinsky’s table is equally notable for what is left unaddressed. First,

Kletzinsky makes no mention of how he compiled or created the table. Although it is

likely that he collected formulas from commercial flavor manufacturers rather than

developing them himself, it is unknown how generally these formulas were used among

flavor manufacturers, or, alternately, how local or particular they were to one town or

region.119 What is certain, however, is that the process of developing these formulas did

not begin with an analysis of the chemical components of fruits. It started with a

recognition of the sensory qualities of organic chemicals. Manufacturing chemists

worked empirically with available organic chemicals, combining and diluting them,

mixing and sniffing, until they obtained recognizable, and pleasurable, results.120

Kletzinsky’s table also did not explain the process of actually making these

mixtures: how to select chemicals in order to ensure that they were of proper purity or

quality, what order they should be combined in, or what type of instruments should be

used to do this. Nor did it explain anything about usage: what foods or beverages these

could be added to, the quantity of flavoring that should be used in different products, how

119
There is some evidence that formulas may have varied regionally and internationally.
For instance, an 1866 article in the London Chemist and Druggist (reprinted in the
American Druggist’s Circular and Chemical Gazette) notes that the artificial fruit
essences produced by German manufacturers in the Zollverein department “differ
considerably from those met with in British commerce.” The substance of this difference
is left unexplained. “The Composition of Some Artificial Fruit Essences,” Druggist’s
Circular and Chemical Gazette, (Jan 1866).
120
Roberts 1995.
39
the mixtures should be stored. All of these factors, as manufacturers and users of

synthetic flavors were beginning to recognize, had an effect on a flavoring’s quality and

utility. Ultimately, by presenting different fruit flavors as combinations of a limited set of

related chemical compounds, Kletzinsky’s table had a static and closed quality. Aside

from glycerine, it made no attempt to describe the role that each of the components

played in the ultimate composition, and thus had limited utility on its own as a tool for

creating novel flavors, altering existing ones, or incorporating new materials.

The contexts where Kletzinsky’s formulas appear give some indications of how

different groups of flavor-makers might have put these formulas to use. For instance, in

his Monograph on Flavoring Extracts, Harrop replicates Kletzinsky’s formulas (without

attribution) but also provides variants for a few flavors: pineapple, strawberry, and

raspberry. Harrop’s alternative formulas are simpler versions with fewer components. For

instance, Harrop’s second raspberry flavor includes only three of the thirteen chemicals

included in the first formula, which reproduces Kletzinsky’s original.121 Harrop did not

explicitly address the differences between alternative formulas for a single flavor, or the

contexts for which each was best suited. However, he implicitly provides a key for the

interpretation of the flavor formulas. In his explanation of his strawberry flavors, he

writes that butyric and acetic ethers “form the base, although the combination may be

added to almost without limit.”122 In other words, by building on a standard chemical

foundation that provides the “sensible core” of a flavor, the practical chemist can invent,

improvise, add nuance, capitalizing on the multiple sensory possibilities available in each

121
Harrop 1891: 78-9. Also changes the relative proportion of these ethers to each other.
122
Harrop 1891: 77.
40
chemical to achieve desired effects and inflections, while still maintaining a resemblance.

Harrop ends with the valediction, “license is given to figure for yourself, provided you

are able.”123

Nearly thirty years after Harrop’s monograph, Kletzinsky’s table is reproduced in

the 1919 edition of the Scientific American Cyclopedia of Formulas, a compendia of

miscellaneous recipes for manufacturing household goods, where it is included alongside

15,000 formulas for things such as glues, embalming fluids, and varnishes, and

descriptions of the symptoms of poisoning by sewer gas, among many other things.

Although Kletzinsky's table remained more or less unchanged from its first appearances

in chemistry and pharmacy journals, its meaning had changed; its standing in the world

had dropped. By the twentieth century, its formulas were no longer cited in professional

literature, except with caution or derision. Erich Walter, in his 1916 Manual for the

Essence Industry, wrote: "In the course of time, the public has come to look with disfavor

on the artificial fruit flavors formerly employed, and in the formulas which follow no

attention will be paid to such imitations." (He then went on to supply his own formulas

for imitation fruit flavors.) The 20th edition of the U.S. Dispensatory (1918) was the first

to demur from including Kletzsinky's formulas, referring readers looking for that

information to previous editions.

The persistence of Kletzinsky's table is one of the signs of the expanding

commercial need for synthetic flavor additives, which could perform functions in factory-

produced foods that “genuine” flavors could not. The diminishing status of Kletzinsky’s

123
Harrop 1891: 85.
41
formulas, however, indicates something else: a widening divide between flavor amateurs,

following standard formulas, and flavor professionals, the kind of workers who would

“go figure for [themselves].” This marks the opening of a rupture at the beginning of the

twentieth century between "practical chemists" who mix up flavors and fragrances,

among many other things, and specialized chemical workers (affiliated with newly

established firms specializing in flavor and fragrance materials) who claim a particular

kind of expertise with aromatic materials, an expertise that is both scientific and sensory.

Against Formulas: The Specialization of Flavor Creation

Flavor manufacturers were not merely supplying a market that required flavor

additives, they were creating it — in part by distinguishing their synthetic specialties

from the kind of products one obtained when following published formulas. Even as they

used Kletzinsky’s table as a base for their synthetic formulations, flavor manufacturers

and users improvised, customized, and improved upon the formulas to better adapt them

to desired applications and specifications, and to produce unique and distinctive effects.

Formulas, in fact, were something that flavor manufacturers began to openly

disparage, in the interests of protecting their own share of the market by discouraging the

users of their products from attempting to make their own. “The preparation of a

satisfactory extract is not by any means the simple matter than most soda water bottlers

think it to be, and a good deal of money has been lost by people starting in to

manufacture extracts on the strength of some formulas that have been purchased or given

to them,” lectured the 1921 catalog from Warner-Jenkinson, a major supplier of bottlers’
42
extracts and other beverage-making supplies. “A formula in extract-making means

nothing except trouble, unless the compounder of the extract has an intimate knowledge

of the chemistry of the bodies he is handling…. Hence, a formula should only be

considered by the chemist as the basis on which to build.”124 The true work of the flavor

compounder was not following existing formulas, but developing new mixtures.

The increasing specialization of flavor-making after the First World War was

illustrated in a pungent, purplish essay titled "The Formulist," which appeared in the

February 1921 Ungerer's Bulletin: A Symposium of Aromatics, a bimonthly compendium

of editorials, news, and gossip published by Ungerer & Co., a New York City firm

dealing in synthetic perfume and flavor materials. "The Formulist" is a moral fable of the

aromatic materials business, where the eponymous figure is ultimately contrasted with

the "real creative perfumer or flavor maker."125

"The Formulist," we are told, "is he who, on a day in the far dim past, has

inherited, achieved, or had thrust upon him a formula. On that... eventful day our

Formulist entered the valley of self-satisfied contentment and ceased forever to function

as a builder and producer." The Formulist's career is subsequently spent assiduously

protecting his cryptic recipe, like a mystic whom illumination has visited only once.

"There is nothing more to be done," intones the narrator, "but to guard jealously the

precious scrap of paper containing the clue to the sublime odor or flavor of his; to make

124
Warner-Jenkinson Company, Bottlers’ and Ice-Cream Makers’ Handy Guide, (St.
Louis, Mo.: Warner-Jenkinson, 1921): 77-8. A. W. Noling Hurty-Peck Collection of
Beverage Literature, Shields Library Special Collections, University of California Davis
(hereafter cited as Noling Collection, UC Davis).
125
F.N. Langlois, “The Formulist,” Ungerer’s Bulletin, February 1921.
43
his sacred mixes in guarded seclusion; and to carry on pompously in his self-assigned

role as creator of the magnum opus."

The author of the fable (F.N. Langlois, of the United Drug Company, Boston)

identified two major faults with the ways of the Formulist. First, in taking his formula as

perfect and complete, the Formulist shut out new research developments in chemistry,

including new materials, that could enhance his formula's sensory qualities, decrease its

production costs, or improve its utility. Second, the Formulist's hermeticism precluded a

proper market orientation. As a secretive recluse, the Formulist was incapable or

unwilling to work with others in the flavor and fragrance company, to admit that other

realms of knowledge were involved in shaping a commercially viable product.

Advertising men, salesmen, "the container and label artist" — all these professionals

contributed to the success of a new flavor or fragrance product. By refusing to share the

details of his formula with them, or integrate their reports about consumer needs or

desires into his working process, the Formulist doomed himself to obscurity and his

product to obsolescence.

In contrast: "Your real creative perfumer or flavor maker moves with the times.

He rotates with his market. The development of one great success acts as an incentive to

a series of accomplishments. If he cannot improve the odor or the flavor he casts about

for a more agreeable color for it. He smells or tastes his formula with the nose or palate

of the outsider. Approaching from that direction, he appreciates the inevitable fact that

the world eventually tires of perfection itself. He borrows a leaf from the experience of

44
the cigar maker, who knows that there is a certain important section of his public which

prefers a new good smoke to an old better one."

This is an early description of the role of the specialized flavor maker within the

flavor company, negotiating between the material requirements of manufacturers, the

sensory possibilities of chemicals, and the sensual desires of consumers. The implication

is that a successful flavor could not merely reproduce static, timeless nature. The

successful flavor also must reflect consumer tastes, expectations, and, especially,

fashions. In other words, the flavor maker was in a fashion business, one that must

constantly produce novel sensations, new variations for a public hungry for untasted

fruits, unsampled pleasures, both low delights and high ones. The real creative flavor

maker appreciated the inevitable fact that the world eventually tires of perfection itself.

There is no perfect. There is only the pluripotent new, perpetually refreshed by the stream

of newly discovered synthetic organic chemicals.

This is not to suggest that flavor makers worked freestyle, without formulas, only

using their senses for a guide. If anything, proprietary formulas gained increasing

importance among flavor manufacturers as they represented the accumulated skill of their

specialized workforce, and were treated as significant company assets. For instance, a

1927 obituary for Dr. Rudolph Pabst, chemist and owner of the Reading Extract

Company in Reading, Pennsylvania, notes that his formulas were willed to his son.126

Bernard Polak, who headed Polaks Frutal Works, kept his formulas secure with a

126
“Dr. Rudolph Pabst, Chemist,” obituary, American Perfumer and Essential Oil
Review, October 1927.
45
personal, hand-written ‘code book,’ which assigned alphanumeric values to different

compounds.127 At a time when flavor and fragrance companies had access to otherwise

unknown materials and processes, such secrecy could protect a company’s advantage, as

well as their investments in research and development. But these formulas were not seen

as definitive, absolute, or sufficient for success. In the hands of creative flavor makers,

they were tools, not final products — subject to adaptation, alteration, and innovation.

III. “Twentieth Century Raw Materials”: Synfleur

Scientific Laboratories and the Formation of a
Scientific Flavor and Fragrance Industry

On an April afternoon in 1908, dapper, consumptive Alois von Isakovics lectured

before an audience of Columbia University students about the chemistry of synthetic

perfumes and flavors. Isakovics was the founder and chief chemist of Synfleur Scientific

Laboratories, one of the first U.S. companies

to specialize in the manufacture of synthetic

aromatic materials. In a lecture suffused

with odorous demonstrations, Isakovics

outlined the distinctive chemistry of flavor

and fragrance molecules, while making the

127
Bernard Polak, Code Book, [1938] Bernard Polak and Polak’s Frutal Works, Inc.
Collection, 2008.044, Series 1, Chemical Heritage Foundation, Othmer Library Special
Collections.
46
case for the synthetic production of these substances. According to Isakovics, with

research and careful attention to chemical purity, it would be possible to produce

synthetic versions of flavors and fragrances that not only rivaled but surpassed their

natural counterparts in terms of sensory qualities, performance, and use value.

The first decades of Isakovics’ company provide an exemplary story of

specialization within this branch of the chemical industry. In the first years of the

twentieth century, Isakovics transformed his company, Herbene Pharmacal, a small,

urban firm producing a variety of retail goods, including proprietary medicines and

perfume specialties, to Synfleur Scientific Laboratories, a company that produced

perfume and flavoring materials for manufacturers — to whom they offered not only

reliable and high-quality chemical materials, but also customized, exclusive flavors and

fragrances, as well as expert advice on manufacturing processes, formula development,

and business practices. Synfleur’s shift to a primarily intrabusiness orientation —

supplying other manufacturers with specialized components, rather than selling

household extracts directly to consumers — became the model that would define the

contours of the flavor and fragrance industry in the new century.128 It also signaled a

sharp turn away from the flavor industry’s association with pharmacy or proprietary

128
Regina Lee Blaczszyk has written extensively about intrabusiness relationships,
especially those focused on design and development processes in the production of
consumer goods. Her scholarship has highlighted the key role of these sorts of “fashion
intermediaries” in the development of the modern consumer economy, and offers a way
to bridge the divide between consumer-oriented and producer-oriented histories. Regina
Lee Blaszczyk, Imagining Consumers: Design and Innovation from Wedgwood to
Corning, (Baltimore: JHU Press, 2000); Blaszczyk, The Color Revolution, (Cambridge:
MIT Press, 2012).
47
medicines, and a turn towards innovation and specialization driven by chemical research

and scientific expertise.

From Herbene Pharmacal to Synfleur Scientific

Laboratories

Alois von Isakovics was born in Prague in 1870, the son of a Judge Advocate

General in the Austro-Hungarian army. Although he resisted his father’s entreaties to

pursue a military career, he showed an entrepreneurial bent from a young age. The stamp-

collecting business he started as a boy grew large enough that he needed to employ

several schoolmates to help with correspondence and filling orders. His education, in

Vienna, “comprised the regular curriculum of a young man of good European family,”

according to one obituary written by a friend, though other accounts claim he studied

chemistry at the university level.129 In any case, his formal education seems to have

ended at the age of sixteen, when he left Europe for the United States.

Two years later in New York, he met Mary Upshur, a seventeen-year-old student

whose background stood in sharp contrast to those of the recent immigrant; her family

129
William Dreyfus, “Alois von Isakovics,” Industrial and Engineering Chemistry 9
(July 1917): 716. His obituaries in the American Pharmaceutical Association journal and
in the American Perfumer and Essential Oil Review claim that he came to the U.S. after
completing the course in chemistry at the University of Vienna. Image of Isakovics used
here from his obituary in Metallurgical and Chemical Engineering (July 1917): 44.
48
traced its roots to seventeenth-century Virginia.130 They were engaged within a year, in

July 1889. It would be a long engagement. When they did get married in July 1895, six

years later — and three years after Isakovics became a naturalized citizen — it was

apparently with the blessing of her family. The Reverend Doctor Houghton, who

officiated the ceremony at the Church of the Transfiguration in New York City, had also

presided over her parents’ nuptials.131

Synfleur advertising material and stationery boasted that the company was

founded in 1889, but most company materials skim over its first decade. 1889 was the

year of Alois’ engagement to Mary, and perhaps this is a recognition of the crucial role

she played in helping him build the business. I have found no record of Synfleur’s

existence or activities before the early 1890s, when, doing business as Herbene or

Herbene Pharmacal, the company sold proprietary medicines, perfumes, bandages, and

other druggists’ sundries through the mail.132

An 1892 notice placed by a Toronto drug wholesaler in the advertising pages of a

Canadian humor magazine touts “golden Herbene Gems” as a “sure cure” for

“nervousness, general debility, and all female complaints,” but the majority of

advertisements for Herbene that I have located sell not products, but opportunities.133

Crammed between notices for morphine cures, weight-loss pills, and clandestine

130
According to Nina [von Isakovics] Allen, great-great-granddaughter of Alois and
Marie, she may have been a student at Parson’s at the time, studying fine arts. Personal
communication.
131
Dreyfus 1917.
132
Synfleur Scientific Laboratories stationery dating from 1905 does note that it is the
successor to Herbene Scientific Laboratories.
133
[Tarbox Bros.] “Golden Herbene Gems,” advertisement, Grip 38 (January 2, 1892): 2.
49
abortifacients, Herbene’s solicitations sought agents through two- or three-line

advertisements in the back pages of small journals — including those targeting African-

American and female readerships — offering to supply perfumes, household goods, and

other sales items “on credit” with “expenses paid.” “150 per cent profit,” “big profits”

were promised. The curious were invited to write to the “old and reliable Herbene Co” at

a P.O. Box at Station L, in New York.134

Station L was the uptown post office branch in Harlem, near the building on East

121st Street where the company’s manufacturing laboratory was located. Otherwise, there

is little definite information about the company’s operations in the 1890s and early 1900s,

although a 1904 judgment ordering Herbene to pay an outstanding debt of $627 to

Antoine Chiris Co., the American branch of a venerable Grasse essential oil and perfume

company, indicates that the company was using this firm’s products as components of

their specialties.135 A 1902 notice of incorporation — at least ten years after Herbene

began running advertisements, and thirteen years after it was allegedly founded —

134
I’ve only been able to locate a handful of these, the earliest advertisement dating from
1894, in a newsprint journal called The Golden Rule. Other advertisements appeared in
the back pages of Ladies’ World (1896); The [Baltimore] Afro-American (1896); and the
[Washington, D.C.] National Tribune (1898). The language in all examples was similar.
As examples of the kind of advertisements Herbene shared space with, the following
appeared in the 1898 National Tribune: “LADIES When Doctors and others fail to
relieve you, try S.R.&Sw. IT never fails. One full treatment free.” Other advertisements
in the 1896 Ladies’ World included: “Opium or Morphine Habit Cured at Home. Trial
Free. NO Pain. Address Compound Oxygen Ass’n. Ft. Wayne Indiana.” and “Fat Folks
reduced in weight — safely, surely, speedily. Trial Bottle Free. Chase Remedy Co.
Chicago.”
135
“Notices of Judgment,” New York Tribune 28 September 1904. [Judgment was filed
27 September, 1904]
50
claimed $60,000 in capital, and named three directors: Alois, Mary, and Effingham L.

Holywell, of Brooklyn.136

Sometime around 1903, due to Alois’ worsening pulmonary tuberculosis, he and

Mary moved their family and the company to Monticello, a Catskills town in Sullivan

County, New York, about ninety miles northwest of the city. This move also marked a

change in the company’s name and business model. Now dubbed Synfleur Scientific

Laboratories, the company tacked away from producing low-status retail goods sold over

the mail by commissioned agents. Instead, the company addressed itself directly to

manufacturers of soaps, toilet goods, perfumes, confectionery, and other consumer

products, offering fragrance and flavor materials of the highest quality, informed by the

very latest scientific research. As a 1905 spring catalogue put it, Synfleur was in the

business of providing “Twentieth Century Raw Materials for Progressive

Manufacturers.”137

136
“New Corporations,” Paint, Oil, and Chemical Review 34.8 (August 1902): 28. I
haven’t been able to learn anything definite about Holywell — though he (or a son of the
same name) may have been a real estate lawyer. He doesn’t come up in any other
reference I’ve found to Isakovics or Synfleur. As this is around the time that Herbene
relocated upstate — and as Herbene Pharmacal continues to be listed in druggists’
suppliers directories into the 1910s, he may have taken over the company from Alois and
Marie. [A 1903 report from the Annual Meeting of the Manufacturing Perfumers’
Association lists Holywell as Secretary and Counsel of Herbene, Marie as Treasurer,
Alois as President and General Manager.] There also appears to have been a lawsuit filed
by Holywell against Herbene in 1905. “The Courts,” Brooklyn Daily Eagle (Oct. 4,
1905): 3.
137
[Synfleur Scientific Laboratories], “Synfleur Materials Wholesale List,” Spring 1905,
Included as an enclosure in a letter from Alois von Isakovics to Harvey Wiley, May 3,
1905, Food Standards Committee, Correspondence and Reports, 1897-1938, Records of
the Food and Drug Administration, Record Group 88, National Archives at College Park,
MD.
51
 The 1905 catalogue emphasized their commercial orientation: “we have no retail

list. We sell to manufacturers only.” But Synfleur was also selling a particular kind of

intrabusiness relationship, one that would give clients access to the expert knowledge of

its founder. The 1905 list offered “complete research laboratories… at the service of our

friends in the manufacturing industry. Expert advice on manufacturing problems,

improvement of processes, and working formulas…. Our friends can freely submit their

ideas or working formulas for suggestions… and we will cheerfully supply any legitimate

manufacturer who is a consumer of our materials, with detailed working methods,

insuring the best possible results at the lowest possible cost.”

Essentially, the company was selling not only a standard set of goods, but also

expert knowledge and technical advice. Clients were promised that their inquiries would

“receive the careful, personal attention of Mr. Alois von Isakovics.” Underwriting this

business proposition was a wager on the central role that research and development

would play in the flavor and fragrance industry. Materials and processes in the field were

changing rapidly, and manufacturers needed a partner who would stay abreast of “the

latest work and the highest improvements in science.” Synfleur had made an investment

in specialized research facilities, and was offering to share its benefits while implying

that manufacturers could not replicate these results on their own. This was important, as

this new business model also required substantial trust on the part of the manufacturer,

who was asked to share confidential and proprietary information about formulas and

52
processes with Synfleur. Synfleur’s 1905 wholesale list made repeated assurances that all

inquiries would be kept “strictly confidential.”138

In 1910, Synfleur’s advisory and consulting operations were further formalized.

Announcements and advertisements in trade journals such as American Druggist and

Pharmaceutical Record and American Perfumer and Essential Oil Review trumpeted

that the company had established a special department in its laboratories dubbed the

“Synfleur Manufacturing Service,” which would “furnish practical up-to-date suggestions

for the users of Synfleur products as to the best means of perfecting their perfumes, toilet

waters, sachets, flavoring essences, and toilet specialties of all kinds,” by “placing at their

disposal a wide experience based on careful research work by a staff of chemists who

have for many years been scientifically trained in this field.”139 Synfleur would work with

manufacturers to develop exclusive custom formulations, “proprietary combinations” that

would “give them a proprietary odor or flavor of their own.”140

"Thousands of private formulas of well-known manufacturers were entrusted to

his honor in the hope, seldom disappointed, that his competent staff of synthetic perfume

and flavoring material experts might find a way to improve upon the established formula

by giving a successful touch of Synfleur excellence to the finished product,” wrote

Isakovics’ son-in-law and successor at Synfleur, Luis de Hoyos, in an obituary for his

138
“Synfleur Materials Wholesale List” 1905.
139
“Synfleur Perfume Materials,” advertisement, American Druggist and Pharmaceutical
Record (August 22, 1910): 124.
140
Synfleur, advertisement, Era Druggists Directory, 1912.
53
chemist mentor and employer. “Whatever the origin of the formula was, whether of his

own compiling or the private property of his client, the secrets of the perfume world were

safe” with him. He went on to assure readers that these values were "so firmly impressed

upon the efficient clannish Synfleur force who have together labored for many years by

our beloved leader, that it grew to be an all-pervading feature of our business policies."

He vowed that no member of what he calls their "business family" would ever "prove a

traitor to these most sacred principles of a very unique and singularly eccentric

business.”141 Even though Isakovics was gone, the company would continue to honor its

relationships and maintain the confidences of its clients.

Synfleur and the Flavor Chemical Industry

Synfleur was one of a number of U.S. companies that began to specialize in flavor

and fragrance materials at the beginning of the twentieth century.142 Some, like Synfleur,

had roots in pharmacy. Dodge & Olcott, one of the most important American flavor and

fragrance companies during the first half of the twentieth century, began in 1798 as an

141
Synfleur, advertising insert, American Perfumer and Essential Oil Review, July 1917.
142
For profiles of the companies involved in the flavor and fragrance trade in the late
nineteenth and early twentieth centuries, see Wayne E. Dorland and James A Rogers, The
Flavor and Fragrance Industry, (Mendham, Nj: Wayne E. Dorland, 1977): 171-246. For
a history of the production of aromatic chemicals in the United States around the time of
the first World War, see William Haynes, American Chemical Industry: A History, Vol
III: The World War I Period, (New York: Van Nostrand, 1945): 327-339.
54
apothecary shop on Pearl Street, in lower Manhattan. Until the 1880s, it was best known

as an importer of pharmaceutical products, chemicals, and essential oils, as well as

surgical instruments, perfumes, cosmetics, paints, and sundries.143 It was not until the

1890s, when Francis Dodge (1868-1942) took the helm of the company, that the firm

shifted its focus to manufacturing flavor and fragrance chemicals.144 Like most

nineteenth-century Americans who wished to pursue chemistry as a scientific vocation,

Dodge had traveled to Germany to earn his doctorate, studying under Victor Meyer in

Heidelberg, where he had distinguished himself by being the first chemist to obtain

citronellol from rose oil. On returning to the U.S. in 1891, he joined the family business,

and redirected its focus to specialty chemical manufacturing, including the production of

synthetic aromatic chemicals, natural isolates, and essential oils.145

Other American flavor and fragrance companies were founded as branches of

European essential oil and aromatic chemical firms. For instance, Fritzsche Brothers was

founded in 1871, by three German emigrants, in association with Schimmel & Company,

of Leipzig, one of the major European producers of essential oils, natural isolates, and

143
Founded by Robert Bach, the company went through several name changes in the
early 19th century, and was not known as Dodge & Olcott until 1861. Around 1811,
Robert Bach and his sons founded a distillery in Brooklyn, where they produced whiskey
and other alcoholic products. This was separate from the drug business, but by 1848,
Bach’s “celebrated alcohol and pure spirits for perfumers” was listed among the
merchandise. The Story of An Unique Institution: Dodge & Olcott, Inc. 1798-1948. (New
York: Dodge & Olcott, 1948); Gabriel Sink, “A Tribute to the Oldest American Flavor
and Fragrance House,” Perfumer & Flavorist 17 (January/February 1992): 37-9.The
Story of an Unique Institution… 1948: 12-3.
144
Starting in the 1860s, Dodge & Olcott did manufacture methyl salicylate, oil of
wintergreen, in a plant in Bayonne, NJ, but this was the company’s only synthetic
product prior to Francis Dodge’s leadership. Haynes, History of the Chemical Industry,
331.
145
Haynes, History of the Chemical Industry, 331.
55
synthetic aromatic materials.146 Until the First World War, American companies were

often reliant on European suppliers for chemical intermediaries, technical advice, and

sometimes financial backing.147

Like Synfleur, Dodge & Olcott, Fritzsche, Antoine Chiris, Van Dyk, and other

pre-war flavor and fragrance companies began to build and maintain dedicated

laboratories near their manufacturing plants. Although these were mainly control

laboratories, they employed specialized chemists, and performed some basic research.148

For instance, Dr. Clemens Kleber, who headed a laboratory in Clifton, New Jersey for

Fritzsche Brothers, published what was perhaps the earliest flavor chemical analysis of a

fruit, determining that amyl acetate was present in bananas.149

The increasing organization and professionalization of the flavor industry is also

indicated by the incorporation of the Flavor and Extract Manufacturers Association

(FEMA) in 1909, an industry trade group that represented the business and political

interests of flavor makers. Formed partly as a response to the 1906 Pure Food and Drug

Act, which brought federal regulatory scrutiny and heightened consumer distrust of

flavoring additives, FEMA’s initial agenda aimed to restore public confidence in their

members’ products.150 This involved working with regulators to show that adulteration

146
Sink 1992: 38.
147
Haynes 329.
148
Haynes 331.
149
Clemens Kleber, “The Occurence of Amyl Acetate in Bananas,” American Perufmer
and Essential Oil Review 7.10 (December 1912): 235.
150
“Flavor & Extract Manufacturers’ Association: The First 75 Years,” Perfumer &
Flavorist 9.3 (June/July 1984): 57-8. For a comparable case study in the organization of
the canning industry, see Anna Zeide, “in Cans We Trust: Food, Consumers, and
56
was rare, improving the quality of flavoring products, and actively combating media

accounts that grouped their members with “Adulterators, Food Poisoners, and Drug

Dopesters.”151 Promoting and sharing scientific research among its members —

emphasizing that the industry was on a sound scientific footing — was both a strategy

and a goal of the group.

By the 1920s, the synthetic flavor and fragrance industry was on firm ground,

with domestic manufacturers supplying most of the synthetic organic chemicals

consumed in the United States.152 The U.S. Tariff Commission noted that business was

booming for manufacturers of synthetic aromatic chemicals, who supplied the raw

materials for the flavor and perfume industries. “Progress has been made in overcoming

the former prejudice against synthetic aromatic chemicals, and the most important factor

in this result has been the successful and systematic development of quality products.

American manufacturers of these products have not neglected that essential unit of their

business, namely, the research laboratory, and the industry has consequently been placed

upon a stable and scientific basis.”153 In the1920s, new raw materials, new processes, and

an expanding and diversifying array of new food and beverage products on the market

continued to drive the growth of specialized flavor and fragrance companies.

Scientific Expertise in Twentieth-Century America,” PhD Diss, UWisconsin-Madison,

2014: 67-82.
151
“Report of the Proceedings of the Fourth Annual Convention of the Flavor & Extract
Manufacturers Association,” American Perfumer & Essential Oil Review 8.4 (June
1913): 86-93, quote at 89-90.
152
U.S. Tariff Commission, Census of Dyes and Other Synthetic Organic Chemicals
1923, Tariff Information Series 34, (Washington: Government Printing Office, 1924):
113.
153
U.S. Tariff Commission 1924: 113.
57
Flavor Chemicals as Progressive, Scientific Materials

In order for manufacturers to be persuaded of the value of the services offered by

Synfleur, they first had to be convinced that aromatic materials were scientific materials,

which required expert knowledge and specialized skills. Starting in 1910, Synfleur was a

monthly advertiser in the American Perfumer and Essential Oil Review, the leading trade

journal for the flavor, fragrance, and cosmetics industries. Synfleur’s advertisements

most frequently took the form of four-page inserts, conspicuously printed on pink

cardstock paper, and featured Isakovics’ writings, which advanced a research-and-

development focused business ideology. “Science is necessarily progressive,” lectured a

typical advertisement, from 1912, “and only manufacturers that apply science actively in

their business, that take advantage of the latest research work, can hold their own with the

competition.”154 Yet, as he wrote in an advertisement the following year, “so many

manufacturers do not correctly understand the materials they are using. A man that is not

acquainted with the nature of the products he handles every day, cannot appreciate

quality, cannot take advantage of new ideas, cannot apply materials intelligently.”155

Isakovics reprinted his entire chapter on “Essential Oils, Synthetic Perfumes, and

Flavoring Materials,” from a textbook on Industrial Chemistry, as a special sixty-seven (!)

page supplement for American Perfumer and Essential Oil Review readers in 1914.156

154
Synfleur advertisement, Era Druggists Directory 1912.
155
Synfleur advertisement, American Perfumer and Essential Oil Review, 1913.
156
Synfleur advertisement, American Perfumer and Essential Oil Review, 1914; Alois
von Isakovics, “Essential Oils, Synthetic Perfumes, and Flavoring Materials,” in Allen
58
Manufacturers had to be educated and informed about the scientific basis of the field

before they could properly appreciate the quality of Synfleur materials.

Isakovics’ insistence that flavor and fragrance materials were scientific and

progressive — dynamic materials of the future — was essential to establishing their value

and virtue in a marketplace that remained suspicious of synthetics. Accomplishing this

also meant establishing “Alois von Isakovics,” himself, as an expert in the emerging and

still weakly defined field of chemical research that dealt with the properties of aromatic

materials.157 Isakovics’ professional reputation, and his company’s prospects, was

intimately connected with the status of the substances that he manufactured and sold.

Isakovics deliberately cultivated an image as a man of science, an authority and

an expert on matters related to the chemistry of aromatic materials, even as his

commercial interests called his disinterestedness into question. He labored to make both

himself and synthetic chemicals respectable. He accumulated memberships in scientific

societies and business trade groups, building a network of close relationships with others

involved in chemical research and the chemical industry. A 1905 letter to Bureau of

Chemistry chief Harvey Wiley refers to a conversation they had at a recent meeting of the

American Electrochemical Society; his company letterhead lists his membership not in

only that group, but also in the American Chemical Society, the Manufacturing

Perfumers Association, the Society of Chemical Industry, and the Verein Deutscher

Rodgers, ed., Industrial Chemistry: A Manual for the Student and Manufacturer, (New
York: D. Van Nostrand, 1914): 766-802.
157
The absence of a stable identity for flavor and fragrance chemistry can be gleaned
from the following circumlocution in one of his obituaries, which called him “a genius in
the particular field of chemistry to which he devoted his talents.” Dreyfus 1917.
59
Chemiker, a list whose ultimate extensiveness is underscored by its terminal “etc.” At his

death, he was also a fellow at the American Association for the Advancement of Science

and the New York Academy of Sciences, as well as a member of the Chemists’ Club, the

American Pharmaceutical Association, the Franklin Institute, and the Royal Society of

Arts, London.158

Isakovics also built his reputation as an expert by associating himself with

academic institutions, giving lectures to university students, and publishing instructional

material on the chemistry of flavors and fragrances.159 That is to say, Isakovics strove to

link his branch of chemistry with professional scientific training and academic research,

and actively sought to encourage students to enter the field by portraying it as an

advancing research frontier.

Between 1908 and 1914, Isakovics lectured several times at Columbia University

to students of organic chemistry and pharmacy.160 Isakovics’ presentations were

apparently popular with students. “The lecture was listened to with the closest attention

by the students who nearly filled the large lecture hall,” according to one contemporary

account, “the subject being evidently one of more than ordinary interest for them.”161 A

158
“Synfleur Herald,” 1915. Dreyfus, 1917.
159
He was not the only flavor and fragrance chemist to take this step. Samuel Isermann,
the president of the New York synthetic fragrance and flavor chemical manufacturer Van
Dyk & Co. and the Chemical Company of America, contributed a chapter on Perfumes
and Flavors to H.E. Howe’s textbook Chemistry in Industry (1924).
160
The 1908 lecture is reprinted in a pamphlet published by Synfleur. Alois von
Isakovics, Synthetic Perfumes and Flavors: A Lecture Delivered at Columbia University,
(Monticello, NY: Synfleur, 1908).
161
“Synthetic Perfumes: Lecture by Dr. Alois von Isakovics,” American Druggist and
Pharmaceutical Record (April 13, 1908): 176.
60
notice in the college newspaper promoting Isakovics’ 1912 talk announced that it was

open to all interested students; in any case, it seems that he spoke to an audience that

grew each year. While he had previously spoken in ordinary lecture classrooms, in 1914,

his talk was in the Chandler Lecture Theater, the largest room not only in Havemeyer

Hall, which housed the chemistry department, but in the entire university.162 He kept his

presentations interesting with chemical demonstrations that illustrated to the eye and to

the nose the process of transforming one substance to another, more pleasant and

valuable one — for instance, moth-ball scented napthalene to the ethyl ether of beta-

naphthol, a material used in a perfume known as Bromelia, reminiscent of orange

blossoms. The Chandler Museum at the university later acknowledged several gifts that

he had made of the “fine synthetic perfumes” that he used in his lecture.163

Isakovics presented the chemistry of scented compounds in terms of key

elements, functional groups, and chemical structures. He began by introducing the

chemistry of synthetic perfumes and flavors as a branch organic chemistry — one that

largely concerned molecules comprising carbon, hydrogen, and oxygen atoms, as well as

some nitrogen-containing molecules. Sulfur also entered into the composition of many

odorous substances, but never, he said, in desirable ways.164 The addition of certain

chemical functional groups — including aldehyde, hydroxyl, ketone, alcohol, methyl, and

162
A summary of von Isakovics’ talk can be found in American Perfumer and Essential
Oil Review 9, no 3 (May 1914): 84. For the size of Chandler Hall (and the popularity of
chemistry classes and lectures, see “Alumni and University News,” [Columbia
University] School of Mines Quarterly 24, (Nov 1902-July 1903): 103.
163
“Department of Chemistry,” Columbia University Quarterly 14 (June 1912): 322.
164
The importance of sulfur-containing compounds to flavor chemistry would be
established in the 1950s.
61
ethyl radicals — sometimes, but not always, converted an odorless, flavorless molecule

into one with a strong scent or flavor. This was particularly true when these functional

groups were located at certain points on the molecule. For instance, ortho- and para-

derivatives of benzene were often valuable, while meta- derivatives were frequently

odorless.165 Nonetheless, there were exceptions to all of these rules of thumb, and science

was not yet at a point where the odor of a material, or its value, could be deduced by

knowing its chemical composition and structure.166

This was important, because it provided a prospective research program for

synthetic flavor and fragrance chemistry. While accounts of synthetic flavoring materials

still largely focused on describing the properties of the compound ethers, Isakovics

presented the synthetic production of new scented materials as a continuation of work on

organic synthesis begun by Wohler and carried forward by Berthelot, Liebig, Kolbe,

Perkin, and other chemical luminaries.167 The most important discoveries in the analysis

and synthesis of scented compounds still lay ahead, he told the assembled students; each

year brought new advances in this branch of organic chemistry, and there was plenty of

room for growth. “This is a most fascinating field for the research chemist,” he assured

his young listeners, offering “endless material to the investigator.”168

165
Ortho-, meta-, and para- refer to the position of functional groups around the six-
carbon benzene ring for molecules with two side chains. In ortho- molecules, the two
radicals are connected to adjacent carbons on the ring; the radicals are separated by one
carbon in meta- molecules; and in para- molecules, the radicals are across from each
other on the ring.
166
Isakovics 1908: 6-7.
167
Isakovics 1908: 6.
168
Isakovics 1908: 6.
62
 Isakovics was aware that the reputation of synthetic materials among both

manufacturers and users needed rehabilitation. “Years ago, like everything new,

synthetics had a hard road to travel, because they met a certain amount of prejudice

among manufacturers.”169 The poor reputation was partially earned; the quality of these

materials on the market varied widely. Aromatic materials act on us in extremely small

quantities, delivering scent in minute concentrations. Manufacturers of aromatic

chemicals thus had to be stringently careful in their production, as trace impurities that

escaped chemical detection could be undeniably present to the nose.

The value, and risk, of synthetic aromatics was intimately connected to their

power in small quantities. Minute concentrations of specific substances could not only

condemn a material as unusable, but also distinguish and glorify it. This observation was

central to Isakovics’ model for synthetically producing a high-quality scent or flavor. The

skillful utilization of trace constituents, especially newly developed ones, in multi-

component blends was integral to Isakovics’ conception of a well-made flavor.

From Chemical to Flavor

How did a chemical become a flavoring material? How much of a role did

chemical research play in the development of new synthetic flavoring materials? As

noted, in the nineteenth century, the components of “artificial fruit essences” were esters,

169
Isakovics 1908: 10.
63
most often synthesized from fusel oil and related chemicals.170 Other types of materials

were also used in flavorings, including essential oils, botanical extracts, and tinctures,

prepared from spices, roots, leaves, fruits, and other botanical materials. Although these

were not, properly speaking, synthetic chemicals, their production depended on chemical

techniques and technologies, including distillation, expression, and extraction using

alcohol or other solvents. Further, buyers and users of essential oil utilized chemical

methods to detect adulterations, verify claims about identity, and assess value.

Determining boiling point, measuring specific gravity, or adding reagents that reacted in

certain ways with known adulterants supplemented organoleptic (ie, sensory) evaluations

of essential oils. Books such as Ernest J. Parry’s Chemistry of Essential Oils and

Artificial Perfumes became essential texts, providing tables of physical and chemical

constants for various commercially important substances, as well as instructing readers in

techniques of analysis, including newer methods such as refractometry.171

Some new flavoring materials were introduced to the market due to analytic

research conducted within the essential oil industry. Flavorings have always been closely

linked to fragrances and perfumes, connected by raw materials, craft processes, and

technologies, as well as shared cultural meanings. As the essential oil and perfume trades

industrialized in the nineteenth century, they began producing aromatic materials on a

170
“Compound ether” is a synonym for esters, organic compounds comprising an oxygen
atom bonded to an alcohol radical and an acid radical. Compound ethers with a fatty acid
radical were known to have a fruity smell. List some, say where they were sold.
171
Ernest J. Parry, Chemistry of Essential Oils and Artificial Perfumes, (Scott,
Greenwood & Co., London and D. Van Nostrand, New York, 1899). Parry’s Chemistry
of Essential Oils was reprinted at least four times in the subsequent two decades, in
expanded and revised editions that reflected ongoing research in the field.
64
large scale, at costs suitable for use in mass-market goods.172 They also began producing

new materials — perfume isolates (compounds isolated from “natural” essential oils) and

synthetics that claimed to reproduce valuable constituents identified in essential oils, such

as citral, piperonal (“artificial heliotrope”), and geraniol (“artificial rose”), from cheaper

raw materials. This followed the pattern that Isakovics had described in his Columbia

University lecture: analysis followed by synthesis.

These novel perfumery synthetics contributed to a transformation of the sensory

qualities of perfumes, as well as to the transformation of cultural and social meanings

which diffused with these molecules from the modern, perfumed body.173 Meanwhile,

many of these new synthetic fragrance molecules also found uses in flavorings, though

their meanings and associations varied in these different contexts, and their common

presence in perfumes and flavorings was likely unsuspected by consumers. 174

172
As Eugenie Briot notes, the shift from perfumery as an artisanal trade to one that
utilized industrial manufacturing processes did not result in decreasing prices for
perfumes, but saw a rise in prices even as their use became more widespread. She argues
that this is a result of deliberate marketing strategies by nineteenth century perfumers,
who aspired to associate their goods with luxury even as they became more widely
accessible. See Eugenie Briot, “From Industry to Luxury: French Perfume in the
Nineteenth Century,” Business History Review (Summer 2011): 273-294. See pp 279-283
of that article for a survey of some of the technical innovations (including use of steam
power, the vertical integration of flower farms with factories, and the adaptation of
machines from other industries (such as pharmaceuticals, soap making, and distillation)
for use in perfume material factories. See also: Geoffrey Jones, Beauty Imagined: A
History of the Global Beauty Industry, (Oxford: Oxford UP, 2010), especially chapter
one.
173
See Jones, 20-9; also, Maksym Klymentiev, “Creating Spices for the Mind: The
Origins of Modern Western Perfumery,” The Senses & Society 9.2 (2014): 212-31, and
Luca Turin, The Secret of Scent, (New York: Ecco, 2006).
174
For instance, Houbigant’s Fougere Royal, an important masculine scent introduced in
the early 1880s, established synthetic coumarin as one of the three key basenotes of the
65
 Vanillin, one of the most commercially important flavoring synthetics, is an

exception to the general pattern by which new materials became available to the flavoring

industry. Although vanillin had been identified in vanilla beans in the 1850s, its correct

empirical formula, molecular structure, and synthesis emerged not from further analysis

of vanilla beans, but from basic research into the chemical structure of glucosides. In

1874, Ferdinand Tiemann and Wilhelm Haarmann were studying the composition of the

glucoside coniferin in the laboratories of August Hofmann at the University of Berlin,

when they obtained a substance that they later confirmed to be vanillin. Tiemann and

Haarmann partnered with fellow chemist Karl Reimer to manufacture synthetic vanillin

from coniferin. The Haarmann & Reimer factory in Holzminden is often celebrated as the

birthplace of the synthetic flavor and fragrance industry.175

In the twentieth century, new materials emerged alongside the development of

new synthetic processes. In 1904, Georges Darzens, a French chemist who headed the

research laboratory of L.T. Piver, a Parisian perfumery company, described a method for

fougere, a new, modern family of fragrances. Coumarin, a compound originally identified

in tonka beans, was first synthesized by Perkin from coal tar in 1868, was an important
component of vanilla flavors as well as flavorings added to tobacco products. Methyl
anthranilate — a compound most of us now will associate with the musky purple of
artificial grape — was first isolated and identified as a key component of neroli (orange
blossom) essential oil, and subsequently in other flower oils as well, by Schimmel, in
Lepizig. Linalyl formate, which was sold as artificial oil of bergamot or petit grain, was
used in formulas for synthetic peach, apricot, apple, and quince flavorings soon after its
commercial introduction. Jones, 23. Patricia de Nicolai, “A Smelling Trip Into the Past:
The Influence of Synthetic Materials on the History of Perfumery,” Chemistry and
Biodiversity 5 (2008): 1137-1146.
175
The process by which vanillin became a viable synthetic alternative or adjunct to
vanilla beans, and a source of vanilla flavor, is quite a bit less direct than this summary
implies. See Nadia Berenstein, “Making a Global Sensation,” 2016, especially pp. 405-
410.
66
synthesizing aldehydes and ketones which now bears his name.176 One of the first

commercial products produced via the Darzens reaction was ethyl

methylphenylglycidate, which, along with its homologue, ethyl phenylglycidate, are

described as having a strawberry-like aroma. This compound was sold under the name

“Aldehyde C-16” (although it was not an aldehyde, and did not contain 16 carbon atoms.)

Another important new addition to the series was so-called “peach aldehyde,”

undecalactone, first synthesized by Russian chemists Shukov and Shestakov in 1908.

This was produced and sold under a variety of trade names, including Persicol and

Pescol, as well as “Peach Aldehyde” or Aldehyde C-14, although it, too, was not an

aldehyde. One 1916 catalog from a New York essential oil and synthetic chemical dealer

listed the substance under the name, Aldehyde C-14, noting “similar products are sold in

the market as Persicol and Pescon,” before going on to say that their product was

“absolutely pure” and guaranteed to contain “no foreign bodies or matters.” The

catalogue recommended it for use in flavoring extracts, as well as in talcum powders and

creams. The description concluded: “It gives new odors,” and praised its stability and

lack of reaction with acids and alkalies.177

These synthetic ‘aldehydes’ marked a significant shift in the chemical market for

flavoring materials. While the “compound ethers” used in flavorings in the nineteenth

century bore no verified relationship to the fruit they were intended to suggest, many

176
Pierre Laszlo, “Georges Darzens (1867-1954): Inventor and Iconoclast,” Bulletin for
the History of Chemistry 15/16 (1994): 59-65.
177
George V. Gross & Co. “Essential Oils and Synthetic Chemicals,” [catalog] (New
York, 1916): np. Hagley Museum and Library, Delaware.
67
chemists understood that the reaction process that produced these esters could occur as a

result of fruit ripening.178 But with these new synthetic aldehydes, lactones, and ketones,

as one flavor chemist remarked later in 1949, “here then was a really new development,

for now the synthetic chemist had developed compounds with flavors similar to those of

natural origin, but of vastly greater flavoring power.”179

Although companies such as Synfleur and Van Dyk manufactured these materials

domestically before the war, most of the new synthetic aromatics were manufactured on a

very small scale, in laboratory glassware.180 The First World War, and its disruption of

trade networks with Europe, spurred the growth and diversification of an American

synthetic chemicals industry, including the production of pharmaceuticals, fertilizers, and

petrochemicals.181 It also drove the domestic manufacture of a wider range of synthetic

flavor and fragrance chemicals, including materials which had previously been imported

from Europe.

Sometimes, the production of an important new flavoring material could be a

matter of happenstance, emerging not from directed chemical analysis or exact chemical

knowledge but from a close attention to, and capitalization upon, the sensory qualities of

chemical materials. A sterling example of this can be found in the story of Fries’ peach

178
See, for instance, Clemens Kleber, “The Occurrence of Amyl Acetate in Bananas,”
American Perfumer & Essential Oil Review 7.10 (December 1912): 235.
179
David E. Lakritz, “Development of Flavors,” Drug and Cosmetic Industry 65
(December 1949): 723. Lakritz was the chief chemist at Florasynth, a synthetic flavor and
fragrance manufacturer.
180
Paul Z. Bedoukian, “The Perfumery Aromatics Industry in the United States,”
American Perfumer & Aromatics 70 (November 1957): 31.
181
Kathryn Steen, The American Synthetic Organic Chemicals Industry: War and
Politics, 1910-1930, (Chapel Hill: UNC Press, 2014).
68
flavor, as recounted by James Broderick, a flavorist whose career began in the late

1930s.182 Fries’ peach had been the “target for peach” when he entered the industry.

During the war, Fries’ had a government contract to process castor oil. During

processing, “something went wrong and a powerful odor of peach developed. They

repeated the processing exactly and again developed a peach aroma.”183 At the time, they

were unaware that the peachy component in the reaction mixture was a gamma

undecalactone; even without analytic knowledge of the identity of the compound in

question, they used this substance as the basis of their peach flavor, which gave them an

unmatchable edge over competitors until the lactone in question became commercially

available. Another compound used in Fries’ admirable peach flavor was also derived via

a similarly inexact process. “For reasons we never ascertained,” Broderick writes, using

his customary first-person plural, “a strong cheese had been soaked in alcohol and placed

in the basement near the furnace.” Months later, the cheese gave off an estery-fruity

peach scent. Although neither of these compounds would pass contemporary quality

control procedures, their use showed an open-mindedness to the sensory potentials of

materials, the chemical improvisations necessary to achieve new effects. “The modern

flavorist might — the flavor researcher most certainly would — think this strange,”

Broderick agrees. “But in the days when there were no lactones, no hexenols, no

pyrazines, no raspberry ketone, etc., the flavorist had to resort to various modifications to

182
James J. Broderick, “Reflections of a Retired Flavorist Before He Forgets: Peach,”
Perfumer & Flavorist 17.1 (Jan/Feb 1992): 35.
183
Broderick 1992: 35.
69
achieve desired nuance.”184 The virtues of a well-made flavor, he concludes, derive from

the skillful use of synthetics.

The Virtues of the Synthetic

During his lecture at Columbia University, Isakovics posed a question that he

knew to be on the minds of his listeners. “Why is it necessary to make substances by

synthesis on a commercial scale, when these same bodies may be found in nature?”185

His answer to this sheds light on how organic chemistry had reshaped the contours of

commerce.

First, chemical analysis permitted a re-calculation of the sources of value within

natural substances, such as essential oils. Nothing in nature is pure or unmixed. An

essential oil comprises many different compounds: some are valuable, some are useless,

and some are actually undesirable (such as certain terpenes, which take on unappealing

aromas when oxidized). Further, all methods of producing essential oils inevitably altered

the sensory qualities of the blossom or plant. Something was always altered or lost.

Finally, nothing about these substances was certain. The quantity and quality of different

essential oils varied from year to year, as did prices on the market. In all of these

situations, “synthesis comes to the relief of the manufacturer,” offering ways to reliably

produce in pure form and at a stable price “the active constituents imparting either odor

184
Broderick 1992: 35.
185
Isakovics 1908: 7.
70
or flavor, in the most concentrated form, readily soluble, always of the same strength,

free from by-products or objectionable constituents.”186

This reveals a very different view about the sources of value, and the meaning of

purity, in aromatic materials than the ones established in the Bureau of Chemistry’s

flavoring extract standards, which acknowledged only materials of exclusively botanical

origins as ‘pure’ and ‘standard.’187 Isakovics argued that the value of aromatic materials

derived not from their botanical origins (and the types of labor involved in their

cultivation and production, as well as the cultural narratives that follow them from bloom

to bottle), but from definite and identifiable molecules. Where the Bureau of Chemistry

prized purity of origins, Isakovics instead advocated for purity of substance —

compounds that efficiently delivered sensory effects unencumbered by useless,

insensible materials. These were two rival versions of progressivism, the first supporting

an absolute distinction between the products of nature and those of industry, the second

celebrating the triumphs of industrial science in surpassing nature’s boundaries and

improving nature’s processes.

The Improvement of Nature

By now, working with synthetic flavors meant gaining a mastery of specific

chemical materials and processes. Increasingly, it also came to include a recognition of

186
Isakovics 1908, 8-9.
187
U.S. Department of Agriculture, Standards of Purity for Food Products (Circular 19),
Washington D.C. (June 26, 1906): 13-5. This subject is discussed at greater length in
Chapter Two.
71
the multisensory and multimodal aspects of flavor perception. Creating successful flavors

came to require not only perfecting a specific aromatic formula, but also considering the

role that flavor played within particular foods, in interaction with other food components.

It also came to mean explicitly taking into account the desires, preferences, and

expectations of people as consumers, and the ways that flavor could influence and inform

those preferences and desires.

Beginning with Erich Walter’s 1916 Manual for the Essence Industry, textbooks

related to flavor chemistry and manufacturing routinely opened with a chapter addressing

the physiology and psychology of flavor perception — topics that had been only

haphazardly considered, if at all, in earlier works.188 In Walter’s model, “taste” comprised

both the four “basic” sensations perceptible by the tongue (sweet, sour, bitter, and salt),

as well as what he called the aromatic taste. These sensations, he explained, were all

responses to specific kinds of chemical stimuli. For instance, sourness indicated the

presence of acid. The aromatic taste responded to volatile substances, such as terpenes, as

well as a category of substances that he described as “extractives,” non-volatile

compounds present in cell sap, which were associated with bitterness. “The taste is not

itself a substance, but is a special property of substances… a phenomenon of energy

recognized by our nerves,” he explained, somewhat eccentrically. “It is for this reason

that it is possible to transfer the taste, or flavors, to our foods and beverages.”189 In other

188
Erich Walter, Manual for the Essence Industry, first edition, (New York: John Wiley;
London: Chapman & Hall), 1916.
189
Walter 1916: 3.
72
words, to study taste was to study the effect of certain chemical compounds upon the

human nervous system.

Although Walter’s model of taste did not seem to have been widely adopted, its

presence in a book on flavor manufacturing — one written by a self-described “beverage

specialist,” and containing chapters on the use of flavoring essences in non-alcoholic and

alcoholic beverages, confectionery, and other foods — reflects the broadening scope of

the field. People who worked with flavors were beginning to systematically examine the

multi-sensory and multi-modal aspects of flavor perception, and to link these to bodily

processes, such as appetite and physiological stimulation. Take, for instance, the writings

of Melvin De Groote, who worked on three fellowship projects related to flavor

chemistry and flavor manufacturing at the Mellon Institute for Industrial Research, where

he gained a reputation as an authority on these subjects.190 A chemical engineer by

training, De Groote was also an ardent advocate for the importance of chemical industrial

research in the food and beverage industry, and frequently contributed articles to industry

trade journals on the subject.

190
De Groote worked on Fellowships 38, 39, and 90. Fellowships 38 and 39, which ran
consecutively from November 1920 to November 1923, were funded by the Pittsburgh
Brewing Company and the Research Extracts Corporation, and focused on the production
of emulsion flavors for non-alcoholic beverages. Fellowship 90, which ran from
November 1917 to November 1920, was funded by Procter & Gamble and related to
glycerine, an important component of emulsion flavors and an alternative to alcohol as a
flavoring menstruum. The fellowship contracts can be found in the Mellon Institute
Collection 0000.42, Carnegie Mellon University Archives.
73
 De Groote began one such article, “Chemical Research In Beverage Ingredients,”

by posing the question: what makes consumers decide to purchase a soft drink?191 Was it

for the calories? This was surely not the driving motive, as there were many more

economic sources of energy. Was it to satisfy thirst? Again, even though a soft drink

fulfills this function, a parched citizen can reliably quench her or his thirst more cheaply

by other means. “Regardless of other properties,” De Groote asserted, soft drinks “are

sold primarily because they delight the sense of taste.”192 With this statement, De Groote

definitively associated the value of the drink with its added flavor — which meant that it

was critical to get that flavor right.

“Many bottlers wonder why they are never successful in compounding a flavor

according to some direction found in a trade journal or in a handbook of formulas,” he

wrote. By thinking of flavoring extracts as “trade secrets,” they missed the point: modern

flavors required vast amounts of technical and scientific knowledge, including knowledge

of new materials and processes. Compounding them properly also demanded extensive

practical experience, which, when combined with some degree of innate ability. Those

few who combined knowledge with experience and a certain degree of innate ability

developed “a sort of sixth sense that means in reality an intuitive genius or at least the

sense of intuitional creation.” He continued:

191
Melvin De Groote, “Chemical Research in Beverage Ingredients,” Beverage Journal
58, no 5 (July 1922): 50a-h.
192
De Groote 1922: 50a.
74
 “The bottler should remember that… in the purchase of an extract he is
not paying so much for the materials employed or the cost of manufacture,
but rather for that incommunicable technic of the specialist. The expert is
placing his product in their hands for their use is evidence of years of
difficult, tedious, and laborious apprenticeship, together with the
intuitional creative genius of the skilled and gifted aromatician, certified
by scientific knowledge of the most modern advances in modern aromatic
chemistry…. The manufacturer of bottlers’ extracts must in the future sell
his product, not on the basis of a closely-guarded trade secret, but rather
on the basis that it is the finest product that science and art can
produce.”193

The flavor maker’s product was not combinations of chemicals — it was

specialized knowledge, material expertise, and creative skill. The peculiar conjunction of

artisanal craftsmanship, modern scientific knowledge, and “intuitional creative genius”

sums up the unique identity of the workers who developed and formulated flavorings for

the rapidly growing food industry. Drawing on models of the past (such as

apprenticeship), this new kind of skilled worker was nonetheless a future-oriented

participant in modern industrial processes, not bound to protect guild secrets, but

empirically creating new knowledge and new sensations, “certified” by science. Although

this particular form of expertise was not widely recognized — as the absence of a stable

professional appellation for these workers indicates — by the end of the First World War,

the skilled flavor maker was known to be more than a chemical mixer or a formula

follower by those industries that needed his (and, at the time, it was almost always “his”)

193
De Groote 1922: 50g-h.
75
services. Meanwhile, the sensational mixtures he produced were becoming increasingly

familiar and prized by the ever-growing population of Americans who consumed the

products of his creative scientific labor.

TTT

76
 Chapter 2
A Flavor You Can’t Forget: Genuine,
Imitation, and the Meaning of “Natural”
Flavor

“At the very first sip, you are happily conscious of the miracle of it,” ran a 1923

newspaper advertisement for NuGrape, a deep-purple-hued grape-flavored soda that was

rapidly becoming a best-selling beverage at a time when consumption of carbonated “soft

drinks” was increasing due to Prohibition’s restrictions on the hard stuff. The promised

“miracle” was that “all the flavor of the Vineyards” — “the aroma, tang, sunny splendor

of wide vineyards, and the perfume of growing, ripening grapes!” — were present in each

bottle, in each sip, of NuGrape.1

“It is no mere echo,” the advertisement assured. “It is Reality — as if you had

plucked a cluster of purple Concords and were pressing their amber juice between your

lips. NuGrape is the liquid flavor of Concords, livened, given champagne-life, by the

secret NuGrape process —it leaps and glistens in the glass with the glow of health.”

Indeed, the “secret process” that transformed grapes into NuGrape also seemed to cast

enchanting effects over the palates, bodies, and lives of its drinkers. “What a zest it

provides for other things,” the advertisement murmured. “The most languid hours are

brightened and made more endurable.”

1
“All the Flavor of the Vineyards in this Bottle,” advertisement, Atlanta Constitution
(April 17, 1923): 7.
77
 No other grape soda tasted quite like NuGrape, according to advertisements,

posters, songs, and other promotional material that appeared first in Southern states and

then around the nation in the 1920s and 1930s, and no other grape soda offered the same

refreshment and pleasure. It was, according to its slogan, “a flavor you can’t forget.”

That’s why buyers were warned to “keep a sharp eye on the NuGrape Bottle,” and make

sure that it had three embossed rings around its neck. “It is our three-ringed trade-mark

guarantee of the REAL THING — and there are many imitations.”2

2
“All the Flavor of the Vineyards…” 1923.
78
 “All the Flavor of the Vineyards in this Bottle,” Atlanta Constitution, (April 17, 1923): 7.
Digitized on www.newspapers.com (https://www.newspapers.com/image/26969317). Downloaded Oct 31,
2016.

In a world of imitations, its makers insisted, NuGrape was the original. But

according to federal regulators charged with enforcing the 1906 Pure Food and Drugs

Act, NuGrape was itself an imitation — of grape juice and genuine grape flavors. In

1925, regulators took action against the NuGrape Company of America, charging it with
79
violating the law by falsely presenting itself as “composed in whole or in part of the juice

of the natural fruit of the grape… thereby tending to mislead the purchasing public as to

the quality of its product and to stifle and suppress competition.”3 The company

conceded, and agreed to prominently print the following admission on all of its bottles

and in all of its advertising: “Imitation Grape — Not Grape Juice.” Later that decade,

NuGrape altered its formulation in an attempt to place itself on the right side of nature.

With the help of “Merchandise No. 25,” a flavoring product devised by Fritzsche

Brothers, a New York flavor and fragrance manufacturer, NuGrape claimed to derive its

flavoring qualities exclusively from grapes themselves. When regulators challenged

NuGrape’s claims to authenticity, the company fought back — ultimately losing the case

after details about the production of “Merchandise No. 25” were revealed in federal

court.

What was the relationship between the flavor of grapes and the flavor of

NuGrape? What made NuGrape an imitation in the eyes of the law, and why did it matter

to regulators, to the company, and to consumers whether it admitted as much? Could a

bottle of NuGrape be the “real thing” — and yet also be an imitation? How was the line

between “genuine” and “imitation” defined and policed, and on what grounds was it

contested?

3
U.S. Federal Trade Commission, “Complaint No. 1199: Federal Trade Commission v.
The NuGrape Co. of America. Charge: Unfair Methods of Competition,” Annual Report
of the Federal Trade Commission for the Fiscal Year Ended June 30, 1925, (Washington,
DC: Government Printing Office, 1925): 179-80.
80
 Since the passage of the Pure Food & Drug Act in 1906, federal agents, armed

with evidence produced by chemical laboratories, had intervened to ensure both the

wholesomeness and transparency of the food supply.4 The law is considered a landmark

for public health, keeping putrid beef, watered-down “swill” milk from filthy urban

dairies, and dangerous patent medicines out of the national food and drug supply. The

law also endeavored to protect citizens by combating consumer fraud: prohibiting

commercial misrepresentations, such as the substitution of imitation for genuine goods.

The law’s chief concern was to prevent goods of lesser value from passing themselves off

as “better than they actually were.”5 The premise was that imitations and substitutions

were inherently less valuable, and of lower quality, than goods designated “genuine.” The

Bureau of Chemistry, a scientific agency within the US Department of Agriculture, was

the official arbiter of these disputes, tasked with interpreting the law and using the

methods of analytic chemistry to make and enforce determinations not only about the

4
The political and social history of Progressive era food legislation in the US has been
extensively documented, including by James Harvey Young, Pure Food: Securing the
Federal Food and Drugs Act of 1906 (Princeton: Princeton UP, 1989); Lorine Swainston
Goodwin, The Pure Food, Drink, and Drug Crusaders: 1879-1914 (Mcfarland, 1999);
and Bee Wilson, Swindled: The Dark History of Food Fraud, from Poisoned Candy to
Counterfeit Coffee, (Princeton: Princeton UP, 2008). The British version of this story can
be found in Michael French and Jim Phillips, Cheated, not Poisoned? Food Regulation in
the UK, 1875-1938 (Manchester, 2000). Alessandro Stanziani authoritatively chronicles
food regulation in France in his Histoire de la qualité Alimentaire: XIXe-XXe Siécle
(Seuil: Paris, 2005). For an account that compares how different types of scientific and
medical expertise shaped regulations in the US and in German-speaking countries in
Europe, see Uwe Spiekermann, “Redefining Food: The Standardization of Products and
Production in Europe and the United States, 1880-1914,” History and Technology 27
(March 2011): 11-36.
5
Peter Barton Hutt and Peter Barton Hutt II, "A History of Government Regulation of
Adulteration and Misbranding of Food," Food Drug Cosmetic Law Journal 39 (1984): 2-
73. The prohibition against a food being made to seem “better than it actually is” came
from an influential model law published in the 1880 Sanitary Engineer, and the phrase
was often cited by pure food advocates in their definitions of food adulteration.
81
presence or absence of chemical entities, but also about identity, authenticity, meaning,

and value. While regulators dictated that even a drop of synthetic flavoring material

relegated a food or beverage to the lesser-value status of “imitation,” flavor companies

and the food manufacturers who used their products argued that synthetic flavorings were

not only safe, but beneficial. Rather than fraudulent concoctions that dishonestly masked

unsavory goods, these manufacturers argued that flavor additives were progressive

products of scientific research. They added value, increased quality, and made entirely

new categories of manufactured products possible and accessible.

What was ultimately at stake in these contested debates about naturalness,

authenticity, and identity were questions of value. How should the value of a food be

determined, and whose expertise should matter in making these determinations? This

chapter begins by considering the 1906 Pure Food Law and the terms under which the

state regulated flavoring additives and manufactured foods that used these products. A

system of regulatory standards imposed a strict distinction between botanically derived

flavorings and products of synthetic chemistry, requiring foods and beverages which

included any of the latter to prominently disclose their status as “imitation” or

“compound” on packages and labels. Chemists — working on behalf of federal and state

governments — were tasked with enforcing the law, but designating the difference

between “natural” and “imitation” proved to be far from clear-cut. Rather than a neutral

requirements that increased market efficiency and transparency, I show how these

regulations simultaneously presumed and imposed a quality and value hierarchy that

placed “genuine” products above synthetic “imitations,” one that did not necessarily align

82
with the way these products were made, used, and experienced, and which was only one

of multiple ways of measuring food’s value that emerged during this period.

I then turn to the case of NuGrape, tracing the history of the beverage and the

changing flavoring materials that gave it its grapiness, alongside the political, economic,

and cultural forces that defined its value and its meaning. Often, the interpretation of new

synthetic materials is folded into a binary with some prior natural material, where the

synthetic material is taken as an inferior substitute for a scarce natural resource.

However, whether celluloid, oleomargarine, vanillin, or synthetic grape flavor, the

“substitute” material often has virtues, and affords possibilities, that the “natural” lacks

(and vice versa). It is, moreover, embedded in a different network of raw materials,

producers, consumers, and calls into being different kinds of expertise.6 What I argue

here is that synthetic flavor additives were not only used as (and understood to be)

cheaper substitutes for the scarce but genuine things of nature, but were called into

necessity by the large-scale production of “everyday luxuries” for mass consumer

markets. Synthetic flavors did not merely make it possible to deliver a version of an

existing sensory experience to broader group of consumers, but instead delivered new

kinds of experiences, and new kinds of value, to consumers whose lives, bodies, and

senses were being transformed in modernity. Against the interpretation of the imitation as

a less-virtuous, imperfect substitute for the “real thing” — NuGrape as less-than grape —

6
On the case of celluloid, see Robert Friedel, Pioneer Plastic: The Making and Selling of
Celluloid, (Madison: UWisconsin Press, 1983) and Jeffrey L. Meikle, American Plastic:
A Cultural History, (New Brunswick: Rutgers UP, 1995.) On oleomargarine, see: Ruth
Dupree, “‘If It’s Yellow, It Must Be Butter: Margarine Regulation in North America
Since 1886,” Journal of Economic History 59.2 (June 1999): 353-71. On vanillin, see,
Nadia Berenstein, “Making a Global Sensation,” History of Science 2016.
83
I instead propose that the effect produced by NuGrape represents a new thing coming into

being. In this emerging order, synthetic flavorings were technologies of sensory

experience, performing in a new sensual economy, one that valued affective aspects of

experience, intensifications of sensation and emotion, and demanded new forms of

refreshment and new varieties of pleasure.

I. QUALITY, VALUE, FLAVOR: ADDED FLAVOR AFTER

THE 1906 PURE FOOD AND DRUG ACT

The 1906 Pure Food and Drug Act became the law of the land in an era of

growing public concern over the hidden dangers lurking in the nation’s food supply.

Decades of media coverage about food adulteration, especially in women’s magazines,

muckraking exposés about “swill milk” and filthy meatpacking plants, and well-

publicized deaths from “ptomaine” poisoning built political will for the law’s substantial

expansion of the federal government’s regulatory powers. For the lawmakers and the

business coalition whose support for the law was essential in securing its passage, rooting

out commercial fraud was as pressing an issue as preserving public health.7 Testimony

about commercial fraud comprised the majority of evidence presented in Congressional

hearings in support of the bill, as well as in the reports of state regulatory agencies and

the Bureau of Chemistry.8 “What we want is that the farmer may get an honest market

and the consumer may get what he thinks he is buying,” explained Harvey Wiley, the

7
Donna J. Wood, “The Strategic Use of Public Policy: Business Support for the 1906
Food and Drug Act,” Business History Review 59 (Autumn 1985): 403-32.
8
Wood 1985: 408-9.
84
chemist who, as head of the USDA’s Bureau of Chemistry, was one of the law’s chief

architects and most ardent champions.9

Wiley’s summary of the bill’s intent connected a fair market for farmers with a

transparent market for consumers, but left out manufacturers, who transformed raw

agricultural and chemical materials into an expanding range of commercial goods, and

were responsible for an ever-growing share of both the food supply and the national

economy.10 For many reformers, manufacturers were at the root of the problem; their

quest for profits coupled with the increasing distance between consumers and producers

introduced new opportunities for fraud.11 Unscrupulous manufacturers had a willing

accomplice: chemical science. “The development of bacteriological and chemical science

contributed to the upsurge of food, drink, and drug adulteration,” writes historian Lorine

Swainston Goodwin. “Large firms began to employ industrial chemists to develop

deodorants for rotten eggs and rancid butter, dyes to enhance color, agents to alter

flavor… and ways to keep pickles crisp.” For many reformers, she explains, these

products of the chemical laboratory were de facto evidence of fraudulence, allowing

9
Harvey Wiley, address to the 1898 National Pure Food and Drug Congress, quoted in
Young 1989: 128.
10
By 1900, a fifth of all goods manufactured in the United states were food products.
Beginning in the late nineteenth century, an ever-increasing share of foods were produced
by large corporations, such as Heinz, Campbell’s, and Nabisco. Harvey Levenstein,
Revolution at the Table: The Transformation of the American Diet (New York: Oxford
UP, 1986): 30-43; Helen Zoe Veit, Modern Food, Moral Food: Self-Control, Science,
and the Rise of Modern American Eating in the Early Twentieth Century, (Chapel Hill:
UNC Press, 2013): 44.
11
Goodwin 1999: 48-9.
85
profit-hungry manufacturers to produce and sell low-quality goods while consumers

remained unsuspecting.12

Advances in chemistry were responsible not only for a growing number of

synthetic preservatives, flavorings, and other food additives, but also entirely novel food

products, such as oleomargarine from the meatpacker’s scraps and “glucose” from corn

starch — products whose economic legitimacy was questioned by rival agricultural

industrial interests, but whose safety was rarely seriously disputed. There was widespread

evidence of the use of chemical additives and substitute substances, but also little

evidence that these were harmful, except to the bottom line of established manufacturers

of butter, honey, and other products perceived to be in competition.13 Most advocates for

reform, however, did not intend to outlaw the use of all chemical additives or synthetic

products in the food supply. By and large, reformers, legislators, and pro-regulation

manufacturers agreed that there was a place in the market for low-quality and substitute

foods, and that poor consumers should have access to cheaper, albeit inferior, food items.

Meanwhile, for manufacturers who used preservatives, flavorings, and other additives,

the presence of these substances was not evidence of fraudulence, but represented

attempts to improve the eating quality of their products by technical means.

“Transparency” was a fraught question for these new kinds of products. Canned

vegetables and meats, boxes of biscuits, condiments in glass jars, candy in tins, soda in

bottles: processed foods were packaged foods, and interposed a layer of opacity between

12
Goodwin 1999: 49.
13
Young 1989: 66-92, 104-5;
86
the consumer and the thing to be consumed, foreclosing the possibility of direct sensory

examination. Furthermore, the proof of the pudding was less and less likely to be found

in the eating. In a market of sealed containers, distant producers, and unfamiliar additives

and processes, how could consumers know not only what their food was, but what it was

worth?

For turn-of-the-century consumers, the food market was shot through with

uncertainty and risk. Economic and business historians have described the Pure Food and

Drug Act as an effort to reduce this risk, and resolve informational asymmetries in order

to increase market efficiency.14 According to these scholars, chemists and other technical

and scientific experts were authorized to make official determinations about a food’s

identity and contents — and thus, by implication, its value — determinations that

consumers were no longer equipped to make. In other words, when consumers cannot

detect whether they are being “cheated,” official chemists must step in to make the

determination on a material, rather than a sensory, level. It has thus been argued that

regulations were necessary to restore consumer confidence in fundamental food quality,

even as the consumers’ susceptibility to the sensible aspects of these distinctions

eroded.15

This line of reasoning presumes that judgments about quality are made against

pre-existing norms, that differences in quality are determinable by experts using agreed-

14
See, for instance, Wood 1985; Marc T. Law, “How do Regulators Regulate?
Enforcement of the Pure Food and Drugs Act, 1907-1938,” Journal of Law, Economics,
and Organization 22.2 (2005): 460-86. Marc T. Law, “The Origins of State Pure Food
Regulation,” Journal of Economic History 63 (December 2003): 1103-1130.
15
Law 2003: 1116.
87
upon methods, and that the meaning of these differences is self-evident. Recently

scholars have begun to push back against these assumptions, historicizing not only the

meanings of quality but also modes of negotiating and adjudicating disputes about the

quality of foods.16 As Alessandro Stanzioni has observed in his study of the origins of

French food regulation, “quality is not an objective, ahistorical category,” it is a contested

category, whose meaning not only changes over time, but also reflects different regional,

political, social, and economic beliefs and interests.17 Whether a chemical additive

constituted an “adulteration” or an “improvement” — whether it was an “imitation” or an

“innovation” — was anything but self-evident, especially at a time of rapid changes in

food production, distribution, and consumption. Nor was the meaning of “pure” or

“natural” simple to determine. In the case of milk and other dairy products, for instance,

some physicians and other reformers opposed pasteurization, on the grounds that it would

stall efforts to improve the sanitation of dairies. “Pasteurization, they emphatically

repeated, could make milk safe, but it could not make dirty milk clean.”18 The craving for

the “purity” was not necessarily a demand for nature’s raw materials, unaltered.

Furthermore, the authority of scientific experts to adjudicate these disputes was

not a given. Scientific authority had to be laboriously and contentiously established, a

task often complicated by the ambiguous and indeterminate results produced by

16
Jerome Bourdieu, Martin Breugel, Peter Atkins, "'That Elusive Feature of Food
Consumption:' Historical Perspectives on Food Quality, a Review and Some Proposals,"
Food & History 5, no 2 (2007): 247-266.
17
Alessandro Stanziani, “Negotiating Innovation in a Market Economy: Foodstuffs and
Beverage Adulteration in Nineteenth-Century France,” Enterprise and Society 8 (June
2007): 375-412.
18
Kendra Smith-Howard, Pure and Modern Milk: An Environmental History Since 1900,
(New York: Oxford UP, 2013): 33.
88
analysis. 19 As Benjamin Cohen has pointed out, when it came to enforcing regulations —

for instance, distinguishing what could call itself butter from what could legally not use

that name — the context and consequences were both social and scientific. Analytic

chemists were “not just detectors of chemical impurities; they were participants in a vital

cultural arbitration” that sought to disentangle the authentic from the imitation, the

deceptive surface from the genuine interior.20

As governments took a more active and interventionist role in regulating the food

and drug supply, the stakes in the debates over how these distinctions were to be made

grew more significant.

Flavor Additives and Food Standards

At the time when the market itself was becoming a central presence in not only

the economic, but also the social and cultural lives of Americans, rising concerns about

food adulteration and quality reflected a growing anxiety about the limitations of market

forces to equitably distribute rewards and secure virtuous outcomes.21 Can the fair market

value of food be established by market forces alone? Flavor additives and other chemical

19
Spiekermann 2011. In a comparable case study examining the adoption of the
hydrometer as the standard tool for determining alcoholic proof (and assessing duties),
William Ashworth has illustrated how the scientific authority of the instrument and the
power of the state were contested by merchants who had previously relied on sensory
expertise to classify sprits. William J. Ashworth, “‘Between the Trader and the Public’:
British Alcohol Standards and the Proof of Good Governance,” Technology and Culture
42 (January 2001): 27-50.
20
B.R. Cohen, “Analysis as Border Patrol: Chemists Along the Boundary Between Pure
Food and Real Adulteration,” Endeavor 35 (2011): 66-73.
21
Susan Strasser, Satisfaction Guaranteed: The Making of the American Mass Market,
(New York: Pantheon, 1989): 125.
89
products and processes aggravated doubts on the matter. Reformers worried that the

production and price of foods could be deranged by the superficial and specious appeal of

chemically altered and enhanced goods. Because synthetic flavor additives erased the

perceptible difference between actual and apparent value, these chemical materials

complicated the equation for determining the actual worth of foodstuffs. The law’s

strategy for protecting consumers from this type of fraud was by prohibiting the sale of

foods defined as “adulterated” or “misbranded.”

Adulterated food was defined to include contaminants that posed threats to health

(such as rotten or diseased meat, or ingredients known to be harmful), as well as various

sorts of material manipulations that were perceived to affect food’s value. In the latter

cases, a food was adulterated, and thus outlawed in interstate commerce, if “any

substance has been mixed and packed with it so as to… injuriously affect its quality or

strength,” “if any substance has been substituted wholly or in part” for the food item, “if

any valuable constituent… has been wholly or in part abstracted,” or “if it be mixed,

colored, powdered, coated, or stained in a manner whereby damage or inferiority is

concealed.”22

“Misbranding” concerned the claims made on the package about the food’s

identity or contents. Foods were misbranded if they were imitations “offered for sale

under the distinctive name of another article,” or “labeled or branded as to deceive or

mislead the purchaser” about its contents or identity by including “any statement, design,

22
Federal Food and Drugs Act of 1906, Public Law 59-384, 59th Cong., 1st sess. (June
30, 1906), Section 7, “Adulterations.”
90
or device regarding the ingredients or the substances contained therein” which were

“false or misleading in any particular.” The statute’s regulation of misbranding allowed

for two exceptions, which made it possible to bring novel types of manufactured food to

market. First, products could be sold if they were plainly and clearly marked as such,

using terms such as “imitation,” “compound,” or “blend” to indicate their distinction

from the standard article. Second, what was known as the “distinctive name” provision

allowed manufacturers to sell their products under unique, coined trade names.23 Court

decisions would establish that this provision protected products such as “Bred Spred,” a

fruit-flavored jam-like spread, from having to label itself “imitation jam,” as well as

products with better-known distinctive trade names, such as “Coca-Cola.”24

The law simultaneously put the contents of food and how it was represented under

review, primarily by attending to and policing a distinction between goods of greater or

lesser value. The definition of adulteration and misbranding as commercial fraud

presumed a stable set of common standards of identity and value for certain foods, which

impostor foods undermined by chemical legerdemain and unsuspected substitutions. Yet

there were no stable or agreed-upon ways of determining the “valuable constituents” of a

food, nor of unequivocally defining “injurious” changes to quality and strength. Chemical

presences and absences could be registered, but their meanings and their effects on food’s

23
Federal Food and Drugs Act of 1906, Section 8 “Misbranding.”
24
The standard established by the court was that the name of the product must be
“either… so arbitrary or fanciful as to clearly distinguish it from all other things, or one
which by common use has come to mean a substance clearly distinguishable by the
public from everything else.” See Suzanne White Junod, “Food Standards in the United
States: The Case of the Peanut Butter and Jelly Sandwich,” in David F. Smith and Jim
Phillips, eds. Food, Science, Policy and Regulation in the Twentieth Century:
International and Comparative Perspectives, (New York: Routledge, 2013): 167-88.
91
ultimate value were underdetermined. At the very least, the authority of chemical analysis

to distinguish between pure and adulterated depended on the existence of a materialist

account of food’s identity: what a certain food must contain to properly represent itself as

such.

In 1897, the Association of Official Agricultural Chemists (AOAC), the

professional organization of regulatory chemists working in state and federal

government, began developing and publishing food standards that defined common foods

in terms of their contents. This was connected with concomitant efforts by the group to

develop standard and uniform methods of food analysis. Generally, the formulation of

food standards involved not only regulators, but also manufacturers and other experts

who were presumed to be authorities on the foods in question. However, to the chagrin of

Wiley, his confrères in the Bureau of Chemistry and state regulatory agencies, and the

AOAC, by the time the Pure Food Law was passed, legal authority for the creation of

these standards had been stripped from the bill. This was due largely to the efforts of

senators sympathetic to the makers of blended whiskey, who believed food standards

would either outlaw their product or force it to be labeled ‘imitation.’25 Until 1938, when

25
The development of “unofficial” food standards before the 1938 Food, Drug, and
Cosmetics Act has yet to be fully chronicled by historians. The most comprehensive
historical accounts of food standards in the US can be found in White Junod 2013, as
well as in Hutt and Hutt 1984. For a fascinating account of postwar food standards that
analyzes the label as information infrastructure, see Xaq Frohlich, “The Informational
Turn in Food Politics: The US FDA’s Nutrition Label as Information Infrastructure,”
Social Studies of Science (2016): 1-27. Angie Boyce’s recent study focuses on debates
around the creation of a standard of identity of peanut butter in the 1960s and 1970s,
which she uses as a case study to examine the relations between experts and lay
consumer activists in the determination and definition of technological artifacts. Angie
M. Boyce, “’When Does it Stop Being Peanut Butter?’ FDA Food Standards of Identity,
92
the Food, Drug, and Cosmetics Act authorized the establishment of mandatory food

standards, these food standards had only “advisory” status, which meant that their

legitimacy could be subject to judicial challenge.26

“Circular 19,” published in June 1906, articulated definitions and compositional

limits for multiple categories of commodities, staple processed foods, and condiments

including meat, dairy, grain, fruit and vegetable products, vinegars, fats and oils; they

also specified standards for twenty-three different flavoring extracts, from almond extract

to wintergreen.27 These and subsequent advisory standards served as a guide for

manufacturers whose foods had to conform to the published definitions (or else be

labeled “imitation,” “compound,” or sold under a distinctive name) in order to avoid

regulatory action against their products, and were meant to protect the interests of

consumers, but were written in “laboratory language,” the terms of art of analytic

chemists, specifying upper and lower limits for various chemically measurable

constituents.28 Standards were exclusionary: components not listed in definitions were

not permitted in standard products.

According to Circular 19, a flavoring extract was “a solution in ethyl alcohol of

proper strength of the sapid and odorous principles derived from an aromatic plant, or

Ruth Desmond, and the Shifting Politics of Consumer Activism, 1960s-1970s,”

Technology and Culture 57.1 (2016): 54-79.
26
Court rulings on the authority of the government’s food standards were inconsistent;
some rulings found in favor of the USDA’s standards to guide enforcement actions,
others rejected regulators’ authority. See Hutt & Hutt 1984: 59.
27
“Standards of Purity for Food Products” 1906: 13-15.
28
This is in contrast with later, ingredient-based standards, imposed after the 1938 Food,
Drug, and Cosmetics Act. White Junod 2013: 168.
93
parts of the plant, with or without its coloring matter, and conforms in name to the plant

used in its preparation.”29 Flavoring extract standards, attempted to set minimum quality

requirements by specifying two things: botanical origins — traceable to a plant or its

leaves, roots, or seeds — and minimum “proper” flavoring strength. For instance, a

product calling itself “vanilla extract” guaranteed that it derived its vanilla flavor

exclusively from vanilla beans and that at least 10 grams of beans had been used for

every 100 cubic centimeters of extract.30

At the end of 1906, in response to numerous inquiries, the USDA Bureau of

Chemistry issued further guidance on labeling flavorings that were excluded from the

standards. This included flavorings that included compounds such vanillin and coumarin

as well as “numerous preparations made from synthetic fruit ethers intended to imitate

strawberry, banana, pineapple, etc.”31“Such products should not be so designated as to

convey the impression that they have any relation to the flavor prepared from the fruit.

Even when it is not practicable to prepare the flavor directly from the fruit, ‘imitation’ is

29
“Standards of Purity for Food Products” 1906: 13.
30
The vanilla extract standard’s specification of materials used during the production
process was an exception to the typical wording of these requirements, one that was made
necessary by the availability of synthetic vanillin and the commercial importance of the
vanilla industry. Most extract standards specified minimums in terms of volume of
valuable components in the final product. For instance, cinnamon extract was required to
contain at least 2 percent by volume of oil of cinnamon; oil of cinnamon was defined as
the volatile oil from the bark of Ceylon cinnamon, containing no less than 65 percent
cinnamic aldehyde and no more than ten percent eugenol. See Berenstein “Making a
Global Sensation” 2016.
31
James Wilson, Secretary of Agriculture, Food Inspection Decision no. 47: “Flavoring
Extracts,” U.S. Department of Agriculture Bureau of Chemistry, (December 13, 1906),
Flavoring Extracts General Data, Food Standards Committee, Record Group 88, Records
of the Food and Drug Administration, National Archives and Records Administration II,
College Park, MD.
94
a better term than ‘artificial.’” This designation carried over to the foods that these

“substitutes” were included in. For instance, ice cream made using a synthetic strawberry

flavor could not be legitimately labeled “strawberry ice cream.” Even when there was no

comparable flavoring product — no ‘genuine’ strawberry flavoring that the synthetic

product was competing with or substituting for— synthetics were de facto imitations.

Other standards related to the use of flavorings in food products — such as soft

drinks — were elaborated in subsequent published notices of judgment, USDA circulars,

and department bulletins, as well as in speeches and other communications between the

USDA and trade groups, such as the Flavor and Extract Manufacturers’ Association.32

These attempted to keep pace with the rapidly changing technological, chemical, and

commercial conditions of manufactured foods and beverages as new kinds of products

came on the market. For instance, when “cloudy” citrus-flavored beverages became

popular in the 1920s, regulators moved rapidly to define the legitimate versions of this

product. The Bureau of Chemistry specified that the terms -ade, squash, punch, crush,

32
Guidelines indicating the Bureau of Chemistry’s interpretation of statute can be found
in the published notices of judgment under the Federal Food and Drugs Act, USDA
Circular 21 (Rules and Regulations for Enforcement). Circular 19, “Standards of Purity
for Food Products,” was superceded in 1919 by Circular 136, of the same title. In
addition to standards for flavoring extracts (which remained largely unchanged from
Circular 19), Circular 136 contained standards for soda water flavors. Bureau of
Chemistry Bulletins concerning spices and flavoring extracts include Bulletins 63, 132,
and 152. Official information about regulations and labeling was also regularly published
in trade journals for bottlers, druggists, soda fountain operators, essential oil and
fragrance manufacturers, and others. Flavoring manufacturers also communicated
directly with the Bureau of Chemistry (and later, the FDA) seeking clarification and
advice about appropriate label language. Although the agency demurred from granting
approval to proposed labels, it did offer technical advice and guidance. See: Center for
Food Safety and Nutrition, Office of Nutritional Products, Labeling, and Dietary
Supplements, Record Group 88, Records of Food and Drug Administration, National
Archives and Records Administration II, College Park, MD.
95
and smash could only be used to describe beverages that contained fruit juice; others,

including those flavored with essential oils and essences of botanical origin, must be

labeled ‘imitation.’33

Meanwhile, food officials struggled to develop reliable and standard chemical

methods for distinguishing “true” products from those which must be labeled as

“imitation,” thus providing scientific evidence of adulteration that could carry weight in

the federal courts where these charges were adjudicated. This required not only methods

of identifying the presence of adulterants, but also the analysis of the chemical

components responsible for flavor in foods, spices, and “pure” products. For this reason,

some of the earliest published chemical research into the flavor chemistry of fruits was

performed by USDA researchers.34 Only by determining the actual chemical components

of apples, peaches, and grapes could the presence of synthetic additives be demonstrated.

But even the most painstaking chemical analysis could never quite provide unequivocal

proof of a substance’s status, as the flavor chemistry of natural foods remained largely

unknown. These determinations were particularly fraught in cases where the synthetic

33
J.W. Sale, “Labeling Beverages and Beverage Materials Under the Federal Food and
Drugs Act,” in R.O. Brooks, Critical Studies in the Legal Chemistry of Foods, (New
York: Chemical Catalog Company, 1927): 267. This paper was originally presented at a
meeting of state, local, and federal and food officials in 1924.
34
For instance, see Frederick B. Power and Victor K. Chestnut, "The Odorous
Constituents of Apples," Journal of the American Chemical Society 42 (7) (1920), 1509-
1526; Frederick B. Power and Victor K. Chesnut, "The Odorous Constituents of
Peaches," Journal of the American Chemical Society, 1921: 1725-1740.
96
compound was known to be chemically identical to the molecule found in nature, as was

the case with synthetic vanillin and vanilla extracts.35

In the majority of cases involving flavoring additives and flavored products,

misbranding was a necessary precondition of adulteration. Penalties involved the seizure

and destruction of goods as well as the imposition of fines, generally between $25 and

$100. These relatively small fines did not always “deter the careless or dishonest

manufacturer from continuing the adulteration of his products,” but according to Bureau

of Chemistry officials, “usually the adulteration of flavors is discontinued when the

manufacturer’s or importer’s attention is called to the matter.”36

The elaborate, growing set of food standards and the legal interrelationship

between adulteration and misbranding show that the enforcement of the Pure Food law

required regulating both language and chemical contents. What concerns and procedures

guided these efforts and shaped how regulatory meanings were articulated and

implemented? As Dr. William Frear, who served as a technical advisor on the

development of food standards, explained to a meeting of flavor manufacturers, standards

35
See Berenstein 2016: 417-8. Discussion of the official methods of vanilla evaluation
can be found in: AL Winton and EH Berry, “The Chemical Composition of Authentic
Vanilla Extracts, Together with Analytical Methods and Limits of Constants,” in Harvey
Wiley and AL Pierce, eds., Proceedings of the 28th Annual Convention of the Association
of Official Agricultural Chemists, USDA Bulletin 152, (Washington: Government
Printing Office, 1912): 146-58.
36
J.W. Sale and W.W. Skinner, “Food Flavors: Their Source, Composition, and
Adulteration: Part VI, Conclusion,” Beverage Journal (October 1922): 50a. This was the
concluding article of a six-part series by Skinner, Assistant Chief of the Bureau of
Chemistry, and Sale, the Chemist in Charge of its Water and Beverage Laboratory. Prior
articles in the series discussed the chemical composition of both “natural” flavoring
materials, such as spices, essential oils, and so on, and the chemical components of
synthetic flavors.
97
must reflect “the generally accepted name in such a way as to make the distinctions the

people ordinarily make between the product under consideration and all other food

substances.”37 In other words, when the agency had to arbitrate between the language

used by the public and the language of food manufacturers, food marketers, and

manufacturing chemists, the language of the “ordinary consumer” was statutorily

definitive. These usages, although in a certain sense arbitrary, had to be made stable,

unambiguous, and precise through the enforcement of these statutes, which relied on

chemical analysis rather than assessments of sensory quality. In this way, presumed

(though contested) distinctions in market value — between “true” and imitation — were

produced and reinforced by the regulations that claimed only to enforce market

transparency.

Why were flavorings of botanical origin privileged over products of synthetic

chemistry? In part, this derived from deeply held cultural beliefs about naturalness, what

Lorraine Daston and Fernando Vidal have termed the “moral authority of nature.”38 Yet

the meaning of “natural” in food had never been self-evident, and was even more in

dispute as the social and geographical distance between food producers and food

consumers increased. For a consuming public that was growing and making less of its

food at home and buying more of it in markets, “purity” and “naturalness” were

increasingly valued. However, we should not take this to signify a demand for nature’s

37
Dr. William Frear, “Standards for Flavoring Extracts: Address before Flavor Extract
Manufacturers Convention, Atlantic City, June 29, 1916,” Simmons’ Spice Mill 39
(September 1916), 1035-6.
38
Lorraine Daston and Fernando Vidal, eds. The Moral Authority of Nature, (Chicago:
University of Chicago Press, 2004).
98
raw materials unaltered. As Kendra Smith-Howard has shown in the case of milk, milk

became “nature’s perfect food” — considered wholesome, safe, and pure — by virtue of

technologies that standardized and centralized its production and distribution, even as its

marketers cultivated a pastoral ideal of dairying that was rapidly vanishing from the

countryside. “Though they credited nature for milk’s purity,” Smith-Howard writes,

“reformers altered the very nature of milk and the cows that produced it.”39 The qualities

that seemed to indicate the “naturalness” of pasteurized, homogenized milk produced in

large-scale dairies, or of creamery butter made in a centralized factory — not only the

absence of disease-causing microorganisms but also of off-flavors or flavor variations —

were the hard-won goods of technoscientific control rather than natural givens.

The naturalness of “pure food” was associated with authenticity, a virtue whose

meaning at this time must be defined within the changing social and economic contexts

of American life. The veracity of representation was an increasingly fraught question in

nineteenth-century America, when the set of local social relations that validated personal

identity began to fray.40 With the growth of the industrial economy, doubts swelled to

include things as well as persons. Is this item what it purports to be or is there a

disjunction between its sensible qualities and its inherent contents? Anxieties around the

disjunctions between essence and presentation swelled during this period of rapid

industrialization and account in part for the widespread perception of chemical additives

39
Smith-Howard 2013: 15.
40
Karen Halttunen, Confidence Men and Painted Women: A Study of Middle-Class
Culture in America, 1830-1870, (New Haven: Yale UP), 1982; Scott Sandage, Born
Losers: A History of Failure in America, Cambridge: Harvard UP, 2005.
99
as a sign of capitalism run amok.41 Whether or not any food additives posed a definite

risk to health, the proliferation of these chemicals was seen as a symptom of a broader

threat to national well-being posed by untrammeled competition in an unregulated

market.42 A pure product, by the definition of Progressive-era reformers, was not only

free from hazardous substances; it was also a product with a morally privileged history,

associated with agricultural production rather than industrial manufacturing.43

What of flavoring strength, the other property defined in the food standards? If

flavoring strength was a virtue when it came to products of botanical origins, it was a

suspect quality in synthetic materials. Many flavoring extracts contained both botanically

derived and synthetically produced substances. Flavoring manufacturers argued that

synthetic compounds only comprised a small fraction of the net contents of a flavoring;

moreover, these chemicals played a functional role in the mixture, serving as “fixatives,”

preserving the original flavor by forestalling flavor loss to volatility, and as intensifiers,

which increased the flavor’s power and strength, increasing its utility to food and

beverage manufacturers. Why, on the basis of two percent of a flavoring extract’s total

content, should the extract and the product it flavored both be labeled ‘imitation’? “It is

conceivable that so little synthetic flavor may be added… that the predominating flavor

of the article is genuine fruit flavor,” conceded J.W. Sale, the chief chemist of the

41
Cohen 2011.
42
Goodwin 1999: 48-51.
43
This is not to say that reformers and the consumers they claimed to represent were
unequivocally opposed to industrial food production. For an account of how moral
sentiments concerning trust and purity were co-opted by a large food manufacturer, see
Gabriella Petrick, "'Purity as Life': H.J. Heinz, Religious Sentiment, and the Beginning of
the Industrial Diet," History and Technology 27 (2001): 37-64.
100
Bureau’s beverage and water laboratory, before continuing: “but as a matter of fact,

owing to the great difference in flavoring power between the natural fruit flavors and

synthetic fruit flavors, the amount of synthetics which are ordinarily used is such that the

predominant flavor of the resulting product is due to the artificial flavor rather than to the

natural flavors.”44 In the context of the food regulations, the efficiency of synthetics — or

the sensitivity of the human sensorium to compounds used in synthetic flavorings — was

not a valued quality.

According to the food standards, a synthetic chemical could never be pure,

regardless of its harmlessness, its pleasant sensory qualities, or its other advantages —

not even a chemical such as vanillin, which was indistinguishable from the compound

found in ‘natural’ vanilla beans. The simple presence of synthetically produced

substances, no matter how small the quantity or how ‘pure’ the rest of the materials, was

enough to condemn a flavoring and the food that contained it as ‘imitation.’ But the

Bureau of Chemistry’s food standards were far from the only way of calculating value

and assessing quality in the American marketplace in the Progressive era.

Calculating the Value of Food and of Flavor

At around the same time that reformers were advocating for regulatory oversight

of the food system another way of calculating the value of food was rising to prominence:

nutritional analysis. The science of nutrition concerned both consumers and consumed,

combining the chemical analysis of foods and the physiological determination of caloric

44
Sale 1927: 268.
101
and nutrient needs of organisms. In the US, nutritional science gained authority by

affiliating itself with progressive political programs, and was deployed to rationalize both

production and consumption to optimize the abilities and capacities of laboring citizens.45

By quantifying the value of foods in terms of calories, macronutrients, and later,

vitamins, nutritional science made it possible to imagine substitutions among very

different kinds of foods.46 As Helen Zoe Veit has written, “By arguing that foods that

seemed superficially very different could be vehicles for the same needed nutrients,

nutritionists transformed food into a variable in a kind of cultural algebra.”47 Foods with

divergent market values (and distinct social meanings), such as rib roast and baked beans,

could be revealed to possess equivalent nutritional value. Maximal nutritional efficiency

was achieved when each person received her or his precise set of nutritional units at the

lowest possible cost. “Of course,” Veit observes, “this supposedly culture-blind

nutritional equivalency was only possible by deemphasizing tradition, habit, and often,

the pleasure of eating itself.”48 A heap of beans may provide the same caloric energy, and

the same quantity of protein, as prime rib, but the sensory experience of the two could not

have been been more distinct.

45
Jessica Mudry, “Quantifying an American Eater,” Food, Culture & Society 9.1 (2006):
43-67; Hamilton Cravens, “Establishing the Science of Nutrition at the USDA: Ellen
Swallow Richards and Her Allies,” Agricultural History 64.2 (Spring 1990): 122-33.
46
Gyorgy Scrinis, Nuritionism: The Science and Politics of Dietary Advice, (New York:
Columbia UP, 2013).
47
Helen Zoe Veit, Modern Food, Moral Food: Self-Control, Science, and the Rise of
Modern American Eating in the Early Twentieth Century (Chapel Hill: UNC Press,
2013): 46.
48
Veit: 51.
102
 How did flavor figure into these calculations? Although some valuable

ingredients had inherent tastes (e.g., sugar’s sweetness), flavor as such was thought to

contribute no “food value,” at least none that was directly measurable in terms of calories

or macronutrients. If a nutritional equivalency between rib roast and baked beans was

established, flavor differences between the two could be factored out. According to

historian Laura Shapiro, the Boston Cooking School, Ellen Swallow Richards’ endeavor

to promote the principles of scientific cookery among working-class women, held the

sensual aspects of eating in low regard. “Cooking-school cookery emphasized every

aspect of food except the notion of taste,” she writes. Students learned meal planning,

marketing, food chemistry, and nutrition. “But to enjoy food, to develop a sense for

flavors, or to acknowledge that eating could be a pleasure in itself had virtually no part in

any course, lecture, or magazine article.”49 The “New Nutrition” prescribed choosing

foods on the basis of macronutrient and vitamin content, not on taste; flavor was an

obstacle to the accurate comprehension of food’s value.50

This is not to say that flavor played no role in nutritional theory.

Psychophysiological research, such as that conducted by Ivan Pavlov, suggested a link

between the psychic phenomenon of appetite and the essentially mechanistic “chemical

laboratories” of digestion.51 Attractive flavor stimulated the appetite which triggered a

cascade of physiological changes — most notably, the preliminary flow of digestive fluid

— which allowed the valuable nutrients in food to be efficiently assimilated. Foods

49
Laura Shapiro, Perfection Salad: Women and Cooking at the turn of the Century,
(Berkeley: University of California Press, 2008): 68.
50
Veit 2013: 51; Levenstein 1986: 72-86.
51
Daniel P. Todes, “Pavlov’s Physiology Factory,” Isis 88.2 (June 1997): 205-46.
103
lacking in appetizing flavor, and meals “bolted down” without enjoyment or interest,

could lead to digestive stagnation and disease.52 “It has long been known that the value of

foods in nutrition does not depend solely upon the quantity of nutriment which these

foods contain, but also… on the ability of the digestive functions to utilize these

nutriments,” explained Harvey Wiley to the readers of Good Housekeeping. “The

influence of flavor… has long been recognized by physiologists as an exciter of the

digestive enzymes, promoting digestion and favoring health.”53 By connecting psychic

phenomena with physiological processes, flavor converted the latent, abstract nutritional

value of foods into utilitarian value, the currency that could build and sustain actual

living bodies.

But flavor’s exciting effects on the body could be taken too far. Food that was too

highly flavored, that was over-seasoned, and that mixed different herbs and spices could

be dangerously overstimulating, awakening appetites that would seek other stimulating

52
This theory of the utilitarian value of deliciousness, that it was necessary for proper
digestion and assimilation of nutrients, was widely promulgated at the turn of the century.
(For instance, the hope of finding an additive that could facilitate digestion and thus
improve national health by improving flavor was one of Kikunae Ikeda’s motives for
developing a process of manufacturing MSG in early 20th century Japan. See Sand 2005:
38.) In the United States, one of the most ardent promoters of this theory was Henry T.
Finck (1852-1926), who had studied experimental psychology and psychophysics in
Berlin, Heidelberg, and Vienna before a long career as a critic for the New York Evening
Post. Finck’s 1913 book Food and Flavor, a fascinating and comprehensive treatise on
the utilitarian virtues of deliciousness, made a case for improved health through
conscientious, Fletcher-influenced gourmandizing, and advocated for the improvement of
American cuisines. “Sensual indulgence,” for Finck, was a “duty,” as appreciating the
flavor of food enhanced national health, happiness, and the capacity for hard work. “The
most important problem before the American public,” he wrote, in all seriousness, “is to
learn to enjoy the pleasures of the table and to insist on having savory food at every
meal.” For Finck, gourmandism assumed the status of moral duty: “the highest laws of
health demand of us that we get as much pleasure out of our meals as possible.”
53
Harvey Wiley, “Life in the Husk,” Good Housekeeping 63.3 (September 1916): 65-6.
104
pleasures, such as alcohol and narcotics. Spices “pamper perverted appetites,” as one

cookbook author wrote in 1917.54 As Veit and others have shown, the proscription

against “strong” flavors and seasonings emerged in part from a resistance by White

Northeastern food reformers to immigrant cuisines as well as to African-American

southern foodways.55 Even though flavor per se was outside of the realm of calculation

when it came to assessing nutritional value, nutrition-minded reformers nonetheless

promoted the idea that certain kinds of gustatory experiences were wholesome and

healthful and others were dangerous and aberrant.

These critiques included artificial flavors and the “adulterated” foods that they

made alluring. The widespread consumption of foods flavored with synthetic “coal tar”

chemicals was credited not only with physical diseases, such as neuralgia, dyspepsia, and

“rheumatic and gouty twinges of nerves and muscles,” but with moral derangements —

with a creeping insensibility that threatened the health of the body and of the nation.56 “A

perverted taste has become so universal that we have lost the exquisite, delicate office of

the palate, and the flavors of pure food are not appreciated by the masses,” ran the

introduction of an 1896 cookbook by a pure food advocate.57 The key word used by many

of these critiques was perversion. “Reasonable gratification of the palate is not

incompatible with health,” one hygienic journal instructed in 1902, “for a healthy taste

54
Quoted in Veit: 130-1.
55
Veit 2013. E. Melanie Dupuis, Dangerous Digestion: The Politics of American Deitary
Advice, (Berkeley: University of California Press, 2015): 54-74.
56
Quote from Ellen Goodell Smith, A Practical Cook Book and Text Book for General
Use: The Fat of the Land and How to Live On It, (Amherst: Carpenter and Morehouse,
1896): 15-6.
57
Smith 1896: 15.
105
will not crave artificial flavoring. It is easy enough to distinguish between the promptings

of a healthy instinct, and the perverted longings of a diseased appetite.”58 Desiring and

surrendering to the appeal of “bad” flavors was both a symptom of a poorly-fed body and

a sign of a perverse and diseased system, and indicated a troubling lack of self-control,

one of the necessary virtues for full democratic citizenship.59

Arguing for the Value of Synthetic Flavors

In the mode of calculation used by reform-oriented nutritionists, home

economists, and their hygienist allies, flavor additives added nothing of value. Their

effect, instead, was inflationary, adding specious allure to food of poor quality or

inflaming appetites beyond the capacity of food alone to satisfy. The purpose of the Pure

Food Law’s “imitation” labels was not only to prevent consumer deception, but also to

protect people from their own appetites and warn them against the deleterious

consequences of the substances they may not be able to help but desire.

Makers of flavoring extracts, and some of the food manufacturers who used their

products, vigorously contested this condemnation of synthetic flavors. For many flavor

and food manufacturers, food reformers’ campaigns for “purity” and against “chemicals”

as flavors was based on an ignorance of chemistry and suggested a rejection of science as

a progressive force for material and moral advancement. “Some so-called scientists act

58
“Sauces,” The Medical Brief 30.2 (February 1902): 230.
59
Veit 2013; Dupuis 2015: 75-94; Chin Jou, Controlling Consumption: The Origins of
Modern American Ideas About Food, Eating, and Fat, 1886-1930, (PhD Diss, Princeton
University, 2009): 35-6.
106
like a bull when a red cloth is waved before them at the mere mention of coal-tar

products,” railed one beverage manufacturer speaking to a convention of state bottlers’

associations, decrying the lecturers and “yellow journalists” who based their invectives

against synthetic flavor and color additives on alleged scientific expertise. For anti-

additive reformers, “the fact that today there are 200,000 chemicals made from coal tar to

relieve pain in the human system, to stimulate industries, and please untold thousands, the

great results obtained by world-renowned chemists stand for naught.” Pure Food

advocates, the manufacturer argued, mistook the meaning of purity and spread false

information. “The resemblance of these absolutely pure chemicals has no more to do with

the raw coal tar than has the candle light with the sun.”60 The beverage manufacturer

countered the pure food movement’s claim to scientific and chemical authority with a

parallel claim on nearly identical grounds: chemical knowledge deployed for the

progressive ends of human and national advancement. The dispute between these two

factions was a dispute over the legitimacy of these rival claims, one that played out

within the community of chemists as well as in the public sphere.

Although there was broad support for the Pure Food law among flavor and food

manufacturers, who for the most part wanted uniform federal regulations and standards to

protect themselves from competitors’ fraudulent misrepresentations, many took issue

with the regulations’ suspicion toward synthetic substances. Alois von Isakovics, whose

Monticello, New York company Synfleur manufactured synthetic flavors and fragrances,

penned a pair of furious letters to William Frear, head of the Food Standards Committee,

60
H.C. Schrank, “Pure Food Laws,” Paper read before the conventions of the Wisconsin
and Illinois State Bottlers’ Associations, American Bottler 29.4 (April 1909): 20.
107
and Bureau of Chemistry Chief Wiley in response to the exclusion of synthetic materials

from recognition in the flavoring extract standards, especially the exclusion of synthetic

vanillin and other materials from vanilla extract. Isakovics claimed to have spent ten

years analyzing the chemistry of vanilla extract. Synfleur’s vanilla flavoring,

Vanillodeur, included vanillin and other chemical flavoring compounds that he said he

had isolated from extract and then reproduced by synthetic processes. He appealed to

these men as fellow chemists. Addressing Frear, he wrote that by requiring all vanilla

extracts to be made exclusively from vanilla beans:

“You shut your eyes to all advances in modern synthetic chemistry…. Our
product has now been marketed for years and is used by some of the
largest consumers in the country. Yet you step in as a chemist and desire
to kill my interests with one stroke, to prevent all further research along
these lines…. Why should you as a chemist try to hold back advance in
the science instead of encouraging it to your best ability. Our product has
come to stay and you know it. No amount of legislation will compell [sic]
a manufacturer to pay the grower of bean in Mexico five dollars when he
can get the same thing identical in every way made in the U.S.A. for fifty
cents.”61

For Isakovics, the anti-synthetic bias of the standards destroyed the value that had

been created in the synthetic product through his scientific labors and discouraged future

61
Alois von Isakovics to William Frear, May 3, 1905; Flavoring Extracts: 1900-1908;
Food Standards Committee Correspondence & Reports, 1897-1938; Records of the Food
& Drug Administration, Record Group 88; U.S. National Archives at College Park, MD.
108
research. Isakovics recapitulated many of the arguments he had made to Frear in his letter

to Wiley, but made a further personal appeal to a fellow man of science. (Isakovics had a

collegial relationship with Wiley through their common membership in various scientific

societies.) He also described the relative cheapness of the synthetic product as a social

virtue:

“You are a chemist, a professional chemist who always hails with delight
anything new in the science. Yet why discriminate against new work in
this line…. If we can give the manufacturer a product that gives the
identical same flavor as the bean for one tenth the money, it enables the
manufacturer to cheapen the product and the masses can enjoy a good
flavor which if pure mexican bean was used could only be afforded by the
well to do…. I cannot understand why you as a scientific man, as a
progressive and broadminded chemist should oppose any advance in the
science and should compell [sic] the American manufacturer to use the old
fashioned and out of date raw materials.”62

Why should the law discriminate against chemical progress in flavors, when the

government’s agents (rightly) celebrated chemical progress in all other applications?

Progress in synthetic flavors would secure a broader and more equitable distribution of

pleasures, a democratization of delights for an era of mass luxuries. Where pure food

advocates had agitated for “imitation” labeling to protect consumers from deception,

62
Alois von Isakovics to Harvey Wiley, May 5, 1905, Flavoring Extracts: 1900-1908;
Food Standards Committee, Correspondence & Reports, 1897-1938; Records of the Food
& Drug Administration, Record Group 88; U.S. National Archives at College Park, MD.
109
Isakovics argued instead that the “imitation” label actually harmed consumers by giving

them a false impression that the product contained within was of low quality. He

elaborated on this in a subsequent letter to Frear:

"The manufacturer of the chemical product does not have the facilities of
reaching the public…. If you compell [sic] the manufacturer of flavoring
extracts individually to commence to educate the public this will at once
cause a barrier that cannot be overcome. The average person using Vanilla
does not care what it is made from as long as it gives the true Vanilla
flavor. That is all they are interested in. They understand nothing of
chemistry and don't want to know about it and you can't teach them — the
experiment would could [sic] cost ten times as much in advertising as the
total possible sales of the product. You look at the whole question from an
entirely unpractical standpoint. On the one hand you have the man who
has spent many years in research to produce something really good. He
puts it out and has no trouble at all to convince the consumer — the
manufacturer — as to the value of his product. Now you step in and say to
the manufacturer. You must not use these goods. We will not allow it.
What chance have we in that case?"63

Although synthetic products were not forbidden, the label disclosure ‘imitation’

carried a stigma that manufacturers were eager to avoid.64 Consumers labored under the

63
Alois von Isakovics to William Frear, May 9, 1905; Flavoring Extracts: 1900-1908;
Food Standards Committee Correspondence & Reports, 1897-1938; Records of the Food
& Drug Administration, Record Group 88; U.S. National Archives at College Park, MD
64
The term “artificial” seems to have been less stigmatizing, and manufacturers seem to
have preferred it, but through the 1920s, at least, the Bureau of Chemistry favored the
110
belief that ‘imitation’ meant low quality, even if this belief was not sustained by their

experience with the product. Further, the flavor manufacturer — whose direct customers

were food and beverage manufacturers, not the ultimate consumers of flavored products

— was in a particularly tricky position. No matter how excellent the sensory and material

quality of the imitation flavor was, it would still bear the mark of non-standard, and thus

sub-standard, quality.

Regulators’ interposition between the manufacturer and the consumer was not

only a problem for manufacturers of synthetic flavors. Among makers of botanical

flavoring extracts, there was a widespread belief that the food standards distorted the

market by forcing higher-than-standard-quality goods to compete with those that just met

the standard. Many of these manufacturers encouraged consumers to put their trust not in

the imprimatur of the federal government, but in brands.65 For instance, McCormick &

Co., the Baltimore spice and extract company, published an educational pamphlet that

sought to enlighten readers that “purity” as defined by the Pure Food Law was not

necessarily a sign of quality. Under the heading, “Quality v. Purity,” the pamphlet

term ‘imitation.’ In 1922, the National Manufacturers of Soda Water Flavors adopted a
resolution at their annual meeting lobbying for a change to the terms “artificially
flavored” and “artificially colored” on the grounds that the word imitation “applies to all
other ingredients of the beverage as well as to the flavor and color,” ie, might indicate
that it contained saccharin rather than sugar, and “is a disparaging term, giving the public
the impression of cheapness and inferiority,” thus constituting “a hardship and injustice
to the manufacturers of soda water flavors and to the bottlers of soda water.” “Flavor
Manufacturers in Annual Meeting,” Beverage Journal (Nov 1922): 51.
65
This was a common strategy for building trust in (and a market for) unfamiliar
processed food before the passage of food regulations as well. See: Nancy F. Koehn,
“Henry Heinz and Brand Creation in the Late Nineteenth Century: Making Markets for
Processed Foods,” Business History Review 73.3 (1999): 349-93; Koehn, Brand New:
How Entrepreneurs Earned Consumers’ Trust from Wedgwood to Dell, (Boston: Harvard
Business School Press, 2001); Strasser 1989.
111
instructed: “The people have been taught by the laws and the Pure Food propagandists to

believe that the word ‘Pure’ upon a package ensures that its contents are all right.

Nothing can be further from the truth. An article may be Pure and yet of very Poor

Quality.” A Pure Vanilla Extract may be made from low-quality “rank” Tahitian vanilla

beans rather than . “The time is coming when consumers will realize that the important

thing to look for in the purchasing of foodstuffs is not the word ‘Pure’ — but the name of

the reputable manufacturer whose dealings are beyond reproach.”66 Both makers of

synthetic flavorings and those of botanical products were making the same argument, that

the label disclosures imposed by regulators bore little relationship to the actual quality of

the goods in question.

Ultimately, questions of sensory quality were what most sharply distinguished the

chemists working in the laboratories of regulatory agencies from the chemists working in

the laboratories of flavor companies and food manufacturers. While officials from the

Bureau of Chemistry could make credible determinations about chemical presences and

absences, they had little authority when it came to evaluating sensory quality.67 Flavor

66
McCormick & Co, Spices: Their Nature and Growth; The Vanilla-Bean; A Talk on
Tea, pamphlet, (Baltimore: 1915): 28. Smithsonian Libraries Trade Literature Collection.
67
Analytic chemists at the Bureau of Chemistry did, of course, make judgments based on
sensory evaluation in the course of their assessments of different flavoring extracts, but
these were not held as evidentiary when prosecuting charges of adulteration or
misbranding. The one exception to this concerns the standard for “vanilla and vanillin”
flavor, which was based on an organoleptic assessment of the relative flavoring power of
each substance, to ensure that 50 percent of the flavor sensation was attributable to
natural vanilla and 50 percent to synthetic vanillin. (This worked out to equal parts
standard vanilla extract and 0.7% vanillin solution.) See: JW Sale, "Labeling of Flavoring
Extracts" American Perfumer & Essential Oil Review, July 1925. Originally presented at
the 16th Annual Meeting of Flavor Extract Manufacturers’ Association, Chicago, IL, June
24, 1925.
112
manufacturers claimed expertise over both the sensory and chemical aspects of flavoring

materials, arguing for the virtue and value of their products not only by insisting on their

harmlessness and chemical purity, but also by making the case for their integral role in

improving the sensory quality of foods.

C.F. Sauer, the head of the Richmond, Virginia extract company that bore his

name, outlined a typical case for the necessity of added flavorings at the Flavor and

Extract Manufacturers’ Association meeting in 1918. Flavor, he explained to his

colleagues in the business, was “the basis of all foods,” and thus “essential to a great

many institutions,” from hospitals preparing “delicate and tasty foods for the sick” to

food industries with many millions a year in revenue. The important role of flavor in

national life was particularly acute during wartime staple rationing. Sauer argued that

flavoring extracts were “the most concentrated of all foods. They help to make meatless

days a success. They conserve eggs, sugar, flour by stimulating the use of substitutes, as

they make more palatable the somewhat insipid foods” that had replaced familiar items in

the wartime pantry.68 An advertisement for Sauer’s Extracts that appeared elsewhere in

the trade journal that carried his speech underscored this point. Sauer’s Extracts, it read,

are “first aids in conservation… make war-time foods and substitutes tempting.”69

Flavoring extracts were essential components of a rationalized national food system,

efficiently improving quality of foods and quality of life.

68
C.F. Sauer, “Why Flavoring Extracts are Essential Food Products,” Simmons Spice Mill
(August 1918): 1014. Originally presented at 9th Annual Meeting of Flavor Extract
Manufacturers’ Association, Richmond Virginia.
69
C.F. Sauer Company, “First Aids in Conservation: Sauer’s Pure Flavoring Extracts,”
advertisement, Simmons’ Spice Mill (August 1918): 1017.
113
 Flavoring additives were not just useful for making minimally acceptable wartime

foods more palatable; they had a purpose even in high-quality and standard foods.

Recapitulating the Pavlovian argument about the digestive utility of flavor, Sauer

thundered, “I do not… think that I exaggerate when I say that flavoring extracts

contribute to the health of a nation, as health depends on the enjoyment and the ease with

which we digest the food we eat. Few of us realize the part that flavoring extracts play in

our daily life.”70 Implicitly rejecting the distinction made by food reformers between

wholesome “pure” and dangerously overstimulating “impure” flavors, for Sauer and his

colleagues, flavor itself was a virtue. To add flavor was to add value.

According to the manufacturers and users of synthetic flavors, the value of these

products was not only commercial but also social, physiological, and even patriotic.

Countering the scientific authority of reformers and regulators with their own claims to

chemical expertise, they argued that flavoring additives were modern, scientific products,

exemplars of progressive virtues such as efficiency and purity. Rather than adulterants

that harkened back to a risky, dishonest marketplace prior to national regulation,

flavorings were necessities in a modern food system, integral to the new kinds of food

and beverage products made in factories and to the new kinds of pleasures these

delivered. The value of synthetics was located not only in sensory or chemical similarities

with “natural” products, but also in their sensory possibilities and material differences.

The meaning of these differences in raw materials, in methods of production, and in

physicochemical properties cannot be assessed only in terms of their ability to replace or

70
Sauer 1918.
114
substitute for botanical substances, but must also include the new kinds of products, and

the new modes of experience, that they made possible.

Ultimately, manufacturers argued that the proper way of assessing the value of

flavoring extracts was not in terms of material origins or production costs, but in terms of

sensory quality. Or, as a 1921 flavor catalog put it, “In order to arrive at the valuation of

an extract, it should not be regarded as a commodity… but rather should be visualized as

a potential means of producing 10,000 pleasurable sensations.”71 Its value was proven not

in the chemical laboratory, but in the sensory responses (and commercial behaviors) of

consumers:

“A good flavor is an intangible and fugitive thing that is gone almost

before it can be perceived; but the real test by which every flavor should
be judged is, — does it leave a lasting and favorable impression behind it
when the sensation of taste has disappeared? This is the way we judge our
Red Seal Extracts and we can honestly say that each one leaves a pleasant
memory behind; so pleasant indeed that anyone who drinks a bottle of Red
Seal Soda involuntarily craves another so that he can again enjoy the
pleasure afforded by its delicious flavor.”72

Beyond pure and imitation, beyond the calculations of nutritional reformers or the

analyses of regulatory chemists, there was a unique virtue and value to mass-produced

71
Warner-Jenkinson Co., Bottlers’ and Ice Cream Makers’ Handy Guide, (St. Louis:
Warner-Jenkinson, 1921): 80. A.W. Noling Collection, UC Davis.
72
Warner-Jenkinson Co. 1921: 80.
115
pleasures, to the repeatable charms of an expertly crafted flavor, which could deliver its

anticipated delights again, and again, and again, precisely as remembered.

II. NuGrape and Nature: Added Flavor in the New

Sensory Economy

The Bureau of Chemistry’s bifurcation of the flavored world into higher-value

“pure” or “standard” and lower-value “imitation” failed to capture aspects of value

creation, and dimensions of the citizen’s sensual and social relations to the products that

they bought, that were coming into being in the first part of the twentieth century. The

case of NuGrape illustrates both the practical complexities of implementing the Pure

Food law, as well as the ways in which NuGrape’s unique brand value was built upon the

fundamental ambiguity of its relationship to “natural” grapes.

NuGrape, a product of the NuMint bottling company of Atlanta, was introduced

just ahead of the 1921 summer season, and rapidly became one of the most popular

bottled sodas in the United States. By the time it attracted the attention of federal

regulators in 1925, the company was claiming that NuGrape was the second-best-selling

116
five-cent bottled beverage in the world, with more than a 1.5 million bottles sold every

day.73

With the introduction of NuGrape, NuMint may have been looking to capitalize

on the growing market for non-alcoholic “soft drinks” during prohibition, as well as on

an increasing appetite for grape-flavored sodas, as numerous rivals (such as Grapico,

Nehi Grape, and Grape Nip) began to appear on the market, especially in the South and

Midwest. NuGrape’s territory expanded with breakneck speed. Within a year, NuGrape

was being shipped from Atlanta to more than 100 bottling plants throughout the South,

each of which was granted an exclusive franchise over designated territory.74 A

modernized, fire-proof building allowed for expanded production.75 The Atlanta

Constitution featured photographs of “solid carloads” of NuGrape being shipped by

traincar to bottlers, and new trucks were added to the NuGrape fleet.76

Almost as soon as NuGrape came on the market, the company took pains to

establish its unique brand identity — to position itself as the original, most beloved, and

most desirable grape soda pop in a marketplace teeming with lesser imitators. The

73
NuGrape Bottling Co., “At Last — Cincinnati is to Know the Thrill of ‘A Flavor You
Can’t Forget,” [Advertisement] Cincinnati Enquirer, (May 5, 1925): 8.
74
“Infant Atlanta Industry Grows to Giant in a Year,” Atlanta Constitution, (June 25,
1922): 8; “NuGrape Makes a Hit,” The Re-Ly-On Bottler 3.1 (January 1922): 11.
“NuGrape Bottling Co. Perfecting Organization, Growing in Favor,” The [Nashville]
Tennessean, December 2, 1923: 41. A discussion of the beverage industry business
model that distributed flavoring products to independently operated bottling plants, which
were given exclusive rights to distribute a branded beverage over a certain geographical
area, can be found in Elmore, Citizen Coke: The Making of Coca-Cola Capitalism, 32-41.
75
“NuGrape is Showing Enormous Increase,” Atlanta Constitution, (September 10,
1922): 8.
76
“Solid Carload of NuGrape Being Shipped to Many Cities,” Atlanta Constitution,
(April 16, 1922): 8.
117
company invested heavily in advertising, spending more than $3 million on ads between

1922 and 1927.77 These campaigns spread the message that for the drinker in search of

flavor and refreshment, there was no substitute. “The flavor of NuGrape is unmistakable

— there is no other drink whose flavor is even remotely like NuGrape.”78 Advertisements

and jingles instructed consumers to “use your eyes to protect your taste,” by looking for

the three rings embossed around the neck of the genuine NuGrape bottle. The goal was to

build consumers’ exclusive relationship with NuGrape, rather than generate an appetite

for grape-flavored sodas in general.

Spurious grape beverages not only lacked the unmistakable NuGrape flavor, they

were potentially hazardous. “A Pure Beverage is Non-Injurious! What a Substitute Is,

Nobody Knows!” warned one of the earliest newspaper advertisements for NuGrape,

from August 1922. These impostor grape sodas contained “unknown — possibly

dangerous ingredients.”79 But what, exactly, did NuGrape contain? The advertisement

reproduced a letter from an Atlanta chemical testing laboratory testifying to the soda’s

soundness. “The summary of these tests shows your product to be a pleasant and

wholesome beverage in every particular,” read the letter. “It contains nothing that is

deleterious or dangerous to health. Neither does it contain anything prohibited by our

state or federal laws…. In the light of the facts disclosed by this investigation, we have no

77
“An Investment Offering of Unusual Merit! First Public Offering of Stock of NuGrape
Co. of Atlanta,” Greenville News, (May 7, 1929): 12.
78
Nugrape Company of America, “Use Your Eyes to Protect Your Taste,” advertisement,
Atlanta Constitution, July 19, 1922: 4.
79
Nugrape Company of America, “A Pure Beverage is Non-Injurious!” advertisement,
Atlanta Constitution, Friday August 18, 1922: 11.
118
hesitancy in recommending Nu-Grape as a pleasant and wholesome beverage.”80 Rather

than offering assurances based on a disclosure of components, NuGrape’s claim to purity

and wholesomeness rested on the trustworthiness of chemical analysis.

But the chemist’s testimonial avoided mentioning the source of NuGrape’s flavor,

carefully evading any statement about its relationship to grapes or grape juice. Elsewhere,

however, NuGrape’s marketing campaigns were replete with images and language that

drew grape and NuGrape close together. Newspaper ads for Nugrape were often framed

within garlands of grapevines, and featured messages such as: “NuGrape has the same

wonderful flavor of ripe, juicy grapes. You can’t mistake it once you taste the original.”81

The original NuGrape bottle was plumply embossed with a bunch of grapes, which took

on the purple color of the soda contained within. But the proposed dyad of grape and

NuGrape was about more than proving a point of taste; these advertisements sought to

invest NuGrape with the meaning of grapes as well as their flavor. “The exquisite,

delicate flavor of the finest Concord grapes is better duplicated in NuGrape than in any

other beverage, and that is why NuGrape has a flavor distinctively its own,” read a 1923

advertisement, beneath an image of a young woman proffering a platter heaped with

grapes. She wore a ruffled rustic blouse, and a bonnet tied loosely with a ribbon. Over her

right shoulder, neat rows of grapevines receded toward gentle hills; over her left, a

marbled counter spread before a gleaming soda fountain. The ad continued, “There are

many inferior imitations of the winey NuGrape flavor, but none that is so magically

80
Ibid.
81
Nugrape Company of America, “Warning! Beware of Substitutes!” advertisement,
Atlanta Constitution, (June 20, 1922): 2.
119
suggestive — by color, aroma and taste — of real Concords.”82 The advertisement staged

a relationship between NuGrape and nature that took in not only the grapes themselves,

but the bucolic scenario of their cultivation and consumption. The thirst it hoped to spark

was not only for the taste of Concord grapes, but a nostalgic longing for a way of life

increasingly remote from that of the parched soda-drinkers in an urbanizing and

industrializing America, pausing for a drink in the midst of the acceleration all around

them.

So did NuGrape deliver the flavor of Concord grapes, a replica, or something

“distinctively its own”? The vineyards themselves, or a “magical suggestion” of them?

By skillful feats of rhetorical misdirection, these and other advertisements reframed the

question of NuGrape’s relationship to grape by putting NuGrape’s authenticity at the

center. NuGrape itself was the original, the genuine article. NuGrape, not nature, was the

model that other beverages aspired to. Indeed, what NuGrape promised exceeded

anything that grapes alone could offer. “NuGrape showed Nature how to improve the

flavor of the Concord grape.”83 If NuGrape and nature were not identical, it was not

because NuGrape fell short of nature’s model, but because it surpassed it — not only in

the quality of its flavor, but in the extravagance of the pleasures that it promised.

Even if the role of grapes in producing NuGrape was coyly evaded in its

marketing campaigns, the question of NuGrape’s relationship to genuine grapes was of

82
Nugrape, “Their De-Licious Flavor,” advertisement, Reading [PA] Times, (August 21,
1923): 6.
83
Nugrape Company of America, “A Thirst-Hitting True-Grape Flavor,” advertisement,
Atlanta Constitution, (July 14, 1922): 8.
120
concern to the Federal Trade Commission.84 In 1925, the federal agency took action

against the company, accusing it of unfair trade practices. According to the federal

complaint, the product’s name and advertising suggested that the beverage “is composed

in whole or in part of the juice of the natural fruit of the grape, when in fact it is not made

of the juice of the grape,” misleading the public and unfairly competing with beverages

made from actual grape juice. Instead of contesting the charges, NuGrape agreed to

immediately cease and desist from using “any pictorial representation of grapes or grape

vineyards, or any words, pictures, or symbols stating or suggesting that NuGrape is made

from grapes or grape juice,” in its packaging and advertising. The company also agreed to

include “Imitation grape - Not grape juice” in every instance where the word NuGrape

was used.85 In advertisements, this disclosure fit between NuGrape’s trade name and its

slogan, “The flavor you can’t forget!” The NuGrape bottle retained the three rings around

the neck, but the embossed bunch of grapes vanished, replaced by the words “imitation

grape.”

84
Although cases of adulteration and misbranding involving synthetic flavors were most
often brought and pursued by the USDA’s Bureau of Chemistry, the FTC also had
regulatory authority here, although its actions were restricted to cases where a specific
(anonymous) complaint had been filed by a competitor.
85
U.S. Federal Trade Commission, Complaint No. 1199, 1925: 179-80.The agreement
precisely stipulated the size and visibility of this disclosure, requiring that “imitation
grape - not grape juice” appear “in close proximity to the word ‘NuGrape’ and in letters
at least one half as high and one half as wide… and of heaviness of color and style of
lettering which will render them at least equally as conspicuous in proportion to their
height and width” as the brand name of the beverage.
121
How Methyl Anthranilate became Grape Flavor

But if NuGrape did not derive its flavor (exclusively) from Concord grapes,

wherewith was it flavored? The story of the production of synthetic grape flavor in the

early twentieth century demonstrates how the flavor and fragrance industries were bound

by shared material and business networks, while also showing how the latent potentials,

uses, and meanings of materials varied between contexts of use.86

Sometime between 1909 and 1914, Gilbert Hurty — Amherst-trained chemist,

bachelor, bottling company proprietor — was riding an Indianapolis streetcar when he

caught a whiff of destiny in a fellow rider’s perfume. It smelled just like ripe Concord

grapes. “Soon after,” according to a manuscript documenting the history of his company,

Hurty-Peck, “he canvassed all the essential oil houses for perfume materials and finally

discovered that the product he had smelled was methyl anthranilate. He then incorporated

it in a grape flavoring oil and had an outstanding product.”87 Hurty-Peck’s grape flavor

was showcased in Louisville, Kentucky in 1914, at the first exposition of bottlers’

supplies for the American soft drink industry, where it was sold to or sampled by both

small and established bottlers from around the country.88 It was later advertised to those

who sought the “utmost degree of PURITY, STRENGTH, and NATURALNESS,” and

86
In this regard, the chemical used in grape flavorings, methyl anthranilate, can be
described as a boundary object, whose meaning, qualities, and potentials shift as it passes
between these professional and social worlds. See Susan Leigh Star and James R.
Griesemer, “Institutional Ecology, ‘Translations’ and Boundary Objects: Amateurs and
Professionals in Berkeley’s Museum of Vertebrate Zoology, 1907-39,” Social Studies of
Science 19.3 (1989): 387-420.
87
A.W. Noling, History of Hurty-Peck & Company: Its First 50 Years, 1903-1953,
annotated manuscript, (Indianapolis: A.W. Noling, 1968). Noling Collection, UC Davis.
88
Noling 1968: 7.
122
appears to have propelled the company from shaky financial standing to a solid foothold

in the beverage flavor business.89

By the time Gilbert Hurty first sniffed it out, methyl anthranilate was a well-

known material in perfumery, albeit one of recent vintage. In the mid-1890s, essential oil

chemists in Germany had identified the molecule as the characterizing compound in

neroli, the essential oil of orange blossoms, a popular perfume material. The chemical’s

presence was subsequently detected in other fragrant essential oils: jasmine, tuberose,

gardenia, ylang-ylang, and bergamot.90 Synthetic methyl anthranilate had been

commercially available since the end of the nineteenth century, and by 1913, it was

manufactured and sold by numerous chemical suppliers, often listed as artificial neroli.

Hurty-Peck’s grape flavor was not the only one to appear on the bottlers’ market

around this time, and almost certainly not the only one to use methyl anthranilate to

deliver its grape effect. Advertisements for grape flavorings begin to appear in trade

journals such as the American Bottler as early as 1911, with bottlers’ supply companies

such as Sethness, Twitchell’s, Lehman-Rosenfeld, and Warner-Jenkinson highlighting

“Concord grape” flavors among other fruit flavorings that were highly concentrated,

economical, kept indefinitely, and were “absolutely pure.”

The Concord-grape-flavored sodas that these substances produced were destined

for consumers who had been steadily gaining a taste for “unfermented wine,” non-

89
Hurty-Peck Company, “Hurty-Peck Flavors,” advertisement, American Bottler 37
(April 1917): 61. Noling 1968: 7-9.
90
Eduard Gildemeister and Friedrich Hoffmann, Die Ätherischen Öle, (Berlin: Springer,
1899).
123
alcoholic grape juice, a beverage increasing in popularity with the spread of the tee-

totalling sentiments. (For most of the nineteenth century, Americans mainly consumed

fruit juices in the form of home-made hooch.)91 As more states and counties became

“dry” territory, brand-name fruit juices such as Welch’s Concord grape juice were ready

to slake American thirsts, chastely. Secretary of State William Jennings Bryan, one of

prohibition’s bulldogs, notoriously served the company’s purple juice at a 1913 dinner in

honor of the British ambassador. The following year, the secretary of the navy banned

alcoholic beverages on ship, substituting Welch’s for sailor’s customary grog.92

Carbonated beverages — “soft drinks” — had also made themselves into

temperance beverages, in part to shed now-disreputable associations with the narcotic

ingredients and proprietary medicinal purposes of their pasts.93 When the Volstead Act

went into effect in 1919, bottled sodas competed directly with bottled fruit juices, and

cost a fraction of what fruit juice cost to manufacture.94 Both fruit juices and soda pop

claimed to be healthful, wholesome alternatives to spirits.

91
As Andrew Smith explains, fermentation occurred more or less spontaneously in the
temperate American climate. Apple juice became cider, or was distilled as applejack;
pear juice was enjoyed as perry; peach juice formed a cider known as mobby; grape juice
was the raw material for wine and brandies. These alcoholic beverages were produced at
home as well as commercially, and were a means of preservation as much as a source of
intoxicating pleasures. Although it was common knowledge that fermentation could be
stopped or prevented by boiling the juice, there was little apparent interest in or market
for “unfermented wine” and other non-alcoholic fruit beverages, except among a few
religious communities. Smith 2013: 142-3.
92
Smith 2013: 142-3.
93
See Elmore 2015: 111-34.
94
Smith 2013: 143.
124
 Yet although Sethness, Hurty-Peck, and others boasted that their flavorings were

“pure” products of the grape, they also called attention to the differences between their

flavoring extracts and syrups made from grape juice. Twitchell’s Imperial Grape Flavor,

for instance, boasted that it allowed bottlers to produce a carbonated drink “that has the

delightful taste and delicious flavor of Freshly Pressed Grape Juice” without the “cooked

taste” that developed when grape juice was treated to prevent fermentation.95 Sethness

advertised itself as “makers of the grape that keeps,” in contrast to the juice-based syrups

that readily fermented, produced sediment, and altered in color or flavor if kept too long

or under the wrong conditions.96 Crucially, none of these products were flavoring syrups;

these companies were selling a different kind of product — an unsweetened concentrated

flavor. A series of Hurty-Peck advertisements that ran in the 1914 edition of the monthly

trade journal for drugstore operators, The Pharmaceutical Era, explained the value

proposition. Instead of buying prepared fruit syrup at the beginning of the season, and

carrying the risk and burden of that perishable investment, “by the Hurty-Peck method,

you buy only the Real Fruit Flavors, and get your sugar and water as you need it,” adding

the flavor to the syrup as demand required. Thirty dollars’ of Hurty-Peck’s Real Fruit

Flavors could make the equivalent of $200 of “Old Style Prepared Syrup. Think of the

work the $170 saved could do in capable hands!” the advertisement urged.97

95
S. Twitchell Co., “Our Leap Year Proposal: Imperial Grape Flavor,” advertisement
American Bottler 33 (January 1913): 48.
96
Sethness Company, “Yes, Sir, We’re the People Who Put Grape in Concord Grape
Soda,” advertisement, American Bottler 33 (1913): 21.
97
Hurty-Peck & Co., “Turn Your Money Over,” advertisement, Pharmaceutical Era,
July 1914: 41.
125
 The use of synthetic chemicals is not disclosed in any of these advertisements for

grape flavorings dating from the 1910s; no mention is made of whether the products

should be labeled “imitation.” Indeed, quite the opposite. Sethness assured potential

customers that they are “the people who put grape in Concord grape soda,” and that theirs

is “the only absolutely true fruit extract.”98 Hurty-Peck insisted that its “Real Fruit

Extracts contain nothing but the extractive matter of Sound Ripe Fruit without any

additions whatsoever, either for flavoring or coloring.”99 These flavorings were being

sold as “true” fruit flavors, producing carbonated beverages that did not need to be

labeled “imitation.”

By 1919, officials at the Bureau of Chemistry were aware that methyl anthranilate

was being used to produce grape flavors, and commissioned Frederick B. Power, in the

Bureau’s Phytochemical Laboratory, to investigate methods for detecting the chemical’s

presence.100 Adapting techniques developed in the essential oil industry, Power in 1921

outlined a set of steps to determine whether the molecule was present in fruit juice using

beta-naphthol as a reagent.101 Across the country, state regulatory chemists began to test

commercial grape juices as well as grape-flavored sodas and flavoring extracts. When

they found that many juices contained methyl anthranilate, some concluded that the

chemical had been added, and that the juices were misbranded and adulterated. “In

98
Sethness 1913: 21.
99
Hurty-Peck & Co., “You Can’t Reach Big Profits Tied Down By Old Methods,”
advertisement, Pharmaceutical Era, May 1914: 40.
100
Power had previously worked as a chemist at the flavor and fragrance manufacturer
Fritzsche Brothers.
101
Frederick B. Power, “The Detection of Methyl Anthranilate in Fruit Juices,” Journal
of the American Chemical Society 43.2 (1921): 377-81.
126
consequence of these deductions,” Power and his colleague Victor K. Chesnut wrote later

that year, “it has naturally become of much importance to determine whether a pure and

entirely unsophisticated grape juice may not contain small amounts of methyl

anthranilate.”102 In other words, was the presence of methyl anthranilate in commercial

grape juices evidence of adulteration, or was the chemical compound already in the

grapes themselves, just as it had been shown to be present in neroli blossoms and other

floral oils? Power and Chesnut tested a number of grape juices they made in the

laboratory from different varieties of grapes provided by the USDA’s Bureau of Plant

Industry. “The observations that have thus far been made enable us to conclude that

methyl anthranilate is a natural and apparently constant constituent of grape juice,” with

the dark purple juices of Concord grapes richest in the compound. They published their

preliminary results “in order that those engaged in the examination or control of

commercial products may not be led to wrong conclusions respecting their purity.”103

Subsequent research showed that grape varietals of the native Vitis labrusca tended to

contain methyl anthranilate, while European grapes, Vitis vinifera, most often did not.104

USDA researchers also took methyl anthranilate as a proxy for quality, measuring the

quantity of the compound present in grape juice prepared by different methods, and

102
F.B. Power and V.K. Chesnut, “The Occurrence of Methyl Anthranilate in Grape
Juice,” Journal of the American Chemical Society 43.7 (1921): 1741-2.
103
Power and Chesnut “The Occurrence of Methyl Anthranilate in Grape Juice” 1921:
1741-2.
104
Frederick B. Power and Vincent K. Chesnut, “Examination of Authentic Grape Juices
for Methyl Anthranilate,” Journal of Agricultural Research 23.1 (1923): 47-53.
127
noting that its diminishing quantities after storage could explain the deterioration in

flavor in some commercial bottled juices.105

It would seem that Hurty’s recognition of the grapiness of his fellow rider’s

perfume was more than coincidental; the chemical that scented orange blossoms also lent

its aromatic qualities to the Concord, the Scuppernong, and the other foxy Vitis labrusca

varietals of North America. The availability of synthetic methyl anthranilate for grape

flavoring was due to its use in a different sensory and commercial context: floral fashion

perfumery. Methyl anthranilate’s presence in New World V. labrusca, but not in Old

World V. vinifera grapes, may be why the European essential oil and aromatic chemical

supply houses that initially manufactured the chemical did not seize upon the

resemblance and advertise their product for uses in grape flavorings. (Its trade name in

essential oil catalogs was generally synthetic or artificial oil of neroli.)

This is not to say that methyl anthranilate was, naturally and self-evidently, grape

flavor. As noted, not all grapes contain methyl anthranilate, but all grapes do contain

many other substances that contribute to their particular flavor and aroma. Methyl

anthranilate became grape flavor not only by its presence in grapes themselves, but by its

repeated and continued use in grape flavorings, in alliance with other substances, such as

tartaric acid, sugar, and, notably, purple coloring, that were made to signify and reinforce

the sensation of grape-ness, at a moment when Americans were beginning to consume an

expanded variety of grape-flavored things.

105
J.W. Sale and J.B. Wilson, “Distribution of Volatile Flavor in Grapes and Grape
Juices,” Journal of Agricultural Research 33.4 (1926): 301-310.
128
 So if both “true” grape juice and its synthetic imitation contained methyl

anthranilate, then how could regulators distinguish the genuine thing from the pretenders?

The answer they found was to measure the quantity of the chemical present in the

product. Grape juices rarely contained more than two parts per million of the chemical,

and the concentration of the compound decreased significantly during storage.106 Grape-

flavored soda pops contained many times more methyl anthranilate than was found in

even the freshest juices. One state health official in 1923 detected concentrations between

seven and 17.5ppm in four commercial bottled sodas.107 Regulatory chemists developed

normative standards based on quantities of methyl anthranilate detected in grape juices,

and used these calculations to distinguish “pure” from the of the enhancements of the

“imitation.”

Armed with a standard method of quantifying methyl anthranilate, regulators in

the 1920s took action against a number of manufacturers of grape flavorings and grape-

flavored beverages. Sethness, the company which had once touted itself as “the People

who Put Grape in Concord Grape Soda,” plead guilty in 1924 after Bureau of Chemistry

agents found that its Cosco Grape Soda Water Flavor “was an imitation grape flavor,

most of the flavor of which was due to methyl anthranilate,” and contained “little, if any,

grape juice.”108 Hurty-Peck, which had once claimed on its label that its “Superb Brand

106
Sale and Wilson 1926.
107
R.D. Scott, “Methyl Anthranilate in Grape Beverages and Flavors,” Industrial and
Engineering Chemistry: 15.7 (July 1923): 732-3. Scott was a chemist with the Ohio State
Department of Health.
108
U.S. v. Sethness Co. U.S. District Court, Northern District of Illinois, 1924. F.&D.
No. 18576. Adulteration and misbranding of Concord Grape soda water flavor. Plea of
guilty. Fine: $100.
129
True Concord Grape Soda Water Flavor” contained “no artificial flavor,” did not contest

the charges of adulteration and misbranding. A default judgment was entered against the

company, and the thirty-five gallons of flavoring that had been seized were destroyed by

government agents.109 Other companies who were similarly charged generally plead

guilty and paid a fine, or did not contest the charges.110

Indeed, J.W. Sale, the chemist in charge of the Bureau of Chemistry’s Water and

Beverage Laboratory, said in 1924 that enforcement actions had convinced him that

“there are no true grape flavors for bottlers’ use on the market, although there may be

several that are alleged to be of this type.” As far as he had been able to discern, all of the

“so-called grape flavors” on the market were in fact mixtures of grape wine (which

109
U.S. v. 35 Gallons of Superb Brand true Concord Grape Soda Water Flavor. U.S.
District Court, Eastern District of Wisconsin, 1924. F&D No. 18813.
110
See, for instance: U.S. v. 69 Barrels of Grapico Syrup. U.S. District Court, Northern
District of Alabama, 1923. F&D No 17361. After an interstate shipment of Grapico syrup
(“Deliciously Refreshing Grapico Naturally Good Syrup”) was seized by federal agents
in Alabama, J. Grossman’s Sons, the New Orleans company that manufactured the grape
flavoring, pled guilty to adulteration and misbranding, and paid a bond of $4,000 to have
their merchandise released, on the condition that they change the label to read: “Imitation
Grape Syrup Grapico Naturally Good Syrup. Contains Pure Grape Flavor, Artificial
Flavor and Color.” The 1925 case involving the Orange Smash Company, makers of
Grape Nip Concentrate, was an exception to the general pattern of pleading guilty or
failing to contest the charges after government agents had seized allegedly adulterated
and misbranded flavorings. In 1923, federal agents in Maryland seized a quantity of
Grape Nip Concentrate that had been shipped from the company’s headquarters in
Alabama. The label claimed that Grape Nip contained “extract of Ripe Grapes Sugar and
Water & Tartaric Acid [sic],” but analysis by the Bureau of Chemistry “showed that it
was an imitation grape sirup composed in part of sugar, glycerin, and water, artificially
colored with a coal-tar dye, and flavored with methyl anthranilate.” Orange Smash
contested the charges of adulteration and misbranding, and the case came to trial before a
jury, which ultimately found the company guilty. The court imposed a fine of $100. U.S.
v. Orange Smash Co. U.S. District Court, Northern District of Alabama, 1925. F&D No.
19252.
130
provided the alcoholic menstruum for the flavor), methyl anthranilate, and other flavoring

chemicals.111

The use of methyl anthranilate in grape-flavored things was not so much a

reproduction of grape, then, but an intensification —an intensification that had affective

consequences for consumers.

“When a Better Grape Drink is Made, NuGrape Will Make It”

This, then, was the context for the 1925 regulatory action against NuGrape.112

NuGrape was one of several manufacturers of grape-flavored beverages that were forced

to disclose imitation status — although, as appears clear from Sales’s statement, it was

not possible to make a commercially viable grape-flavored soft drink without the use of

synthetic chemical additives.

Nonetheless, the resources of regulators were never sufficient to penalize all

wrongdoers, and it seems evident that some makers of grape sodas continued to falsely

pass their products as statutorily “pure.” In a full-page “Open Letter to the Trade,”

published in the June 1927 issue of the Beverage Journal, NuGrape positioned itself as

an industry leader in “faithful and fair compliance with government rulings,” and railed

111
Sale 1927: 269.
112
The prosecuting agency here was the Federal Trade Commission rather than the
Bureau of Chemistry. The FTC was charged with preventing unfair competition for
goods in interstate commerce, but only took action after a complaint was made. (The
complainant remained anonymous.) As far as I can tell, the two agencies often worked
together in cases regarding the enforcement of the Pure Food Law, and used the same
chemical and commercial findings as evidence. “Imitation Flavors Must Be Designated
as Such,” The Beverage Journal 63.4 (April 1927): 47-8.
131
against competitors who have “attempted to make capital of our action and to hurt our

product by falsely claiming that their product did not have to be labeled ‘Imitation.’”113

Promising a “showdown,” NuGrape challenged manufacturers and distributors of grape

syrups and concentrates to “TELL THE TRUTH ABOUT THEIR PRODUCTS AND

COMPLY WITH GOVERNMENT REGULATIONS,” and the government to compel

compliance “IN FAIRNESS TO US AND THE PUBLIC.” NuGrape, the letter alleged,

“virtually created a market for grape flavored beverages throughout the United States.

Other grape drinks have come and gone, and what temporary popularity they had was to a

large extent due to NuGrape advertising and distribution.” NuGrape had earned its

popularity because of its quality, flavor, and its other investments in building a market;

dishonest competitors were capitalizing on these investments, and making a play for

consumers’ favor not by appealing to their senses (by producing a higher-quality grape

drink), but to their biases against ‘imitation’ products. Although the letter did not directly

criticize the regulations themselves, it undercut them by presenting them as a

disproportionate response: “NuGrape is made from real grape wine and grape products,

with less than one-tenth of one percent artificial flavor, but on account of even this small

percentage of artificial flavor” they were forced to bear the “Imitation” stigma.

Paraphrasing Buick’s then-famous slogan, NuGrape closed the letter with a vow: “when a

better grape drink is made NuGrape will make it.”114

113
NuGrape Company of America, “An Open Letter to the Trade,” advertisement,
Beverage Journal 63.6 (June 1927): 59.
114
Thanks to Anne Boyd for pointing out the source of this slogan.
132
 What was a better grape drink? Would it still be ‘imitation’? NuGrape’s open

letter presented a fair playing field as one where all options were ‘imitation,’ and thus one

where grape sodas would compete on the sensual flavor experience delivered by the

contents of the bottle rather than the false impressions of quality conveyed by its label.

As a Beverage Journal advertisement for a grape flavoring made by the Fonyo Chemical

Laboratories of Chicago put it later that same year: “The Goodness of a Grape Drink

Depends on the Quality and Distinction of the Imitation Flavor.”115 May the best

‘imitation’ win.

But the stigma of “imitation grape” was apparently so acute that in 1929,

NuGrape changed the formula for “the flavor you can’t forget” to evade that designation.

That year, advertisements announced “the Supreme Triumph of the Makers of Nugrape!”

one which “marks the final victory of science over the ancient King of all Fruit Juices…

King Grape Juice.”116

The copy continued:

“World famous chemists have been telling us for years it couldn’t be

done… the difficulties were too great! Fermentation… price of grape
juice… variation in flavor — all these things they said made it impossible
to produce an exquisite, carbonated soda with the flavor of the grape and
sell it for 5c. Nevertheless we’ve done it — by creating and perfecting a
secret new process of concentrating grape juice.”

115
Fonyo Chemical Laboratories, “The Goodness of a Grape Drink…,” advertisement
Beverage Journal (September 1927): 2
116
NuGrape Company, “The Supreme Triumph of the Makers of NuGrape!”
advertisement, [Jackson, MS] Clarion-Ledger, (January 6, 1929): 19.
133
 The advertisement went on to note that, in addition to concentrated grape juice,

NuGrape’s other ingredients were tartaric fruit acid (“which itself is a by-product of

grapes”), pure cane sugar, carbonated water, and “harmless U.S. Government certified

food color, such as is used in making candies, ice-cream and hundreds of other

wholesome food products. These ingredients and no other give NuGrape its wonderful

flavor of the grape and appearance.” This “supreme triumph” meant that NuGrape

bottles, labels, and advertisements were no longer emblazoned with “imitation grape.”

NuGrape was again on the right side of nature.

This restoration was short-lived. Regulatory officials again challenged NuGrape’s

labeling; this time, the company resisted and took the case to court. The subsequent trial

record revealed much about the process of making NuGrape.117 Including this: it was not

the NuGrape Company of America that had created the “secret new process,” but

Fritzsche Brothers, the flavor and fragrance company in New York. Thirty-nine of the

forty gallons of NuGrape syrup were water, sugar, tartaric acid, and certified coloring; the

final, and crucial, gallon was “Merchandise No. 25.”118 What was Merchandise No. 25?

This was “Fritsboro True Grape Aromatics, New Process,” purchased from Fritzsche

Brothers. The base of this was a four-fold grape juice concentrate from California.119 In

117
U.S. Federal Trade Commission Decisions. In the Matter of NuGrape Company of
America. Docket 1576. Complaint, Feb 27 1929; Decision, May 19, 1931. 15 FTC.
118
15 FTC In the Matter of NuGrape… (1931): 118.
119
To make a four-fold concentrate, four gallons of vacuum-distilled juice were reduced
to one gallon of concentrate. Achieving further concentration was technically difficult for
manufacturers at this time; not only did the concentration process risk unfavorably
134
order to achieve an eight-fold concentration — a concentration that would be viable for

use in bottled sodas, but which was technically extremely difficult to produce without

altering or losing flavor — the company testified: “we add aromatic grape concentrate

made from grapes by our own secret process.” The company refused to provide any

additional information to the investigators about this “aromatic grape concentrate,” on the

grounds that these were trade secrets.120 In other words, Fritzsche claimed that

Merchandise No. 25 was a mixture of highly concentrated grape juices. This meant that it

met the USDA’s criteria that it be “derived wholly and without chemical change from

grapes or grape juice,” and so was “entitled to be labeled ‘grape flavor.’”121

However, an analysis performed by USDA regulatory chemists in the Spring of

1930, cited as evidence in the trial, cast doubt on the Fritzsche’s claim that their secret

process used only grapes. “Exhaustive analyses” made by USDA chemists of

Merchandise No. 25, NuGrape Syrup, and NuGrape soda proved that Merchandise No.

25 “is so changed by the removal of certain solids” such as fruit sugars and acids, and by

the addition of alcohol, that “it has ceased to be a pure concentrated grape juice and has

become a grape extract.” NuGrape syrup contained less than 4 per cent grape juice, and

they found it “does not contain the natural fruit or juice of the grape in quantities

sufficient to give it its color or flavor.” NuGrape soda “derives both its color and its

flavor chiefly and substantially” from artificial color and tartaric acid, “both of said

ingredients being added by respondent to Merchandise No. 25 in the production of

altering or losing the volatile flavor compounds in the juice, the sugar and solid contents
of the juice also posed challenges. 15 FTC In the Matter of NuGrape… (1931): 118.
120
15 FTC In the Matter of NuGrape… (1931): 118.
121
Sale 1924: 270.
135
NuGrape syrup.” Tartaric acid was not found in grapes or grape juices, but obtained from

“crude argols, commonly called wine lees, by-products, or precipitates, obtained in the

treatment of grape juice or the manufacture of wine.”122 In other words, even if it was not

found in grapes or grape juices, tartaric acid was, in a literal sense, a “grape product.”

(Indeed, NuGrape argued this point.123)

In the eyes of regulators, however, there was too much distance between grapes

and tartaric acid; what was grape about the grape had been transubstantiated, turned into

a chemical. NuGrape, artificially colored, flavored with materials once derived from

grapes but grapes no longer, is Imitation. The FTC's ruling, handed down in 1931,

required the company to change their labeling and marketing to reflect that the product

"is an imitation, artificially colored and flavored."

What underlies this chemical judgment is a value judgment: that the flavoring

chemical was made, essentially, from garbage — from the wastes of other industries.

Although it dates from a decade later, this October 29, 1941 letter from P.B. Dunbar,

assistant commissioner of Food & Drugs, to the chief of the central regulatory district,

substantially reflects the agency's attitude and policy toward flavoring additives:

122
15 FTC In the Matter of NuGrape… (1931): 118-119. The origins of tartaric acid are
obscurely commemorated on the label of containers of cream of tartar, which often
feature a wooden wine barrel.
123
Before changing their formula, for instance, NuGrape claimed that its flavor was
made: “from real grape wine and grape products with less than one tenth of one percent
artificial flavor.” “An Open Letter to the Trade” 1927: 59.
136
 "Heretofore on products of vague identity offered to food manufacturers
we have felt that the requirement for the labeling of the ingredients by
their most informative names was a means by which the buyer could
determine the worth, if any, of these often glorified addition substances. In
other words, the mere recitation that the product is a few cheap chemicals
and water takes out all the mystery."124

The "products of vague identity" are the flavor additives produced by flavor and

fragrance companies. By requiring flavor additive manufacturers to reveal their

ingredients, regulators at the Food and Drug Administration wanted to demystify these

"glorified" and overvalued additives. For Dunbar and his colleagues at the agency,

flavoring additives were not innovative products developed by skilled workers, but "a

few cheap chemicals and water." Underlying this was a more profound anxiety: that

consumers would not be able to tell the difference between — for instance — grape and

NuGrape unless "Imitation" was prominently branded on the label. But, if there was a

world of difference between the pastoral orchard and the chemical leached from the lees,

then shouldn't that difference reveal itself at first sip? If the distinction between "real" and

"fake" is somehow no longer self-evident, then what were the prospects for the continued

persistence of the real?

In 1932, the year after the FTC’s ruling, NuGrape once again tried to get on the

right side of nature, this time partnering with another brand name to deliver a “real grape

124
P.B. Dunbar, Assistant Commissioner of Food and Drugs, to Chief of Central District,
October 29, 1941; Butter, Butterscotch etc. Flavors; Office of Nutritional Products,
Labeling, and Dietary Supplements, Center for Food Science and Nutrition; Record
Group 88, Records of the Food and Drug Administration; US National Archives, College
Park, MD.
137
drink… deriving its entire flavor and color from Welch’s Grape Juice.”125 The new, new

NuGrape was again touted as a scientific triumph. “After years of expensive research our

labors are rewarded,” read one advertisement, which repeated the identical language used

in 1929, proclaiming a “final victory of science over the ancient King of all Fruit

Juices.”126 Naturalness, then, was finally achieved, but only through intensive scientific

labor and technological innovation. Naturalness was not a return to the once-familiar, but

a new kind of novelty: “never before has there been a drink like this introduced to the

American public.”127

But the new NuGrape did not last. It’s difficult to know exactly what happened;

one account, provided by the son of a local bottling company owner, recalled that the

grape juice fermented in the bottle, destroying the product.128 In 1933, NuGrape

advertisements all but vanished from newspapers; the once-heavily advertised beverage

would not be widely touted until the mid-1950s. Bankruptcy announcements appeared for

NuGrape regional plants in Louisville, Kentucky and Charleston, West Virginia; bottling

machinery, trucks, bottles, and cases were auctioned off.129 The economic depression,

and the repeal of prohibition, clearly had consequences for the bottled soda market, but it

125
NuGrape Company, “Enjoy a Real Grape Drink,” advertisement, Whitewright [TX]
Sun, (May 19, 1932): 8.
126
NuGrape Company, “The Triumph of Beverage Perfection,” advertisement, Kokomo
[IN] Tribune, (June 17, 1932): 32.
127
“The Triumph of Beverage Perfection” 1932.
128
Bill Baab, “TIP Baby’s Father Shares Gary Beverage Co. History,” The Federation of
Historical Bottle Collectors Newsletter (July-Aug 2007): 21. Available at:
http://www.fohbc.org/PDF_Files/GarysBev_BBaab.pdf
129
“Legal Notice,” Charleston [WV] Daily Mail (January 13, 1933): 17; “At Auction,”
[Louisville, KY] Courier Journal (July 16, 1933): 35.
138
is also likely that the repeated pursuits of nature, its attempts to avoid “imitation,”

contributed to NuGrape’s tumble.

The Flavor You Can’t Forget

The NuGrape Twins’ recorded output is tiny: four songs in praise of the Lord, two

in praise of NuGrape.

I got a Nugrape nice and fine

Three rings around the bottle is a-genuine

I got your ice-cold Nugrape

Like NuGrape, the NuGrape Twins hailed from Georgia. But while NuGrape

came into the world in urban Atlanta at the outset of the booming 1920s, Mark and

Matthew Little were born in 1888, in Tennille, a railway stop approximately halfway

between the state capital and Savannah. NuGrape’s rise in the world was much steeper

and swifter than that of the two African American brothers, of whom little is now

known.130 “I’ve Got Your Ice-Cold NuGrape” — recorded in 1926, when the purple

drink’s territory was spreading beyond the borders of the Southern states — reflected the

130
Mark and Matthew Little are recorded, along with their mother Low Little, in the
“United States Census, 1900," database with images, FamilySearch
(https://familysearch.org/ark:/61903/1:1:M3J8-JXW : accessed 17 November 2016).
139
ways that the meanings and powers that bubbled up in this new carbonated sensation

intersected with the daily lives of growing numbers of consumers, reshaping the contours

of sensory and affective experience.

The song is, according to the All Music Guide to the Blues, “a simultaneous hymn

and jingle that advertises the soda as a cure for any earthly or spiritual ailment.”131 One

twin sings in a tinny, determined countertenor, which, at moments, thins to wispiness; the

other provides a shuffling baritone accompaniment, sometimes lagging a beat behind. In

a series of comic verses delivered in a plaintive, sing-songy cadence, the twins described

NuGrape as a tonic that could lift depressed spirits:

When you're feeling kinda blue

Do not know what's ailing you

Get a NuGrape from the store

Then you'll have the blues no more

Pacify the rage of a termagant wife:

If from work you come home late

Smile and 'prise her with NuGrape

131
Burgin Mathews, “Sinners and Saints (1926-1931)/Document,” in Vladimir
Bogdanov, Chris Woodstra, and Stephen Thomas Erlewine, eds., All Music Guide to The
Blues: The Definitive Guide to the Blues, 3rd ed., (San Francisco: Backbeat Books, 2003):
672.
140
 Then you'll sneak through in good shape

Or serve as a love-charm in courtship, a token of ardor otherwise inexpressible:

Sister Mary has a beau

Says he crazy loves her so

Buys a NuGrape every day

Know he's bound to win that way

The seductive, spiritual power of NuGrape derived from the incommensurable

pleasure it produced, a sensation that emerged from its ice-cold temperature — a

differential effect with the summer heat, that, in 1926, would have only recently become

technologically possible for leagues of parched Southern drinkers — from its sweetness,

and, especially, from its distinctive flavor. “A Flavor You Can’t Forget,” was NuGrape’s

slogan, emblazoned on the crimped metal caps of NuGrape bottles, repeated in

advertisements, wall-hangers, and other promotional merchandise.

But the most important lesson of the NuGrape Twins’ song is that only genuine

NuGrape had these powers:

I got a NuGrape nice and fine

Got plenty imitation but there’s none like mine

141
 I got your ice-cold NuGrape

Historians have convincingly argued that brands originated as a means of

overcoming suspicions about canned and packaged foods, and of gaining and sustaining

consumer trust in products that could not be directly examined.132 Through advertising

and other promotional activities, manufacturers such as Heinz and the National Biscuit

Company established direct relationships with consumers. As consumers became more

confident in the safety and reliability of the food supply, brands became invested with

other meanings and values. Advertising and design were powerful technics for creating

needs, lubricating the gears of the mass consumer economy by continually renewing and

replenishing the sources of desire. (Indeed, some early twentieth century advertising

professionals sought to underscore their role in these economic processes by calling

themselves “consumption engineers.”)133 By the 1920s, advertisers accomplished this by

adopting methods and insights from the social sciences and psychology in order to study

consumers themselves — investigating their habits, surveying their preferences, and,

increasingly, probing their motivations.134 Similarly, the design fields and art industries

became affiliated with psychology and other social sciences as they professionalized, in

order to develop solutions that encouraged productive consumption and the “smooth

132
Koehn 2001; Strasser 1989.
133
Marchand 1985: 26.
134
Lawrence R. Samuel, Freud on Madison Avenue: Motivational Research and
Subliminal Advertising in America, (Philadelphia: UPenn Press, 2011).
142
flow” of economic activity.135 Advertising and design worked on the largest possible

scale — messages disseminated through mass media or broadcast over radio-waves,

colors, forms, and features stamped into mass-produced and mass-distributed goods —

but their effects were meant to be intimately felt.

What role did flavor play in this system of stoking and provoking desire? A

product’s flavor comprises part of what David Howes has called the sign-value of a

commodity, and is an experiential index to the system of sensory and social relations in

which it is embedded.136 In other words, a consumer’s relation to and appreciation of

flavor involves both her or his direct sensory experience and also the web of social,

cultural, political, and historical circumstances, through which the flavor’s meaning and

its value are construed, at that moment, for that taster. When food and beverage

manufacturers such as NuGrape began deliberately designing the flavors of their

products, controlling their material constituents and concomitant sensory effects, they

simultaneously sought to shape their meanings.

Just as food companies, and their associated brands, used advertising to build

direct relationships with consumers, they used flavor to cement those relationships. Even

as regulators prosecuted NuGrape for failing to inform consumers that their product was

merely an imitation of grape, NuGrape touted the distinctiveness and originality of the

135
Regina Lee Blaszczyk, The Color Revolution (Cambridge: MIT Press, 2012);
Christina Cogdell, Eugenic Design: Streamlining America in the 1930s, (Philadelphia:
UPenn Press, 2010).
136
Davis Howes, “Marx’s Skin,” accessed at: http://www.david-
howes.com/senses/marxsskin.html. Adapted from David Howes, Sensual Relations:
Engaging the Senses in Culture and Social Theory (Ann Arbor, MI: University of
Michigan Press, 2003).
143
flavor of its beverage. The primary goal of the makers of NuGrape was not for its flavor

to be mistaken for that of Concord grapes. It was for it to be recognized, remembered as

NuGrape. (In this regard, the “three rings around the bottle” might be taken as an

indication of the company’s lack of full confidence in the flavor alone to do this.)

To be clear, none of these things are necessarily more grandiose or remarkable

than what foods could do to bodies in the early modern era, when food could treat and

cure diseases, temper imbalanced humors, and recalibrate one's relationship with the

actual cosmos.137

In the final accounting, however, there is something heavenly about NuGrape. "Is

there no change of death in paradise?" asked Wallace Stevens. "Does ripe fruit never

fall?" "Heaven is a place where nothing ever happens," according to the Talking Heads.

For NuGrape to become "the flavor you can't forget," it must conform itself not to

the flavor of grapes hanging heavy on the bough, but to prior memories of NuGrape. To

the bodily, social, and spiritual array of pleasures, comforts, and gratifications that

affiliate themselves with the sensations that NuGrape provides. Like the unchanging

fruits of heaven, NuGrape must always resemble itself.

All the way from Maine to the Gulf of Mexico

From the Atlantic to the calm Pacific shore

NuGrape is the best friend yet

137
Steven Shapin, “Changing Tastes: How Things Tasted in the Early Modern Period and
How They Taste Now,” Hans Rausling Lecture 2011, Uppsala University.
144
So try a bottle of NuGrape

The flavor you can't forget

145
 Chapter 3
Assembling the Human Instrument: Taste
Panels, Flavor Measurement, and the
Origins of Sensory Science

On Friday, November 12, 1937, listeners tuning in to "Housekeepers' Chat" — a

weekday radio segment produced by the US Department of Agriculture’s Bureau of

Home Economics —were given a glimpse of the agency’s methods for providing reliable

information about food to the public. Usually, these fifteen-minute weekday segments

offered recipes from "Aunt Sammy," household tips, and nutrition and family health

advice for the effective, scientific housewife.1 On this day, listeners were assured that

even at the Bureau's food laboratories, there were still some tasks that had not been

mechanized: "even modern science with all it's [sic] labor-saving machinery hasn't

devised a robot that tastes and smells." When it came to evaluating the flavor of food, the

1. Morleen Getz Rouse, "Daytime Radio Programming for the Homemaker: 1926-1956,"
Journal of Popular Culture 12.2 (1978): 315-327. For broader context on the technical
and scientific aspects of home economics, see Amy Sue Bix, “Equipped for Life:
Gendered Technical training and Consumerism in Home Economics, 1920-1980,”
Technology and Culture 43.4 (October 2002): 728-54, and Carolyn Goldstein, Creating
Consumers: Home Economists in Twentieth-Century America, (Chapel Hill: UNC Press,
2012).
146
host announced, "no one but a human being can judge the flavor of the food human

beings eat.”2

But who were the human beings whose judgments gained the Bureau’s official

scientific imprimatur? These "taste judges" were USDA staff members who were

"regularly employed in other work," but who had demonstrated sound and consistent

judgment, as well as the ability "to analyze their own reactions to what they taste... [and]

express these reactions on the score sheet they are using." The host explained that the

Bureau of Home Economics routinely assembled panels of these specially chosen tasters

to scientifically evaluate how changes in production methods affected the quality of

different foods, including meat, bread, cakes, canned goods, and dairy products. Tasters,

however, were never allowed to know the details of the experiments they contributed

their sensory capabilities to: "if they did, it might possibly influence their judgment,"

undermining the evidentiary validity of their conclusions. "In all of these experiments,"

the chat’s host concluded, "the opinions of taste-testers are really important. Because

flavor and aroma are two of the biggest items in food quality, and so far there is no other

way to judge them."

This chapter considers the consequences of this proposition: "no one but a human

being can judge the flavor of the food human beings eat." With the increasing

industrialization of food production in the 1930s, flavor became an object of scientific

2. "How Does It Taste?" USDA Housekeeper's Chat, Bureau of Home Economics.

Broadcast date: Friday, November 12, 1937, 1. Emphasis in original.
147
and technical concern for food manufacturers and the federal government. Researchers

from multiple disciplines – including chemistry, agriculture, physiology, psychology,

home economics, and food technology – working in different institutional settings,

harboring a variety of motives, found it necessary to develop standard tools capable of

measuring the “organoleptic” qualities of foods: quantifying the various sensory

dimensions of flavor experience, including taste, aroma, texture, and appearance.

The "taste panel" — a small group of individuals (trained, tested, but, crucially

not “experts”) producing sensory judgements in specialized settings under controlled

conditions — first appeared in research and industrial laboratories in the 1930s. By 1950,

the taste panel had become the primary research tool within both government and

industry to measure, compare, and evaluate the sensory qualities of foods, including

flavor, texture, and visual appeal. Rather than a transitory record of subjective individual

preferences, taste panels were expected to produce reliable, stable, and reproducible

information about food’s sensory qualities — the type of data that could be evaluated

alongside, and correlated with, other, new instrumental measurements and determinations

of the chemical and physical properties of foods.

By examining the practices of organoleptic research in their specificities — the

formally articulated methods, associated technologies, social structures, and desired

outcomes — this chapter tracks a major change in the scientific study of flavor during the

1930s and 1940s. As food-science research increasingly revealed the physicochemical

components responsible for the qualities of foods, the sensory aspects of flavor also

became the subjects of systematic study and investigation. The interwar and wartime
148
years marked the convergence of these two modes of research into flavor, which was

increasingly studied in the context of food manufacturing and the sensory changes that

occurred during production. The development of the taste panel as a laboratory

instrument shows us not only how flavor was made into a scientific object during this

period, but also marks the formation of a model of the subjective, tasting self that could

be incorporated into rationalized, industrialized processes of product development and

design.3

The first part of this chapter traces the early days of laboratory taste panels,

beginning in the 1930s until the Second World War. I show how this instrument

developed at the convergence of multiple research programs and needs, shaped by, but

distinct from, both traditional practices of "expert tasters" employed in assessing or

grading specific commodities, and from new polling and statistical sampling methods

from market research.

The second part of this chapter looks at how sensory evaluation, and laboratory

taste panels, rose to prominence in the context of army research at the Quartermaster

Food and Container Institute. The seminal status of the Quartermaster Institute has been

reinforced by historical accounts from some of the scientists who participated in

developing this field — including several who worked at the institute's laboratories in

Chicago, and later, Natick, Massachusetts. These accounts often tend to dismiss work

3. Steven Shapin, "Sciences of Subjectivity," Social Studies of Science 42, no 2 (2012):

170-184.
149
done in the 1930s as the pre-history of the discipline.4 Beginning this story in the 1930s

— when methodologies, research protocols, institutional settings, and disciplinary

identities were still in flux — reveals a great deal about the diverse interests that were

involved in the project of shaping the taste panel into a scientific instrument. The

Quartermaster's Food Acceptance Research Laboratory was dominated by psychologists,

and their preeminence there has in some sense foregrounded the contributions from

psychometrics and psychophysics, while minimizing the contributions from other fields.5

Taking a closer look at the work that preceded the Food Acceptance Research Laboratory

exposes the key contributions of chemists, home economists, and food technologists, not

to mention the technicians, administrative staff, factory workers, and others who

volunteered to serve on panels. Their material technologies, skills, professions, and social

arrangements laid the groundwork for the later claiming of sensory science by

psychologists.

The chapter concludes by following the path from military food research back

into civilian food production. I look at the relatively rapid acceptance of sensory

evaluation methods in industry and non-industry laboratories, and consider the

consequences both for the ways that industrialized foods are made to taste, and the ways

that consumer desires are probed, presumed, configured, and satisfied.

4
Herbert L. Meiselman and Howard G. Schutz, “History of Food Acceptance Research in
the US Army,” Appetite 40 (2003): 199-216; Howard G. Schutz, "Evolution of the
Sensory Science Discipline," Food Technology 52, no 8 (August 1998): 42-6; David R.
Peryam "Sensory Evaluation -- The Early Days," Food Technology 46, no 1 (January
1990): 86-89. A practitioner's account with a more generous view of work in the 1930s is
Rose Marie Pangborn, "Sensory Evaluation of Foods: A Look Backward and Forward,"
Food Technology (September 1964): 63-7.
5. Meiselman and Schutz 2003: 200.
150
I. Testing the Tasters: The Laboratory Taste
Panel Before World War II

During the 1930s, the laboratory taste panel emerged in relation to and in

distinction from two other contemporary methods of evaluating the sensory qualities of

foods: expert tasters and consumer research. Although it is difficult to pinpoint exactly

when the first laboratory taste panel was convened, precursors and related forms were

thick on the ground in the 1930s.6

Expert Palates and the Appetites of Ordinary Eaters

The services of expert tasters had long been called upon by manufacturers and

traders in particular foodstuffs, especially luxury goods. Tea and coffee cuppers, wine

and liquor connoisseurs, vanilla-bean graders: all of these experts assigned grades based

on ritualized organoleptic evaluations of sensory qualities, permitting the market to set

prices based on established standards of relative excellence.7 In the twentieth century, the

6. For an example of the selection and use of a tasting panel in a commercial bakery —
also the earliest use I have found of the term "tasting panel" — see H.C. Moir, "Some
Observations on the Appreciation of Flavor in Foodstuffs," Chemistry and Industry 14
(February 21, 1936): 145-8. For examples of the production of scientific information
about food's sensory qualities for the purposes of commodity research, see for instance,
W.H. Catchcart and E.J. Killen, "Scoring of Toast and Factors Which Affect Its Quality,"
Food Technology 5 (1940): 308. For an account of the sensory sciences in action, see
Larry Owens, "Engineering the Perfect Cup of Coffee: Samuel Prescott and the Sanitary
Vision at MIT," Technology and Culture 45.4 (October 2004): 795-807.
7
A contemporary ethnographic account of the relationship between evaluations of
sensory quality and price-setting can be found in: Sarah Besky, “The Future of Price:
151
evaluation of the sensory properties of foods was extended to commodities, with

wholesale markets employing trained graders to assess the quality of farm products and

assign scores based on properties including flavor, texture, and appearance.8 The 1919

Food Products Inspection Law extended this authority to the USDA, empowering

officials to assess the "quality and condition" of perishable staples such as fruits,

vegetables, butter, and poultry sold in interstate commerce.9 These trained inspectors

evaluated the sensory qualities of foods using formalized procedures and following

published guidelines, which not only dictated the conditions under which evaluations

were to take place, but also described desirable and undesirable sensory qualities, and

assigned specific penalties to the latter.

Expert tasters were not presumed to have been born with exceptional senses.

Their sensory authority was not general, but acquired, and specific to a particular type of

product. "Many professional tasters are people with only normal taste and odor

sensitivities who happened, as boys, to take jobs in tea or coffee blending plants, or

apprenticed themselves to chefs," observed Ernest Crocker, speaking at the “Flavors in

Foods” symposium, held during the 1937 meeting of the American Chemical Society.

"Long years of practice at their art has not sharpened their sensitivities to any appreciable

degree." Instead, "the art of tasting is one of learning how to concentrate on the

Communicative Infrastructures and the Financialization of Indian Tea,” Cultural

Anthropology 31.1 (February 2016): 4-29.
8
For a comparison of different butter standards in use by New York, Chicago, and San
Francisco wholesale grocer associations, see Roy C. Potts and H.F. Meyer, Marketing
Creamery Butter, USDA Bulletin 456, Washington, D.C., February 5, 1917: 17-19.
9. USDA Office of the Secretary, "Rules and Regulations of the Secretary of Agriculture
Under the Food Products Inspection Law of July 24, 1919," Circular 144, Washington
DC: Government Printing Office, 1919.
152
indications of palate and nose, and particularly of learning what to look for as the 'critical'

factor in any article with which one is working."10 Researchers at Cornell studying the

reliability of the judgments made by official milk graders likewise noted that "specialists

attain a high proficiency in the art of tasting, mainly because of a knowledge of what

signs to look for and how to interpret these signs rather than an increased sensitiveness to

stimuli."11 In other words, a taster became an expert by attending to both sensory and

social information, learning established signs of quality rather than refining his or her

own preferences. This skill was only attained after repeated experience with particular

materials in the presence of other experts.12

As the food industry became increasingly concerned with the large-scale

production of novel kinds of foods, both the practicality and validity of the “expert taster”

10. E.C. Crocker, “Measuring Food Flavors,” Food Research 2.3 (1937): 282.
11. G. Malcolm Trout and Paul F. Sharp, “The Reliability of Flavor Judgments, with
Special reference to the Oxidized Flavor of Milk, Cornell University Agricultural
Experiment Station Memoir 204 (June 1937): 40.
12
This bears a close resemblance to the sociology of tasting elaborated by Antoine
Hennion and Gevevieve Tiel. Hennion and Tiel present the taster’s acquisition of
knowledge about the qualities of the things he or she is tasting as an ongoing, reflexive,
and fundamentally social process, where particular qualities are detected, named,
contested, and confirmed by a process of “collective respondence” among a community
of tasters. An implication is that the flavors of a food or wine are “anything but pure and
natural properties” that produce pre-ordained sensory effects that can be universally
determined for all tasters in all conditions; flavors are as historically and culturally
contingent, and socially produced, as bodies themselves. Although Hennion and Tiel’s
subject is the development of communities of taste among “amateurs,” (by which is
meant enthusiasts and connoisseurs, such as audiophiles or vinophiles, rather than naïve
consumers), their observations about how knowledge about taste is produced has clear
resonances with the practices and forms of authority claimed by officially sanctioned
tasters. Genevieve Teil and Antoine Hennion, “Discovering Quality or Performing Taste?
A Sociology of the Amateur,” in Mark Harvey, Andrew McMeekin, and Alan Warde, ed.
Qualities of Food, (Manchester: Manchester UP, 2004): 19-37; Genevieve Teil, "No
Such Thing as Terroir? Objectivities and the Regimes of Existence of Objects," Science,
Technology, and Human Values 37.5 (2012):578-505.
153
approach was called into question. As Rose Marie Pangborn, one of the founders of

sensory science, wrote in a 1964 article about the history of her field: "with the growth of

food processing and the development of many new products came the realization that

there were not enough experts to cover all products, and that it might be statistically

unsound to rely on the judgment of only one or two individuals."13 Food processors had

habitually complained of what they considered the "arbitrary" and unscientific methods

of evaluation used by official food-graders, and they searched for new systems of

quantifying sensory judgments that were more exact, reliable, and generally applicable.14

The pursuit of scientific modes of determination and control over phenomena

once thought not to be susceptible to exact measurement was not unique to the food

industry; it was consonant with a turn towards rationalization, professionalization, and

technocratic authority that transformed many aspects of life in the progressive era and the

interwar years. The new science of acoustics had brought exact methods and

experimental authority to the optimal design of concert halls and the mitigation of noise

pollution.15 Color theorists, industrial designers, and consumer psychologists were

rationalizing, standardizing, and operationalizing the hues of fashions, consumer goods,

13. Pangborn 1964: 63.

14. See, for instance, L. Charles Mazzola, "Grading Food by a Descriptive Method,"
Food Industries 2 (May 1930): 214-5; and "How a Formula for Descriptive Grading Was
Developed," Food Industries 2 (August 1930): 340-44. Mazzola, an erstwhile member of
the research staff at New York Canners Inc. and current general manager of the Genessee
Jam Kitchen in upstate New York outlines a scoring method that relies on sensory
analysis, and proposes a mathematical equation that captures the accelerating decline in
perceived quality as the defects in a product increase in number or intensity — in other
words, a logarithmic scale rather than a linear scale.
15
Emily Thompson, The Soundscape of Modernity: Architectural Acoustics and the
Culture of Listening in America, 1900-1930, (Cambridge: MIT Press, 2004).
154
and architectural spaces, developing schemes of “functional color” that could move

merchandise, increase productivity, and improve well-being.16 Heating and ventilation

engineers experimentally determined measureable, enforceable standards of comfort,

encoded in automatic systems that reproduced, in numberless office buildings, the precise

atmospheric conditions deemed to be optimally pleasant to the normalized (male) body

engaged in white-collar labor.17 Similarly, food engineers and technologists sought

standard methods of measuring and controlling flavor, in hopes of one day developing

optimal standards of quality, so that the sensory qualities of food could be calibrated to

the exact register of consumer desire.

The judgments of expert tasters often failed to coincide with the preferences of

ordinary consumers.18 This meant that the evaluations of specialists were often poor

guides when it came to product development, forecasting, and market analysis.19 But how

could the preferences of “Mrs. Housewife” be ascertained? Food manufacturers needed

reliable information about the preferences and tastes of the consuming public in order to

16
Regina Lee Blaszczyk, The Color Revolution, (Cambridge: MIT Press, 2012).
17
Michelle Murphy, Sick Building Syndrome and the Problem of Uncertainty, (Durham:
Duke UP, 2006): 19-34.
18
See, for instance, Asher Hobson and Marvin A. Schaars, “Consumer Preferences for
Cheese,” University of Wisconsin Agricultural Experiment Station Research Bulletin 128
(October 1935). The experimenters found that across various groups studied — which
included grocery store customers, as well as doctors, nurses, and agriculture students
eating in university dining halls — consumers were resistant to the aged American cheese
graded highest by experts, and in some cases preferred a low-grade cheese with “an
undesirable acid flavor, open texture, and soft body” which was “distinctly objectionable
from a trade standpoint,” but which was chosen more often than higher-scoring cheeses.
The experimenters proposed that the standards of quality, which were used to grade and
price cheeses, “may… not conform wholly to consumer preferences.”
19
Washington Platt, "Rational Methods of Scoring Food Products," Food Industries 3
(March 1931): 109.
155
design and manage the sensory aspects of their products. They turned to another set of

scientific experts for guidance. As historian Sarah Igo has demonstrated, the interwar

period saw a proliferation of attempts to measure, quantify, and statistically analyze the

desires, beliefs, and behaviors of U.S. populations. Survey data served not only as a

crucible for the formation of the mass public, but also shaped private lives and lived

identities.20 Bringing together social scientists, political movements, and industrial

enterprises, consumer research claimed to close the circuit between the forces of

production and the forces of desire, offering manufacturers "measurable opinions" that

could be used to coordinate both assembly lines and advertising campaigns.

As food became a mass-market good, food manufacturers turned to market

research and consumer polling firms to establish their competitive position, guide product

development, and address lagging sales. Meatpacker Swift & Co., for instance, claimed

to have surveyed 100,000 consumers "regarding the flavor, aroma, appearance, or

tenderness of a great variety of foods, including ham, bacon, lard, shortening, butter,

cheese, sausage, meat specialties and many others."21 Controlling this treasury of

preferences meant Swift could point to deficiencies in a product that made it less pleasing

to shoppers than its neighbors in the grocery aisle. Companies and industry groups also

hired pollsters to conduct fundamental research, including large-scale surveys of factors

influencing Americans' food choices. When the American Meat Institute hired Elmo

Roper's polling firm in 1939 to investigate the causes of declining beef consumption, they

20. Sarah Igo, The Averaged American: Surveys, Citizens, and the Making of a Mass
Public, (Cambridge: Harvard UP, 2007).
21. Donald R.G. Cowan, "Developing and Improving Foods by Consumer Testing," Food
Industries 1941: 41.
156
put these findings to work not by changing their products, but by tailoring their

advertising to counter negative perceptions of red meat; they managed to reverse the

trend and increase sales.22 In some cases, large companies — most prominently, General

Foods and Kroger — skipped the middleman and did their own consumer research,

soliciting opinions on new products or advertising campaigns from housewives, and

analyzing the results.23

While food manufacturers continued to seek out and pay dearly for this kind of

direct information about the fancies and desires of Mrs. Housewife, incorporating hard-

won information about "public tastes" into production processes and product

development required the intercession of actual tasters. Tasting panels became a way for

manufacturers to apply knowledge about consumer preferences to the improvement of the

quality of food.

“A New Approach to the Subject of Flavor:” Joining

Chemistry and Psychology at the 1937 American Chemical
Society Flavors in Foods Symposium

By the second half of the 1930s, researchers were attempting to develop standard

laboratory methods that could connect the physicochemical components of foods to

distinct, measurable sensory effects, and associate those experienced effects with

attitudes and behaviors in consumers. The landmark 1937 Flavors in Foods Symposium,

22. Harvey Levenstein, Paradox of Plenty: A Social History of Eating in Modern

America, New York: Oxford UP (1993): 74.
23. Igo 2007: 112; Peryam 1990: 85; Schutz 1998: 43.
157
which took place during that year’s annual American Chemical Society meeting held at

the University of North Carolina in Chapel Hill, was the earliest scientific conference to

take flavor as its subject. The event brought together a diverse group of experts working

on problems of flavor measurement and control — not only other chemists, but also

home economists and physiologists, hailing from industry as well as agricultural

experiment stations and research laboratories. Papers addressed subjects including the

flavor chemistry of raw and cooked meat, butter, and alcoholic beverages, the use of

activated charcoal to remove off-flavors from municipal water supplies and consumer

products, and modern trends in flavoring extract production. Of the ten papers presented

at the Chapel Hill symposium, three were explicitly and primarily concerned with

techniques for measuring sensory responses to food flavors.24

The fundamental question posed by the symposium’s organizers, and engaged

with in some degree by each of the ten papers presented, concerned the epistemic and

experimental basis for a legitimate, objective science of flavor. Ernest C. Crocker and

Washington Platt, who organized the symposium, proposed that flavor science needed to

be fundamentally interdisciplinary: "A new approach to the subject of flavor consists in

attacking several of its many sides simultaneously, but especially the psychological and

24. Articles presented at the American Chemical Society Symposium on Flavors in Foods
(Chapel Hill, NC, April 12-15, 1937) were reprinted in Food Research 2.3 (1937). The
articles that took as their explicit subject methods for measuring sensory responses to
flavor are: Florance B. King, "Obtaining a Panel for Judging Flavor in Foods,"
Washington Platt, "Some Fundamental Assumptions Pertaining to the Judgment of Food
Flavors," and Ernest C. Crocker, "Measuring Food Flavors."
158
the chemical sides."25 Crocker was a pioneer of industrial flavor and odor consulting at

the Cambridge, Massachusetts consulting firm Arthur D. Little. Platt was the head of the

Borden milk company's research laboratories. Both were trained as chemists.

The challenge, as they expressed it, was to find a way to determine the

relationship between chemical presences and embodied experiences. Although this may

have been a new question for the chemists who were posing it, it was not a new problem

for psychological research. In the mid-nineteenth century, a group of researchers (based

largely in Germany) began to investigate methods of measuring and quantifying the

correspondence between objective physical stimuli and the subjective, psychic

phenomena of sensation and perception. Psychophysics, in the words of Gustav Fechner,

one of the field’s founders, proposed to develop “an exact theory of the relation of body

and mind,” one which could be expressed mathematically. While psychophysics began as

a discipline chiefly concerned with the accurate measurement of physical and sensory

magnitudes, of determining and quantifying the limits (perceptual thresholds) and

increments (just-noticeable-differences) of sensory experience, by the early twentieth

century, psychophysical practices had yielded modes of experimental psychology

increasingly concerned with producing an objective account of the qualitative, subjective

experiences of the sensing subject.26

25. E.C. Crocker and Washington Platt, "Editorial Review. Food Flavors -- A Critical
Review of Recent Literature," Food Research 2.3 (1937): 183.
26
F. Nowell Jones, “History of Psychophysics and Judgment,” in Edward C. Carterette
and Morton P. Friedman, eds. Handbook of Perception, Vol. II. Psychophysical
Judgment and Measurement, (New York: Academic Press, 1974): 2-20; Alexandra Hui,
159
 Around the turn of the twentieth century, Edward Titchener – professor of

psychology at Cornell and one of psychophysics’ most prominent American disciples –

elaborated methods of experimental introspection that could produce scientifically valid

accounts of experience. Through attentive and disciplined self-observation, and aided by

laboratory hardware that produced standardized physical stimuli, Titchener claimed that

the trained, observing self could accurately and impartially report on subjective

experience, from which the general structures of consciousness could be deduced.27

Although these methods had largely fallen out of favor among experimental

psychologists by Titchener’s death in the late 1920s, they were to enjoy a sort of

resurgence starting in the 1930s in a different disciplinary realm: as foundations for the

new field of sensory evaluation.28

But there were challenges in applying the methods of psychophysical and

psychological laboratory to flavor research. The first problem had to do with

experimentally defining the stimulus. Psychophysical research into sensory perception

most often concerned sights and sounds; rarely did it dabble in the messier world of the

The Psychophysical Ear: Musical Experiments, Experimental Sounds, 1840-1910,

(Cambridge: MIT Press, 2014).
27
Deborah J. Coon, "Standardizing the Subject: Experimental Psychologists,
Introspection, and the Quest for a Technoscientific Ideal," Technology and Culture 34.4
(October 1993): 757-783; Christopher D. Green, “Scientific Objectivity and E.B.
Titchener’s Experimental Psychology,” Isis 101. 4 (December 2010): 697-721.
28
For an intriguing account of this recovery of psychophysical methods by sensory
science, written by a participant and practitioner, see Howard R. Moskowitz, “How
Psychophysics Changed the Food Business, and How the Food Business Forever
Changed Psychophysics,” International Society for Psychophysics 25 (2009). Available
at: http://www.ispsychophysics.org/fd/index.php/proceedings/article/view/104
160
“lower” stimuli, smells and tastes.29 Auditory and visual stimuli could be represented,

reproduced, and analyzed as energetic waveforms, had agreed-upon standard units of

measurement, and scientists possessed tools that could be used to automatically produce

and measure stimuli of a given intensity. (Helmholtz, for instance, contrived ingenious

devices to reduce auditory stimuli to simple waveforms.) But what were the basic stimuli

or units of flavor sensations? Flavor, as Crocker and his colleagues at the 1937 ACS

symposium well knew, was a multisensory phenomenon. The experience of flavor

involved not only taste and smell, but also “mouthfeel”: chemical sensations (the

coolness of menthol, the pungency of mustard), as well as responses to textural qualities,

such as smoothness, graininess, and unctuousness.30 Moreover, the sensory modalities of

flavor were contested. Crocker, for his part, excluded visual and auditory sensations from

his strict account of the experiential constituents of flavor, but others did not.31

Even if the assembled chemists agreed to limit their scope to the senses of odor

and taste – which were agreed on as the dominant sensory modalities involved in flavor –

29
Edward G. Boring, Sensation and Perception in the History of Experimental
Psychology, (New York: Appleton-Century-Crofts, 1942): 437-8.
30
Crocker “Measuring Food Flavors” 1937: 273-4. Although Crocker describes the
relevant tactile sensations, the term “mouthfeel” is used by Platt. Washington Platt,
“Some Fundamental Assumptions Pertaining to the Judgement of Food Flavors,” Food
Research 2.3 (1937): 238.
31
Visual cues (especially color) were widely recognized as integral components of flavor
perception and sensation by this point. Auditory sensations as flavor factors (eg, the
sound of crunchiness) began to gain recognition later on. On the importance of color to
flavor recognition: H.C. Moir, “Some Observations on the Appreciation of Flavor in
Foodstuffs,” Chemistry and Industry 14 (February 21, 1936): 145-8. On auditory cues:
Rose Marie Pangborn, “Flavor Perception: Relation of Sensory to Instrumental
Measurements,” in DJ Tilgner and A Borys, eds. Proceedings of the 2nd International
Congress of Food Science and Technology, August 22-27, 1966, Warsaw, Poland: 303-
318.
161
defined the basic stimuli of flavor as chemical compounds, and focused their work on

correlating specific molecules with definite sensations, they still ran into trouble.32 The

human sensorium responds to complex combinations of molecules. Further, pure

compounds are rarely encountered in the world, which is filled instead with odoriferous

stews of sensible compounds, whose fluxing concentrations deliver sensory experiences

of varying qualities and intensities. As historian of chemistry Carsten Reinhardt has

described it, the scientific study of smell has been fractured by the problem of defining

the boundaries of the olfactory object.33 Should attention be directed to an analysis of

individual chemical components? Or should it instead focus on understanding the “whole

thing,” the integrated perception of a smell produced by combinations of volatile

molecules? While the latter would be more directly useful for manufacturers and others

who sought to apply this knowledge, it “does not easily enable the scientific aim of

theory building.”34 Some approached this problem by turning to an analogy with vision,

seeking primary taste or odor sensations that could be used as the building blocks of more

32
It should be pointed out that the categorization of smell and taste as the “chemical
senses” — ie, the place of chemists in this discussion — was not at all established at this
point. (It was through professional symposia, such as at the 1937 ACS meeting, that the
foundations of this claim were laid.) Chemists’ authority to turn the study of smell and
taste into an objective science was not uncontested. For instance, Boring (1942) writes:
“Although smell is always said to be one of the two chemical senses, there is no clear
evidence that chemistry will eventually provide the knowledge of the essential nature of
the olfactory stimulus. The mere fact that different substances have different smells and
also different chemical constitutions does not make a smell a chemical sense. Different
substances have likewise different colors and different chemical constitutions, and yet
color vision is for not this reason a chemical sense.” (p. 446-7).
33
Carsten Reinhardt, “The Olfactory Object: Toward a History of Smell in the 20th
Century,” in Ursula Klein and Reinhardt, eds. Objects of Chemical Inquiry, (Sagamore
Beach, MA: Science History Publications, 2014): 321-340.
34
Reinhardt 2014: 321-2.
162
complex experiences.35 Others, such as chemist Marston Bogert at Columbia University

in the late 1920s, pursued theories that linked particular molecular architectures (such as

functional groups) with discrete categories of sensations.36

These experimental and epistemological challenges were compounded by the lack

of a standard vocabulary for describing flavor sensations, especially those related to odor.

Various systems of classification had been proposed over time, ranging from descriptive

Linnaean taxonomies to experimentally derived systems, such as the olfactometrically

derived lexicon proposed at the end of the nineteenth century by Dutch physiologist

Henrik Zwaardemaker, and the spatial representation for smell developed by German

experimental psychologist Hans Henning in 1915.37 Crocker himself, with his erstwhile

35
At the symposium, Crocker described work that he and Henderson (1932) did to
reproduce a more complex taste sensation by combining taste primaries. Crocker and
Henderson attempted to duplicate the taste of monosodium glutamate through
combinations of basic solutions of sour, salty, sweet, and bitter (work they deemed rather
successful, but which would be called into question in the following decade, when
chemists at Arthur D. Little began conducting contract research for International Minerals
and Chemicals, after the company bought a factory manufacturing MSG), but say that
attempts to duplicate odors in terms of fundamentals have been “less successful.”
Crocker 1937: 188.
36
Marston Bogert and Arthur Stull, “Odor and Chemical Constitution in the
Benzoselenazole Group,” American Perfumer and Essential Oil Review 22.2 (April
1927): 63. As has been the case for much basic research on odor and taste, Bogert and
Stull conducted this research in an industrial, rather than academic, laboratory context, in
the laboratories of the American Manufacturers of Toilet Articles. This line of research
would be carried forward, most prominently by John Amoore, at the USDA Western
Regional Research Laboratory, who, beginning in the 1950s, developed a stereochemical
theory of odor that linked the architecture of molecules with specific sensory experiences,
mediated by the shape of olfactory receptor sites, which would accept some molecular
couplings but not others. See Reinhardt 2014.
37
Boring 1942: 437-449. Zwaardemaker developed and used precise quantitative tools of
olfactometry to study the human responses to different odors. Henning mapped the set of
possible human odor responses on the surface of a six-sided, three-dimensional polygon,
163
colleague Lloyd Henderson, had devised a numerical system for describing odor that

proposed to comprehensively describe each extant odor as a four-digit number, indicating

both odor qualities and intensities. The system, which became commercially available as

a kit with odor standards in the late 1940s, attracted some attention, but was never widely

used.38

In addition to the problem of defining and standardizing the stimulus, flavor

researchers had to concern themselves with the subjects, the necessary bodies that formed

the instruments of flavor measurement. How could researchers ensure and confirm that

particular bodies produced accurate, reliable knowledge about flavor, undistorted by the

“personal equation”?39 Psychophysical techniques of experimental introspection

demanded intensive training. “In order to standardize themselves as experimental

observers,” writes Deborah Coon, “psychologists resorted to long and rigorous

introspective training periods… necessary to bring all observers up to a comparable level

of expertise, a standard level of expertise. Only if introspectors themselves were

standardized could they become interchangeable parts in the production of scientific

psychological knowledge.”40 For Titchener, accurately and disinterestedly reporting

whose corners represented what Henning had determined to be the six principal
qualitative classes of odors, and whose planes indicated mixtures of those sensations.
38
For more on this, see: http://nadiaberenstein.com/blog/2014/8/25/is-there-a-dewey-
decimal-system-for-the-library-of-smells
39
Simon Schaffer, “Astronomers Mark Time: Discipline and the Personal Equation,”
Science in Context 2.1 (1988): 115-145. The “personal equation” is the technical term for
differences in time measurement recorded by observers in the same situation, which
became an object of psychophysical study and quantification in the nineteenth century.
40
Coon 1993: 775.
164
internal psychic experiences was a technical skill, one that could be acquired only with

long effort.

But the situations where flavor measurement and control were needed were

profuse and diverse, and often encountered in industrial settings, rather than the closed

chambers of the experimental psychology laboratory. Coon aptly describes the method of

introspection as an “artisanal” method. Part of the reason for its decline in experimental

psychology was that, as the discipline turned its attention to industrial problems, such as

human management and social control, it needed “industrial” methods, such as mass

studies of behavior, that could operate at scale.41 How could psychophysical methods be

adapted to the needs of food manufacturers and allied researchers, who were more

interested in determining the sensory qualities of foods than the structures of

consciousness? Moreover, the judgments of exquisitely trained experts were one of the

things that researchers in both academic and industrial contexts were trying to move

away from. What training, tools, or methods would be appropriate in the scenarios of

flavor research?

The 1937 Flavors in Food symposium at the ACS conference was not the first

time that these issues were raised, but it signified a convergence of expert attention on the

matter. In particular, the symposium organized and crystallized attempts to address these

two experimental problems: the challenge of stimulus definition and control, and the

challenge of forming human tasters into reliable instruments of sensory measurement.

The symposium also underscored the ways that the problems of flavor in industrial food

41
Coon 1993: 760.
165
production (rather than, say, in agriculture) would come to dominate the research in this

field. The small laboratory taste panel, which was first comprehensively described at the

1937 symposium, would emerge as the tool best suited to managing the inherent

experimental problems of flavor measurement, within the context of industrial food

production.

Operationally, the concern with measuring sensory experience led to the

development of experimental methods that not only captured the sensory qualities of

foods, but also the sensory acuity of the humans doing the tasting. Measurements and

records of the sensory acuity of tasters become a defining feature of laboratory taste

panels, distinguishing them from consumer research and expert evaluations. Although the

explicit purpose of taste panels was to measure food’s qualities, the senses of the tasters

who comprised the panels were also captured in researchers' evaluations, measurement,

and scrutiny.

First, Test the Tasters: Laboratory Taste Panels

Florence B. King, a food researcher at the USDA Bureau of Home Economics,

was frustrated. She and her research group were comparing the results of two common

methods of home bread-making: the ‘sponge’ method, which required a fermentation

period between two mixing stages, and the ‘straight-dough,’ single-mix, method. Which

method produced the better loaf?42

42
King 1937: 207-219.
166
 In order to find out, King and her colleagues convened a panel of nearly one

hundred men and women — laboratory workers, statisticians, clerks, stenographers, and

executives – demographically "fairly representative of one consumer group." These

tasters were asked to record differences in flavor, texture, and appeal between breads

made with two different manufacturing processes on scorecards. But there were

irreducible problems. The group was "not sufficiently discriminating" to detect the small

differences between samples, and "only a very small percentage" could duplicate a

previous judgment when given the same sample. Even worse, the sixteen individuals with

prior experience with food tasting performed no better than their inexperienced peers.43

The large panel’s judgements was both inaccurate and unreliable.

It was commonly known that capacities for sensory discrimination varied widely

across the population. Indeed, recent studies had documented the presence of smell and

taste "blindness" among individuals.44 Could choosing tasters with greater sensory acuity

improve the consistency and reliability of results? King outlined a multi-stage process to

cull the panel so that it included only those with the sharpest capabilities. The selection

process began with a questionnaire. The original 96 judges were surveyed about their

age, gender, smoking habits, and susceptibility to head colds, as well as how much bread

they typically ate and whether they had any "prejudices" against the flavor and odor of

bread. Excluding the head-cold-prone and bread-averse left 64 tasters. Next,

experimenters deployed established techniques from experimental psychology —

psychophysical methods of measuring sensory thresholds and perceptual gradations of

43. King 1937: 207.

44. See, for instance, Blakeslee 1936.
167
intensity — in order to measure the basic sensory capabilities of their prospective pool of

judges. Prospective tasters were asked to identify simple solutions by taste, and then to

rank them in order of intensity; they were similarly evaluated on the acuity of their senses

of smell.45 The fourteen best performers were re-tested on the original experimental bread

substances. The results were mixed: this smaller, more acute group was not any better at

detecting differences between the two types of bread. However, the group was more

consistent: better at duplicating previous judgments when re-tested with the same sample.

The significance of King’s paper lay not in her findings about bread qualities, but

in her conclusions about the tasting instrument that she had assembled. A small, select

panel of tasters could provide experimental data about both preferences and sensory

differences that was comparable to that produced by a larger group that was more

demographically "representative" of the general population. King’s paper was also one of

the first in this field to distinguish difference testing — which used the senses as an

instrument for determining sensory properties of foods — from preference testing, which

registered the reactions of the taster, rather than the qualities of the food.46 In her paper,

King outlined a practice that would become standard in taste-panel research:

systematically testing prospective panel members to assess their basic sensory-

45. The test solutions for taste sensitivity were chemically pure solutions of sodium
chloride, sucrose, lactic acid, and caffeine, at different dilutions in water. The sample
scents were benzaldehyde, citral, coffee, menthol, oil of turpentine, and a 10% aqueous
solution of ammonia. Experimenters also tested subjects for their ability to recognize the
scent of a yeast dispersion in water, and a 95% alcohol solution, since these are important
aromatic components of bread. King 1937: 208-210.
46. Pangborn 1964: 64.
168
discrimination capabilities. In other words, her research established a protocol wherein

the tasters were tested, before the food could be.

King's protocol for selecting tasters reflected an epistemic shift in the purpose of

the tasting panel. While earlier users of small panel techniques had suggested selecting

tasters based on the correspondence of their preferences with those of the general public,

King’s selection standard was experimentally determined, normative sensory acuity and

reliability.47 Rather than serving as a small-scale model of consumer behavior, the tasting

panel could be used as an instrument for detecting and measuring sensory qualities

independent of preference. Tasters were expected to act as neutral instruments,

registering the qualities of the food rather than personal reactions.48 In other words,

King’s taste panel was a group that represented and reproduced general human sensory

capacities, rather than human sensory communities. This had important consequences.

While a consumer panel was at risk of becoming less typical and less representative as it

became more "professional," a laboratory tasting panel could potentially improve its

reliability, accuracy, and consistency with experience and training.49 After all, expert

tasters were understood to have acquired their proficiency through practice.

Crucially, however, taste panel members were not being trained to be “expert

tasters” – specialists in particular commodities – but to improve their capacities for

47
For a discussion of the use of small consumer preference panels, see Platt 1931.
48
Pangborn cites King as one of the earliest to disaggregate difference testing from
preference testing. Pangborn 1964: 64.
49. On the hazards of "professionalization" of consumer panels, see Paul Lazarsfeld and
Marjorie Fisk, "The 'Panel' as a New Tool for Measuring Opinion," Public Opinion
Quarterly 2.4 (October 1938): 600.
169
sensory discrimination and reliability more generally. King’s method of using standard

solutions to test sensory acuity were also applicable to training regimens to improve this

skill, and thus to increase the accuracy and reliability of the tasting panel as an

instrument. Dairy researchers at Cornell had concluded that, by prescribing exercises that

improved a taster's capacity to identify and discriminate among “basic” taste and aroma

sensations — using simple solutions representing, for instance, bitterness, saltiness,

sourness, and sweetness at increasing intensities – one could improve a taster's general

"proficiency," meaning accuracy and reliability.50 Crocker addressed this point in his

remarks at the symposium, noting that expert tasters in industrial contexts were not

trained to detect “ultimate sensation elements,” but rather for substances or qualities

“known or believed to be present.” So, for instance, tasters at a processed meat plants

may be trained to taste for vinegar, spice, or smoothness. He suggested “in the training of

flavor judges, to familiarize them with the principle of the more classical sensation

detection as against the more industrial ingredient detection.”51 Calibrating taste panels in

this way was a strategy for producing an instrument that could be standardized across

research contexts and locations, a general tool for sensory measurement rather than one

which reflected local conditions and individual particularities.

Flavor researchers also developed methodologies that, by strictly controlling the

conditions of the experiment and constraining the parameters of the test, buttressed the

validity of taste-panel results, while also further distinguishing the laboratory taste panel

from methods reliant on “expert tasters.” An example of this can be found in the work of

50. Trout and Sharp 1937.

51
Crocker, “Measuring Food Flavors,” 1937: 285.
170
Sylvia Cover, a home economics researcher at the Texas Agricultural Experiment

Station, who in the late 1930s was studying the effect of cooking temperatures on the

palatability of meat.52 The National Cooperative Meat Investigation (NCMI) committee,

an industry group that studied meat quality, had established standards for meat

evaluation. Their expert tasters were asked to judge palatability by grading ten factors –

such as aroma, flavor of fat, and flavor of lean – in terms of intensity or desirability.

Cover’s group of tasters, drawn from staff members at other labs in the Station, had "little

training in subjective tests."53 (Cover makes no mention of testing her tasters’ sensory

acuity prior to using them as judges.) Their understanding of what sensory qualities

comprised each of the NCMI’s factors was evidently vague; asking them to assign scores

would mar her results with fatal inconsistencies and subjective distortions. Cover needed

a method that would be simple enough for these inexperienced tasters to use, while also

producing useful, objective, and reliable results.

Instead of requiring tasters to score all ten factors, she asked her judges to attend

to only one factor: tenderness. Each was given a pair of numbered samples, taken from

the different sides of the same animal, cooked at different temperatures. Blind to the

method of cooking used for each sample, the judges were asked to record only whether

they found a difference, and if so, to indicate which was more tender.54 “By this method,”

Cover wrote, “differences are easily detected and recorded by the judges and the results

52
Sylvia Cover, "A New Subjective Method of Testing Tenderness in Meat -- The
Paired-Eating Method," Food Research 1.3 (May-June 1936): 287-295.
53
Cover 1936: 293.
54
In later iterations of this research, the score sheet was modified to allow judges to
indicate the degree of difference – none, slight, or decided. Cover 1936: 289.
171
of the judgments may be interpreted with little doubt as to the actual differences

involved.”55 The objectivity and reliability of the results were obtained by rigorous

control of both the sample and the instrument. The sample varied in only one factor

(temperature); the human instrument measured only one dimension of experience

(tenderness).

Cover’s techniques bore strong resemblance to the method of paired comparison,

and the determination of least noticeable difference — both with roots in the

psychophysical laboratory.56 It shared with these earlier experimental techniques similar

strategies to limit subjective interference and obtain scientifically valid results: tight

control over experimental conditions and disciplinary control over the operation of the

human tasters. Both of these concerns would remain central to laboratory taste panel

research. But Cover differed from later researchers in her relative lack of concern for the

influence of social factors. For instance, tasters were permitted to chat while tasting, as

long as they did not know which of their samples represented the same experimental

conditions.

55
Cover 1936: 289.
56
Cover does not trace her “paired eating” technique to experimental psychology, but
calls it an adaptation of a method used in nutritional science. She cites a 1930 paper that
studied the effects of specific nutritional deficiencies by feeding animal pairs diets that
were identical but for the nutrient (eg, vitamin B, cysteine) under investigation. The
somewhat tortuous feats of adaptation necessary to suit this technique to Cover’s own
research, and the multiple disciplinary fields crossed by these experimental techniques,
demonstrates the nonstandard routes by which these standard psychophysical methods
entered sensory science. H. H. Mitchell and Jessie R. Beadles, "The Paired Feeding
Method in Nutrition Experiments and Its Application to the Problem of Cystine
Deficiencies in Food Proteins." Journal of Nutrition 2.3 (January 1, 1930): 225-243.
172
 For Cover, biases could be managed by another strategy that would be used by

flavor researchers to secure the objectivity of taste panel experiments: statistical control

over the results.57 Cover used simple statistical methods – binomial and chi-square

techniques – to eliminate aberrant data and produce results that seamlessly reflected

aggregate acts of tasting.58 Later researchers would apply statistical methods not only to

validate the accuracy of the flavor measurements, but also to monitoring the performance

of individual tasters. For instance, by the use of "control charts," a technique imported

from industrial process engineering that uses statistical calculations to identify judges

whose performance was inconsistent, skewed, or unreliable. This information could then

be used to investigate the cause of unreliability — whether it was because of a health

issue, or a deficiency of training, or because of some fundamental problem with the

design of the testing conditions.59 By these acts of statistical maintenance, the taste panel

could be trusted to remain a standard instrument.

57. Christopher Phillips provides a detailed examination of how statistical methods were
used in the sensory evaluation of wine in the postwar; he demonstrates that the statistical
processing of taste panel results manufactured a collective objectivity from the
aggregation of subjective reports. Christopher J. Phillips, “The Taste Machine: Sense,
Subjectivity, and Statistics in the California Wine World,” Social Studies of Science 46.3
(2016): 461-481. For the definitive account of how techniques of quantification, such as
statistics, gained validity, authority, virtue, and social power in modernity, see Thomas
Porter, Trust in Numbers.
58
Sylvia Cover, “Some Modifications of the Paired-Eating Method in Meat Cookery
Research,” Food Research 5 (1940): 385.
59
Sophie Marcuse, "An Application of the Control Chart Method to the Testing and
Marketing of Foods," Journal of the American Statistical Association 40.230 (June
1945): 214-222.
173
The Psychophysics of Quality Control: Taste Panels in

Industry

Laboratory taste panels were not only used in basic research at agricultural

research stations, but also in industry, where they were applied to both quality control and

product development. In the late 1930s, management at Joseph E. Seagram & Sons

Distillers, in Louisville, Kentucky, became disenchanted with the results they obtained

from professional tasters. Seagram, one of the largest producers of alcoholic beverages,

needed a system for ensuring that the sensory qualities of their blended whiskies

remained consistent from batch to batch – a tremendously complicated sensory and

chemical question – as well as methods for developing improved blends.60

In the late 1930s, Seagram management put Edward H. Scofield, a psychologist

whose doctoral work investigated the classic psychophysical phenomena of taste

thresholds, in charge of the research department at their Louisville plant.61 Under

Scofield’s leadership, the “poorly defined methods employed by the traditional taste

artists” were dumped, and their “sniff, sip, snort, and spit technique[s]” replaced by a

psychological program that put the measurement of quality on a sound scientific basis.62

His program combined rigorous experimental control with the use of trained and

disciplined subjects in order to measure sensory qualities and correlate them with

preferences, producing “data possessing the properties of discriminability and

60
David R. Peryam, “Sensory Evaluation: The Early Days,” Food Technology (January
1990): 86-9.
61
Peryam 1990: 87.
62
H.F. Willkie and E.H. Scofield, “Some Factors Influencing Determination of Relative
Preferential Values of Distilled Alcoholic Beverages,” Institute of Food Technologists,
1941 Proceedings, (Champaign, IL: Garrard Press, 1941): 204, 208.
174
reproducibility.”63 The primary psychophysical method he used was that of paired

comparison: depending on the experimental situation, tasters were asked to identify

which of two samples they preferred, or, in difference tests for quality control, to indicate

whether they perceived a difference. In order to secure the validity of these results,

Scofield made sure that all variables that appeared to have an effect on taste judgments

— the temperature of the sample, its alcohol content, and color — were made consistent

across samples, and that tasters consumed identical quantities of each sample for each

evaluation. He designed laboratory equipment that allowed for the automatic control of

many of these variables, thus rendering the testing system both more reliable and more

efficient.64 In order to ensure that tasters produced judgements that accurately reflected

perceptual experience, unclouded by subjective biases, they were allowed only twenty

seconds of judgment time per pair. “The employment of a long-time interval merely

allows the observer to confuse himself,” Scofield wrote. “This results in sheer guesswork

and later self-contradiction.”65 For this same reason, tasters were not encouraged to

identify the type of beverages which they had expressed a preference for during testing,

in order to avoid “the development of fixed ideas which almost invariably accompany

identification.”66

Scofield also used techniques from experimental psychology to define and

measure quality factors that had previously been tacit. For instance, “lightness” and

“heaviness” were often used to describe alcoholic beverages, and clearly influenced

63
Willkie and Scofield 1941: 208.
64
Willkie and Scofield 1941: 206.
65
Willkie and Scofield 1941: 204.
66
Willkie and Scofield 1941: 204.
175
quality judgments, but there was little agreement as to what exactly these terms referred

to. Yet, Scofield reasoned, “if such properties actually exist they must be measurable.”

After much research, it was determined that heaviness and lightness were descriptors of

flavor intensity. But how could flavor intensity be measured? Scofield employed a classic

psychophysical procedure, the method of limits. Tasters were presented with a series of

whiskey-water mixtures, in which the concentration of whiskey increased by discrete

increments, and were asked to indicate the sample where the flavor of whiskey was just

perceptible. This threshold concentration was defined as the lightness value of the

whiskey. Once lightness was made measurable, it could then be correlated with

preference using paired sample comparisons.67

In Scofield’s difference tests and preference tests, tasters reported on only one

factor, such as odor, taste, or color. The integrity of this monofactoral analysis was

vouchsafed by a rigorous control over the conditions of tasting, attending to the ambient

environment and physiological limits of the body, as well as the standardized conditions

of the sample. When a quality, such as ‘lightness,’ seemed perilously vague and ill-

defined, it was made exact and measurable. Rather than relying on the sensory skill and

committed effort of tasters to report reliably on sensory experience, Scofield developed

strategies and deployed technologies to engineer maximal control into the experimental

design of his tasting protocols.

67
Willkie and Scofield 1941: 207.
176
The State of the Laboratory Taste Panel Just Before the
War
By the early 1940s, small panels of selected and trained tasters were used in

diverse institutional settings: industry research and development laboratories and quality

control facilities, agricultural experiment stations – especially in home economics

research into the effects of cooking and preparation methods – and in psychological and

psychometric laboratories studying human sensory physiology. Across these settings,

three features had come to define the laboratory taste panel as a standard, reliable

instrument. First, panel members were selected based on assessments of sensory acuity

using standard samples and procedures. Second panel members were trained in general

techniques of sensory evaluation, which reflected an expectation that the panel was not an

ephemeral entity, but would serve in an ongoing and recurrent role.

Finally, researchers used experimental testing methods that restricted taste panel

considerations to one sensory factor, and constrained output to schematized and

statistically analyzable forms. Whereas expert tasters judged quality by assessing

multiple sensory factors (for instance, evaluating aroma, tenderness, and color in meat),

laboratory taste panels were expected to register differences or degrees of intensity along

only one sensory dimension. Early users of taste panel methods, such as Frances King

and Sylvia Cover, attempted to achieve this by instructing their judges and tailoring

response forms to minimize ambiguity. Scofield, at Seagram, sought to produce

monofactoral sensory data via experimental design. He placed his tasters within highly

controlled experimental systems engineered to ensure that the human senses attended to

only one factor at a time.

177
 The Quality Control program at Seagram would serve as the direct model for the
protocols at the Quartermaster Food Acceptance Research Laboratory. As will be seen,
the evolution of sensory evaluation methods at the Quartermaster involved increasing the
control over both tasters and the experimental spaces where their sensory labor occurred.

II. The Importance of Tasty Rations: Food

Acceptance Research in the US Army

When is a chocolate bar not a chocolate bar? When it is Field Ration D, the

emergency ration developed in the late 1930s by the Hershey Chocolate company for the

US Army Quartermaster’s Subsistence Research Laboratory (SRL). Intensively

engineered by Hershey’s chief chemist to meet the anticipated needs of a mobile army

deployed in combat zones around the globe, Field Ration D was no ordinary chocolate

bar. Super durable, it would not melt at temperatures below an infernal 120F. At six

hundred calories per four-ounce bar, it provided a dense caloric payload in a pocket-size

package. A triad of these, in poison-gas-proof wrappers, was the standard issue for a

day’s field rations. More than a quarter-billion bars were shipped and stockpiled overseas

between the attack on Pearl Harbor and D-Day.68

There was another important way in which Field Ration D was unlike ordinary

chocolate bars: Field Ration D was not designed to taste good — “just a little better than

a boiled potato,” was how Colonel Paul Logan, head of the SRL, (allegedly) put it. He

68
Anastasia Marx de Salcedo, Combat-Ready Kitchen: How the U.S. Military Shapes the
Way You Eat, (New York: Current, 2015): 86-7; Kawash 2013: 222-3.
178
even suggested adding kerosene powder to “throw the product off flavor.”69 Col. Logan

worried that making the emergency ration too tasty would impair its functionality, as

soldiers would glut themselves on chocolate rather than sticking to a regimented feeding

schedule. This was not perversity on Col. Logan’s part, but reflected the priorities of the

military at the time: nutritive value, stability, and utility outranked acceptability in the

design and development of field and emergency rations.70

As heaps of abandoned and discarded military rations accumulated in war zones,

the problem of acceptability rose to the fore. The uneaten rations were not only a waste of

money and material; the situation had real consequences for military preparedness.

Improperly fed soldiers were underperforming soldiers. Morale, an attitudinal factor that

psychologists associated with victory on both the homefront and the front lines, was also

strongly correlated with ration satisfaction.

The Quartermaster’s Food Acceptance Research Branch, founded in 1944 as a

division of the SRL, signaled the recognition of the functional importance of good flavor

to military readiness, national advancement, and even human survival. The methods,

purposes, and scope of “food acceptance” research encompassed physicochemical

research on food and flavor, taste panel determinations of the sensory qualities of foods,

and the study of human behavioral responses to those qualities. But the goal of this work

was not simply to identify the conditions and qualities that divided what would be eaten

69
Alissa Hamilton, “World War II’s Mobilization of the Science of Food Acceptability:
How Ration Palatability Became a Military Research Priority,” Ecology of Food and
Nutrition 42 (2003): 327.
70
Hamilton 2003: 330.
179
from what would not. As one of the division’s scientists explained in 1957, the ultimate

criterion of food acceptability was not consumption alone, but “‘consumption with

pleasure’ — we might say, ‘the nutrition of body and soul.’”71 In other words, the goal of

food acceptance research was not to determine the lowest threshold of palatability, but to

discover the factors that influenced desire, renewed appetite, and increased satisfaction.

Trained taste panels played a central role in the Food Acceptance Branch’s

research protocols. However, the war reoriented sensory research toward the evaluation

of new kinds of food products. Pre-war taste panel research on the sensory qualities of

foods had typically focused on familiar fare — bread, meat, milk, canned vegetables, and

fruits — items which, though they may be somewhat changed by processing, had a pre-

established record of acceptability. Wartime conditions altered the objects and objectives

of flavor research. “For the first time in history,” observed W. Franklin Dove, the first

chief of the Food Acceptance Research Branch, about the diet of soldiers during the war,

“large groups of men lived for long periods of time solely on commercially produced and

processed foods.”72 The food substances that concerned the military were often anything

but familiar, and sometimes unprecedented: dehydrated milk, eggs, and potatoes,

hydrogenated fats, soy oils, vitamin-enriched flours. The question was not how to meet

71
Francis J. Pilgrim, “The Components of Food Acceptance and Their Measurement,”
American Journal of Clinical Nutrition 5.2 (1957): 171.
72
W. Franklin Dove, “Developing Food Acceptance Research,” Science 103.2668 (Feb
15, 1946): 188.
180
some given (if arbitrary) standard of quality, but to shape these new substances into

appetizing forms.73

As a center of coordinated research, the Quartermaster division was critical in

articulating the basic research modalities that would be mobilized to study the problem of

flavor in food. The Food Acceptance Research Branch included statisticians,

physiologists and psychologists studying sensory thresholds and attitudes, researchers

studying physical and chemical components of food quality, and home economists and

food technologists working on experimental cooking techniques.74 In particular, food

research at the Quartermaster connected organoleptic testing methodologies to ongoing

anthropological studies of regional and national food habits, as well as to psycho-

physiological studies that delved into the mechanisms of appetite, thirst, hunger, and

satiety.75

One discipline would come to predominate at the Quartermaster Food Acceptance

Research Branch: psychology. This would profoundly affect the shape of the emerging

field of sensory science. Although prewar home economists, chemists, and food

technologists who used taste panels may have tested their tasters’ sensory acuity, their

focus was the accurate measurement of sensory qualities in foods, not the determination

73
Backer argues that the work of the SRL and Food Acceptance Branch was shaped by
(and produced) a normative ideology of “American food,” reconstituting novel
substances into forms that reflected “standard” American habits. Backer 2014: 51-87.
74. Dove “Developing Food Acceptance Research” 1946: 189.
75. George Gelman and Charles S. Lawrence, "Foreword," in Quartermaster Food and
Container Institute for the Armed Forces, Committee on Food Research Conference on
Food Acceptance Research, Quartermaster Corps Manual QMC 17-9 (Washington, D.C.:
War Department, Office of the Quartermaster General, 1946): 5-6.
181
of psychic states or attitudes in tasters. The Food Acceptance Research Branch sought to

link the measurement of sensory qualities of food with both physicochemical components

and behavioral and affective outcomes; the taster was the subject as much, if not more,

than the thing tasted. “The observer is the key, and not the product,” explained one

prominent researcher, speaking at a 1953 Quartermaster-sponsored symposium on food

acceptance testing methods. “To state this another way, when evaluating a food product,

it is human behavior and not succotash, bologna, or dehydrated milk that is being

investigated.”76 Psychology would be the primary disciplinary orientation of those

involved in this new field of study, especially psychometric and psychophysical

approaches to sensation, perception, and preference.

The Quartermaster’s Food Acceptance Research Branch should also be

understood as part of a broader national program of food research that ultimately sought

to reshape food habits in order to most effectively utilize available resources to fill known

human needs. Food acceptance research drew from and complemented other wartime

research on food habits and nutrition also supported by the National Academy of

Sciences’ National Research Council: the Food and Nutrition Board, which coordinated

biochemical and physiological research on nutritional needs, and the Committee on Food

Habits (CFH).77 Led by anthropologist Margaret Mead, the CFH studied food

76
Dean Foster, “Purpose and Scope of the Conference,” in David R. Peryam, Francis J.
Pilgrim, and Martin S. Peterson, eds. Food Acceptance Testing Methodology: A
Symposium Sponsored by the Quartermaster Food and Container Institute for the Armed
Forces, Palmer House, Chicago, October 8-9, 1953, (Washington: National Academy of
Sciences, National Research Council, October 1954): 3.
77
George Gelman and Charles S. Lawrence, "Foreword," in Quartermaster Food and
Container Institute for the Armed Forces 1946: 5-6.
182
consumption patterns and attitudes toward food from the perspective of cultural

anthropology. Its goal was not merely descriptive, but advisory: to guide the development

of government food policy, and, in particular, “mobiliz[e] anthropological and

psychological insights as they bear upon the whole problem of changing food habits in

order to raise the nutritional status of the people of the United States and ultimately of

other people of the world.”78 As Amy Bentley has observed, this was a form of

“democratic social engineering,” that aimed to change behavior by “voluntary” rather

than compulsory means.79 While the CFH pursued this by developing a deep

understanding of the ideologies and cultural structures that guided Americans’ eating

habits, the Quartermaster’s Food Acceptance Research Branch’s program instead zeroed

in on the consuming body, its sensations, drives, affects, and behaviors.

The Social Architecture of Taste Panel Research at the

Quartermaster Food Acceptance Research Branch
Taste panels were used throughout the Food Acceptance Research Branch

laboratories, both as an instrument of research and a subject of study. Army scientists

attempted to refine methodologies and procedures in order to be able to quickly evaluate

products, determine food preferences, and assist in product development and

improvement. The first chief of the new branch, Dr. W. Franklin Dove, a biologist from

the University of Maine, had a background in studying human and animal food

78
Quoted in Amy Bentley, Eating for Victory: Food Rationing and the Politics of
Domesticity, (Urbana: University of Illinois Press, 1998): 25. See also: Brian Wansink,
“Changing Eating Habits on the Home Front: Lost Lessons from World War II
Research,” Journal of Public Policy and Marketing 21.1 (Spring 2002): 90-99.
79
Bentley 1998: 27.
183
preferences using psychophysical techniques. 80 Dove, who headed the Branch between

1945 and 1949, was instrumental in establishing the small trained panel as the premier

functional unit for determining the acceptability of a food.

Dove naturalized the small, trained panel’s origins, inscribing both the authority

and necessity of the tasting panel within a narrative about humanity’s historical

relationship to food production. Prior to the industrialization of agriculture, he wrote, "the

family taste panel passed judgment upon many characteristics conceded important in

today's scientific panels," providing a set of judgments about the flavor of different crop

varietals, cooking methods, storage practices, and keeping qualities, that "came... to

shape the pattern of agriculture in every region."81 The rise of commercial agriculture and

the industrialization of food production not only severed the direct connection between

grower and consumer, but also substituted new values for old when it came to making

decisions about production. For instance, family seed-stock was replaced by varieties

developed for disease resistance and high yield; home-canned and preserved foods were

replaced by standardized commercial products. This system had conferred numerous

benefits, including the efficient, centralized production of more, and more nutritious

foods. But, he said, “we have left out the relationship — we have left out the connecting

link between the living subject (the consumer) and the stuff of life (food) he lives on: that

80
Dove is perhaps best known today for his experiments surgically manipulating the horn
buds on the brows of immature bulls so that they would grow up to be one-horned
creatures, artificial unicorns. See Dr. W. Franklin Dove, "Artificial Production of the
Fabulous Unicorn: A Modern Interpretation of an Ancient Myth," Scientific Monthly
42.5 (May 1936): 431-6.
81
W. Franklin Dove, "Developing Food Acceptance Research," Science (February 15,
1946): 188.
184
link is acceptability.”82 In this new industrial food system, there was no clear route by

which “unorganized” consumer knowledge about food preferences could exert influence

on food production. “Now is the time,” Dove urged, “for the essence of the family taste

panel, now lost, to be returned — not as it was, but in a modern scientific form.”83 With

the obsolescence of the family taste panel, the scientific taste panel had to take its place

— playing the same role the family taste panel once did, but rather than operating from

below, at the level of the disaggregated household, it now operated from above, inserting

experimentally produced knowledge about taste, flavor, and acceptability into

technoscientific planning and decision-making processes concerning agriculture, food

manufacturing, storage, and consumption.

What did it mean to bring a “modern scientific form” to the taste panel? Dove

formally outlined the elements of what he called the "Subjective-Objective Approach" to

measuring food acceptability, stabilizing and elaborating many of the previously

mentioned psychophysical techniques of taste-panel selection and methodology into a set

of standard principles and practices.84 The central psychophysical technique used by

Dove was the method of paired comparisons, where tasters were given two samples and

asked either whether there was a detectable difference between them, or whether one was

preferable to the other. Studies of difference were only in support of studies of

preference, the ultimate goal of research.

82
Dove 1946: 188-9. Emphasis in original.
83
Dove 1946: 189. Emphasis in original.
84
W. Franklin Dove, “Food Acceptability: Its Determination and Evaluation,” Food
Technology 1.1 (January 1947): 39-50.
185
 Dove created a dedicated facility for taste panel evaluations at the Food

Acceptance Research Branch in Chicago, based, in part, on observations of the panel

room at Seagram.85 However, design of the Food Acceptance Research laboratory in

Chicago set a new standard for these spaces, providing a pattern for other research

facilities, and shaping the atmospheric, architectural, and social conditions under which

the sensory labor of trained tasters would take place.

The architecture of the sensory evaluation facility was designed to permit

maximum experimental control over testing conditions and subjects, as well as the

efficient, routinized management of panel activities. The room included five isolation

booths, each with a wall hatch that opened into the adjoining sample prep room, so that

researchers could deliver the samples with a minimum of human contact, as well as its

own food-disposal unit and water fountain for mouth-rinsing between sample pairs.86

Walls, table tops, and other features of the space were colored a "natural gray, which

does not add color to the foods."87 The isolation booths excluded social sources of bias as

well as possible sources of distraction, allowing the taster to devote her or his undivided

attention to the task of sensory discrimination. In their rigorously controlled austerity, the

booths also provided the warrant for experimental replicability in other laboratories, with

equivalently equipped spaces. The panel testing facility was "entirely air-conditioned,"

85
Meiselman and Schutz 2003: 200.
86
Five booths are prescribed as maximally efficient, as two series of tests will provide
“the ten records required of a carefully selected group of judges,” and one operator can
effectively attend to five subjects at a time. Dove warns against a single booth, which
would make it impossible to test a food such as soup, which must be offered at the same
temperature to all tasters. Dove 1947: 45.
87
Dove 1947: 45.
186
and had its own ventilation system to eliminate any atmospheric contaminants. The room

also gave experimenters some operational flexibility. Dove's innovation was to install a

system of spotlights in each of the individual tasting booths at his lab, with "three degrees

of natural light and two degrees of colored lights (red to blue), plus control of intensity."

This allowed the experimenter to control and alter the apparent color of foods,

augmenting or eliminating differences, thus segregating judgments based on color from

those based on other aspects of flavor.88

Tasting booth at the Quartermaster Food Acceptance Research Branch. Note the hatch in
the wall for the impersonal conveyance of samples from experimenter to tasting subject,

the pencil atop the standard evaluation form, and the faucet and bowl for rinsing the
mouth between tasks. Image source: L.L. Thurstone, “Psychophysical Methods,” in T.G.
Andrews, ed., Methods in Psychology, (New York: John Wiley & Sons, 1948): 155.

88
Dove 1947: 45. In a footnote, Dove credits the Cleveland General Electric Company
with assistance in developing this lighting scheme, which was based in part on his
experiments using red lighting to prevent cannibalism among experimental animals.
187
 The sensory evaluation laboratories were designed to extract reliable sensory

information from tasters, and certify its scientific validity, by a system of external

controls and disciplining procedures. The artificial conditions of the room – its silence,

neutral palette, piped-in and odorless atmosphere – created a scenario where the taster

seated in the booth was stripped (as much as possible) of the distorting scrim of social

relations that came between the basic perceptual response to a food and her or his

awareness of that response.89 The architecture of the room aspired to form the taster into

a sensing machine, not a human eating but a taster tasting (then spitting and rinsing),

neutrally registering binaries of difference or preference between samples that were

designed to vary along only a single vector of sensation.

The epistemology of food acceptability at the Quartermaster Food Acceptance

Research Branch was also indicated by the sensory laboratory’s location within a

networked complex of other laboratories, research spaces, and technical facilities.

Acceptability research, Dove emphasized, "is a cooperative venture whereby

technologists, commodity specialists, and packaging specialists join in the discussion of

the purpose, plans, and conduct of the experiment."90 For instance, the sensory labs had

a close working relationship with the nearby physical-chemical laboratory, where the

same foods whose qualities were being studied in the sensory laboratory were analyzed

89
For a critical discussion of the consequences of this artificiality on the validity of food
and flavor research, from the perspective of food science but informed by anthropology,
see Jacob Lahne, “Sensory Science, the Food Industry, and the Objectification of Taste,”
Anthropology of Food 10 (2006). See also: David Howes, “The Science of Sensory
Evaluation: An Ethnographic Critique,” in Adam Drazer and Susan Küchler, eds., Social
Life of Materials: Studies in Materials and Society, (London: Bloomsbury, 2015).
90
Dove 1947: 45-6.
188
chemically, and where researchers could obtain the standard chemical solutions used to

test tasters’ sensory thresholds. The work of the sensory labs was also linked to that of

research groups studying vitamin and mineral content of foods, bacteriological

conditions, and packaging materials and design.91

The constant circulation of information, discussion, and results between the

“subjective” sensory laboratories and the spaces where “objective” physicochemical and

biochemical research took place were supposed to create cycles of rectification by which

the results of both subjective and objective forms of research would be brought closer to

true and useful knowledge. “Alternate movement is essential to progression,” Dove

instructed, unleashing his inner Hegelian. “So too in science, the alternate emphasis upon

the subjective and then the objective will in the end fuse into one process whereby all

unessential objective tests and all incoherent subjective responses will be exfoliated and

fall into discredit and disuse.”92 The “Subjective-Objective Approach” elaborated by

Dove claimed to produce more reliable and relevant information about both subjective,

sensory effects and their objective, material causes by treating the two forms of

knowledge as fundamentally interdependent. Although Dove helmed the Quartermaster

Food Acceptance Research Branch for only four years, his leadership helped to establish

the study of the sensory qualities of food as a legitimate scientific field, and to position

the chosen, trained taste panel operating within specially designed, rigidly controlled

conditions as the standard instrument for that work.

91
Dove 1947: 45-6.
92
Dove 1947: 44.
189
Measuring Pleasure: The Hedonic Scale

Dove’s system of difference-preference testing measured acceptability and

preference only indirectly, in relation to a system of comparative relationships. His

expectation was that, as test series were repeated, results could then be compiled into

“Tables of Experience,” from which basic attitudes toward foods could be deduced.93 But

could investigators experimentally measure preferences directly? Attempts to accomplish

this resulted in the Quartermaster Food Acceptance Research Branch’s other major

contribution to sensory evaluation: the hedonic scale.94

The hedonic scale was developed under the leadership of David Peryam, who was

brought on to head the Food Acceptance Research Branch in 1949 after Dove left the

military and returned to the academy. Peryam, a psychologist, remained at the

Quartermaster until 1957, when he left to found Peryam & Kroll, an influential consumer

testing and market research firm. Peryam had been in charge of quality control at

Seagram, where he had worked closely with Scofield in developing a psychological

program for flavor evaluation and management.95 His tenure at the Quartermaster not

only helped build “the largest collection in the world of researchers working on both

93
Dove 1947: 48.
94
David R. Peryam and Francis J. Pilgrim, “Hedonic Scale Method of Measuring Food
Preferences,” in “The Methodology of Sensory Testing: Contributions to the Institute of
Food Technologists Symposium in Pittsburgh, Sponsored by the Committee on Taste
Testing and Consumer Acceptance for the 17th Annual Meeting of the IFT,” Food
Technology 11 (September 1957): 9-14.
95
Meiselman and Schutz 2003: 200.
190
theoretical and applied areas in food acceptance, appetite, and hunger,” but definitively

established the centrality of psychology to the field of food acceptance research.96

Peryam and his colleague Frank Pilgrim, a psychologist and chemist whom he

hired to head the psychophysiological division, insisted that the hedonic rating scale was

not an entirely new tool, but a special application of a psychometric technique that had

been in wide use since the nineteenth century.97 (The psychometric and psychological

testing of soldiers was particularly well-established in the Army, where it had been used

since the First World War to test intelligence, personality, and other capabilities.98) The

Quartermaster’s earliest study using a hedonic scale preceded Peryam’s arrival at the

Quartermaster; in 1947, a seven-point scale was used as part of a field survey of soldiers

to determine preference for different menu items. The scale was shelved until 1949, when

researchers returned to it in search of a method of evaluating preference under the more

controlled conditions of the laboratory.99

The hedonic scale presented liking as a continuum, a vertical gradient whose nine

intervals ranged (in the final, validated scale) from “dislike extremely” to “like

extremely”; its midpoint was indifference (“neither like nor dislike”). Tasters were asked

to “show your reaction” to a food by checking the point on the scale that “best describes

your feeling about the food.” The rating scale, and the language used to designate its

96
Meiselman and Schutz 2003: 200.
97
Peryam and Pilgrim 1957: 9.
98
Daniel Kevles, “Testing the Army’s Intelligence: Psychologists and the Military in
World War I,” Journal of American History 55.3 (1968): 565-581; Michael Sokal, ed.
Psychological Testing and American Society, 1890-1930, (Rutgers UP, 1987).
99
Peryam and Pilgrim 1957: 9.
191
intervals, was further refined in collaboration with L.L. Thurstone’s psychometric

laboratory at the University of Chicago. The goal was to develop a scale where “no one

would question that the successive intervals are in the proper ordinal position, and where

all subjects understand and use the intervals in about the same way,” — ie, one that

minimized any ambiguities around lexical meanings, and that smoothly conformed to

subjects’ own understandings of the degrees of affective response.100

More important, possibly, than the scale, were the printed instructions given to the

tasters prior to the evaluation session. The tasters who participated in hedonic scale

testing were not tested, selected, and trained judges, as in taste panel evaluations, but

larger groups, generally totaling around forty individuals. The standard instructions cast

the taster as a self-defining, autonomous, authoritative subject, who was providing a

valuable service to the experimenters:

“You will be given several servings of food to eat and you are asked to say
about each how much you like or dislike it. Use the scales to indicate your
attitude by checking at the point which best describes your feeling about
the food. Keep in mind that you are the judge. You are the only one who
can tell what you like. Nobody knows whether this food should be
considered good, bad or indifferent. An honest expression of your personal
feeling will help us to decide. Take a drink of water after you finish each
sample and then wait for the next.”101

100
Lyle V. Jones, David R. Peryam, and L.L. Thurstone, “Development of a Scale for
Measuring Soldiers’ Food Preferences,” Food Research 20 (1955): 512-20.
101
Peryam and Pilgrim 1957: 10.
192
 What investigators were aiming to capture was a sort of stimulus-response to food

that preceded judgment: “the emotional aspects of mental life as opposed to the

intellectual.”102 Ideally, the scale would yield the basic cognitive units of “like” or

“dislike.” The instructions were intended to “encourage [the taster] to report his

immediate naive response without any conscious effort to remember or to judge.”103 This

was a restatement of the method of introspection used by experimental psychology at the

turn of the twentieth century. However, rather than obtaining objective accounts of the

structures of consciousness from subjects rigorously trained to be good

phenomenologists, Food Acceptance Branch researchers relied on a combination of

experimental design and positive disciplinary procedures to extract reliable data.

Underlying the design and deployment of the hedonic scale was “the theory that it is the

uncomplicated response which determines pleasure in eating and governs the formation

of attitudes and future preference choices.”104 As was the case with the design of the

sensory-evaluation facilities, the assumption here was that there was a fundamental

human response to food quality that operated outside the realm of social relations. The

hedonic scale went further, explicitly locating this response prior to conscious reflection.

The affective responses to food that it sought to record and quantify were akin to instincts

102
Peryam and Pilgrim 1957: 9.
103
David R. Peryam and Norman F. Girardot, “QM Pins Food ‘Likes’ and ‘Dislikes’ with
Advanced Taste-Test Method,” Food Engineering 24.7 (July 1952): 59.
104
Maynard Amerine, Rose Marie Pangborn, and Edward Roessler, Principles of Sensory
Evaluation of Food, (New York: Academic Press, 1965): 372.
193
or reflexes; they were completely unlike the fully considered, educated tastes of the

connoisseur.105

Peryam and Pilgrim cautioned that the hedonic scale was not a measure of

acceptance, but of preference — which was, however, strongly correlated with, and could

be used to predict, acceptance.106 In other words, the hedonic scale should not be used as

a tool to measure of one individual’s pleasurable responses, but as a device to study

“human behavior potential” in aggregate, a future-oriented forecasting tool designed for

the problems of mass feeding.107

III. Food Acceptability and the Postwar Military-

Food-Industrial Complex

The problem of food acceptability, observed Quartermaster Captain R.O. Raub in

1946, would become increasingly important in a "peacetime Army... because the average

soldier will have increasing opportunities to decide which foods he will consume and

which ones he will refuse."108 As the distinction between “soldier” and “consumer”

eroded within the military, the civilian food system also came to show the stamp of the

105
I suspect that this has deeper connections to contemporary trends and ideologies in
psychology, but following those leads is outside of the scope of this dissertation, and will
be pursued in future research.
106
Peryam and Pilgrim 1957: 14.
107
Peryam and Pilgrim 1957: 12.
108
Captain R.O. Raub, "Food Acceptability Tests in the Army," in Quartermaster Food
and Container Institute for the Armed Forces 1946: 12.
194
army’s research.109 Food technologies such as freezing and dehydration, and chemical

additives such as MSG, that had been key to the production of wartime rations, found

continued use in processed foods after the war.110 Meanwhile, the processes by which a

new ration component was developed in the Quartermaster and a new frozen TV dinner

was developed in a private food company came to resemble each other more and more.

The postwar food industry readily adopted the sensory evaluation procedures and

practices formalized at the Quartermaster. By the early 1950s, laboratory taste-panels,

and the psychophysical and psychometric methods that had been refined in the army’s

Food Acceptance Research branch, had become standard tools used in the development

of new consumer products, quality control procedures, as well as in basic research

conducted in non-industry laboratories at the USDA and university food science and

technology programs.111

One reason for the rapid acceptance of sensory evaluation methods was the

circulation of scientists among Army labs, industry positions, and academic

appointments. The researchers who passed through the Quartermaster Food Acceptance

109
Marx de Salcedo’s recent book provides a detailed account of the connections
between military and industry, and links various consumer products (energy bars,
lunchables) directly to army food research. Marx de Salcedo 2015.
110
Backer documents the intimate relationships between the food industry and the
military in his dissertation.
111
Dove noted that the taste panel could also be a source of valuable information for food
marketers and advertisers. In addition to a record of differences and preferences, taste
panel research also produced "a record of words that express... differences" -- the
language that judges themselves used to describe the distinctions they sensed. These
records could "supply the advertising bureaus with substantial gustatory appeal to
supplement the more apparent eye appeal" of foods. Army research thus had implications
not only to for how food was made to taste, but also how food was sold to American
consumers. Dove 1947: 50.
195
Research Branch between 1948 and 1957 went on to careers in both industry and

academy, disseminating the Branch’s methodology and philosophy, and helping it

become standard in the field.112 Quartermaster funds also supported external research at

physiological and psychological laboratories studying taste and smell at Florida State, the

University of Chicago, and the University of North Carolina, among other sites.113

Further, sensory evaluation practices were publicized in conference proceedings and

scientific publications, including a 1947 Quartermaster-published bibliography on the use

of taste panels in palatability testing, comprising about 400 titles, and available on

request, without charge.114

In the late 1940s and early 1950s, the material and social infrastructure of

sensory evaluation was assembled in food and flavor industry research and development

facilities, as well as in the growing network of non-industry laboratories. Tasters were

recruited, tested, and trained for service on taste panels; dedicated rooms were outfitted

with isolation booths, special lighting, and sophisticated climate control systems;

researchers and technicians prepared samples, operated the human tasting-instrument to

assess sensible qualities, and analyzed the results. Even as best practices and standard

methodologies continued to be developed and debated, by the early 1950s, the laboratory

112
Meiselman and Schutz 2003: 204. Quartermaster veterans went on to direct market
and field research programs for companies including Coca-Cola, Pillsbury, and Lipton.
Schutz went on to UC Davis. David Peryam and Beverly Knoll found their own sensory
evaluation consumer research firm.
113
Meiselman and Schutz 2003: 201.
114
Bureau of Human Nutrition and Home Economics, Agricultural Research
Administration, Sensory Methods for Measuring Differences in Food Quality: Review of
Literature and Proceedings of Conference, Agriculture Information Bulletin No. 34,
(Washington, D.C.: U.S. Department of Agriculture, August 1951): 49.
196
taste panel was widely accepted as a reliable instrument in food and flavor research, and

its system of disciplined human tasters was credited with providing objective information

about food qualities, detectable differences, and preferences.115 A 1952 article in Fortune,

reporting on the new scientific techniques that were “taking the guesswork out of flavor,”

described the extensive sensory evaluation procedures that had recently been adopted by

four of the largest food and beverage companies: Heinz, Nabisco, General Foods, and

Seagram, in order to develop products that “meet the public taste and maintain flavor

uniformity.”116

The rapid spread of taste-panel testing is notable because sensory evaluation

demanded substantial investments: of personnel, time, and square footage. In many cases,

available facilities were retrofitted to meet the new requirements of food research. In

1951, Helen Moser, a food technologist at the Northern Regional Research Laboratory in

Peoria, described converting an 11x16 foot windowless storage room into a taste panel

room equipped with four isolation booths and a separate preparation area. Panel members

entered the room from the corridor, and sat down in one of the booths, which triggered a

light in the adjoining preparation area. Researchers transferred heated samples of soybean

oil to panel members through sliding hatches in the back of each booth, ensuring there

115
For an account of standard practices and ongoing issues in sensory evaluation, see
Mildred M. Boggs and Helen L. Hanson, "Analysis of Foods by Sensory Difference
Tests," Advances in Food Research 2 (New York: Academic Press, 1949): 219-258. Best
practices in sensory testing continued to be discussed and debated among professional
groups, such as the Institute of Food Technologists and the American Society of Testing
and Materials, into the late 1960s.
116
“What has Happened to Flavor?” Fortune 45 (April 1952): 130-3, 146-52.
197
was no contact between the person in the preparation area and the panelist.117 The room

was kept at a steady temperature of 78°F and 40 percent humidity.118

Floor plan of taste panel room and preparation and distribution of samples, from Moser et al,
"Conducting a Taste Panel for the Evaluation of Edible Oils," Food Technology (March 1950), p.
106

At large food companies, laboratory taste panels and sensory evaluation

procedures were integrated into research and development and quality control programs,

where they were used to study problems such as flavor changes during storage, strictly

ensure flavor consistency, and develop new products and lines. The Fortune article

117
Helen A. Moser, H.J. Dutton, C.D. Evans, and J.C. Cowan, "Conducting a Taste Panel
for the Evaluation of Edible Oils," Food Technology 4 (1950): 105-109. Dawson and
Harris, "Sensory Methods for Measuring Differences in Food Quality," 87.
118
Bureau of Human Nutrition and Home Economics 1951: 87.
198
describes panel testing procedures to evaluate the detectability of formula changes at

Nabisco, includes a photograph of Jell-O tasters working under red lights in individual

booths at General Foods’ Central Laboratories in Hoboken, and explains the quality

control system at Heinz. “Hourly samples from all of Heinz’s twelve factories are

shipped daily to the Pittsburgh ‘organoleptic’ department, run by Marie Pierkowski. She

makes sure products do not vary from one factory to another,” by using triangle tests and

other psychophysical methods to ensure standard qualities.119

At General Mills, sensory testing facilities built at the company’s central research

laboratories outside of Minneapolis in the early 1950s were used intensively. A 1953

feature in the company’s newsletter, Progress Thru Research, claimed that the taste panel

rom was in use nearly eight hours a day as tasters and other experts worked “under

controlled conditions to develop tastier food products for your dinner table.”120 General

Mills’ taste panel facilities were designed to maximize both experimental control and

efficiency. An advanced HVAC system controlled both temperature and humidity; an

ozone lamp handled the “big job of destroying odors” that wafted in from the surrounding

area or that lingered from previous tastings.121 Windows were blacked out to exclude

changeable natural illumination; a carefully designed lighting system allowed for a range

of flexible possibilities, including color filters. There were eight isolation booths,

“separated by partitions to eliminate conversation and reduce any other distractions

which would interfere with the important business at hand.” But “in a manner of minutes

119
“What has Happened to Flavor?”: 131.
120
Gloria Gershun, “Taste Testers,” Progress Thru Research 7.4 (Summer 1953): 7.
121
Gershun 1953: 7.
199
these private booths can be folded into wall cabinets,” to make room for conference

tables and open discussion which were necessary components of flavor profile evaluation

(to be discussed in Chapter Five). In addition to the sensory evaluation room itself, the

sensory laboratory also included “fully-equipped modern kitchen” and preparation center,

which shared space with a working area for record keeping, telephoning, and “other

detail operations which keep taste panel work running smoothly.”122

The details of General Mills’ tasting laboratory sheds light on the considerable

labor and substantial investment that were required to operate these facilities. For food

manufacturers that did not have the resources to install and maintain their own sensory

evaluation facilities, contract-consulting laboratories advertised and offered a range of

organoleptic-testing services.123 In the late 1940s, established chemical consulting firms

and contract laboratories, such as Arthur D. Little, Inc. in Cambridge, Foster D. Snell,

Inc. and Wallerstein Laboratories, in New York, and Food Research Laboratories, in New

Jersey, began offering sensory evaluation as part of their portfolio. These companies had

their own testing rooms, highly trained tasters, and other resources, such as libraries of

odor samples and flavor and fragrance materials. Sensory evaluation and testing was

increasingly seen as a necessary part of product design and development, not only for

foods and beverages, but for an expanding range of consumer products — from cosmetics

to rubber tires to refrigerators. The varied criteria of sensory quality, “too elusive to be

122
Gershun 1953: 6.
123
L.C. Cartwright and Robert A. Nanz, "Flavors Improved, Sales Boosted Through
Organoleptic Tests," Food Industries 20. 11 (November 1948): 1608-9; “Human
Analyzers,” Chemical Industries 67 (November 1950): 721-2; Arthur D. Little Papers
[MC579] MIT institute Archives. Box 7, Folder 39. History: Flavor Laboratory [memo].
200
caught in the analytical control laboratory,” one article on the subject explained, “can

make or break a product.” But organoleptic control, provided by sensory panel testing,

“can make it,” providing the key to commercial success. 124

Managing the Human Instrument

Ultimately, a taste panel is an assemblage of human beings, and this presented

unique challenges to the experimenters who had to manage these sometimes reluctant

instrumental components. Finding the right people to serve as members of a taste panel

took logistical and experimental labor. In designing experiments or planning tasting

sessions, researchers had to be mindful of the sensing capacities of human bodies —

including how the senses may be affected (or not) by environmental and experimental

conditions. Researchers were dealing not only with the tasters' senses, but also their

perceptions. Just as they had to accommodate the intractable requirements of bodies, they

also had to concern themselves with mental states, such as attitudes and motivations.

Finally, the task of managing tasters had different implications depending on the site of

research and the relationship between the panel members and the researchers. In an

industrial setting where panel members were often factory employees, the utilization of

the tasting panel could be more coercive than in USDA research facilities or private

research laboratories. All of these considerations required investigators to use various

inducements, coercive tactics, and surveillance of performance to obtain usable results.

124
“Human Analyzers” 1950: 722.
201
 Researchers also needed to separate the able from the merely willing, eliminating,

when possible, individuals with limited discriminatory capacities or sensory deficits. But

as an individual's sensory capabilities varied from day to day, this meant that the

screening and evaluation process was ongoing. Researchers obtained two kinds of data

from taste panel experiments: a record of the perceptible sensory qualities of foods, and a

record of the performance of individual tasters. Monitoring the latter was necessary to

assure the panel’s adequate function; "checking should be frequent, preferably every

day.“125 Tasters were asked to abide by certain behavioral restrictions, such as refraining

from smoking or eating for several hours prior to tasting.126

Researchers were also conscious of the need to arrange the conditions of the test

to prevent compromising each taster's sensory acuity. Tasters could be fatigued by the

presentation of too many samples, or at a too-rapid pace; results could also be

compromised if stronger-tasting samples were introduced before more subtly flavored

items. Investigators in the field had long been aware of physiological research

demonstrating that the sensate body had physical limits, and that as the senses became

fatigued, they became less responsive to stimuli and less capable of distinguishing

differences.127 Taste-panel experiments had to be designed to respect these limits, and to

provide recuperative accommodations, such as mouth rinses, to preserve discriminatory

capacities throughout the duration of the test.

125
Bureau of Human Nutrition and Home Economics 1951: 81.
126
Trout and Sharp (1937) quote one dairy plant manager, who prohibits not only
smoking and heavy meals prior to tasting, but also the consumption of chewing gum,
cough drops, "or other strongly flavored materials." Trout and Sharp 1937: 43.
127
Trout and Sharp 1937.
202
 The management challenge takes on a new aspect when one considers the

conditions of the labor required from taste panel members. At commercial companies and

in research laboratories, taste panelists were essentially volunteers, extracted from other

professional obligations and responsibilities to perform this function.128 As General Mills

explained in Progress Thru Research, the personnel who served on its taste panels

weren’t “casual guinea pigs; they’re hand-picked observers who are whisked away to a

spanking new laboratory equipped with modern conveniences to help them concentrate

on the job at hand.”129 Volunteers might include chemists, bakers, food engineers,

packaging experts, and other employees who were involved in distinct research and

development work at the company’s laboratories. “As an added feature,” the company

added, “taste panel participants work in a comfortable room which increases their

efficiency as objective observers.”130

How much of a pleasure should the sensory labor of taste testing be?131

Throughout the literature, the importance of maintaining a "comfortable" panel room is

emphasized, but rarely elaborated, beyond the stipulation that the room should be quiet

128
This differentiates the situation of these workers with that described by Simon
Schaffer, in his comparable account of the management of astronomical observers at
Greenwich. As volunteers, taste-panel members retain some power in the labor
arrangement. See Schaffer 1988.
129
Gershun 1953: 6
130
Gershun 1953: 6.
131
Some projects were certainly less pleasant than others. One Quartermaster Food and
Container Institute investigator, attempting to determine the reason why fish was so
loathed in army mess halls, observed that the popularity of her research section "fell
several degrees when tasters found that they were launched on a long-term fish program."
Marion Bollman, "Influence of Food Preparation Methods on Acceptance in the Army,"
in Quartermaster Food and Container Institute for the Armed Forces: 17.
203
and "free from distractions."132 This did not, however, ensure undistracted panel

members. L.C. Cartwright, of Foster D. Snell, Inc., a New York contract laboratory that

offered organoleptic evaluation services, observed that calling panel members away from

"their usual jobs may result in mental block. Panel members who are usually good may

be immersed in a piece of work which is interrupted by the judging and may give

judgments out of line on that occasion. They may be careless because they want to get

back to the job." His solution was accommodation. "We try to fit panel members into the

sessions most convenient for them."133 Although some judges may have found taste panel

duties to be a nice change of pace, it is evident that others were more ambivalent about

the task.

Mildred Boggs and Helen Hanson, of the USDA Western Regional Research Lab,

expressed an increasingly common sentiment when they wrote: "it is generally agreed

among those who direct research doing difference tests that the attitude of the judges is of

132
The recommendation that the sensing subject be provided with comfortable
surroundings may come from laboratory practices in experimental psychology, especially
those studying the basic structures of sensation and consciousness. For instance, in his
1898 textbook, Primer of Psychology, E.B. Titchener stipulates that the experimental
psychologist studying the structures of consciousness through introspection must "be
comfortable" in order to obtain access to pure sensations, images, and feelings untainted
by personal meaning. "Do not begin to introspect till all the conditions are satisfactory;
do not work if you feel nervous or irritated, if the chair is too high or the table too low for
you, if you have a cold or a headache. Take the experiment pleasantly." He also advised
that investigators "stop working the moment that you feel tired or jaded." Titchener's
manual of laboratory practice is cited by several of my sources, despite the claim by
Christopher Green that his methods had fallen into disrepute among psychologists by
Titchener's death in 1927. Unlike the researchers in my account, however, who attempted
to elicit information about the senses of others, Titchener was prescribing this (easeful)
disciplinary regimen to the experimenter, who was his own subject. See Christopher D.
Green, "Scientific Objectivity and E.B. Tichener's Experimental Psychology," Isis 101,
no. 4 (December 2010): 697-721.
133
Bureau of Human Nutrition and Home Economics 1951: 70.
204
great importance to the success of the experiment." Maintaining the proper attitude

among panel members, they explained, meant balancing two competing needs: the need

to stimulate judges' interest in the experimental project — and sustain it — in order to

ensure conscientious performance, and the need to avoid introducing potential sources of

bias, "which may result when there is too much knowledge about the problem under

investigation."134 Panel members must be interested, but not wise; trained, but not

knowledgeable. The researchers were to remain the experts in this scenario, not the

tasters.135

One way to maintain interest was to share experimental results with judges after

the experiment was completed. Helen Moser, of the Northern Regional Research

Laboratory, remarked that the practice in her lab was to allow each judge to learn the

identity of the samples, and to compare his or her tasting results with others, as soon as

he or she had left the panel room. "This opportunity for comparing his scores helps to

maintain an interest in the judging," she said, adding, "we also bribe the judges with

cookies at this period."136 Boggs explained, "We find that our tasters like to be right, they

like to be consistent and reproducible, so they will take advantage of every solitary bit of

information they can garner. We therefore do not give them much information in

advance, but keep up their interest by giving them the full results of every experiment

134
Boggs and Hanson 1949: 239-40.
135
In Daston and Galison's model of "trained judgment" as the 20th-century version of
objectivity, the researchers -- not the tasters -- would be the objective observers here,
utilizing their own tacit expert knowledge to derive objective results from the subjective
mesh of responses provided by the tasters.
136
Bureau of Human Nutrition and Home Economics 1951: 87-88.
205
after it is finished, as well as their own individual performance in the test."137Moser’s and

Bogg's comments show that experimenters deliberately used social and interpersonal

dynamics in their relations to their tasting subjects to improve the instrumental

performance of the taste panel. Competitive feelings among judges would inspire them to

put forth their best effort, and provide a motivation for continual improvement. We find

that our tasters like to be right. Cookie-bribes could be effective in rewarding and

sharpening those instincts.

In research settings, where judges were drawn from staff at a college or from

adjoining laboratories, experimenters had to accommodate the scheduling and

professional needs of panel members, just as panel members were asked to abide by the

abstentions and other practices required by researchers. When taste panels were deployed

in industrial settings, the power dynamics could be less egalitarian. David Peryam, who

worked at Seagram prior to heading the Quartermaster Food Acceptance Research

Branch through the 1950s, described his tactics for managing tasters in a distillery’s

quality control department:

"Motivation is important. To make the system work, the observer must

consider each unknown pair a challenge.... At the end of each test the
observer is told the results and a continuous record is kept of each
observer's percentage correct. He knows that the penalty for falling
significantly below the performance level of the group is banishment from

137
Bureau of Human Nutrition and Home Economics 1951: 67.
206
 the attractive laboratory job to the comparative Siberia of the bottling
lines."138

Rather than fully eliminating the "human equation" from the taste panel,

experimenters utilized (or perhaps manipulated) human motives, desires, and drives as a

means of eliciting the best results from their laboratory tool.

CONCLUSION: TAKING THE MEASURE OF TASTE

The taste panel, a laboratory instrument for measuring flavor, also made flavor

measurable. In other words, the instrument defined the boundaries of the object — the

thresholds of human sensory perception — and the conditions under which any

discovered difference might be taken as meaningful. It endowed flavor with a complex

materiality, registering its multi-sensory, psychological, and social dimensions even as

experimenters attempted to control and constrain which factors it measured.

But what kind of instrument was a taste panel? It comprised multiple,

heterogeneous parts: human bodies; dedicated and designed spaces with technologies for

illuminating, deodorizing, controlling climate, and excluding social influence; utensils for

food preparation and consumption; standardized paperwork. Managing a taste panel

138
David R. Peryam, "Quality Control in the Production of Blended Whiskey," Industrial
Quality Control (November 1950): 19. This paper was originally presented at the
Baltimore Section meeting of the American Society for Quality Control on November 15,
1948, while Peryam, a former Seagram quality control staff member, was employed at
the Calvert Distilling Company in Baltimore. By 1950, when the article was published,
Peryam had succeeded Dove as Chief of the Food Acceptance Division at the
Quartermaster Food and Container Institute.
207
demanded the coordinated efforts of various groups of scientists and technicians, as well

as the cooperation of the humans that provide the detecting function of the instrument.

Experimenters understood the taste panel to be a kind of scientific instrument or tool,

which, like any other scientific apparatus, had to be consistently calibrated, and had

discernible limits of precision and accuracy.139

Further, the taste panel must also be understood as one component of a broader

laboratory ensemble. It became increasingly common practice to correlate taste panel

output with instrumental readings from a battery of laboratory machines — colorimeters,

tenderometers, shortometers, and other instruments that measured texture, viscosity,

shear, and other physical properties — as well as a growing number of chemical tests for

measuring food qualities and constituents.140 Although these devices provided useful

results, and were sometimes more efficient and simpler to operate than panel tests, many

researchers continued to believe that "a physical or chemical method may be superior to

an organoleptic method in precision but not in accuracy."141 That is, the human senses

were the most reliable guide to detecting qualities in food, which machines or chemical

processes might not be able to register or measure. Further, machines and chemical tests

could not provide a measure of "over-all quality" — only indices and correlates.142

As a laboratory technology, the taste panel operated across several different

categorical divides: between human being and instrument, expert and non-expert, the

139
Platt 1937: 243.
140
Bureau of Human Nutrition and Home Economics 1951: 105-6.
141
Bureau of Human Nutrition and Home Economics 1951: 106.
142
Bureau of Human Nutrition and Home Economics 1951: 106.
208
sensory and the semiotic, and the laboratory and the field. The taste panel mediated

between and joined together disciplines, professions, and institutions concerned with food

flavors, the sensory qualities of food that contribute to beliefs about its value. Methods

and techniques were shared between different kinds of laboratories — basic research,

product development, quality control — at different sorts of institutions — government

agricultural experiment stations, military research centers, food factories, commodity

research institutes.

Along with the methods and material accoutrements of sensory research, the food

industry also adopted its premises and purposes. First among these was the notion that

both the sensory qualities of food and the human responses to those qualities were

measurable. By accepting the accuracy of the human instrument to register the qualities

of foods, they also accepted the idea that human responses, behaviors, and preferences

were more than merely personal, and could be understood as universal and objective, at

some deep level.

The spread of the laboratory panel and the science of sensory evaluation in the

food industry indicated a renegotiation of the division of authority about the qualities of

food. The proper personnel to organize and conduct the work of sensory evaluation were

not commodity experts, the “expert tasters” of coffee and whiskey and butter and cheese,

but scientific and technical workers whose authority derived from psychology and

statistics. Further, as flavor chemistry became an increasingly established subfield of

chemical research, it would be joined with scientific practices of food acceptance

209
research and sensory evaluation, allowing for the connection between physicochemical

properties of flavor and psychological aspects of behavior.

The science of flavor is never only about the qualities of foods; it also comes to

require the study and surveillance of sensing bodies. The conflation, or perhaps

confusion, of these two objects of scientific investigation will come to fuel critiques of

the food industry’s methods and ambitions. As the sciences of sensory evaluation are

applied to the purposes of enhancing acceptability, are food companies becoming better

at giving consumers the choices that they desire, or are they honing their abilities to

manipulate the sensible qualities of foods (and other things) in order to exploit

irresistible, subconscious instincts — provoking complex, elemental hungers that only

they can satisfy?

210
 Chapter 4
Fresh, Easy, New: Postwar Technologies
of Food and Flavor

Leaving their ration cards behind and entering an unprecedented era of prosperity,

postwar consumers began to spend more money on food than ever before. In 1941,

Americans spent $20 billion on food. In 1953, “to the stupefaction of just about everyone

who thought he understood the food market,” in the words of Fortune magazine, food

spending topped $60 billion.1 Only a fraction of this increase could be explained by

population growth and inflation. The larger cause was readily identified: consumers were

buying far more processed and packaged “convenience” foods, and paying more for

them.2

Why were Americans buying more processed foods, and spending more for them?

Historians of the American postwar period typically weave the ascendancy of processed

food into the complex tapestry of social, technological, economic, and cultural changes

that shaped American life during these decades. Rising incomes and a growing white

middle class; suburbanization, with its attendant sociotechnical menagerie of automobile,

refrigerator-freezer, television, supermarket; a regressive ethos of female domesticity

paired with an outsized faith in the goods of technological progress, and aggressive

1
“Fabulous Market for Food,” Fortune, October 1953, 137-9; quoted in Kellen Backer,
World War II and the Triumph of Industrialized Food, PhD Diss, University of
Wisconsin-Madison, 2012.
2
Harvey Levenstein, Paradox of Plenty: A Social History of Eating in Modern America,
(New York: Oxford UP, 1993): 109.
211
marketing campaigns for convenience foods that rang both those bells.3 But few, if any,

accounts of this period confront, head-on, the contradiction between the growth of the

food industry and the reigning, received wisdom about the poor quality of its products.4

Weren’t the 1950s a gastronomic nadir? Wasn’t postwar processed food just plain lousy?

Historians writing about the history of processed foods tend to wax dismal when

it comes to the flavor of these products. Laura Shapiro describes postwar processed foods

as reflecting “culinary values bred in the factory — blandness and uniformity, interrupted

by sudden jolts of novelty,” which nonetheless “became pleasing to many appetites,

3
On the social and cultural history of the postwar, see: Lizabeth Cohen, A Consumer’s
Republic: The Politics of Mass Consumption in Postwar America, (New York: Knopf,
2003); Elaine Tyler May, Homeward Bound: American Families in the Cold War Era,
(New York: Basic Books, 1988); Karal Ann Marling, As Seen on TV: The Visual Culture
of Everyday Life in the 1950s (Cambridge: Harvard UP, 1994). Tracy Deutsch has
argued that the shift to supermarkets and mass-retailing changed not only the kinds of
foods that were widely available, but also produced a new kind of passive, disengaged,
depoliticized consumer. See Tracy Deutsch, Building a Housewife’s Paradise: Gender,
Politics, and American Grocery Stores in the Twentieth Century (Chapel Hill: UNC
Press, 2010). For scholars who have probed the tensions, contradictions, and
complications in middle-class women’s social roles in the postwar, the modern kitchen
well-stocked with convenience foods has emerged as particularly fraught territory. See:
Ruth Schwartz Cowan, More Work for Mother: The Ironies of Household Technologies
from the Open Hearth to the Microwave; Erika Endrijonas, “Processed Foods from
Scratch: Cooking for a Family in the 1950s,” in Sherrie A. Innes, ed. Kitchen Culture in
America: Popular Representations of Food, Gender, and Race, (Philadelphia: UPenn
Press, 2001); Katherine J. Parkin, Food is Love: Advertising and Gender Roles in Modern
America, (Philadelphia: UPenn Press, 2011).
4
One important exception can be found in the work of Rachel Laudan, a historian of
technology and of food, whose work scrupulously avoids declensionist narratives about
the state of food in the present, both by unseating the myth of an idealized “natural” food
past innocent of technoscience, and by making a serious accounting of both the
technological systems and the human labor that are necessary for all food production. See
Rachel Laudan, “A Plea for Culinary Modernism: Why We Should Love New, Fast,
Processed Food,” Gastronomica 1.1 (Winter 2001): 36-44.
212
while subtleties of flavor and texture lost their importance.”5 For Shapiro, this was not a

case of sudden-onset loss of discernment. She argues that the prior half-century of

factory-made food had a debilitating effect on American tastes, literally reshaping

consumer appetites and rendering them more complaisant. “During the first decades of

the twentieth century, millions of American palates adjusted to artificial flavors and then

welcomed them; and consumers started to let the food industry make a great many

decisions on matters of taste that people in the past had always made for themselves.”

And while Shapiro’s sensitive account of American postwar cooking and eating deftly

undermines the notion that consumers readily and passively accepted industrialized

foods, she nonetheless concedes that their acts of resistance were not on the grounds of

taste. “There wasn’t much the food industry could do to repel a nation that was already

stirring chopped tomatoes and pickles into Strawberry Jell-O for a Red Crest Salad.”6

This narrative sets the stage for an enlightened rump of Europeanized experts —

Julia Child, James Beard — to reeducate the American palate, and to reintroduce real,

“authentic” habits of cooking and eating, a mission that would be carried forward by

Alice Waters, the Slow Food movement, all the way to the locavore foodies of the

present day. Although Shapiro meticulously documents the differences between industry

and media representations of convenience foods and how middle-class American women

actually cooked and ate, she never examines her premise that processed foods were

5
Laura Shapiro, Something from the Oven: Reinventing Dinner in 1950s America, (New
York: Penguin, 2005): 56-7.
6
Shapiro 2005: 57.
213
inherently worse than other food options. In this narrative model, then, mass American

tastes, deranged by the food industry, are always in need of a redeemer.

Other historians have explained the lousiness of postwar processed food by

concluding that flavor was of little concern to the postwar food industry. Mark Schatzker,

describing the A&P’s Chicken of Tomorrow contest, which sought to breed bigger, more

efficient broiler chickens, asks: “How did these miracle chickens taste? No one knows.

The judges didn’t measure flavor. The point of the contest… was to create a chicken that

looked like a wax model.”7 Taking a similar tack, Harvey Levenstein laments that “the

so-called advances” in food processing after the war “were in economics of production,

not in taste. It was widely acknowledged that in practically all spheres taste had been a

casualty of processing.” When big business did acknowledge consumer preferences, it

was to disdain them; “food industry moguls had a generally low opinion of consumers’

taste buds,” he states.8 Further, there was little that was actually new in the “new and

improved” foods of the postwar; indeed, he argues, the food processing industries

“consistently ranked near the bottom in the proportion of sales invested in research and

7
Schatzker, Dorito Effect, Chapter 2. Schatzker’s insistence that the food industry
doesn’t care about “flavor” rests on a rigid nutritional and moral distinction between
“real” and “fake.” The substance of his argument is that the processes of industrialization
depleted the authentic flavor of “real” foods (meat, vegetables, fruits), while adding
synthetic, substitute flavor to processed foods; in this way, he says, the food industry uses
our innate, evolutionary attraction to flavor against us.
8
Levenstein 1993: 110-1.
214
development.”9 The pretense of novelty was just another aspect of these ersatz products’

sham appeal.10

At best, the Jell-O salads and TV dinners of the era get the nostalgia treatment —

evoked by garish reproductions in coffee-table books whose prose drips with fond

sarcasm, the lifeblood of kitsch. To know that it was all awful, and that we should laugh,

is to reassure ourselves not only of our own gastronomical sophistication, but also of the

integrity of our personal standards of taste. The earnest and deluded homemakers of the

1950s! Serving up that tasteless muck to eager, suit-clad husbands, to smiling,

wholesome children — and thinking that it’s good! When we distance ourselves from the

caricatured food of that era, we also distance ourselves from the possibility that we might

be similarly susceptible (misled by advertising, by the food industry, or by our own

unreliable appetites) to finding trashy food delicious. By expressing disgusted amusement

at the food of the 1950s, we perform a pantomime of “knowing better,” inoculating

ourselves against the destabilizing anxiety that we may not recognize bad food for what it

is.

9
Levenstein 1993:111.
10
Historians of technology have produced more nuanced accounts of the interplay
between technoscientific innovations and consumer appetites in food product
development in the postwar. See J.L. Anderson, “Lard to Lean: Making the Meat-Type
Hog in Post-World War II America,” in Belasco and Horowitz, eds. Food Chains: From
Farmyard to Shopping Cart, (Philadelphia: UPenn Press, 2009), 29-45; Paul Josephson,
"The Ocean's Hot Dog: The Development of the Fish Stick," Technology and Culture 49,
no 1 (January 2008): 41-61.
215
 Approaching postwar food from the perspective of flavor research and flavor

science tells quite a different story. Food manufacturers were well aware that the sensory

qualities of foods were affected by every aspect of food production, and, in the postwar,

were exquisitely concerned with improving the flavor of their products. Increasingly,

food manufacturers believed that flavor was the factor that made the difference between a

successful product and a flop. After the war, a growing and diverse group of experts

contributed to the knowledge, practices, materials, and technologies that shaped how food

was made to taste.

The food industry’s fixation on flavor during this period may have escaped the

notice of many previous scholars because the dynamics, disputes, controversies, and

challenges of shaping the sensory qualities of foods were largely addressed either

internally, before finished products made their way to supermarket shelves, or in the

context of intrabusiness relationships with producers of additives, packaging materials,

and processing machinery, where food manufacturers were the clients and customers. In

this regard, the food industry’s “investment” in research and product development —

which Levenstein dismisses as paltry — cannot be properly calculated without

acknowledging the substantial investments made by auxiliary industries and businesses

that served and supplied food manufacturers, the heterogeneous network of entities that

underwrote the integrity of the ‘food chain’ and comprised the totality of the food

production system.11 Although this chapter’s focus is on flavor manufacturers, one

11
For more on the notion of the “food chain,” and the heterogeneous networked assembly
of producers, manufacturers, institutions, technologies, knowledge, and capital that
216
should also consider the contributions of other chemical companies (that developed other

food additives meant to improve sensory qualities such as texture, preserve the

appearance of food, or forestall decay, as well as plastics and other packaging materials

that enhanced stability and improved shelf-life), companies that built processing and

filling machinery that preserved food qualities during manufacturing, packaging

companies, trucking and shipping companies, and manufacturers of commercial freezers

and refrigerators — not to mention the federal government, which, through the USDA

Agricultural Research Service, the US Army Quartermaster Food & Container Division,

and other scientific entities, undertook research directly intended to address problems

faced by the food industry related to the qualities of food. Food manufacturers were thus

the beneficiaries of huge public and corporate investments in technology, infrastructure,

and research all along the food chain, which helped underwrite and make possible the

development of new products, and which reflected the considerable attention devoted to

enhancing and improving the sensory qualities of foods which reached consumers.

What I hope to demonstrate in this chapter is not only that flavor mattered to the

food industry, but also how it mattered. By illuminating both the challenges and

opportunities that flavor offered to food manufacturers, and the role that flavor

companies played in developing flavor solutions for the food industry and its ramifying

comprise it, see Belasco and Horowitz, eds. Food Chains: From Farmyard to Shopping
Cart, (Philadelphia: UPenn Press, 2010).
217
consumer markets, I hope to provide a fuller picture of both “industrial taste” and its

meanings in postwar America.12

This chapter asks what research and development looked like at flavor companies,

using this question to examine the relationship between the flavor and food industries, as

well as the consequences of these investments for the way that foods were made to taste

in postwar America.

I begin by considering the commercial context for the increasingly close

relationship between food manufacturers and flavor companies in the postwar. What

factors drove food companies to become more “flavor conscious,” and to find technical

and material solutions in the flavor industry and its products? How did flavor companies

strategically leverage their research and development operations to integrate themselves

12
The concept of “industrial taste” as a set of qualities produced by industrial processing,
and distinct from the sensory possibilities of homemade foods, comes from Gabriella
Petrick, “The Arbiters of Taste: Producers, Consumers, and the Industrialization of Taste
in America, 1900-1960,” PhD Diss, UDelaware, 2006. Petrick examines the emergence
of this set of qualities in the early twentieth century food industry, categorizing it as the
“good enough” flavor that emerged as the result of a compromise between food safety
and food quality in canning and other processing. Another aspect of the story of
“industrial taste” can be found in Amy Bentley’s Inventing Baby Food: Taste, Health,
and the Industrialization of the American Diet (Oakland: University of California Press,
2014). Bentley argues that the flavors of processed foods are something that consumers
must develop an appetite for, and that the “early consumption of commercial baby food
may have helped to prime Americans’ palates for the highly processed industrialized
products that have contributed to our health problems today.” (p.6). While both of these
conditions — the negotiations between flavor and safety, the effects of familiarity and
exposure on shaping appetites and preferences — form important parts of the story of
processed foods, this chapter instead considers the deliberate design and development of
flavors and other sensory qualities of foods.
218
into the food product development and manufacturing process? By examining the

contours of this intrabusiness relationship, I show that flavor companies positioned

themselves as expert interpreters of both chemical materials and consumer markets,

savvy not only to the possible uses for the rapidly expanding list of available synthetic

flavor chemicals but also to the commercial potential of new kinds of products.

I then turn my attention to flavor additives themselves. Flavor additives are

technologies, deliberately designed artifacts whose complex composition reveals a

convergence of chemicals with diverse material, sociocultural, and scientific “life

histories.” How did particular chemicals come to be entangled with each other, brought

together to deliver certain sensory effects? What purposes were these technologies

designed to serve? How were they deployed in consumer products?

My attempts to answer these questions reveals the intricacies and breadth of

“flavor research” in the postwar period. As the flavor industry developed its technical

capabilities, it invested not only in the improvement of the sensory qualities of flavors

(the formulations of creative flavorists), but on the enhancement of flavor performance in

foods. Focusing on flavor performance meant considering factors related to the utilization

of flavoring materials during manufacturing — factors such as dispersability (how

uniformly a flavor could be distributed through a food matrix), reactivity with other

compounds in a food, and ability to withstand processing conditions — as well as to the

stability and durability of flavors in finished packaged foods. Often, optimizing flavors

for food manufacturing entailed attending to the material components of flavoring

products that were not, strictly speaking, “flavor chemicals.” Synthetic solvents,
219
emulsifying agents, vegetable gums, and related materials played an increasingly

important role in the production of flavoring products and processed foods, with chemical

companies such as Dow and Atlas Powder supplying these compounds to flavor and food

manufacturers. As such, flavoring additives were not only participants in, but

beneficiaries of, what historian Suzanne White has dubbed the “chemogastric

revolution,” the increasingly close association between the food and chemical industries

in the postwar U.S.13

My first case study concerns postwar pineapple flavor. I trace the dynamic set of

relationships among agricultural research, the production of petrochemical

intermediaries, and the utilization of these sources of chemical materials and knowledge

by flavor manufacturers. I then place this supply-side story in the context of market

demand, examining the cultural and social causes of growing pineapple-appetite among

postwar consumers.

But there’s more to the picture. Understanding the material substance of flavor

additives — the capabilities and affordances of these products as wholly designed

objects, i.e., not just what a flavor “tasted like” but how it was expected to perform — is

a crucial but overlooked part of the story of how foods were made to taste in the postwar

period. I include two stories here: one successful, one less so. First, the development of

encapsulated flavors, “spray-dried” flavor powders that were crucial components of

processed foods including cake mixes, frozen foods, and beverage mixes. Encapsulated

13
Suzanne White, “The Chemogastric Revolution and the Regulation of Food
Chemicals,” in Seymour H. Mauskopf, ed. Chemical Sciences in the Modern World,
(Philadelphia: UPenn Press, 1993); 322-55.
220
flavors “locked” volatile flavor compounds within stable, non-reactive containers of

vegetable gums and other chemical components, protecting them from the effects of time

and environment until the moment of preparation or consumption. The development and

widespread acceptance of these products profoundly shaped the sensory capabilities of

processed food in the postwar. Second, I examine a case where research and development

failed to realize commercial success. Givaudan’s Aerosol Research Laboratories

positioned the company as an industry leader, and a central node for the network of

manufacturers concerned with the development of “push-button” foods. Despite

substantial investments by Givaudan and others, the product category flopped with

consumers.

In my concluding story, I look at the production of flavorings for “nationality”

specialties, examining how the flavor industry facilitated a strategic shift from mass

markets to market segmentation.

I. “Your Flavor Problem Is Our Flavor Problem:”

Research and Development in the Flavor Industry

“Nothing Sells Like Flavor”

Supermarkets differed from earlier grocery stores not only because they were

organized around ideas of self-service, branded goods, and volume sales, but also

because of the dazzling variety of products that they carried. Cultivating the appearance

of limitless abundance was the defining style of the supermarket, as well as its business

221
strategy. In Allen Ginsberg’s 1955 poem, “A Supermarket in California,” the poet trails

an earlier bard of American plenitude, Walt Whitman, down the aisles of a Berkeley

supermarket, passing peaches, avocados, and “brilliant stacks of cans,” fancy artichokes

and “every frozen delicacy,” families and possible angels, sustained (but also depleted)

by it, possessing it all without consuming it, “never passing the cashier.”

Which came first, the supermarket, or the dizzying array of products to occupy

every inch of shelf space in these replete, orderly, vertiginous emporia? According to

Progressive Grocer, a trade magazine that compiled industry statistics, while a “good”

food store in 1928 might have stocked some 800 different items, by 1946 this number had

swelled to 3,000, and continued to climb. In 1955, a typical “well-stocked supermarket”

could be expected to carry 5,000 different items.14 By 1962, this number topped 6,000.15

Although this increase also reflected the inclusion of non-food items — such as cleaning

products, housewares, and toiletries — within the standard scope of supermarket goods, a

substantial portion was due to new food product lines, “primarily convenience foods

characterized by built-in maid and chef service.”16 New types of products appeared —

such as frozen foods, cake and mixes, diet foods, ethnic specialties, and new kinds of

baby food. Older, established brands also expanded their offerings with new sizes,

products, and flavors. While grocers welcomed these new packaged products, in part

because of their higher prices and greater margins, they also acknowledged that the

hypertrophic expansion of inventories could not continue forever. As was noted by

14
Facts in Grocery Distribution: Published as a Service to the Food Industry by
Progressive Grocer, the Magazine of Super Markets and Superettes (1956 edition): 6.
15
Facts in Grocery Distribution (1963 Edition): 3.
16
Facts in Grocery Distribution (1955 edition): 6-7.
222
Progressive Grocer: “Many retailers… are finding that inventory of a store cannot

expand indefinitely and as a result there was a greater weeding out of poor sellers than

ever before.”17

In the postwar era, merchandisers began attending to the sensory environment of

the supermarket, attuning its qualities in accordance with psychological research about

behavioral impulses and drives, such as that popularized by Ernst Dichter’s Institute for

Motivation Research.18 These strategies of sensual persuasion were thought to be

particularly effective on women. "Leaders in the supermarket business... deliberately

targeted what they saw as women's base physical desires,” writes historian Adam Mack,

“contending that female consumption derived not from rational calculations, but rather

from irrational 'impulses' encouraged by sellers who knew how to manipulate the female

sensory apparatus."19 Colorful displays, artful lighting, spacious floor-plans, softly piped

in music, strategies of odor control and design: the supermarket itself became an

invitation to desire and to buy, a plea made not through explicit advertising, but

implicitly, through sensory design.

Food manufacturers also began to understand flavor in terms of its psychological

and affective appeals to the consumer. A 1947 article by A.D. Hyde, General Mills’ vice

17
Facts in Grocery Distribution (1955 edition): 7.
18
Adam Mack, “'Speaking of Tomatoes': Supermarkets, the Senses, and Sexual Fantasy
in Modern America,” Journal of Social History 43:4 (Summer 2010): 815-842; Lawrence
R. Samuel, Freud on Madison Avenue: Motivational Research and Subliminal
Advertising in America, (Philadelphia: University of Pennsylvania Press, 2011). For a
comparable case study, where companies drew on psychological research to “put color to
work” to move merchandise by appealing to unconscious motives and drives (especially
those of women), see Blaszczyk 2012: 215-64.
19
Mack 2010: 817.
223
president for research, spotlighted the increasing importance of flavor in product design

and development.20 In Hyde’s account, supermarket aisles were the battleground in a war

of all against all, as products contended with each other for consumer favor in a

marketplace constrained by the inevitable limitations of human appetite. “A vegetable

soup,” Hyde observed, “must vie for the consumer’s dollar not only with other vegetable

soups, but with every other food.”21 In this overheated marketplace, where consumers

were free to choose between growing numbers of appealingly packaged items lining the

wide, well-lit aisles of self-service supermarkets, he contended that a product’s success or

failure depended largely on its flavor.

But it was no longer enough for flavor merely to be “appealing,” Hyde warned. It

also had to be unique and different, adding “a new ‘note’ to the ‘symphony’ that modern

families demand in their meals. If a new product tastes exactly like a dozen or so other

established foods, the housewife will have little incentive for buying it.” Further, the

flavor should be distinctive: easily identifiable and memorable. A flavor that was “readily

recognized” and “conjured mentally” had “inestimable value,” serving as “a built-in

trade-mark which will invariably be identified with its brand name and its producer.”22

The direct relationship between flavor and sales was reflected in the marketing

campaigns of flavor and fragrance firms. Flavor companies had long advertised the

advantages of their products as: cost, uniformity, compatibility with manufacturing

20
A.D. Hyde, “How General Mills Develops Its New Food Specialties,” Food Industries
19 (October 1947): 110-12, 212-15.
21
Hyde 1947: 212.
22
Hyde 1947: 213.
224
processes, and a sensible naturalism impossible to achieve with only natural materials.

Starting in the 1940s, and through the 1950s and beyond, they also touted the direct

relationship between flavor and sales in advertisements that proliferated in pages of trade

journals such as Food Engineering, Food Technology, and Food Product Development.

“Nothing Sells Like Flavor!” was flavor and fragrance manufacturer Fritzsche Brothers’

slogan in the late 1950s. A 1959 advertisement for Polak & Schwarz featured a

photograph of a woman, holding a child by the hand, in front of shelves full of different

cake mixes; she reaches for a box of cake mix perched on the very top shelf, high above

her head. “A sale is made,” the tagline read, “thanks to P&S flavors.”23 The message here

was that flavor could reliably connect with consumers to move a product, overcoming

disadvantages, such as poor placement in a grocery store, that may be beyond the

manufacturer’s control.

A 1956 advertisement for the flavor and fragrance firm Dodge & Olcott vividly

dramatized flavor’s role in winning customers:

“Your customer goes to the store and brings your food product home.
Packaging, promotion or impulse-buying may account for this first-time
sale. But you haven’t really sold her — not yet! You’ve just contacted her.
Only quality food with unique taste-appeal can be counted on to bring her
back again and again — and keep those registers ringing. Flavor goes out
of the store with your customer — it goes to the table and becomes in
essence your personal ‘door to door’ salesman. The final impression this

23
[Polak & Schwarz], “A Sale is Made,” [advertisement], Food Technology 13 (February
1959): 6.
225
 salesman creates decides the ultimate fate of your product. Let the D&O
Flavor Laboratories make your silent salesman, FLAVOR, the best you’ve
ever had!”24

Flavor was an agent that extended the food manufacturer’s control over the

customer to the most intimate realms of private life. While the inducements of

advertising, merchandising, and sales promotions ended at the border between public and

private, flavor crossed the ultimate threshold, carrying the manufacturer’s influence not

only into the home, but into the body itself, reliably yielding subsequent behavioral and

economic outcomes (bringing her back again and again, keeping those registers ringing.)

Flavor, then, was no longer chiefly a problem of standardization and quality

control for food manufacturers. It had become the “silent salesman,” the factor that could

make the difference between a product’s success and failure. Further, flavor design and

development was recognized as the domain of scientific experts, who had a mastery not

only of production processes and market conditions, but also of the growing number of

chemicals available to extend shelf life, and improve the texture, appearance, and flavor

of foods.25 Although large companies, such as General Mills, maintained flavor research

and development divisions in-house, smaller companies had to seek out these services

elsewhere.

24
[Advertisement, Dodge & Olcott], American Perfumer & Aromatics 67 (April 1965):
61.
25
White 1993.
226
 These commercial conditions, as well as the growing acceptance of flavor’s deep

role in shaping consumer behavior, set the stage for an increasingly close relationship

between food manufacturers and flavor companies. As we have already seen, since the

beginning of the twentieth century, flavor companies had offered direct technical

assistance and expertise to users of their products. In the postwar era, flavor companies

continued to tout their investments in research, and expanded laboratory and

manufacturing facilities that allowed them to produce not only an expanding variety of

flavor effects, but to offer flavor additives in new material forms with new performative

capabilities.

“Your Flavor Problem Is Our Flavor Problem:” Research

and Development in the Flavor Industry
After the war, flavor manufacturers increasingly emphasized technical assistance

and in-house research programs, as well as specialized product lines intended for specific

applications. This is reflected in the organization of flavor catalogs and price lists. Prior

to the Second World War, companies that manufactured both flavors and fragrances —

such as Dodge & Olcott, Givaudan-Delawanna, and Fritzsche Brothers — tended to

publish catalogs that included merchandise in both categories. Generally speaking,

flavors took a back seat to perfume products and essential oils, which commanded higher

prices and more catalog pages. After the war, these companies and others began to

publish separate catalogs flavors and fragrances, which not only accommodated an

expanded selection of flavoring products, but also allowed for a more acute targeting of

specific groups of flavor users.

227
 “Your flavor problem is our flavor problem,” Givaudan-Delawanna’s 1949

catalog assured manufacturers. “Let the Givaudan Flavor Research Laboratories assist

you in its solution.” Givaudan, a venerable fragrance and flavor firm with corporate

headquarters in Switzerland, had manufactured aromatic materials and products at its

Delawanna, New Jersey facility since 1924; however, this was only the company’s

second catalog devoted exclusively to its flavoring products. Although the flavors listed

in the catalog “have been carefully created for specific purposes, your product or

manufacturing process may require special study. Aided by years of experience and a

wide range of raw materials, our Flavor Research Laboratories — with its technical sales

staff and skilled chemists and technicians — will thoroughly investigate your product, in

order to develop the flavor ideally suited for your needs.”26

In other words, Givaudan was offering to place their research and development

capabilities at the service of food and beverage manufacturers. In the late 1940s and

1950s, many flavor and fragrance companies redoubled their commitment to flavor

research and development — expanding research facilities, hiring new personnel,

building dedicated laboratories for specific product applications (such as spray-dried

flavors and flavors for aerosol foods), and assembling in-house taste panels to evaluate

materials and products. For companies such as Givaudan, where fragrance materials had

long dominated, this reflected a bet on the continuing growth of the market for flavoring

26
Givaudan Flavors, Inc. “Flavors: Catalogue Number 2,” (New York, NY: Givaudan,
1949): [interior back cover]. Smithsonian Libraries, Trade Literature Collection,
Washington, D.C.
228
additives, driven in part by new types of food products that would need specialized

flavorings.

Givaudan’s increasing investment in flavors is evident through the 1950s and

1960s. Givaudan hired three new flavor chemists in 1952 — younger, American-born

flavorists James Broderick, Earl Merwin, and Jerry DiGenova — to supplement their

existing staff of two older, European-trained flavor chemists.27 Beginning in 1953, the

company began publishing, on a more or less quarterly basis, the Givaudan Flavorist, an

eight-page newsletter for the beverage and food manufacturers that were its clients. The

Flavorist described the latest research in flavor chemistry, promoted particular product

lines, and made a sustained case for the professional and scientific status of the flavor

industry, its complexity, and the importance of leaving flavor problems to the experts

rather than handling them in-house. “The diverse nature of flavors requires the full-time

energy of many flavor and allied specialists,” was explained in an article about the flavor

of coconut, published in 1954. “It is our purpose in The Flavorist to keep our readers,

who are forced to relegate flavor-development to a secondary role, abreast of the

development and trends in the field.”28 This is a recapitulation of a familiar promise that

had been made since the early twentieth century, but with the intensification of

27
It should be noted that these younger flavorists did not feel that the company was fully
behind them, or fully invested in scientific flavor research and development, at this point
in the 1950s. Broderick soon left the company, followed a few years later by Merwin;
DiGenova remained at Givaudan for the remainder of his career, eventually becoming
vice president of the company’s creative laboratories. Further information about this can
be found in Chapter 7. [E.S. Merwin], A Short History of the Flavor Industry, prepared
for the Society of Flavor Chemists and the Chemical Sources Association, 1994: 47.
28
“The Flavor of Coconut,” Givaudan Flavorist 2.1 (1954): 2.
229
technoscientific control over all aspects of food manufacturing, it gained even more

force.

In an another article appearing later that year, the Flavorist described the

coordinated network of scientific labor that distinguished the modern flavor and

fragrance company from the flavoring supply houses of the past, which, by its account,

relied on closely guarded secret formulas and performed little chemical research.

“Today’s aromatic material organizations reek of laboratories and eager young men fired

with the zeal of their college inheritance – pushing away the secrecy and romance, doing

things scientifically.” Developing a flavor at Givaudan required the work not only of

flavor chemists, but also of the organic research laboratory, toxicological laboratory,

analytic laboratory, and control laboratory, as well as an “elaborate purchasing

department who have world-wide connections for spotting new materials and sources.”

Meanwhile, “in the background are the chemists who process the intermediates, the

engineers who develop new equipment and the maintenance staff who keep the

equipment and processes rolling.”29 As powerful, new analytic chemical technologies —

such as gas chromatographs, mass spectrometers, and infrared spectrometry — became

available, flavor companies invested in these machines. A two-page photo-essay

appearing in a 1958 issue of the Flavorist featured images of male and female

technicians, garbed in white lab coats, at an array of instruments necessary to make a

“Mona Lisa” in the flavor lab — including a gas chromatograph, a Beckman recording

spectrophotometer, a multiple reflux assembly, infrared recording spectrophotometer, and

29
“Ol’ Doc,” Givaudan Flavorist 2.3 (1954): 3.
230
“the latest model refractometer” — taking care to also point out that it was not the

machines alone, but the combination of advanced instrumentation and specialized skill

that made it possible to make a flavor masterpiece.30

Givaudan upgraded its flavor laboratories and testing facilities again in 1959 to

include a “testing kitchen which would be the envy of any housewife,” staffed by home

economists who used flavor formulations in candies, baked goods, and other foods.

Flavorists’ efforts were assessed by both trained taste panels and consumer panels who

“evalute[d] the effectiveness of the flavor in the finished media.”31 By investing in these

types of facilities and procedures, flavor companies like Givaudan hoped not only to

close the gap between their products and ultimate consumer market acceptance, but to

interject themselves even more deeply within food manufacturers’ product development

process. As a 1968 Arthur D. Little, Inc. report on the flavor industry explained, long-

term success meant having a “particular flavoring formulation locked-in to the final

product formulation,” which would almost certainly oblige a manufacturer to continue

relying on the company for the flavoring. This led companies such as Givaudan to invest

increasing resources on technical services, and research into applications and consumer

30
“A Mona Lisa in the Making,” Givaudan Flavorist 6.2 (1958): 3-4. The claim that
expertise in flavor involves not only instruments but also specialized skills is explored at
length in Chapter 7.
31
“New Laboratories, New Testing Kitchen for Flavor Development Work at Givaudan,”
Food Technology 13 (May 1959): 48.
231
responses — work that would have formerly been conducted by food or beverage

manufacturers.32

Givaudan’s expansion of its research facilities was not unique, but part of an

industry-wide trend. In 1953, Dodge & Olcott touted that its new building on Varick

Street in lower Manhattan housed a product development department as well as nine new

flavor and fragrance laboratories, including organic synthesis and analytical laboratories

and dedicated laboratories equipped to study technical flavor problems in processed

foods, confectionery, beverage, and pharmaceuticals. The company also devoted 5,200

square feet of floor space in the building to its flavor compounding laboratories,

adjoining a 4,500 square foot area for perfume compounding on the second floor.33 The

same article describing Dodge & Olcott’s new facilities also noted that the company

“tests its new flavors through an employee ‘taste panel.34’” A series of advertisements for

Fritzsche Brothers, in Food Technology in 1951 and 1952, spotlighted the various

“branch[es] of the food field” — baked goods, frozen desserts, salad dressings, luncheon

meats — that had benefited from the work of the company’s Flavor Research

Laboratories “to develop improved ingredients for tickling the consumer’s palate.35” This

was the company’s basic message to food manufacturers: “Whatever your food product

32
Arthur D. Little, Inc., “The US Flavor Industry: Report to The Andrew Jergens
Company,” (March 26, 1968): 29. AW Noling Beverage Literature Collection, University
of California, Davis.
33
“Seeking Sweet Smells,” Chemical & Engineering News 31 (December 28, 1953):
5350.
34
“Seeking Sweet Smells”: 5350.
35
See, for instance, in Food Technology 5 (1951) and 6 (1952): Baked goods, in January
1951: 18; Frozen desserts, March 1951: 25; salad dressings, May 1951: 28; and
frankfurters, June 1952: 40. Quote from March 1951.
232
— whatever the flavor problem it involves — it is more than probable that our

laboratories have done the basic research that will enable us to supply a quick solution to

your needs.”36

Accelerating a process that had begun in the interwar years, in the postwar

decades, flavor manufacturers expanded and diversified their research and development

capabilities, making it possible to provide more targeted technical support to food and

beverage manufacturers, and increasingly focusing their business on developing

specialized flavoring formulations for specific needs rather than the production of

commodity flavorings.37

II. The Design and Development of Flavor

Technologies
New Flavomatics for the Flavor Industry: Making a Better
Pineapple Flavor

Flavor manufacturer’s postwar claims to the value added by their specialized

workforce of flavorists was corroborated by the increasing material complexity of flavor

work, which necessitated the mastery of an ever-increasing number of ‘flavomatics,’

chemical compounds with potential use in foods. The identification of new compounds in

nature, and the synthesis of entirely novel chemicals, for use in flavorings and fragrances

36
[Fritzsche Bros.] “It’s Only as Good as Its Flavor,” Food Technology 6 (June 1952):
40.
37
Arthur D. Little, Inc. “The US Flavor Industry” 1968. AW Noling Beverage Literature
Collection, University of California, Davis.
233
exploded with technologies of analysis, such as gas chromatography and mass

spectrometry. The use of these compounds in foods was only nominally curtailed by the

1958 Food Additives Amendment, which required chemical additives to prove their

safety before being permitted in the food supply.38 The rapid expansion of available

synthetic flavoring materials preceded the introduction of analytic instruments. Although

new instrumental technologies speeded the pace by which promising new molecules were

isolated and identified, the pattern of producing synthetic molecules for new flavor

effects had been set decades earlier.

Some new flavoring materials and commercial formulations were drawn from

basic research into the chemistry of foods, including at the USDA. For instance, in the

early 1920s, chemists in the Bureau of Chemistry analyzed the chemical constituents of

ripe apples, identifying a handful of esters and alcohols.39 In addition to being published

in scientific literature, this information was made available as a public patent for

“synthetic apple-oil.”40 There is evidence that at least one company— Fritzsche Brothers

— used this as the basis for its own apple flavor formulation in the 1920s and 1930s.41

38
On the 1958 law and its effects on the flavor industry, See Patrick van Zwanenberg and
Erik Millstone, “Taste and Power: The Flavouring Industry and Flavour Additive
Regulation,” Science as Culture (2014): 1-28. Discussed further in Chapter 7.
39
Frederick B. Power and Victor K. Chesnut, “The Odorous Constituents of Apples,”
Journal of the American Chemical Society 43.7 (July 1921): 1725-1739; Frederick B.
Power and Victor C. Chesnut, “The Odorous Constituents of Apples II: Evidence of the
Presence of Geraniol,” Journal of the American Chemical Society 44 (1922): 2938-2942.
40
Patent 1,366,541, Frederick B. Power and Victor K Chesnut, “Synthetic Apple-Oil,”
(January 25, 1921); Patent 1,436,290, Frederick Belding Power and Victor King Chesnut,
“Improved Synthetic Apple Oil,” (Nov 21, 1922).
41
[Letters] Between J.N. Farley, Farley Confections, Chicago, and E.K. Nelson, Senior
Chemist, Food Research Division, Bureau of Chemistry and Soils, USDA, Dec 8, 1937 -
Jan 4, 1938. [National Archives RU 88, Records of the Food and Drug Administration
234
The Second World War also drove basic research into the flavor chemistry of spices and

essential oils, as disrupted trade caused shortages, spurring a search for synthetic

substitutes. 42

New flavor materials entered the food supply not only as synthetic replicas of

compounds identified in nature, but also as entirely novel substances, with no known

natural analogues. As materials, techniques, and knowledge relating to flavor chemistry

passed between scientists working in distinct institutional contexts — agricultural

research in academic or government laboratories, flavor research and development in

private industry laboratories— they put this knowledge to work in different ways. The

development of imitation pineapple flavors after the Second World War provides a vivid

illustration of this. Forces on both the supply side (the flux of available chemical

intermediaries and research funds), and on the demand side (the cultural milieu, or

“market opportunities,” that the flavor would inhabit), shaped how a particular set of

molecules came to be bound together and associated with the taste of pineapple in the

years after the war.

Correspondence and Reports, 1897-1938, Box 44]. It is unclear whether the apple
flavoring Farley acquired from Fritzsche (‘Fritzsbro Arome-Apple’) was the one whose
formula was based on the public patent; in any case, Farley found it quite unsatisfactory.
“We think these are some of the old fashioned ether flavors as they do not seem to have
the characteristics for which we are looking,” he complained to Nelson. Notably, the
1921 and 1922 patents based on the USDA’s apple research contained mainly esters
(which were sometimes referred to as “ethers”), as these had been the compounds that the
investigators had been able to identify.
42
“Chemistry Supplies Synthetic Food Flavors as War Curtails Imports of Exotic Herbs,”
New York Times, (May 16, 1942): 9; J.N. Taylor, “Sales of Synthetic Savors,” Journal of
Chemical Education 1944; “Black Pepper and Cardamom Replacements Announced,”
American Perfumer (April 1950): 307-8.
235
 “Pineapple” was one of the earliest synthetic flavors. In the 1850s and 1860s,

ethyl butyrate and other esters generally performed the role of pineapple in candies and

beverages. (It is highly probable that, for most Western consumers well into the twentieth

century, the pineapple flavor of these esters was more familiar than the acid tang of the

prickly fruit itself, which was more often consumed canned than fresh.) With time,

pineapple flavor formulations began to include a growing list of chemical compounds. By

the late 1930s, the allyl esters had become popular in synthetic pineapple flavors — in

particular, allyl caproate, which was sometimes sold under the name “Pineapple

Aldehyde.”43 None of these molecules had been uncovered by basic research into

pineapple flavor chemistry.

In the early 1940s, Dr. Arie Haagen-Smit, a biochemist in the William G.

Kerchikoff Laboratories of the Biological Sciences at the California Institute of

Technology, undertook a study of the flavor chemistry of pineapple at the request of the

Pineapple Research Institute, an industry group sponsored by eight Hawaiian pineapple

companies. The companies, which grew, processed, and canned much of the pineapple

sold in the United States, had come to believe that fundamental knowledge about the

chemical constituents of pineapple flavor could be used to improve breeding, cultivation,

43
Although allyl caproate was an ester, not an aldehyde, this trade name reflects a
naming convention in the flavor materials market that dates to the 1910s, if not earlier.
Particularly potent synthetics, which provided a characteristic note at a low
concentration, were dubbed “aldehydes,” possibly to obscure their true molecular
composition and, at least initially, prevent rival companies from producing them. Hence,
‘peach aldehyde’ is a lactone, ‘strawberry aldehyde’ is an ester.
236
and canning processes.44 Caltech, at the time, was a center for research on plant

biochemistry, and Haagen-Smit was known for his pioneering work on plant growth

hormones.45

Haagen-Smit and his colleagues started with six thousand pounds of the fruit,

from which they distilled a few ounces of “volatile product which had the typical

pineapple smell.” After distilling off ethyl alcohol and acetaldehyde, which comprised

the majority of the solution, they used techniques of chemical microanalysis,

fractionating the remaining grams of solution to identify “the substances more specific

for the pineapple flavor.” They found that the mixture consisted of various known ethyl

and methyl esters of acids, as well as a previously unknown sulfur-containing compound

which they identified as methyl beta-methylthiolpropionate.46 They confirmed this

identification by synthesis.

“While our research was not intended as a means of obtaining a better artificial

pineapple flavor,” wrote Haagen-Smit, “the results of our analysis would naturally lead to

improved flavor formulae. For, after isolating the flavor principles, and determining their

structure, it was possible to reconstruct the flavor chemically.”47 Haagen-Smit, whose

44
Arie J. Haagen-Smit, “The Chemistry of Flavor,” Engineering and Science Monthly 12
(January 1949): 5.
45
James Bonner, “Arie Jan Haagen-Smit,” National Academy of Sciences Biographical
Memoirs vol 58 (Washington, DC: National Academy Press, 1989): 191-2.
46
A.J. Haagen-Smit, Justus G. Kirchner, Clara L. Deasy, and Arthur N. Prater,
“Chemical Studies of Pineapple (Ananas sativus Lindl). II. Isolation and Identification of
a Sulfur-Containing Ester in Pineapple,” Journal of The American Chemical Society 67,
no 10 (1945): 1651-2.
47
Arie J. Haagen-Smit, “The Chemistry of Flavor,” Engineering and Science Monthly 12
(January 1949): 5.
237
earlier work had focused on the effects of endogenous chemicals on plant growth,

understood flavor molecules not as commercial end-products, but as the outcome of

metabolic processes within the plant. His interest was in the development of flavor

molecules from chemical precursors as the fruit grew and ripened. For him, this basic

chemical knowledge had a practical application, as it could substitute “for the subjective

scale of grading used at present in the fruit industry. In this way, the effects of

climatological factors, changes in agricultural methods, and the results of breeding

experiments may be investigated.”48 In other words, for Haagen-Smit and his colleagues,

the flavor chemistry of pineapple would primarily be applied to growing and selecting

tastier pineapples, with methyl beta-methylthiolpropionate and other compounds serving

as material indices to flavor quality.

How was this knowledge put to use in the flavor industry? One of the tasks of

flavor company research departments was to review the scientific and chemical literature,

staying abreast of discoveries that may yield commercial applications in food and

beverage flavorings. Soon after the publication of Haagen-Smit’s article, flavor and

fragrance manufacturers began synthesizing the sulfur-containing molecule, and

experimenting with it in pineapple flavor formulations. One such firm was F. Ritter &

Company, of Los Angeles. Ritter specialized in supplying aromatic materials — essential

oils, natural isolates, and synthetic chemicals — to other companies in the flavor

industry, rather than selling finished flavorings to food manufacturers. In the years after

48
Haagen-Smit et al. 1945: 1646.
238
the war, Ritter maintained a productive chemical research program, focused on the

synthesis of novel flavor and fragrance compounds.49

In a 1949 article reviewing dozens of newly available aromatic chemicals with

“odor and flavor promise,” Abraham Seldner, research director at Ritter, cited Haagen-

Smit’s recent discoveries about the flavor chemistry of pineapple.50 Chemists at Ritter

had synthesized methyl beta-methylthiopropionate, and had been assessing its potential

by adding small quantities to pineapple ester blends, resulting in “an imitation pineapple

reproduction closer than any others previously attempted.”51 However, Seldner and his

team did not use Haagen-Smit’s chemical identifications as a blueprint for their

laboratory recreation of “natural” pineapple flavor. “It has been many years since flavor

and perfume chemists have limited themselves to the reproduction of chemical bodies

found in nature,” he wrote. “Many modifications in flavor and odor can be worked out in

the laboratory by synthesizing materials not known to be present naturally.”52 In the case

of pineapple flavor, chemists at Ritter had created several new molecules that could

enhance to the flavor of pineapple or extend its shelf-life. Among them, Seldner

recommended two “outstanding pineapple modifiers” the company had developed —

49
Alexander Katz and Abraham Seldner, “California Essential Oil Development,”
American Perfumer & Aromatics 57 (May 1951): 357-60; “New Aromatics to Enhance
Fruit Flavors,” American Perfumer & Aromatics 55 (January 1950): 45; Katz, “Newly
Developed Flavoring Aromatics,” American Perfumer & Aromatics 68 (September
1956): 66-70; Katz, “Newly Developed Flavoring Aromatics,” Perfumer & Essential Oil
Review 48 (March 1957): 131-4.
50
Abraham Seldner, “New Aromatics for Flavoring and Perfume Industries,” American
Perfumer & Essential Oil Review 54 (October 1949): 295-6.
51
Seldner 1949: 295.
52
Seldner 1949: 295.
239
allyl phenoxyacetate and allyl cyclohexanepropionate, both of which Ritter could supply

in commercial quantities.53

A 1957 article from the Givaudan Flavorist about pineapple flavors sheds a bit

more light on how flavorists worked with new knowledge and new materials to formulate

flavorings. When Haagen-Smit’s identification of methyl beta-methylthioprionate was

reported, “it was hoped by all who read these papers that here at last was a ‘pineapple

aldehyde’ which could be produced synthetically and which actually gave the pineapple

flavor its nature,” a single chemical key that could cracking the sensory riddle of

pineapple.54 But although it was useful in very small amounts, “it was evident that this

chemical was not the key to the natural flavor of pineapple.” Further, “the instability of it

chemically and organoleptically limited its use.” On the other hand, allyl

cyclohexaneproprionate, one of the aforementioned pineapple modifiers, had never been

found in nature, but had proven its usefulness pineapple flavors. “These two modern

flavomatics” — the nature-identical synthetic, and the unprecedented molecule — “have

been added to the repertoire of the flavor chemist,” the article continued, “and have

enabled him to produce a more accurate synthetic version of pineapple flavor.” There was

no simple formula or single compound that was the key to a successful pineapple flavor;

the accuracy of the reproduction was not dependent on its molecular indistinguishability

from the original.55 The article concluded with a plea: “The creation of flavors should be

53
Seldner 1949: 296.
54
“The Flavor of Pineapple,” Givaudan Flavorist 5.2 (1957): 2.
55
Indeed, there was no unitary “pineapple” flavor. The article observed that “actually
pineapple is known by two flavors” — canned pineapple, and fresh pineapple. “Each
240
left to those who not only have the necessary training, but also have at their disposal the

varied raw materials and the research facilities to accomplish the desired end product.”

Professional flavorists, and the flavor companies that employed them, were necessary to

make chemical knowledge and materials into “safe, modern instruments for giving your

products distinctive taste appeal.”56

But where did this “constantly increasing… greater variety of aromatic

chemicals,” these new modern flavomatics, come from? These new materials were

intimately bound up with the shift from coal to petroleum as the primary feedstock for

organic synthesis just prior to the Second World War, and the concomitant growth of

petrochemicals and their products — polymers and plastics — during the war and after.57

Before the war, chemists in the flavor industry had generally used coal-tar-derived

chemicals, such as toluene, benzene, and naphtalene, as the basis for many of their

synthetic processes. After the war, an expanding range of available chemical

intermediaries broadened the molecular scope of synthetic possibilities for flavor

manufacturers.58 “When an intermediate is developed and priced to fit into the plastics

field,” explained Seldner in 1949, “it almost automatically qualifies for use in the

aromatics industry.”59 Indeed, “the constant stream of new intermediates being developed

flavor requires its own particular combination of flavor materials and each has
ingredients which do not occur in the other.”
56
“The Flavor of Pineapple” 1957: 2
57
Ralph Landau and Nathan Rosenberg, “Succesful Commercialization in the Chemical
Process Industries,” in Rosenberg, Landau, and David C. Mowery, eds. Technology and
The Wealth of Nations (Palo Alto: Stanford UP, 1992): 90-98.
58
Alexander Katz, “The Development of Aromatics,” American Perfumer & Aromatics
68 (July 1956): 31-2, 35.
59
Seldner 1949: 295.
241
by the chemical industry” were “perhaps the largest single source of new aromatics.”60

For instance, allyl phenoxyacetate and cyclohexanepropionate, the new pineapple flavor

enhancers Seldner had recommended, were both produced by esterification from

methallyl alcohol, a petrochemical sold by Shell, among other companies.61 No longer

“expensive laboratory curiosities,” as the production of these intermediaries was scaled

up for plastics and other large chemical industries, their cost went down for all users,

including flavor manufacturers.62

On the demand side of the equation, a hunger for the flavor of pineapple was

likely sharpened by the postwar fascination with Hawai’i and the South Pacific islands.

The pineapple had been closely linked with the Hawai’i since the first decades of the

twentieth century, when the Hawaiian Pineapple Growers Association began an

aggressive and sustained advertising campaign to promote the canned fruit among

American consumers. Their stated goal was to “make the word ‘Hawaiian’ mean to

pineapple what Havana meant to tobacco,” and soften the reputation of the prickly,

eccentric fruit (which was considered tough, stringy, and bitingly acidic by early

twentieth-century consumers) by associating it with the lush imagined pleasures of the

Pacific island paradise.63 Pearl Harbor, and the American military’s actions in the Pacific

theater, turned these geographies into sites of intense and conflicted interest. After the

war, even as they remained heavily militarized zones, Hawaii and the South Pacific

60
Seldner 1949: 295.
61
Harry Cohen, “Foods, Flavors, and Aromatics,” American Perfumer 56 (July 1950):
47-49.
62
Seldner 1949.
63
Quoted in Gary Y. Okihiro, Pineapple Culture: A History of the Tropical and the
Temperate Zones, (Berkeley: University of California Press, 2009): 144.
242
absorbed cultural longings for a pre-Atomic place of redemption and primitivist

replenishment, one which found expression in spectacles such as Thor Heyerdahl’s 1947

Kon Tiki voyage, as well as in consumer fads such as backyard luaus, Tiki drinks, aloha

shirts, hula girl iconography, and exotica music.64

In this context, when manufacturers expanded food product lines to include new

pineapple flavors — with items such as pineapple-flavored Royal Gelatin (introduced

1948), Borden’s coconut-pineapple ice cream (1949), Reiss’s “Pineapple Confetti” ice

cream (1957), and Jell-O’s pineapple cream instant pudding mix (1960) — they were

capitalizing on the pineapple’s social and sensual association with exotic indulgence:

pineapple-appetites that were the poignant reconfigurations of Cold War geopolitical

anxieties as well as the cravings of postwar prosperity. They were also relying on the

generative and creative capacities of America’s allied chemical industries, which

promised not only “better things for better living,” but also delivered these goods along

with an increasingly sophisticated and intensively designed array of sensory effects,

chemical mnemonics for absent fruits, intensified pleasures as fantastic as they were real.

64
Kaori O’Connor, “The Hawaiian Luau,” Food, Culture & Society 11.2 (2008): 166;
Andrew Cowell, “The Apocalypse of Paradise and the Salvation of the West: Nightmare
Visions of the Future in the Pacific Eden,” Journal of Cultural Studies 13.1 (1999): 138-
60; Janet Borgerson and Jonathan Schroeder, “The Lure of Paradise: Marketing the
Retro-Escape of Hawaii,” in Stephen Brown and John F. Sherry, Jr., eds. Time, Space,
and the Market: Retroscapes Rising, (London and NY: Routledge, 2003): 219-236;
Elizabeth Buck, Paradise Remade: The Politics of Culture and History in Hawaii,
(Philadelphia: Temple UP, 1993).
243
SPRAY-DRIED FLAVORS: SOLVING THE PROBLEM OF FLAVOR

LOSS

The volatility of flavor — its tendency to “bake out,” to fade over time, to vanish

in the wind — had long been a matter of concern for food manufacturers and consumers.

A 1935 Housekeeper’s Chat — the radio program produced by the USDA’s Bureau of

Home Economics — offered the Depression-era “thrifty housekeeper” scientific experts’

advice on storing and cooking foods to preserve and develop “good natural flavors.”65

Calling Americans “a careless and wasteful race when it comes to flavor,” the program

lamented: “Every day we let millions of dollars’ worth of taste leave our kettles in steam,

or go down sink drains, or be spoiled by too much heat or too long cooking.” Expensive

seasonings and sauces were then required to “pep up… abused foods.” The smart,

scientific housekeeper could save money and improve the quality of her family’s diet by

taking steps such as keeping fresh fruits and vegetables in a cool place and cooking them

only briefly, or searing meat before adding it to roasts or stews. In this presentation,

flavor was not an abstract quality of foods, but a material resource to be conserved by

technical means. This Bureau of Home Economics’ advice reflected an increasing

economization of flavor, a growing tendency to cast its value in cash terms.

Food manufacturers also recognized flavor loss as an expensive problem, and

sought technoscientific solutions on a vaster scale than those prescribed by home

economists to the home cook. A 1934 editorial in Food Industries — fittingly titled “Save

the Volatiles”— urged manufacturers to set trained chemists and engineers to the task of

65
"Saving Flavor: Information from the Bureau of Home Economics, USDA" USDA
Radio Service Office of Information, Housekeepers' Chat, Monday, January 28, 1935.
244
finding ways to retain “that part of our food which adds to the zest of eating,” the flavors

and aromas that under current production methods were “most certainly being volatilized

and cast into the atmosphere.”66 Worrying that Depression-era social and economic

forces would lead to more home cooking, the editorial asserted that the only way to

sustain housewives’ loyalty to factory-made goods was to “produce better-tasting foods

than can be prepared in the kitchen at home.” And the way to do this was by

technologically surpassing the kitchen, capturing or retaining volatile flavor molecules

with the aid of machines that were anything but domestic — such as “closed vessels

equipped with reflux condensors, or evaporation carried out as fractional distillation,” for

instance. “Those industries which involve cooking, boiling or evaporation, with

noticeable losses of delightful flavors and aromas in the atmosphere, should consider

carefully what the food would taste like if they were retained.”67

The close of the Second World War not only brought an end to food rationing and

a new era of American prosperity, but a host of new technologies and methods in food

processing and packaging that protected (or did less damage) to the flavor of foods.68 For

instance, enhancements in the heat-processing of canned foods, flash-pasteurization of

milk and citrus juices, low-temperature vacuum-drying, and new packaging materials all

minimized the loss or change of volatile, reactive flavor chemicals during food

66
“Save the Volatiles,” Food Industries 6 (November 1934): 485.
67
“Save the Volatiles” 1934: 485.
68
For a compelling account of the role of US military investment in food technology and
food processing in creating the postwar industrialized food landscape, see Kellen Backer,
“World War II and the Triumph of Industrialized Food,” PhD Diss. University of
Wisconsin-Madison, 2012.
245
production.69 Yet flavor loss, and associated changes to food quality, remained a vexing

problem for manufacturers, given the variety of insults endured by food products on their

journey from factory to consumer. The effects of inconsistent storage conditions on the

“eating qualities” of food were a subject of particular concern. General Mills, for

instance, used a temperature- and humidity-controlled cabinet, dubbed the “weather

room,” to simulate the changing climate of a grocer’s shelf — the heat of the busy day,

followed by the coolness of night in the quiet hours after closing — studying the effects

of these conditions on packaging and the sensory quality of its formulations.70

As has been discussed, flavor companies utilized a growing variety of chemical

materials to create additives that gave manufacturers an expanding range of options when

it came to designating the sensory qualities of their products. But what of those products

when they left the factory, and entered the unpredictable conditions of the distribution

chain? When developing new additives, flavor companies attended not only to flavor

variety, but also to performance — integration into manufacturing methods, stability,

durability — producing flavor technologies that helped extend manufacturers’ control

over the sensible qualities of their products until the very moment of consumption.

Consider the cake mix. First introduced in the 1930s, packaged cake mixes

offered reliability and convenience to home cooks, as well as a way for manufacturers to

address lagging sales of flour. The market for these mixes was middling and mostly

regional until 1948, when General Mills introduced its Betty Crocker Gingercake mix,

69
E.C. Crocker, “A Flavorist Views Food Processing,” Food Industries 17 (March 1945):
69-71, 170-4.
70
“Environment by Humidistat and Thermostat,” Progress Thru Research (Fall 1953): 7.
246
and Pillsbury came out with its own boxes of white cake and chocolate cake. Other

national brands — including Swans Down, General Foods, and Nebraska Consolidated

Mills, which sold its mixes under the name of Duncan Hines — soon followed suit, in a

panoply of different colors and flavors.71 Sales of cake mixes more than doubled in six

years, topping $180 million in 1953, and continuing to grow (albeit at a slower rate) for

the remainder of the decade.72

Cake mixes have come to symbolize the compromised conveniences of 1950s

processed foods. The lore that these products were saved from initial poor sales by

reformulating them to require the addition a fresh egg as a sop to the housewife’s guilt

over her lax standard of care in the kitchen, as prescribed by Ernst Dichter’s Institute for

Motivation Research, has entered marketing gospel. (The reformulation is likely to have

had more to do with challenges in producing dried eggs with acceptable flavor.)73 But an

overlooked key to understanding the proliferation of cake mixes and other dry mixes in

the 1950s lies within the flavor industry, and with the concurrent introduction of a new

product category: spray-dried flavors.

First introduced by American flavor companies in the early 1950s, spray-dried, or

encapsulated, flavors, were a key technology for a food system where products were

expected to tarry for increasing lengths of time on supermarket or pantry shelves. Spray-

dried flavors promised to keep flavor from loss and change until the moment of

consumption, playing a central role in shaping the sensory experience of many products

71
A detailed history of the cake mix can be found in Shapiro 2005: 68-73.
72
Shapiro 2005: 73.
73
Shapiro 2005: 75-7.
247
that came to define the postwar pantry: not only cake mixes, but beverage mixes such as

General Foods’ Kool-Aid and Kraft’s Tang, instant soup mixes, instant puddings and pie

fillings, and frozen foods, as well as chewing gums and pharmaceutical products. As with

many of the new products that featured in the postwar food system, the development and

refinement of spray-dried flavors was catalyzed by Army food research.

As Susanne Freidberg has shown, “freshness” is a quality produced and defined

by food system technologies, from refrigerated rail cars to Frigidaires.74 Although the

meaning of “fresh food” has changed with technologies of production and consumption,

its value to consumers, and its association with other virtues such as authenticity,

goodness, and naturalness, has remained. Spray dried flavors capitalized on this

dimension of flavor, making it possible for products such as packaged cake mixes to

deliver the sensual experience of “freshness”— a vivid immediacy and intensity of flavor

— despite the intensive processing necessary to produce them.

Flavored powders had long been used in pharmacy, where they were known as

oleo-sacchara.75 Until the Second World War, there was not much demand for powdered

flavorings. “It should be remembered that it was the ardent desire of the Quartermaster

Corps to provide our Armed Forces during World War II with ‘luxury’ foods, in which

category such items as flavored beverages, pancake sirups, candy, pastry, and desserts

may be placed, not only for the nutritive well-being of our Armed Forces but also for

maintaining their morale at a high level under the most adverse conditions,” wrote flavor

74
Susanne Freidberg, Fresh: A Perishable History, (Cambridge: Harvard UP, 2009).
75
Jacobs 1947: 201.
248
chemist Morris Boris Jacobs in 1951. “Flavoring powders and tablets were very useful in

the preparation of the aforementioned foods.”76

Although beverage powders — home mixes that, when combined with water,

produced a colorful, fruity beverage — had been available before the war, consumer

familiarity with the product likely increased during wartime, when these mixes became a

standard component of field rations as they made foul-tasting water more palatable.77

After the war, when the exigencies that required them were tempered, these were

products in search of an application. In the late 1940s, the D&O’s Flavor Department

prepared a bulletin for food and beverage manufacturers on the subject of powdered

flavors, including formula sheets and other advice on production methods, costs, color,

packaging, and retailing, as well as guidance on meeting state and federal labeling

requirements.78 “From our own survey we believe that there is a great potential market

for home drink concentrates. We hope that this bulletin will help the Food Industry to

develop this market.” 79 It was clear that the company was not angling to supply a well-

established need, but to promote and facilitate the growth of what it hoped would be

emerging market.

76
Morris Boris Jacobs, “How to Make Flavoring Powders,” American Perfumer and
Essential Oil Review 57 (May 1951): 391.
77
Franz A. Koehler, Special Rations for the Armed Forces: 1946-1953, QMC Historical
Studies, Series II, no. 6, 1958.
78
Dodge & Olcott Flavor Department Bulletin 130, “Suggestions for Preparing Flavor
Bases for Summer Beverages, Home Drink Concentrates, Nectars, Etc.” (New York:
Dodge & Olcott, n.d. [Late 1940s?]). AW Noling Beverage Literature Collection, UC
Davis.
79
Dodge & Olcott Bulletin 130: Introduction [np].
249
 Prior to the early 1950s, powdered flavors like the ones described above were

fussy and a difficult sell, necessitating special packaging and formulating practices to

ensure that they retained their integrity when they reached consumers. For example,

instructions for using D&O’s Cosmo line of imitation flavors to prepare a sugar-based

summer drink mix warned that the resultant product was “prone to absorb moisture even

with the best packaging, and should be protected as much as possible. Should be disposed

of as soon as possible after making, keeping stocks at a minimum. Sells best in a

transparent package.”80 Even if the flavor was good when tested at quality control, the

chemical changes that occurred between factory and consumer might produce a less-than-

desirable impression.

Making flavors in powdered form was a challenge. Typically, liquid flavorings

were combined with a dry adsorbent material such as sugar, dextrose, lactose, or

cornstarch, in a powder mixing machine, and then dried on drum rollers. Depending on

the weather and other factors, the process could take anywhere from ten minutes to as

much as an hour, or even longer.81 This batch process had numerous disadvantages,

including flavor loss, oxidative deterioration and rancidity, clumping and caking, and

limited production capacity.

Worse, volatile flavor compounds continued to dissipate even after the powder

was dry. A patent filed in the early 1940s by two employees of General Foods, makers of

Jell-O, describes the extent of the flavor attrition with powders made using this process.

80
Dodge & Olcott Bulletin 130: 1.
81
J.M. Wenneis, “Making Dry Flavors by Spraying,” American Perfumer (December
1955): 48.
250
Making flavor powders by the standard method “requires the use of as much as four to

ten times the amount of flavor actually needed in the product at the time of consumption

in order to allow for the loss occurring during marketing,” they wrote. “Even with this

precaution, the rate of flavor loss is so great that such products are not infrequently

entirely devoid of flavor when prepared for use by the consumer.”82 Their proposed

solution — encapsulating flavor molecules in a colloidal gelatin matrix, which was then

topped with a protective film (the patent suggested cellophane or polyvinyl alcohol)

permeable to water but not to flavor compounds, and dehydrated before being

comminuted to form a dry powder — was effective in retaining flavor, but had limited

applications due both to the high cost and physical properties of gelatin.83

The crucial step to producing functional powdered flavors without flavor loss

involved the adaptation of an existing technology: the spray dryer. Spray dryers were

mechanical dehydrators that used high heat to convert liquids into powders. A fine mist

of a liquid — such as milk, or fruit juice, or a chemical solution —was sprayed into a

whirling flow of hot air in a large cylindrical drying chamber, where moisture evaporated

swiftly, often within a few seconds, avoiding most thermal damage. The resulting powder

funneled down the cone-shaped bottom of the chamber, where it was collected, cooled,

and packaged. Spray dryers had been used since the early twentieth century to

82
A.G. Olsen and Edward Seltzer [General Foods Corporation], “Preparation of
Flavoring Materials,” U.S. Patent 2,369,847. Filed Dec. 6, 1941. Granted Feb 20, 1945.
83
James J. Broderick, “Blazing the Trail to… Superior Powdered Flavors,” Food
Engineering (November 1954): 83.
251
manufacture pharmaceuticals, chemicals, and powdered milk.84 But until the Second

World War, “spray dryers were… considered novel with limited application,”

commented one engineer in the 1950s. “Spray drying was tried only when other methods

of drying had failed.”85

The war renewed interest in food dehydration, which got a boost from Army

Quartermaster research and investment.86 Dehydrated foods offered various advantages to

Army planners: reduced volume and weight, which meant increased portability, as well

as an extended shelf life, without the metal required in canned foods. The War

Department worked with the Department of Agriculture and the War Production Board to

increase dehydration capacity in US factories, while the Quartermaster promoted and

conducted research to improve dehydration processes and increase quality.87 Spray

drying was particularly well-suited to the production of dried milk and dried eggs, staples

of the new army subsistence canteen. After the war, spray dryers were flexible,

automated, and compatible with other continuous operation processes in food processing

facilities, allowing manufacturers to produce larger quantities of better-quality

84
See, for instance, US Patent 1512776A, for a “Drying Apparatus,” assigned to Gerald
Lough on October 21, 1924.
85
Ralph Reeve, of spray dryer manufacturer Bowen Engineering, quoted in Lyne S.
Metcalf, “Spray Drying of Flavors,” American Perfumer & Aromatics (April 1955): 65.
86
Backer 2012: 81.
87
Backer 212: 82-3.
252
dehydrated foods with minimal labor costs.88 Spray drying was used to dehydrate

vegetable purees for soups and baby foods, as well as citrus and grape juices.89

When it came to the production of flavorings, spray drying was not just a more

efficient, scalable version of batch-drying. Spray-dried flavors were materially different

from earlier forms of powdered flavors, in which flavor chemicals were adsorbed by or

mixed with a dry material such as dextrose. Strictly speaking, spray-dried flavors were

dehydrated flavor emulsions, homogenized colloidal mixtures of flavor chemicals and

non-alcoholic emulsifying agents such vegetable gums, gelatin, and starches such as

sorbitol. “Emulsion flavors” had been available since the end of the First World War. At

that time, the increasing price of ethyl alcohol — as well as burdensome record-keeping

requirements and other restrictions concerning its use, exacerbated by the passage of the

Volstead Act — drove a search for suitable substitute media, including glycerine,

vegetable gums such as gum acacia and gum tragacanth, and other relatively odorless and

flavorless materials, that could produce safe, stable flavorings for various practical

applications.90 As emulsion flavors provided some advantages for certain product

88
Metcalfe 1955: 65.
89
Saul Blumenthal, Food Products, (Brooklyn: Chemical Publishing Company, 1947):
857-8, 665, 913.
90
Melvin De Groote conducted extensive research on glycerine and flavor emulsions at
the Mellon Institute in the late 1910s and early 1920s, under research fellowships funded
by Procter & Gamble, the Pittsburgh Brewing Company, and the Research Extracts
Corporation. His research, which provided a detailed record of the process and materials
necessary to manufacture stable, quality emulsion flavors, was published in the American
Perfumer & Essential Oil Review, The Spice Mill, reprinted, and reported widely in other
trade journals in 1920. See, for instance, Melvin De Groote, “The Manufacture of
Emulsion Flavors,” American Perfumer, 1920.
253
applications, such as beverages, flavor companies continued to manufacture and sell them

even after the repeal of Prohibition.

When emulsion flavors were fed into a spray dryer, an effect was produced that

would later be termed “encapsulation.” Flavor compounds were enveloped within a thin,

protective capsule of the emulsifying agent.91 What this meant, functionally, was that

volatile, unstable flavor molecules were guarded against loss and change, until the dry

mixture was combined with water or another fluid, or sheared apart by mechanical

pressure, breaking the capsule and releasing the flavor back into the realm of sensibility.

This, then, was the promise: an encapsulated flavor persisted undiminished for the

duration of its purgatory on the shelf, in order to deliver its full flavor payload to the

consumer at the moment of consumption.

Spray-dried flavors had actually first been produced before the war, in the mid-

1930s, by A. Boake, Roberts & Co (ABRAC), a venerable British essential oil and

aromatic chemicals firm. The company had come upon the process by chance, while

searching for ways to utilize the excess capacity of a spray dryer (purchased to dry

extracts of saponin, a botanical extract used to add a foamy “head” to bottled

beverages).92 The process was never patented.93 Although ABRAC’s Drydex flavor

powders had seen some success in the UK and Europe, war interrupted production and

exports, and it appears that there was little awareness of the spray-drying process or the

91
Gary A. Reineccius, “The Spray Drying of Food Flavors,” Drying Technology 22.6
(2004): 1289-1324.
92
“Origin of Powdered Flavors,” American Perfumer (March 1956): 62-3.
93
“Origin of Powdered Flavors” 1956: 63.
254
product in the United States flavor industry prior to the late 1940s, when manufacturing

was resumed and ABRAC’s Drydex flavors were first marketed in this country.94 In the

postwar, the applications of this technology were evident. An item in Food Engineering

announcing the introduction of ABRAC’s Drydex flavors to the US noted, “with this type

of flavoring material, the shelf life of ready mixes, so far as flavor is concerned, can be

extended greatly.”95

American flavor companies soon jumped in, and began producing their own

spray-dried flavors. When Van Ameringen-Haebler introduced its line of spray-dried

‘Sealva’ flavors to readers of the trade journal Food Engineering in the early 1950s, its

advertisements took care to differentiate these new products from earlier flavor powders:

“Sealva processed flavors appear physically as powders, yet in reality they are minute

droplets of pure flavor individually hermetically sealed.” Claiming that flavor oils

showed no change to their “pure fresh character” even after a year’s time, the company

assured food manufacturers that “Sealva flavors are protected against the ravages of

oxidation and atmospheric change.”96 The result was “really ‘sealed-in’ flavors that defy

time.”97

What this meant for manufacturers, advertisements explained, was an

enhancement of the possibilities of flavor performance in their products. In gelatin

desserts, Sealva Lime flavor “outlasted” competing flavors in shelf-life tests. In boxed

94
“New Dry Flavorings Marketed by Britons,” Food Industries 20 (May 1948): 144.
95
“New Dry Flavorings Marketed by Britons,” 1948: 144.
96
[Advertisement; Van Ameringen-Haebler, Sealva Flavors] Food Engineering 26,
February 1954, 143.
97
Ibid.
255
chocolates, Sealva flavored mints “maintain their original strength and do not permeate

other confections.” In pharmaceuticals, Sealva Orange provided “a pleasant permanent

taste mask in powders and tablets.” And the capstone: “Sealva fruit flavors have made

possible revolutionary new products in Prepared Mixes and Desserts.”98 Foods that

contained Sealva flavors delivered sensory experiences to consumers precisely as

intended, in their full extension and power.

Within the decade, most of the other major flavor companies were selling their

own lines of spray-dried flavors. The 1957 volume of Food Technology included

advertisements for multiple lines of these products, including Sealva, Felton’s Felcofix,

Fritzsche Brothers’ Aromalok, Givaudan’s Permaseal Flavor Crystals, and Polak Frutal

Work’s Flav-o-lok.

Advertisements for spray dried flavors in food industry trade journals dramatized

themes of protection and security. A 1953 advertisement for Florasynth’s “Entrapped”

flavors featured an illustration of a visibly anxious man in a gray flannel suit, looking on

as a masked burglar filched a segment of an orange. “Is there a FLAVOR THIEF in your

house?” the advertising copy asked. Without the assurance of Florasynth’s special spray

drying process, “you may have a flavor-thief and not know it. He steals the vital elements

of flavor and strength.”99 Along the same lines, a 1957 advertisement for Felton’s

Felcofix flavors depicted a white-jacketed scientist at the massive circular door of a bank

98
[Advertisement; Van Ameringen Haebler, Sealva Flavors] Food Engineering June
1953, 104.
99
[Florasynth], “Is there a flavor thief in your house?” [advertisement] Food Technology
7 (June 1953): 13.
256
vault, within which floated grapes, cherries, raspberries, and strawberries. “The flavor is

LOCKED-IN!” assured the headline, urging readers to insist on Felcofix flavors “for the

Safety of your products.”100 This invocation of “safety” was not a reference to consumer

health, but to the integrity of the food’s sensory qualities when it reached the consumer.

Only when flavor was reliably safeguarded, could manufacturers fully realize their

investments in this aspect of their products.

Felton Flavors. “The Flavor is LOCKED-IN!” Advertisement from Food

Technology 11, May 1957: 55.

100
[Felton], “The flavor is locked-in!” [advertisement] Food Technology 11 (May 1957):
55.
257
 Making effective encapsulated flavors required more than an emulsion and a

spray drier. It necessitated expert knowledge and precise control over the chemical and

physical properties of all components of the flavor system. Flavor companies invested in

special research programs and production facilities to improve the quality and

performance of their spray-dried flavor lines, and distinguish them from competitors’

products. In a 1954 article, James Broderick, a flavor chemist at Givaudan, described the

two-year research program undertaken by the company’s flavor research and analytical

laboratories to develop a proprietary emulsifying matrix, one that produced stable,

soluble, and economical spray-dried flavors.101 A 1956 article in Food Engineering about

Norda’s spray drying operation explained the “ticklish problem” in designing flavor for

spray drying. A single flavor might be comprised of twenty or thirty chemical

compounds, each with different structural properties and physical constants, including a

range of boiling points. Although the total loss of flavor materials during spray drying

was typically under five percent, low-boiling compounds were disproportionately

affected, potentially leading to an unbalanced final product. “Laboratory research and

pilot plant testing are therefore requisites for initial compounding a flavor,” explained

Food Engineering. “Such study is necessary if the final dry flavor powder is to contain

the flavor ingredients in exactly the proportion required — regardless of the evaporation

rate of any and all of the flavoring constituents.”102 Production variables, such as droplet

size and dryer temperature, could also have significant effects on the ultimate sensory

quality of spray-dried flavors.

101
Broderick 1954: 83-4.
102
“Advances in Spray Drying Improve Flavor Quality,” Food Engineering 28 (February
1956): 77-8.
258
 As these examples show, flavor companies such as Givaudan and Norda were not

only conducting research into aromatic chemicals, but also into materials and machinery

corresponding to flavor delivery and performance. That is, flavor was part of a system of

material relations within a food product, intended toward a more precise orchestration of

ultimate consumer experience.

Spray-drying promised flavor that “defied time,” that could persist on the shelf,

delivering its full sensual payload to the consumer only at the moment of consumption.

Flavor encapsulation can be considered as a method of preservation, in the same category

as other technical interventions intended to extend edibility, such as curing, canning,

pasteurization, and freezing.103 In other words, it was a technology for controlling time —

and it operated both by forestalling time’s deleterious effects on food quality, as well as

by extending the manufacturer’s control over the sensory qualities of food over the

duration of its temporal life.

But spray-dried flavors were also a technology that, like other forms of

packaging, did more than preserve sensation — they permitted new intensifications,

mobilizations, and commodifications of experience.104 In their enlightening history of

packaging technologies, Gary Cross and Robert Proctor describe the container’s

evolution from a means of storing sensual surplus to a mechanism for re-engineering the

103
Ronald Versic, “Flavor Encapsulation: An Overview,” in Sara J. Risch and Gary A.
Reineccius, eds. Flavor Encapsulation, ACS Symposium Series 370, 1988, p. 2.
104
Gary S. Cross and Robert N. Proctor, Packaged Pleasures: How Technology and
Marketing Revolutionized Desire, (Chicago: UChicago Press, 2014).
259
scale and scope of sensory experience, “optimizing” sensations for bodies increasingly

calibrated to receive these new intensities. By understanding spray-dried flavors in the

context of other forms of modern packaging, we can see how these technologies liberated

flavors from adherence to the boundaries of the natural, and inserted them in “new worlds

of sensory access, speed, and intensity.”105

Spray-dried flavors promised to extend food manufacturers’ control over the

sensory qualities of their products, in pursuit of an ideal scenario where the flavor never

changed, where a company’s investment in flavor value never depreciated.

A Taste of Failure: Aerosol Foods

Investment in research and development did not always lead to commercial

success; consumers proved themselves resistant to new food technologies’ promises of

convenience, novelty, and stylish modernity, if they failed to yield calculable advantages

in the context of existing technosocial frameworks of food consumption.106 Givaudan’s

Aerosol Research Laboratories offer a case study in a technology’s failure to find a place

in consumers’ grocery carts and daily habits, despite sustained and coordinated efforts

among container, chemical, and food manufacturers to develop, improve, and promote

pressure-packaged food products.

105
Cross and Proctor 2014: 14.
106
Ruth Schwartz Cowan, “The Consumption Junction: A Proposal for Research
Strategies in the Sociology of Technology,” in Wiebe E. Bijker, Thomas P. Hughes, and
Trevor J. Pinch, The Social Construction of Technological Systems: New Directions in
the Sociology and History of Technology, (Cambridge: MIT Press, 1987): 261-80.
260
 The aerosol, or pressurized, container is yet another example of an existing

technology given a boost by wartime investment in research and production, later adapted

for the consumer market in peacetime. Patents for pressurized packaging, containing both

the product and the propellant necessary for its expulsion, date back to the 1860s. In the

1930s, patents were granted for spray-nozzle systems using dimethyl ether, a liquefied

gas, as a propellant. A range of products, including lacquers, rubbers, insecticides,

fertilizers, fire extinguishers, and cosmetics, became available in these containers, which

allowed the user to apply a steady stream or mist of the product over a certain area. In

addition to ease of use, these containers could deliver sensory benefits as well; eau de

cologne, sprayed from a pressurized container, produced a cooling effect on the skin

because of the expansion of the added condensed gas.107

During the Second World War, soldiers deployed overseas, especially in the

Pacific theater, were pestered and sickened by mosquitos and other stinging insects.

USDA research into insect control led to the development of a powerful portable aerosol

dispenser, the “insecticide bomb,” a heavy metal canister topped with a spray valve,

which expressed a mist of bug-killing liquid, propelled by dichlorodifluoromethane.

“Although relatively cumbersome,” observed one textbook on aerosols, “they were

accepted with relief by members of the Armed Services who were thus introduced to a

method of packing which was destined to become a significant feature of peacetime

life.”108 The government’s research became the basis of a public patent, and the heavy-

107
A. Herzka and J. Pickthall, Pressurized Packaging (Aerosols), 2nd ed. (New York:
Academic Press; London: Butterworths, 1961): 2-5.
108
Herzka and Pickthall 1961: 6-7.
261
gauge dispensers came onto the market in 1945, subsequently ushering in lighter,

cheaper, easier-to-use canisters that, by the middle of the next decade, were used to pack

a rapidly expanding range of substances, including roach killers, spray paints, room

deodorizers, shaving cream, sun-tan lotions, athlete’s foot remedies, and “Christmas

snow.”109 According to a survey conducted by DuPont, one of the largest manufacturers

of chemical propellants, in 1955, just a decade after their widespread commercial

introduction, 91 percent of American families bought and used aerosol products, which

saw sales of nearly $200 million.110 In 1961, boosted in part by the outrageous growth in

sales of hairsprays and laundry starches, retail sales of non-food aerosols topped one

billion dollars.111

In this climate of galloping growth, food in disposable aerosol containers seemed

to be a potential moneymaker. One of the first widely successful aerosol products was

whipped cream, which effortlessly emerged from the canister fluffy and aerated with

nitrous oxide. Thirty million containers of whipped-cream topping sold in 1949, the

product’s first year on the market.112 Although the rate of growth of whipped-cream

topping subsequently slowed, savvy market watchers saw huge potential profits at the

intersection of two rapidly expanding postwar industries, aerosols and processed

109
Ibid; “Markets: Aerosols Reaching New Highs,” Chemical & Engineering News,
(December 19, 1955): 5518.
110
“Markets: Aerosols Reaching New Highs,” 1955: 5518.
111
“Markets: Aerosols Crack Billion-Dollar Mark,” Chemical & Engineering News,
(May 21, 1962): 36-7.
112
“Aerosols and the Food Industry, Part I: Beverages,” The Givaudan Flavorist 1
(1957): 1.
262
“convenience” foods. They envisioned a dawn of “push button” cuisine, where “entire

meals… can be oozed forth by a gentle push on a few cans.”113

Aerosols posed multiple unique challenges for food manufacturers, who had to

confront problems that makers of non-food aerosols did not. The container, the nozzle,

and the propellant all shaped the sensory qualities of the product, which had to be

specially formulated to suit these packaging conditions. “You just can’t put an existing

[food] product in a can,” said one aerosol industry expert in the 1960s. “The product must

be born in the can.”114 But the can also had to be born for the product; both the propellant

dispensing valves and the nozzle had to be designed to allow for the easy and complete

dispersal of foods that varied in composition, viscosity, and reactivity.115 New delivery

systems had to be developed to suit these needs, such as the Mira-Flo “free piston”

container, developed in the early 1960s by the American Can Company’s Aerosol

Division, and the Sepro “bag-on-valve” system, which came out of the Continental Can

Company’s research laboratories. These containers completely separated the propellant

from the product by means of a polyethylene diaphragm, minimizing chemical activity

113
Sam Dawson, “Aerosol Sprays of the Future to Do Almost Anything,” Toledo Blade
[Associated Press], August 12, 1964, 55.
114
Earl V. Anderson, “Food Aerosols,” Chemical & Engineering News (May 2, 1966):
94.
115
Aerosols perfectly staged the intimate interrelationship between container and the
thing contained that emerged in the postwar period: processed food’s dependence on its
package, which, in the case of aerosol food, was integral to its very identity.
263
between food and propellant, and making it possible to pressure-pack high viscosity

foods such as cake frosting and cheese spreads.116

Further, while makers of products such as aerosol hairspray and roach killer used

light, powerful fluorinated hydrocarbons as propellants, food manufacturers were limited

in their choice of propellant. Propellants for food aerosols were required to be nontoxic,

but they also had to be odorless and tasteless. They also had to allow the product to be

dispensed in a form and consistency compatible with consumer expectations, and to

permit complete evacuation of the container’s contents.117 Until 1961, when the FDA

approved Dupont’s Freon C-318 for use in foods, manufacturers were restricted to three

gases: nitrous oxide, carbon dioxide, and nitrogen.118

Givaudan made an early bet on the future of aerosol foods. The company began

working on aerosols in the late 1940s, setting up the first laboratory in the flavor industry

to study the problems of flavoring pressurized foods. In 1959, the company expanded its

aerosol research capacities, making its Aerosol Laboratory a centerpiece of its new

research facilities and headquarters in Manhattan.119 "It is quite evident that the aerosol

container will be widely used in the food industry in the not too distant future," predicted

116
John J. Sciarra, “Types of Aerosol Systems,” in Sciarra and Leonard Stoller, eds., The
Science and Technology of Aerosol Packaging, (New York: Wiley, 1974): 50-5; Gordon
L. Robertson, Food Packaging: Principles and Practice, (New York: CRC Press, 1998):
201-2.
117
Jerry Di Genova, “Food Aerosols,” The Givaudan Flavorist 2 (1963): 2; Herzka and
Pickthall 1961: 177.
118
“FDA Approves Freon Propellant for Foods,” Chemical & Engineering News
(October 16, 1961), 26.
119
“New Laboratories, New Testing Kitchen for Flavor Development Work at
Givaudan,” Food Technology (May 1959): 48.
264
a 1957 article in the Givaudan Flavorist. “The huge success of this packaging medium in

other fields can certainly be duplicated, if not bettered, in the food field, and we are

prepared to work closely with food manufacturers to achieve this end."120

Givaudan was certainly not alone in predicting a bright future for push-button

cuisine. Canning companies, including American Can, Continental Can, and Crown Cork

and Seal, developed new aerosol-ready food containers, and also sponsored symposiums

on food aerosol technology and marketing.121 DuPont, which manufactured the only

fluorocarbons approved for use in foods, actively promoted the development of new food

aerosols by formulating recipes in the company’s food laboratory, creating new

packaging concepts, refining filling techniques, and conducting extensive market

research on food aerosol products, which it eagerly shared with food manufacturers.122

Givaudan fashioned itself as both a knowledge-broker and necessary intermediary, with

its Aerosol Laboratory poised to coordinate among the various industries involved in the

production of pressurized foods. “Every known type of can, valve and seal is available,”

assured an article in the Givaudan Flavorist, “and the aerosol laboratory flavor-chemist is

in constant contact with the various manufacturers so that all new components parts can

be obtained even before they are actually released for sale.”123 The laboratory also had

supplies of “every commercial propellant and mixtures of propellants” as well as new

120
“Aerosols and the Food Industry, Part I: Beverages,” 1957: 3.
121
Anderson 1966: 99.
122
Anderson 1966: 94.
123
“Come With Us to Manhattan: A Visit to Our New York Headquarters,” Givaudan
Flavorist 4 (1959): 5.
265
propellant chemicals that were still in development, such as Dupont’s Freon gases.124

Specially designed equipment allowed the company to perform accelerated shelf-life

testing, which meant that the Givaudan laboratory personnel could superintend every

stage of product development.

This coordination was integral to the company’s core business strategy, as it

positioned itself to be the go-to source for flavorings for these new products, a necessary

point of passage for any of the industrial actors seeking success in this category. Rather

than supplying one component for a pre-existing product — a component that could be

replaced by a competitor — Givaudan’s Aerosol Laboratory positioned the company to

be centrally involved in the new product development process, partnering with food

manufacturers from the outset. This strategy was not limited to Givaudan or to aerosols,

but reflects a broader trend in the evolution of the relationship between the flavor and

food industries at this time. New kinds of highly processed foods necessitated the

experience and technical skills of flavor chemists, and so the companies that employed

them had an interest in promoting the adoption of these products and their success in the

marketplace.

Articles in the Flavorist promoted Givaudan’s expansive vision for food aerosols

as a transformative product category, even as pressure-packed products continued to

flounder in the market.125 In 1963, Jerry Di Genova, the administrator of the company’s

124
“Come With Us to Manhattan” 1959: 4-5.
125
Five articles with aerosols as their explicit subject appeared in the Flavorist between
1957 and 1966. Many other articles mentioned the aerosol laboratory in passing, or the
266
flavor labs, restated an oft-repeated prediction that “new foods will be created just for this

packaging medium as was the hair fixative [ie, hair spray] among the non-food aerosols;

or perhaps food combinations that have been previously packaged separately will now be

pressurized together.” He offered suggestions: “Why not a cream cheese jelly mix ready

to put on bread or crackers? Or perhaps specially prepared baby formulas ready to mix

with water or milk? We can only guess what food aerosol research will bring.”126 Indeed,

“the application of aerosols to foods is almost limitless, controlled only by the

imagination of the food technologist and the stability and formulation limitations of some

individual products.” The category’s continuing lack of success was due not to technical

incompetence or high prices, but rather, a “lack of imagination on the part of processors

and marketers” and a “failure of key executives among major food marketing firms to

visualize their products in pressurized packages.” A final article on the subject, in 1966,

continued to insist that food aerosols were a revolutionary product category, “completely

revising and changing methods of eating which have gone on for centuries.” Although the

promise of food aerosols had yet to be realized, although numerous aerosol food product

launches had crashed and burned, and although consumer resistance to novelty and

higher prices still needed to be surmounted, Givaudan continued to insist that “the far

distant future presents possibilities which are unlimited. The future for food aerosols is,

indeed, bright.”127

company’s specialized aerosol flavors for beverages, pharmaceuticals, and other

products.
126
Jerry Di Genova, “Food Aerosols,” Givaudan Flavorist 2 (1963): 2, 4.
127
“Food Aerosols: The Present and the Future,” Givaudan Flavorist 2 (1966): 8.
267
 Ultimately, despite innovations in containers, products, and propellants, aerosol

foods never lived up to the high expectations of industry boosters. Even as the variety of

non-food aerosol products expanded, with sales racing higher and higher through the

1960s, food aerosols continued to languish, with some short-term faddish successes, but

repeated product failures.128 In 1966, Chemical & Engineering News published an

extensive investigation of the disappointing sales and uncertain future prospects for food

aerosols. The technical challenges of designing containers and valves, formulating

products, and successfully marketing them, combined with higher costs to the consumer,

imposed steep barriers to success that only the largest food companies seemed to have the

resources to tackle. Besides, there remained a resistance among consumers toward food

in a packaging form that still suggested insect repellant. Instead of revolutionizing food

production, aerosols found uses in certain marginal products, such as spray cheese. A

1971 news article quoted an aerosol valve manufacturer’s lament that whipped cream had

so far been the only real success for food aerosols. “All the other attempts to use aerosols

for foods have pretty much petered out for one reason or another.”129 With the growing

scientific consensus around CFCs damage to the ozone layer, and subsequent state and

federal regulations phasing out non-essential uses of the chemicals to package aerosol

products by 1979, the prospects for aerosol products of all types dimmed.130

128
These failures included some products touted by Givaudan in issues of the Flavorist:
Whisp, a spray vermouth that was one of the first food products to use Freon C-318; Pet
Milk’s “Big Shot” aerosol chocolate milk flavor; and Sizzl-Spray, an aerosol barbeque
sauce that corroded cans.
129
“Aerosol Foods Sought,” Beaver County Times [UPI], July 16, 1971, 4.
130
“Hard Times Hurt Aerosol Industry,” Chemical & Engineering News (May 19, 1975):
8; “Two Federal Agencies Rule on Fluorocarbons,” Chemical & Engineering News (Nov
268
 The push-button future that aerosol foods promised, one of effortless

convenience, never arrived, despite the best efforts (and substantial investments) of its

advocates and promoters. Because its qualities were profoundly affected by every aspect

of the product and container, flavor was an integral consideration in the development of

new kinds of foods. Even though the category of aerosol foods failed to launch,

Givaudan’s central and coordinating role in the research, development, and manufacture

of push-button-cuisine was typical of the flavor industry’s role in the creation of new

kinds of food products.

A World of Flavors: Frozen Food Specialties and Consumer

Appetites at the Twilight of the Mass Market

The 1950s, the golden age of the consumer mass market, was also its twilight.

The proliferation of brands, products, and buyers resulted in tremendous economic

growth, but also intensifying competition and declining margins. Even as the middle class

expanded and broadened its contours, the mass market that supplied these consumers

with the material accoutrements of postwar prosperity seemed on the verge of imploding,

a victim of its own supersaturated density.131

29, 1976): 5; “Phaseout Set for Fluorocarbon Aerosols,” Chemical & Engineering News
(May 16, 1977): 4.
131
An authoritative account of the decline of the mass market and the rise of market
segmentation can be found in Lizabeth Cohen, A Consumer’s Republic: The Politics of
Mass Consumption in Postwar America, (New York: Knopf, 2003): 292-344. Thomas
Hines provides a portrait of the cultural style of the consumer mass market in its heyday
in Populuxe (New York: Knopf, 1987).
269
 In the second half of the 1950s, a new commercial strategy began to be

articulated: market segmentation. Rather than vying for the dollars of the averaged

American, some business strategists descried untapped potentials on the market’s fringes,

profits that could be realized by attending to buyers and desires hitherto excluded from

normalized models of middle-class preferences. As Lizabeth Cohen explains, “the move

from mass to segmented markets promised greater, steadier profits through expanding the

pool of potential consumers: a wider variety of products, each tailored to a specialized

population, would create more buyers in total and less cutthroat competition to win

them.”132 This strategy provoked a substantial reimagining of the American populace.

Advertisers and marketers, drawing on consumer psychology and recent sociological

research, began to depict the American public as an aggregate of discrete psychographic

segments, each driven by its own particular motivations, each seeking to gratify its own

incommensurable needs. The normative white middle-class housewife, long the primary

object of concern for market researchers and food company executives, was joined in the

pantheon of target audiences by teenagers, African-Americans, white ethnic

communities, and other discrete demographic tranches, to form the big, variegated pie of

the segmented American consumer market.133

132
Cohen 2003: 298.
133
The normative “Mrs. Consumer” was herself a construction, an abstraction forged
from social scientific research in the interwar decades, and fleshed out by advertisers,
marketers, and industrial designers who shaped their messages and products around her
imagined needs. A case study that illustrates how the ideal servantless middle-class
housewife was configured into, and produced by, the design standards for consumer
technologies can be found in: Shelley Nickles, "'Preserving Women': Refrigerator Design
as Social Process in the 1930s," Technology and Culture 43.4 (October 2002): 693-727.
270
 This kind of market segmentation, Cohen points out, was also made

possible by changes in technologies of production. While the move to mass production in

the early twentieth century was driven in part by a quest for economies of scale, by the

mid-1950s, new manufacturing technologies, including new ways of managing

information, had reduced the size of the minimum efficient manufacturing unit, and

lowered the bar for the development and introduction of new consumer products. “More

and more manufacturers,” she writes, “faced with crushing mass market competition and

armed with new technological capabilities, would embrace small batch production as

their salvation.”134 In other words, frozen food manufacturers could charge a premium for

specialty products, and retain more of those profits, as production costs decreased for

batch-manufactured goods.

Shane Hamilton has described how the shift from mass to segmented market

played out in the frozen food industry.135 At the beginning of the 1950s, frozen foods

were the fastest growing segment of the food business, as manufacturers supplied quality,

low-cost staples to an American mass market — albeit a market whose needs were rather

narrowly conceived, to coincide with those of the white, middle-class suburban

housewife. By the end of the decade, growth had slowed, leading to a reconsideration of

market strategies. Frozen food manufacturers “abandoned the idea of selling a cross-class

staple product to the ‘average’ American,” and instead developed products and strategies

to appeal to market segments, including working-class urban black communities, ethnic

134
Cohen 2003: 306.
135
Shane Hamilton, "The Economies and Conveniences of Modern-Day Living: Frozen
Foods and Mass Marketing, 1945-1965," Business History Review 77 (Spring 2003): 33-
60.
271
and religious populations, and affluent consumers.136 In the early 1960s, for instance,

Birds Eye began touting a line of “Southern vegetables,” including okra and collards.

Elsewhere in the frozen food aisle, “Noah Zark” kosher frozen pizza nestled against

frozen knishes, while petite frozen onions in cream sauce made their pitch to well-heeled

shoppers willing to pay a premium for luxury foods.137

The flavor industry played a crucial role in supporting and enabling this shift from

the high-volume production of mass market goods to the batch production of specialties.

This is especially evident in the frozen food aisle. As has been discussed, in the postwar

decades, flavor companies developed additives that were designed to be readily

integrated into existing manufacturing processes, reducing the product development costs

borne by food manufacturers and simplifying the expansion of existing product lines and

the development of new ones. The 1956 Dodge and Olcott (D&O) flavor catalogue, for

instance, was divided into sections that discussed different product lines developed for

canned foods and condiments, frozen foods, diet foods, pet foods and animal feeds, oral

care products, among others. A glance at D&O’s Spisorama dry soluble flavors

developed for the frozen food industry (“one of the newest and most versatile flavoring

developments of many years”) reveals the variety of cuisines and dishes represented: in

addition to Spisorama seasonings for frozen “Bar-B-Q Beef” and fish sticks,

manufacturers could purchase flavors for frozen “Kosher specialties” (“blintzes, knishes,

baked stuffed cabbage, etc.”), Italian foods such as lasagna, ravioli, and eggplant

136
Hamilton 2003: 35.
137
Hamilton 2003: 56-9.
272
parmigana [sic], Mexican and Southwestern foods such as tamales, enchiladas, and chili

con carne, and Chinese foods including egg rolls and chop suey.138

Even as marketing experts classified the nation’s various appetites in order to

guide product development strategies, the fragmentation of consumers along these same

lines was not a given outcome. This is particularly evident in the case of “nationality

specialties,” a newly created category of canned, frozen, and other processed foods that

popularized certain “ethnic” dishes and styles, particularly from Chinese, Mexican,

Italian, and Jewish cuisines. The 14-ounce “Mexican-style” frozen dinner (enchiladas,

“Spanish rice,” beans, and chili, along with a “little container of hot sauce in each

package”) manufactured by the Circle T Meat Company of Dallas was intended to

nourish more than just Mexican-American households; Damiano’s complete line of

frozen Italian specialties offered “a touch of old Italy” to those whose bloodlines bore no

trace of Naples or Sicily. (Indeed, the company’s frozen manicotti with meat sauce was

described parenthetically as “Italian cheese blintzes,” offering a helpful guide for cross-

cultural noshing.)139 Chun King’s canned Mushroom Chow Mein (“in Flavor-Guard

138
[Dodge & Olcott], The Changing World of Food, Reprinted from 1955 and 1956
issues of the D&O News, the monthly publication of Dodge & Olcott, (New York: Dodge
& Olcott, 1957): 12-14. [From A.W. Noling Collection, UC Davis Special Collections];
[Dodge & Olcott] “1956 Dodge & Olcott Reference Book and Catalogue of Flavors and
Seasonings,” (New York: Dodge & Olcott, 1956): 63.
139
A description of the Circle T “Mexican-style” frozen dinner, and an advertisement for
Damiano’s line of frozen Italian specialties, can be found in Quick Frozen Foods
(December 1957): 86.
273
Divider-Pak®”) was a “hearty meatless dish” advertised under a headline promising

“Oriental Food” as a “New Mood for Lenten Meals.”140

“While mass production is the economic law of the land,” the 1956 D&O catalog

counseled, “specialization is the accepted merchandising formula and it is toward this

twin goal… development of an individual product that can be produced and sold to a

mass market… that the creative talents and energies of the D&O Flavor Chemists and

technical staff are projected.”141 In other words, specialty flavors were not designed to

satisfy only niche appetites. They were developed for, and marketed towards, a broader

cross-section of Americans — a reconstitution of the mass-middle as a group that was

increasingly oriented toward personalized products and consumption as a form of self-

expression and differentiation.142

“Zooming sales of National Specialties in canned, frozen and dry mix or

combination form indicate a vastly increased American interest in the culinary delights of

other peoples,” ran an advertisement touting D&O’s specialty flavors in a 1957 issue of

the trade journal Food Technology. “But here indeed the flavor must be right! If it isn’t,

despite the lure of exotic names and places, Mrs. Housewife will not buy a second time.

The general popularity of National Specialties can often obtain your first sale… the

140
“American-Oriental Food News,” [advertisement], Detroit Free Press (March 26,
1957): 28.
141
[Dodge & Olcott] “1956 Dodge & Olcott Reference Book and Catalogue of Flavors
and Seasonings,” (New York: Dodge & Olcott, 1956): 7. [AW Noling Collection, UC
Davis]
142
Thomas Frank, The Conquest of Cool: Business Culture, Counterculture, and the Rise
of Hip Consumerism, Chicago: UChicago Press, 1997.
274
second and all thereafter can only be insured by a quality product that lives up to the

family’s highest expectations.”143

For food manufacturers and flavor companies in the postwar, “getting the flavor

right” meant something other than what we now know as cultural “authenticity,” one of

the most highly valued and fiercely contested attributes that a food can claim to

possess.144 In the case of “Chinese, Italian, and other foreign specialties,” these cultural

styles, and historic cuisines, were rendered as additives that could be applied at will to

similar basic ingredients, producing the sensory experience of variety, difference, spice,

and novelty.145 In other words, the version of these “traditional” dishes offered by the

frozen food industry was distinct from the “originals” — just as other factory produced

foods had always been different than home-made versions. But if reference to ‘authentic’

or original models was not the goal when designing the sensory qualities of flavor

143
“National Specialties Rate High in the Changing World of Food,” [advertisement],
Food Technology 11, August 1957: 17.
144
There is a vast historical, sociological, and anthropological literature on the meaning
of authenticity in cuisine, with culinary authenticity generally defined either as a
(dynamic) principle that constitutes cultural coherence, or as a valuable designation
constantly in threat of being co-opted by commercial forces. See, for instance, Lisa
Heldke, “But is it Authentic? Culinary travel and the Search for the ‘Genuine Article,’” in
Carolyn Korsmeyer, ed. The Taste Culture Reader: Experiencing Food and Drink,
(London: Berg, 2007); Josee Johnstone and Shyon Baumann, Foodies: Democracy and
Distinction in the Gourmet Foodscape, (New York: Routledge, 2014); Meredith E.
Abarca, “Authentic or not, It’s Original,” Food and Foodways 12 (2004): 1-25.
145
Describing the expansion of ethnic fast food chains in the 1970s, Warren Belasco
notes that most “ethnic” dishes developed for mass consumption were based on familiar
dietary staples — ground beef, chicken, fish filet, cheese — and contained few unfamiliar
ingredients. He also asserts that companies favored conservatism in spicing, in particular,
to avoid unsettling the taste buds of children, an important growth market. Warren
Belasco, “Ethnic Fast Foods: The Corporate Melting Pot,” Food and Foodways 2 (1987):
1-30.
275
additives and frozen meals, what guided product development decisions? How did flavor

and food manufacturers know whether they had gotten the flavor “right”?

The flavor industry’s investment in research, and in the development of new

processes and products, made possible some of the commercial triumphs of the postwar

era. The flavor industry obliged food processors with new kinds of products, expanding

the scope of what was possible in packaged foods, as well as shaping consumer

expectations about how these foods should taste. The following chapter examines the

sensory tools that were used to describe and measure the “flavor profile” of foods, to

captivate the appetites of American eaters.

276
 CHAPTER 5
Designing Flavors for Mass Consumption:
The Flavor Profile

Television viewers in early October, 1952, tuning into the latest episode of the

science fiction anthology series, Tales of Tomorrow, would have witnessed a powerful

fable about the perils and promises of food technology.1

“Substance X” opens in the interior of a working-class Queens apartment.

Brassy, blonde Salena Marshall, an employee at a cannery, is hanging her nylons up to

dry in the living room when she is interrupted by a knock on the door. It is Jerry

Carmichael, a “business consultant” representing certain unnamed food industry clients.

He has a proposition for her.

Salena, it turns out, is the only person in the entire country with a living relative in

Whitman City (population 89), a “small, rural community” on the Gulf of Mexico —

“Small rural community!” Salena squawks. “Ha! That place is a dump!” — where she

was born in 1926, and left behind without a backward glance in 1943. To the

consternation of the food industry, the people of Whitman City have recently stopped

buying their products, and have responded with hostile secrecy to any inquiries from

outsiders. “What are they living on?” Carmichael asks. “If they’ve developed or

1
Tales of Tomorrow, “Substance X,” Season 2, Episode 7, Written by Frank Felitta,
ABC, October 3, 1952. Featuring Vicki Cummings as Salena, James Maloney as
Carmichael, Charlotte Knight as Salena’s mother, Cora, and Will Kuluva as Samuel.
Tremendous gratitude to Mark Martucci, who shared recordings of this and other
difficult-to-locate Tales of Tomorrow episodes from his personal collection.
277
discovered a new food, we must know about it.” Carmichael implores Salena to return

home, find out what the town is living on, and abscond with a sample of this “new food”

for analysis. She accepts the mission reluctantly, only upon Carmichael’s promise of a

big payout.

The shabby Gulf town has become even more derelict in her absence. Her former

home is in extreme disrepair — cupboards cobwebbed, broken, and bare, and her mother

a frail, distracted specter who at first seems not to recognize her, and then responds with a

maudlin excess of sentiment at her return. Salena asks for something to eat. Her mother

says they must see Samuel, a “great scientist,” for provisions… but Samuel’s figure

already darkens doorway. He has a round face, a dark mustache, a high, furrowed brow,

and in a plummy voice he gives his blessing: Salena may be fed.

Her mother offers Salena a quivering cuboid of pale loaf that has the appearance

and apparent density of angel food cake. Salena recoils from the “filthy junk” that her

mother calls ‘manna’ but that Samuel calls ‘Substance X.’ “I said I was hungry for food;

not that!” Her mother assures her that it is food, “whatever kind you want.”

Samuel: “Steak or a roast. Pork, lamb, or veal. Or strawberries in

December. Caviar or kale. It’s anything you have ever tasted
before.”

Salena (taking a handful): “Yeah! It is a steak. A big, thick, juicy

steak…. Strawberries!”

Samuel: “It’s sweet, it’s sour, it’s anything you want it to taste like.
It’s everything and anything you want it to be.”

278
 The next scene finds Salena and Samuel in the brick building where he churns

seawater into Substance X through a low-slung contraption of gears and alembics. (A

stack of boxes in the corner bear the familiar labels of Domino’s Sugar, Kellogg’s, and

Lipton’s; when pressed, Samuel admits that as the lone remaining “control” in the

experiment, he refrains from partaking of his creation.) Substance X, Samuel tells Salena,

is a combination of “minute [aquatic] plants” and “other chemical ingredients which I

add to it.” It is these chemicals that transform Substance X from a nutritious slurry to a

delicious one. He explains the principle to Salena:

“When you eat something, the taste buds under [sic] the tongue
send an impulse to taste centers in the brain. The brain in turn
identifies the food as sweet, sour, bitter, or salty. Now, Substance X
just reverses the procedure. It is the brain that sends the suggested
impulse to the taste buds. So you see, in that way, Substance X is
able to taste like anything the brain remembers.”

In other words, the flavor of Substance X is not an inherent quality of the

substance, but derives instead from memory and desire. The chemicals added to it create

flavor not by triggering definite sensory effects, but by reconfiguring the human

sensorium to conform perceptions to appetites.

“You know if you put this stuff in a box, you could make a million dollars!”

Salena burbles. But Samuel is not interested in money. He has a utopian vision of

abundance, where the problem of subsistence is solved, as people feed on the nutritious,

cheap, and plentiful Substance X. Indeed, since beginning their new regimen, the

279
townspeople’s “physical condition has improved tremendously, and they have become

immune to all diseases.”

However, there was an unanticipated consequence of Substance X’s technological

perfection. “So long as there’s plenty of Substance X to be had for the asking, why

work?” Samuel laments. The town has lost its ambition and incentive, the labor of its

citizens has become “callous and slipshod, their habits have degenerated to those of

animals.” Samuel implores Salena to join him, and somehow use her influence to

“rebuild their confidence… bring them out of this mental torpor… [and] restore their

sense of pride.” She refuses his request, flees his laboratory and her hometown,

plundering a hasty handful of Substance X in her suitcase as evidence.

The program then cuts to an urbane restaurant back in New York City, where

Salena finishes recounting her story to Carmichael, the business consultant. This sets up

the tale’s cruelest twist. As Salena tucks into her steak, murmuring, with genuine relish,

“Why, doesn’t that look wonderful! You can give me the good old-fashioned kind every

time,” a dissonant chord strikes. Her face registers disgust and disbelief. She spits it out,

shoves her plate from the table. It tastes like poison. Having once sampled Substance X,

she can no longer tolerate anything else. The episode ends with her return, in tears, to

Whitman City, crying: “Mamma! I’m hungry!”

“Substance X” captures an ambivalence that thrummed throughout the Cold War,

an ambivalence that flared into acute anxiety when technologies produced consequences

280
that proved to be more catastrophic than the problems they claimed to solve. This was a

moment when dazzling displays of postwar abundance coexisted with escalating

warnings about population growth outstripping available resources, when limitless

progress and total annihilation were both latent in the same shaking atom.

Samuel’s communitarian ideals of food for all, and the food industry’s ambitions

of total market domination, converge upon the same thing: flavor. “Substance X” reflects

an ideal of flavor that was just beginning to be articulated. Beyond the synthetic replica

that transgresses nature’s limits of season and supply — “strawberries in December” —

“Substance X” expresses the dark fantasy of an ideal flavor experience that conforms not

to the given outlines of nature, but to the desires of eaters, by acting upon the intimate

mechanisms of sensation and perception themselves.

How Should Our Food Be Made to Taste?

As discussed in Chapter 4, the war’s end marked a new era of “flavor

consciousness” among the manufacturers of America’s processed and packaged foods

and beverages. “All food processing,” concluded one textbook on the subject, “must

necessarily be governed by the flavor of the marketed product. The consuming public

will eat anything it likes regardless of the price, but will not eat anything it does not like

even though its food value is higher and its cost is lower.” Nutrition, cost, value,

281
convenience, were all ultimately secondary considerations. “’The taste’s the thing.’”2

Flavor joined other technics of consumer persuasion and inducement — advertising,

packaging, retailing — as a means by which food manufacturers courted that elusive

goal: repeat sales.

But what principles should guide a company’s flavor design and development

decisions? What should processed foods be made to taste like? What made a flavor good?

In 1953, Loren “Johnny” Sjöström and Stanley Cairncross, two chemists working

in the Food and Flavor laboratories of Arthur D. Little, Inc. (ADL), a venerable

Cambridge, Massachusetts contract research and consulting firm, published a paper in the

journal Food Technology that promised to resolve these disputes over matters of taste by

technical means.3 Some food products, the authors observed, were not only more popular

than their competitors, but were runaway best-sellers, outselling their two nearest rivals

combined. Cases of exemplary “flavor leadership” could be found throughout the

supermarket: from condiments to gelatin desserts, there was one product whose sales

figures (and, presumably, taste appeal) eclipsed all the rest. Just as there were personality

characteristics that “naturally” suited some men for the leadership of corporations or of

nations — traits and habits that could be identified and cultivated — Sjöström and

Cairncross alleged that there were sensory qualities that made some foods stand apart

from their peers in their ability to gratify the desires of mass consumers.

2
Kenneth M. Gaver, “Unit Operations and Processes, Part I,” in Morris B. Jacobs, ed.
The Chemistry and Technology of Food and Food Products, Vol. II, (New York:
Interscience, 1944): 5.
3
L.B. Sjöström and S.E. Cairncross, “What Makes Flavor Leadership?” Food
Technology 7 (1953): 56-8.
282
 The tool that could determine the “common denominator[s] of quality” shared by

all market-leading foods was the one that Sjöström and Cairncross had helped develop:

the flavor profile. First introduced in the late 1940s, the flavor profile was both a

technology of flavor measurement and a powerful tool for flavor design — one that

claimed the unique ability to detect and predict the qualities that would make a flavor

successful among consumers. Produced by a specially selected, highly trained sensory

evaluation panel, a flavor profile was understood to be an accurate, comprehensive record

of a substance’s subjective sensory qualities. Rapidly adopted in both industry and

academy, it came to form a key part of what Steven Shapin has described as the “vast

complex of technical resources that help shape not just our alimentary environment, but

also practically everything that is commercially formulated, designed, and marketed.”4

The context of the flavor profile’s creation, and the particular set of problems that

it was meant to address, illuminates not only a crucial moment in the industrialization of

the food system, but also the means by which the subjective qualities of food came under

the aegis of technical control and design. The flavor profile configured the sensory and

material qualities of flavor into a new kind of scientific object, and provided a framework

for understanding (and attempting to manage) the actions of sensible materials upon

sensing subjects. This chapter begins by detailing the formation of the flavor profile as a

standard tool of flavor measurement and design. I situate the emergence of the flavor

profile within a matrix of manufacturers, consultants, chemists, sensory scientists,

consumers, and chemical materials in postwar America. I then examine how the flavor

4
Steven Shapin, “The Sciences of Subjectivity,” Social Studies of Science 42.2 (2011):
179.
283
profile’s model of successful flavor has shaped not only the way foods are made to taste,

but also configured a particular set of relations between sensible goods and sensing

subjects into the design of things.

“A Concept of Flavor and a Method for measuring

it:” The Origins of the Flavor Profile

When Sjöström and Cairncross introduced the flavor profile as “a new approach

to flavor problems” at the tenth annual meeting of the Institute of Food Technologists

(IFT) in 1949, they were addressing an audience that had been grappling with the

challenges of studying flavor for nearly two decades.5 Since the 1930s, food researchers

in government and industry had labored to find methods to objectively determine and

measure the sensory qualities of foods. Flavor was understood as a multisensory

phenomenon, a complex perceptual effect arising not only from the activities of the

chemical sensors on the surface of the tongue and within the olfactory system, but also

appearance, texture, consistency, and oral sensations such as the “coolness” of menthol or

the astringency of an unripe persimmon. While instruments such as colorimeters and

tenderometers could measure some aspects of a food’s sensible qualities, no tool

approached the sensitivity of the human chemical senses of smell and taste when it came

to detecting the compounds responsible for aroma, the vast majority of which, at that

5
The paper was delivered on July 13, 1949 at the tenth annual meeting of the IFT in San
Francisco. It was subsequently published in Food Technology, the IFT’s monthly
scientific journal. S.E. Cairncross and Loren Sjöström, “Flavor Profiles: A New
Approach to Flavor Problems,” Food Technology 4 (1950): 308-311.
284
point, remained unidentified.6 A specialized community of “expert tasters,” such a coffee

cuppers and master distillers, provided judgments of sensory quality and value, as did

official food graders, but their expertise was generally restricted to a single type of

commodity, and there were persistent doubts about the reliability of their reports.7

Starting in the late 1930s, many laboratories began using a small “trained panel”

of tasters to produce knowledge about flavor and odor qualities. These tasters were

screened and tested to exclude the anosmic and the frequently congested, and to establish

normative levels of sensory acuity. Adapting methodologies from psychometrics and the

psychophysical laboratory, researchers developed a set of standard procedures and

practices that were designed to extract reliable, reproducible information about the

sensory qualities of foods from the subjective, unconfirmable perceptions of human

beings. Tasters conducted their evaluations silently, in isolation booths within

temperature and atmosphere controlled rooms, tasting standardized samples delivered

(ideally) through wall-hatches that foreclosed any contact between experimenter and

subject. They followed routinized procedures, and recorded their responses on

standardized forms.8 By rigorously controlling experimental conditions, scrupulously

6
The introduction of powerful analytic instrumental technologies, such as gas-liquid
chromatography and mass spectroscopy in the mid-1950s, would vastly expand the
number of known volatile chemicals. Prior to those technological breakthroughs, the
isolation and identification of volatile chemicals in foods demanded meticulous, labor-
and material-intensive research, and comparatively few groups took it on. See Chapter 6.
7
H.F. Willkie and E.H. Scofield, “Some Factors Influencing Determination of Relative
Preferential Values of Distilled Alcoholic Beverages,” Institute of Food Technologists,
1941 Proceedings, (Champaign, IL: Garrard Press, 1941): 203-8; Rose Marie Pangborn,
“Sensory Evaluation of Foods: A Look Backward and Forward,” Food Technology
(September 1964): 63-7.
8
See Chapter 3.
285
excluding social and atmospheric contaminants, disciplining the sensory labor of tasters,

and subjecting data to statistical processing, researchers operated the taste panel as a

laboratory instrument, one which used the human senses as a tool of measurement.9

But while taste panel methods proved adequate for assessing thresholds of

difference and vectors of preference, they offered little information about the content of

experience. They were well-suited for quality control purposes — maintaining a

consistent production standard — but less useful for problems of product development,

improvement, and design, which often involved determining and evaluating the multiple,

interrelated chemical and perceptual changes that could occur with a single alteration to

the composition of a food product.

These were the scientific methods at the disposal of Sjöström, Cairncross, and

their ADL colleagues in the 1940s, when they began contract work for two clients facing

quite different flavor problems: one involving multivitamins; the other, monosodium

glutamate (MSG). In 1947, Upjohn, a Michigan pharmaceutical company, hired ADL to

help make their multivitamin tablets less repulsive. The pills tasted awful, both bitter and

9
For a critical discussion, from the perspective of food science, of the consequences of
the highly controlled, laboratory conditions of taste testing upon the epistemological
claims of sensory research, see Jacob Lahne, “Tasting in Context: Consumer Sensory
Perception of Vermont Artisan Cheese,” PhD Dissertation, University of Vermont, 2014.
For an anthropological perspective, see David Howes, “The Science of Sensory
Evaluation: An Ethnographic Critique,” in Adam Drazer and Susan Küchler, eds. Social
Life of Materials: Studies in Materials and Society, (London: Bloomsbury, 2015). For a
detailed examination of the use of statistical methods in the postwar sensory evaluation of
wine, see: Christopher J. Phillips, “The Taste Machine: Sense, Subjectivity, and Statistics
in the California Wine World,” Social Studies of Science 46.3 (2016): 461-481.
286
sour, and smelled like a composite of “solvent, old gelatin, fish oil, and yeast.”10

Experimental work had shown that certain odorants could mitigate one or more of the

noisome aromas, but combinations of additives affected the sensory qualities of the

vitamins in complex ways. Choosing the right mixture of odorants at the most effective

levels demanded iterative work that daunted existing taste panel methods.

The second client was International Minerals and Chemicals Corporation, a large

fertilizer and agricultural chemical business.11 In the early 1940s, International Minerals

purchased the Amino Products Company of Illinois, one of the first domestic

manufacturers of MSG.12 In order to develop a market strategy for this relatively

unknown — in the US, at least — food additive, International Minerals commissioned

ADL to conduct basic research into the chemical.13 The dynamic, multisensory effects

10
Loren B. Sjostrom, “Introduction: Case Work,” in The Flavor Profile, (Cambridge:
Arthur D. Little, 1972): n.p. [3], Arthur D. Little Collection, MIT Institute Archives and
Special Collections, Cambridge, MA [hereafter cited as ADL Collection, MIT], Series 6,
Box 11.
11
For a history of International Minerals, see Thomas M. Ware, So Little Soil… So Little
Time: The story of the International Minerals & Chemical Corporation, Newcomen
Address delivered at a National Dinner of the Newcomen Society in North America,
(Newcomen Society: New York, 1967).
12
Albert E. Marshall, “History of Glutamate Manufacture,” in Quartermaster Food and
Container Institute for the Armed Forces, Flavor and Acceptability of Monosodium
Glutamate: Proceedings of the Symposium, (Chicago: Food and Container Institute,
1948): 1-14.
13
ADL conducted multiple studies of MSG production for International Minerals,
including manufacturing processes, fundamental research of its effects on food flavors,
and investigations of its potential uses as an additive that could reduce the amount of
pepper or salt in foods. For ADL’s work on MSG manufacturing processes, see ADL
Report C-57319, “Report on Amino Products Company to International Minerals and
Chemical Corporation,” December 2, 1942, and ADL Report C-57505, “Report on
Monosodium Glutamate to International Minerals & Chemical Corp,” August 22, 1945.
For fundamental research on MSG’s flavor and its effect on various kinds of foods, see
ADL report C-57634, “Report on Flavor Studies Related to the Use of Monosodium
287
that the addition of MSG produced in the flavor of some foods — even at subthreshold

levels — was not effectively captured by taste panel methods.

Both of these flavor problems demanded new approaches to sensory evaluation.

While borrowing many features of laboratory taste panels, ADL researchers adapted,

adjusted, and improvised to develop a procedure that suited their needs.

A flavor profile is a descriptive, semi-quantitative, multisensory record of a

product’s sensible qualities, produced by a small, specially selected, and highly trained

panel of sensory evaluators. 14 Panel members alternate individual tasting sessions under

standardized and rigorously controlled conditions — “the same number of sniffs for

aroma, the same number and size, of bites or sips for flavor”— with sessions of open

discussion where impressions are shared, compared, and validated.15

During individual tasting sessions, panel members produce a comprehensive list

of the “detectable factors” in a flavor — specific, identifiable aroma and taste ‘notes’

such as the citrus in a cola, or the bitterness in a beer, as well as textural factors, color,

and other sensible properties. Each panelist also records the order in which these factors

Glutamate to International Minerals and Chemical Corporation,” July 16, 1948. Other
ADL reports related to this contract work can be found in: ADL Collection, MIT, Series
4: Technical Reports.
14
Comprehensive descriptions of the Flavor Profile method can be found in Jean F. Caul,
“The Profile Method of Flavor Analysis,” in E.M. Mrak and G.F. Stewart, eds. Advances
in Food Research 7 (New York: Academic Press, 1957): 1-40; and in Maynard A.
Amerine, Rose Marie Pangborn, and Edward B. Roessler, Principles of Sensory
Evaluation of Food, (New York: Academic Press, 1965): 377-85.
15
Sjöström 1972: [7].
288
become perceptible, and the relative intensity of each on a numerical scale that ranged

from one to three, with “)(“ indicating a just-perceptible sensation. Finally, using this

scale of intensities, the panelist assesses the “total amplitude” of the flavor, the strength

of the “over-all impression” that the flavor made upon the taster. (More on this in a

moment.)

The final flavor profile reflects a consensus account of a product’s sensible

qualities. This information is communicated in two ways. First, as a tabulated list, using

“common language terms” to name the detectable factors and a numerical scale to report

intensities.16 (Sjöström called this “a word-facsimile,” a verbal reproduction of how a

given substance tasted and smelled.)17 Second, graphically, as a visual schematic, with

aroma and flavor-by-mouth each represented as a semi-circle pierced by radiant lines —

“a sort of pin cushion model,” in the words of Cairncross.18 The total area of the

semicircle indicates amplitude; each ray represents a perceived aroma or taste note,

whose intensity is indicated by its length. A ray barely crossing the semicircle’s perimeter

represents a just-perceptible note, i.e., one of )( intensity.19

16
Cairncross and Sjöström 1950: 308.
17
Sjöström 1972: [6].
18
S.E. Cairncross, “The Effect of Monosodium Glutamate on Flavor,” in Quartermaster,
Flavor and Acceptability of Monosodium Glutamate 1948: 36-7.
19
Caul explained that the use of the word “profile” in naming the flavor profile was
prompted by the use of that term by the New Yorker magazine to describe its feature-
length character studies. The visual diagram came from Stanley Cairncross’s attempts to
adequately explain that flavor comprises both an irreducible element and discernible
notes. “Prompted by the New Yorker personal profiles, one of the originators (SEC) of the
flavor profile held up his hand to aid in describing a flavor. The palm of his hand stood
for the portion of flavor so well blended that separate components were not recognizable,
289
and his fingers represented the notes protruding from that body. A draftsman translated
this idea into the sunrise form of the diagrammatic profile.” Caul 1957: 34.
290
 Flavor Profile response sheet for malt beverage, showing “common language”
terms used to describe aroma, flavor-by-mouth, and aftertaste. From Amerine,
Pangborn, and Roessler 1964: 381.

“Out of our work for Upjohn and Ac'cent,” Sjöström later recounted, “we

developed both a concept of flavor and a method for measuring it.”20 The method of

measuring flavor captured dimensions of flavor experience that psychometric and

psychophysical methods did not, reflecting a new conceptual understanding of flavor as a

scientific object. First, rather than a relational study of a single sensible factor, such as

bitterness or fruitiness, a flavor profile offered a multisensory portrait, one that could

capture dynamic relationships among different sensory modalities with ingredient

20
Sjöström 1972: [3]. ADL Collection, MIT, Series 6, Box 11.
291
changes — the way that the addition of 0.2% MSG to canned peas, for instance,

augmented their fragrance and introduced a buttery aroma, decreased sweetness,

increased saltiness, and produced a pleasing mouthfeel.21 Second, a flavor profile

comprehended flavor perception as a temporal experience. Tasters considered the process

of ingestion as a sequence of sensory events, attending to the order of appearance and

disappearance of different perceptual elements and the dynamic flux of intensities.

The most radically new aspect of the flavor profile’s “concept of flavor” came

under the term “amplitude.” While psychometric and psychophysical methods used taste

panels as analytic instruments, whose tasters properly responded only to definite,

distinguishable sensory stimuli, the flavor profile proposed that the perceptual experience

of flavor operated both analytically and synthetically. According to the flavor profile’s

creators, the total experience of a flavor comprised both “perceptible factors” — the

distinguishable ‘notes’ that could be identified by a trained taster — as well as an

“overall impression,” an integrative response to an “underlying complex of factors not

separately identifiable,” which constituted the “basic character” of the flavor.22 In a flavor

profile of coffee, for instance, bitterness, sourness, astringency, and bouquet were

21
Arthur D. Little, Inc. “Report on a Study of the Flavor Characterizations of Certain
Food Products Containing Mono Sodium Glutamate to International Minerals &
Chemical Corporation,” Report C-57634, April 1, 1949: 16. ADL Collection, MIT, Series
4.
22
Jean F. Caul, “The Profile Method of Flavor Analysis,” in E.M. Mrak and G.F.
Stewart, eds., Advances in Food Research 7 (New York: Academic Press, 1957): 2. This
aspect of the flavor profile was explicitly indebted to gestalt psychology. For more on the
influence of gestalt social psychology on the development of the flavor profile, see
Arthur D. Little, Inc. “The Dynamics of the Flavor Profile,” [booklet], (nd/1970s): [np/2],
ADL Collection, MIT, Series 6, Box 11, Folder 1; Loren B. Sjöström and Benjamin B.
Fogler, Oral History Interview, ADL History Luncheon, January 26, 1976, 43-4. ADL
Collection, MIT, Series 7.
292
individually sensed as distinct factors; the unidentifiable chemicals, “which do create a

flavor impression, for without them coffee is not coffee,” accounted for the brew’s

amplitude.23 Bread, whether a homemade, hand-kneaded boule or mass-produced loaf,

produced a common experience of breadiness; however, the home-baked loaf might

register an amplitude of 3, while the mass-produced one might muster a )(.24 Amplitude

was not an additive summation of all the intensities of recognizable notes.25 Instead, it

was a kind of plenitude of experience, “the total breadth or over-all impression of a

flavor.26” A high-amplitude flavor gave the impression: “‘There’s a lot there.’”27

The Profile’s Proliferation and The Growth of ADL’s

Flavor Consulting Business

The immediate response to the flavor profile method appears to have been

positive. Sjöström reported that ADL received more than a thousand requests for reprints

after his and Cairncross’ presentation at the 1949 IFT meeting.28 Flavor profile evaluation

and panel training also became cornerstone ADL services, establishing the firm’s

authority in flavor and other sensory consulting and driving the rapid expansion of its

Food and Flavor division.29

23
Sjöström 1972: [4]
24
Caul 1957: 13.
25
J.F. Caul, S.E. Cairncross, and L.B. Sjostrom, “The Flavour Profile in Review,”
Perfumer and Essential Oil Review 49 (March 1958): 133.
26
Caul, Cairncross, and Sjostrom 1958: 133.
27
Caul 1957: 36.
28
Sjöström 1972: np [3].
29
“History — Flavor Laboratory” Memo from Jacqueline D. Knowles to Kay Manion,
March 2, 1955, ADL Collection, MIT, Series 7, Box 1 [Folder 39]. “70th Anniversary
293
 Founded in 1886 as a consulting chemical engineering laboratory, by the middle

of the twentieth century ADL had grown to become one of the nation’s largest

independent contract research organizations.30 The Cambridge, Massachusetts company

offered a global roster of corporate, military, and governmental clients an increasingly

diverse range of services, including basic chemical and physical research, operations and

systems research, product development and testing, and management consulting.

The company had provided research and consulting services related to odor and

flavor problems since the 1920s, efforts that were generally led by E.C. Crocker, an

eccentric chemist who was known for his acute, diagnostic sense of smell and his lifelong

quest to create a numerical system of odor classification.31 But it was not until the

Report,” [memorandum, 1956], 26. ADL Collection, MIT, Series 7, Box 1 [Folder 1].The
latter report claims that the ADL Flavor Laboratory had had “quadrupled” in size since its
founding in the late 1940s. When ADL added a new three-story wing to Acorn Park, its
principal Cambridge research center, in 1956, almost half the space in the new facility
was dedicated to food and flavor technology research.
30
The standard history of Arthur D. Little, Inc. is E.J. Kahn, Jr. The Problem Solvers: A
History of Arthur D. Little, Inc. (Boston: Little, Brown, 1986). David C. Mowery
considers ADL’s place in the history of industrial research in: “The Relationship Between
Intrafirm and Contractual Forms of Industrial Research in American Manufacturing,
1900-1940,” Explorations in Economic History 20 (1983): 351-374. For a discussion of
the role that Arthur D. Little and his company played in promoting “the gospel of
industrial research” during the progressive era, see David Jerome Rhees, “The Chemist’s
Crusade: The Rise of an Industrial Science in Modern America, 1907-1922,” PhD Diss.,
University of Pennsylvania, 1987. A discussion of ADL as a pioneer in management
consulting can be found in Christopher D. McKenna, “The World’s Newest Profession:
Management Consulting in the Twentieth Century,” Enterprise and Society 2.4
(December 2001): 673-9.
31
For Crocker’s role in organizing the landmark 1937 American Chemical Society
Symposium on Flavors in Foods, see Chapter 3. For a portrait of Crocker, including a
discussion of the development of the Crocker-Henderson System of Odor Classification,
see: Robert Yoder, “The Man with the Million-Dollar Nose,” Saturday Evening Post
224.13 (September 29, 1951): 27, 110-12.
294
development of the flavor profile method that ADL formally established its Food and

Flavor Division as a formal entity within its organization.

When in 1956 ADL added a new three-story wing to Acorn Park, its principal

Cambridge research center, almost half the space in the new facility was dedicated to

food and flavor technology research.32 This included pilot plant facilities for testing new

processes for food and packaging production; a state-of-the-art analytical chemistry

laboratory, which by1960 would feature instruments for high-vacuum distillation, gas

chromatography, ultraviolet, infrared, and mass spectroscopy, and freeze drying; an odor

test room, completely lined with polished aluminum, where “micro quantities of odorous

materials may be examined in the range of 0.0001 ppm;” and specially designed,

atmosphere-controlled panel rooms for flavor profile evaluation.33 “The new facilities

give tangible recognition to the place [the Food and Flavor Division] has established for

itself in providing clients with such services as product development and improvement,

quality control and evaluation, industrial problem solving, training of taste and odor

panels, and pilot consumer acceptance studies,” beamed an internal ADL memorandum,

which celebrated the ascendancy of the company’s flavor related business at the

seventieth anniversary of the company’s founding.34

ADL used flavor profiles in its work with a motley group of corporate clients, on

problems including new product development, product improvement, the investigation of

32
Arthur D. Little Inc., “70th Anniversary Report,” [memorandum] (1956): 26, ADL
Collection, MIT, Series 7, Box 1.
33
“ADL and the Food Industry,” [brochure], Arthur D. Little, Inc. (May 1960): [np/8].
ADL Collection, MIT, Series 6, Box 11, Folder 7.
34
“70th Anniversary Report,” 1956: 26.
295
off-flavors and off-odors, and the evaluation of packaging materials. For instance, in

research for the Dr. Pepper Company, ADL used flavor profiles to evaluate the rapid

flavor changes that occurred to the eponymous beverage within the first three days of

bottling, testing both traditional cork bottlecap liners and new vinylite seals. It found that

despite the material used as a seal, “the characteristic fruitiness of the fresh beverage is

lowered along with a change in the delicateness of blending to produce a resulting

product that is thin, consisting primarily of strong benzaldehyde with weak notes of

vanillin.” Based on this sensory diagnosis, the ADL group identified the likely culprit as

“item #9,” a proprietary flavoring component that contained fruit juices, which rapidly

lost flavor, and offered several suggestions for improving flavor stability.35

In other cases, ADL combined profile evaluation with an analysis of the

competition to deliver specific product recommendations, as in a 1951 report to the

Bristol-Myers Company regarding a two-year study conducted on its Ipana toothpaste.36

Ipana, Bristol-Myer’s signature dentifrice, had been the best-selling toothpaste on the

market before the war, but formulation changes due to wartime shortages had diminished

the product’s minty-spicy appeal and added unpleasant “weedy, garbagey” and “rancid”

notes.37 ADL’s goal was not simply to restore Ipana to its prewar glory, but to put it on a

stronger competitive footing against its chief rival, Colgate, which surpassed it in postwar

sales. ADL created profiles of Colgate, Ipana, and more than 140 Ipana components and

35
R.L. Swaine, Arthur D. Little, Inc. “Report on Preliminary Studies of Dr. Pepper
Beverage to Dr. Pepper Company,” ADL Report C-57956, March 24, 1949: 4-5. ADL
Collection, MIT, Series 4.
36
Arthur D. Little, Inc. “Report on Flavor Studies of Ipana to Bristol-Myers Company,”
ADL Report C-57965, March 21, 1951. ADL Collection, MIT, Series 4.
37
“Report on Flavor Studies of Ipana” 1951: 18.
296
experimental formulations. ADL approached the challenge of “reblending” Ipana by

developing an optimal Ipana flavor profile that maintained the product’s familiar ‘old-

time’ qualities while besting Colgate in the areas where it fell short. An ideal Ipana would

demonstrate “high amplitude, strong flavor impact, good foaming, mouthfilling

properties, sweet spicy spearmint flavor, low to moderate bite, low bitterness, and a

pleasant aftertaste.”38 In other words, flavor profiles were not just diagnostic tools to

identify sources of problems in products, but also prescriptive, pointing to particular

kinds of solutions for commercial problems.

ADL emphasized that the flavor profile had uses beyond food and beverages; it

could be applied to any problem of sensory design in product development. A 1960

brochure claimed that flavor profiling had been successfully deployed in the evaluation

and development of packaging materials, appliances such as coffee makers, freezers, and

refrigerators, and consumer products such as kitchen deodorizers.39 The diverse purposes

for which Profile panels were mustered can be gleaned from the 1957 personnel file of

Anne J. Neilson, a chemist who joined ADL’s Food and Flavor division in 1949 and

remained one of its key employees until her retirement in 1991. By 1957, Neilson had

served on or led Flavor Profile panels for more than a dozen contracts, working on

projects that included: evaluating natural gas odorants for the American Gas Association,

assessing the effect of different containers on orange juice flavor for the Container

Corporation of America, working with the F. & M. Schaefer Brewing Company to

38
“Report on Flavor Studies of Ipana” 1951: 7.
39
“ADL and the Food Industry,” May 1960. [ADL Collection, MIT, Series 6, Box 11,
Folder 7.]
297
improve the flavor of their beer, advising Quaker Oats on a new pancake flavor, and

helping Wallace Laboratories develop a palatable liquid form for their pioneering anti-

anxiety drug, Miltown.40

Neilson was also active in ADL’s Flavor Profile panel training program, which

began sometime in the early 1950s. This program aided the dissemination of the flavor

profile and its philosophy by training flavor profile groups at client companies.41 ADL’s

flavor profile training program was initially a year-long process, comprising lectures,

workshops, demonstrations, and assignments, although the group was considered capable

of producing flavor profiles after about six months of training. Training curricula could

be customized to the needs and problems faced by particular companies. For instance,

Neilson’s employment file records that her work with Goodyear Tire & Rubber Company

focused on training a panel to perform odor evaluation of films — such as Pliofilm, a

new material Goodyear was introducing as food packaging.42 By the 1970s, ADL was

offering a range of course options, “tailored to emphasize the product or products of

greatest importance to you,” from a four-day short course, to three-month, six-month, and

twelve-month programs.43 Longer programs prepared a panel to handle any kind of

assignment, while shorter courses prepared panels to work on a narrower set of problems

40
Arthur D. Little Experience Record: Anne J. Neilson. May 16, 1957. ADL Collection,
MIT, Series 1, Box 7.
41
I have found very little information about the cost of ADL’s services. However, one
1962 source claims that ADL charged companies $15,000 to train a four- to six-person
group. W.R. Young, “Cracking the Secret Riddle of Flavor,” Life (November 23, 1962).
42
Arthur D. Little Experience Record: Anne J. Neilson. May 16, 1957. ADL Collection,
MIT, Series 1, Box 7.
43
“Flavor Profile Training Programs,” ADL Food & Flavor Section, [pamphlet] n.d.
[1970s]. ADL Collection, MIT, Series TK.
298
and products. The shortest course was geared “for those who want to sharpen their ability

to communicate in precise flavor terms.”44

By 1961, ADL had trained 55 flavor profile groups at 34 different companies,

including multiple groups at some organizations — for instance, six at General Foods.45

The number of groups trained would more than double by the end of the 1960s, and

would reach 250 by the end of the 1970s.46 Flavor profile panels operated at major food

companies, including Campbell’s, General Mills, Schaefer Brewing Company, and

Beech-Nut, and at flavor companies including Givaudan and McCormick.47

This does not reflect the full extent or contexts in which flavor profile groups

operated in American, or indeed global, industry. ADL had no proprietary claim to the

technique, and in fact actively encouraged its use, adaptation, and adoption by others.48

As Sjöström recounted in a 1976 oral history, “We wanted to come out and tell the

profession that it was a usable tool. We didn’t want to hold back and we wanted other

44
[Arthur D. Little, Inc.] “Profiles of Success,” Food and Agribusiness Memorandum 16
[n.d., 1970s]. [ADL Collection, MIT, Series 6]
45
Irving T. McDowell, “ADL’s Panel Training Program,” ADL Review (December
1961): 22. [ADL Collection, MIT, Series 3]
46
A May 1969 ADL publication claimed that the firm had trained 118 groups at 59
different companies, including firms in Canada, Europe, and South America.
“Distinguished Flavors,” FYI [ADL internal newsletter] (May 1969): 2. [ADL Collection,
MIT, Series 3]. A booklet likely dating from the late 1970s claimed that ADL had trained
250 groups at 120 companies. “The Flavor Profile,” [nd, late 1970s?]: 9. [ADL
Collection, MIT, Series 6, Box 11, Folder 1.]
47
“How Important is Flavor?” The Givaudan Flavorist 1966 (no. 3): 2.
48
Pfaffman and Schlosberg, “An Analysis of Sensory Methods for Testing Flavor,”
Report to Quartermaster Food and Container Institute for the Armed Forces, Chicago,
1953.
299
people to try and develop it.”49 Indeed, Foster D. Snell, a rival contract research and

consulting firm, began advertising a lightly modified version of the Flavor Profile as its

own organoleptic panel method soon afterwards.50

Panel Selection and Training: The Production of

Intersubjectivity

On what did the flavor profile method rest its claim to authority? First, by an

insistent control over the experimental conditions of evaluation. The profile panel worked

“in a laboratory environment,” explained one ADL brochure, “one that is free of

extraneous odors and has a consistent temperature, a consistent size of samplings, and

consistent procedures of tasting.”51 All of these things served as material and procedural

corroborations of accuracy and reliability, and underscored the flavor profile method’s

technoscientific credentials.

49
Loren B. Sjöström and Benjamin B. Fogler, Oral History Interview, ADL History
Luncheon, January 26, 1976, p. 45. ADL Collection, MIT, Series 7.
50
L.C. Cartwright and P.H. Kelley, “Sharper Flavor Ratings with Improved Profile Test,”
Food Engineering 23 (September 1951): 71-3, 215. “Human Analyzers,” a 1950 article in
Chemical Industries describing the use of new scientific techniques in organoleptic
testing, focus on Foster D. Snell, but also naming additional consulting firms offering the
techniques, including ADL, Wallerstein Laboratories, and Food Research Laboratories.
Two years later, Fortune ran a long article entitled “What Has Happened to Flavor?”
surveying the uses of organoleptic panels at major food companies, which featured
ADL’s flavor profile among its examples. “Human Analyzers,” Chemical Industries 67
(November 1950): 721-2; “What Has Happened to Flavor?” Fortune 45 (April 1952):
130-3, 146-52.
51
“The Flavor Profile” [promotional leaflet, n.d. (early 1960s?)] ADL Collection, Series
6, Box 11, Folder 1.
300
 However, rather than statistical certainty as a measure of objective sensory reality,

the flavor profile aimed for objective truth through intersubjectivity.52 Steven Shapin,

drawing on Richard Rorty, defines intersubjectivity as “the achievement… of ‘unforced

agreement,’ of coming to free and practical interactional assent about what is, from

another point of view, private to the experiencing and knowing subject.”53

Intersubjectivity foregrounds the social aspects of sensory phenomena, and posits sensory

knowledge as the social confirmation of private experience — a knowledge produced

primarily through engaged dialogue that connects the external objects of sensation with

specific perceptual effects. A flavor profile panel, then, was a scientific tool that was also

a social entity, one that needed to perform collaboratively in order to operationally

achieve the intersubjectivity that would serve as the warrant for the validity of its results.

If the isolation booth was the signature furniture of psychophysical and

psychometric methods, the flavor profile’s hallmark was the roundtable. Achieving

sensory consensus was a deliberate social dramaturgy, coordinating private, individual

tasting sessions with multiple periods of open discussion, which were conducted seminar-

style with panel members participating as equals. As Dr. Jean Caul, one of the flavor

profile method’s creators at ADL, explained, “the procedure of obtaining a profile might

be regarded as analogous to the production of a stage play. First the actors are selected;

each studies his part; then there are rehearsals which lead up to the dress rehearsal; and

finally, there is the performance of the play” — a performance whose outcome was a

52
For an in-depth consideration of the use of intersubjectivity in sensory science, see
Jacob Lahne, dissertation.
53
Steven Shapin, “The Sciences of Subjectivity,” Social Studies of Science 42.2 (2011):
176.
301
flavor profile.54 For this reason, casting “actors” for roles on a flavor profile panel

carried higher stakes than selecting tasters for panels whose members tasted in

isolation.55

Only certain kinds of people were considered qualified for flavor panel work.

Prospective panel members were tested first on basic sensory capacities; exceptional

abilities were not required, but those with anosmias and other sensory deficits were

screened out. Prospective panelists were then interviewed to assess intelligence, attitude,

and personality. 56 The key measure of intelligence was articulateness — the ability to

speak fluently, with precision and confidence, about sensory experience. Interest in the

work was also important, as an interested panel member would perform her or his

sensory labor more attentively, carefully, and effectively. (“Then, too, there are

detrimental attitudes that regard smelling and tasting work as effeminate and unworthy of

scientific training,” Caul observed. “These attitudes and opinions are ferreted out in the

interview.”)57 Most crucial, however, was personality. Timid personalities were

contraindicated. “It is not satisfactory to have a panel member who will join the majority

despite his own findings,” wrote Caul. “His personal integrity must counteract the herd

instinct; he cannot be a yes man and still serve as a panel member.” Domineering

54
Caul 1957: 29.
55
Although these sensory tests appear to have varied somewhat between locations, the
account given by Caul (1957) is typical. Prospective profile panelists were tested to
ensure that they could differentiate between and recognize basic taste factors (sweet,
sour, salty, bitter), screened for anosmia using an Elsberg olfactometer, and tested for
odor recognition of 15 common odors and five rarer ones. In the latter test, prospective
panelists were considered acceptable if they performed in the median range. Caul 1957:
15-17.
56
Caul 1957: 17.
57
Caul 1957: 17.
302
personalities were also excluded, as they upset the calibrated egalitarianism of panel

work.58

The particular personal qualities deemed essential for flavor profile panelists, and

the means by which the method produced and constituted its intersubjective results as

trustworthy, objective, and valid, reflect not only the contingencies of the corporate

environment, but also deeply historical and political investments about the nature of

scientific knowledge, and the social conditions and individual qualities necessary to

“establish the facts.”59 Singling out authoritarian or conformist personalities as

problematic to the democracy of the flavor profile panel echoes contemporary research in

the social sciences and psychology, which, in the postwar period, pathologized both the

“closed-minded authoritarian” and conformist personality types as threats to the liberal

social order, while also creating a model of the normative ideal citizen as one who was

autonomous, socially well-adapted, and creative.60

Although each panel had a nominal ‘leader,’ it was emphasized that this person

“does not act as a superior in any way.” This democratic social style, with the citizens of

58
Caul 1957: 17-18.
59
Steven Shapin and Simon Schaffer illuminated the social and political conditions of
scientific knowledge production in restive seventeenth-century Britain in their
cornerstone work, Leviathan and the Air Pump. More recently, Donna Haraway has
examined the formation of the “modest witness” in twentieth century laboratories. Steven
Shapin and Simon Schaffer, Leviathan and the Air-Pump: Hobbes, Boyle, and the
Experimental Life, (Princeton: Princeton UP, 1985); Donna Haraway,
Modest_Witness@Second_Millenium. FemaleMan©_Meets_OncoMouse®: Feminism
and Technoscience, (New York and London: Routledge, 1997.)
60
See Jamie Cohen-Cole, The Open Mind: Cold War Politics & The Sciences of Human
Nature, (Chicago: University of Chicago Press, 2016).
303
the panel acting both independently and collaboratively, was fundamental to the flavor

profile method’s claims to truth and authority:

Experiments have shown that an ‘autocratic’ style (as in the classical

experimenter role) produces a group behavior which can be irresponsible;
lacking in initiative; and with the members experiencing very little
enjoyment in carrying out their tasks, and manifesting hostility toward
both their autocratic leader and toward one another. A democratic style of
leadership, on the other hand, in which the leader merely facilitates group
activity and decision making… produces markedly different effects, even
on the very same group of individuals.61

The flavor profile method’s egalitarian mode is explicitly differentiated from the

implicit coerciveness of other social relations in the sensory laboratory: namely,

psychometric and psychophysical panels where the taster was required to instrumentalize

her or his senses to provide sensory data for the experimenter to analyze.62 Flavor panel

members, on the other hand, were entrusted with producing both the raw data and its

interpretation. While experimenters who used psychometric or psychophysical panels

compelled accurate sensory labor by introducing factors such as competition,

surveillance, and discipline, flavor profile panelists were induced to do their best work by

the social dynamics of the panel itself.63 “As is the case in medical diagnosis and

61
Arthur D. Little, Inc. “The Flavor Profile,” [Booklet] [nd/late 1970s]: 12. ADL
Records, MIT, Series 6, Box 11, Folder 1.
62
See Chapter 3.
63
“The ADL Profile approach thus rests upon role relationships which are models of
what contemporary social psychologists advocate in the place of the traditional subject-
304
treatment,” the ADL booklet on the subject explained, “profiling involves professional

skills and experience.”64

In other words, this was not routine work, but highly trained work — performed

by a select group of professionalized individuals. Although disqualifying sensory deficits

were rare among prospective panelists, far fewer made the cut when it came to

personality. Irving McDowell, writing in 1972, estimated that of the two thousand

candidates ADL had interviewed for profile panels, only one-third were found to be

acceptable.65

Once selected, flavor panel members underwent an intensive training program,

which was customized to the needs of each client, and was designed to last as long a full

year.66 Six or so chosen trainees began their journey toward the “mastery of [this] new

language” with a four-day course at ADL headquarters superintended by instructors from

the Flavor Laboratory.67 They attended lectures and demonstrations on the physiology

and anatomy of taste and odor perception, basic flavor chemistry, and best practices for

organoleptic evaluation. The four-day session concludes with guided panel sessions, led

experimenter relationship,” elaborated one ADL pamphlet on the subject from the 1970s.
“It is interesting to note that the Profile has been employing such role relationships for
more than 25 years.” “The Flavor Profile,” [nd/late 1970s]: 12.
64
“The Flavor Profile,” [nd/late 1970s]: 12.
65
Sjöström 1972: [12].
66
Irving T. McDowell, “ADL’s Panel Training Program,” in The Flavor Profile,
(Cambridge: ADL, 1972), ADL Collection, MIT, Series 6, Box 11.
67
“The Flavor Profile,” [nd/late 1970s]: 11.
305
by ADL panel leaders, where the group produced “rudimentary flavor profiles of

products.”68

After this immersive introductory session, the trainees returned to their company

laboratories with a two-month long work program, a series of assignments profiling

products of escalating difficulty. “The products selected for analysis in this part of the

program are chosen for their simplicity and because they present certain flavor

experiences and problems.”69 Each trainee was assigned to be group leader at least once

during this period, to familiarize him or herself with the responsibilities of the panel

leader and “the cooperation” the leader “requires.” Concurrently, an experienced ADL

panel evaluated the same products, and a month into the work-study program the ADL

group visited the trainees to compare results, correct errors, and monitor each trainee’s

progress and assess suitability for the panel leadership position.

Upon completing the two-month work program, trainees once again returned to

ADL for a three-day advanced course, where they learned more sophisticated techniques

and trouble-shooting methods. Again, trainees are rotated in the leadership role, “under

the close scrutiny of our panel leaders, who evaluate their qualities of leadership.”70

Trainees returned to their company with a four-month advanced work program,

this time evaluating products “of increasing complexity that are germane to the interests

68
McDowell 1972: 12.
69
McDowell 1972: 13.
70
McDowell 1972: 13.
306
of their company.”71 Monthly visits from ADL staff continued. At the conclusion of the

advanced work program, around the half-year mark, the panel leader was chosen by

ADL.

During the final six months of the program, the flavor profile panel became

operational and fully situated in the context of its company. ADL staff continued their

monthly visits, but in this case, they were focused on acting as “liaison between the panel

and management. Working with the panel leader, we help him to understand and meet the

objectives of management. Working with representatives of management, we help them

to understand the needs and function of the panel.”72 Flavor panel members were almost

always employed in other (white collar) roles at the company, and their service on panels

was additional labor that had to be accommodated in their work schedules.73 During this

lengthy post-instructional guidance period, ADL staff were particularly conscientious

about guiding the panel toward a presentation of results that could “mak[e] the data

useful for managerial decisions.”74

The ultimate composition of the panel was not expected to reflect any particular

demographic segment of the consumer base or of the population at large. Flavor panel

members were selected and trained to maintain a scientific disinterestedness, to detect

sensory qualities rather than express personal preferences. Their sensory labor was

privileged and professional, but scrupulously non-elite.

71
McDowell 1972: 13-4.
72
McDowell 1972: 14.
73
“The Flavor Profile,” [nd/late 1970s]: 12.
74
McDowell 1972: 14.
307
Persuasive Profiles

The lead article in the May 1969 issue of FYI, the ADL employee newsletter,

described the carefully planned banquet that kicked off the company’s two-day Flavor

Orientation Program, created the previous year.75 Gourmandizing executives,

representatives of the food, pharmaceutical, chemical, and packaging industries, were

served up the finest continental cuisine, and then interrogated about it:

“What flavor character notes did you detect in the crepes de lise?…How would

you describe the aroma of Kahlenburg soup?”76

At first, the article observes, the participants “are limited to responses of ‘mmm’

and ‘delicious’; but by the end of the two-day session they can judge flavor character

notes to be ‘woody’ and ‘burnt,’ with something of the precision of articulate ADL staff

members.”77

These industry representatives were not being trained in the flavor profile method,

but rather “made more aware of the importance of flavor and odor in product

development” over a series of lectures, workshops, and demonstrations led by senior

members of ADL’s Food and Flavor Section “who have had considerable experience in

new product development and in teaching flavor appreciation.”78 Despite the luxe cuisine

75
“Distinguished Flavors,” FYI [ADL Newsletter] (May 1969):1-2. ADL Collection,
MIT, Series 3.
76
I.e., crabmeat crepes with cream sauce and Austrian dumpling soup. “Distinguished
Flavors” 1969: 1.
77
“Distinguished Flavors” 1969: 2.
78
“Distinguished Flavors” 1969: 2.
308
sampled, these programs were meant to demonstrate the professional rigor and utility of

the flavor profile method, to corroborate it as an expert system of knowledge production.

This kind of intensive effort at enlistment shows that the truth value and utility of the

method had to be deliberately and conscientiously promoted. The profile method was

promoted among fellow research scientists at professional conferences, symposia, and in

publications. But the target of many of these demonstrations were not fellow research

scientists, but corporate managers. After all, a flavor profile panel represented a

significant and sustained investment. Its substantial costs must be justified; its advantages

made plain.

It is crucial to keep in mind that the method was developed by a consulting

company for use in commercial contexts. For this reason, it is important to understand the

flavor profile not only as a scientific tool, but also as a rhetorical device, which made its

pitch to specific audiences: managers in competitive consumer-oriented industries where

sensory qualities mattered, that were searching for technoscientific means to assess and

mitigate the risks inherent in product design and development.

First, the flavor profile method facilitated certain key operations in the product

research and development process. For instance, it permitted market analysis to be

integrated into the profiling process, allowing for comparisons between competing

products. A flavor profile was also expected to have durability, to maintain its meaning

over time in order to serve as a reference point for future iterations, and to forestall

sensory drift in production. This was important, as food processors were increasingly

concerned with maintaining standard product quality over time.

309
 Significantly, however, the flavor profile gained buy-in because of its

effectiveness as a tool for communicating sensory experience to organizational decision-

makers, not just technical personnel, within a corporation. As Cairncross and Sjöström

underscored in their paper introducing the flavor profile, the method could provide

management “with greater understanding of their own flavor problems and of alternatives

presented by research and production.”79 Unlike many examples of contested knowledge

production, where specialized jargon is invoked to buttress social and jurisdictional

claims to professional expertise, the flavor profile’s persuasiveness derived, in part, from

its preference for familiar language over technical terminology. Even as panel members

stabilized the intersubjective meaning of sensory descriptors during the flavor profiling

process, the legibility of their results to outsiders likely corroborated the credit they were

given.

This is why the visual component of the profile was central to the method’s

presentation, and indeed, was prominent in discussions of flavor profiles that appeared in

trade and popular media after its introduction. It was a graphical correlate of what was

repeatedly described as the embodied, performative act of elucidating the profile’s

meaning. “Imagine explaining our findings to Upjohn,” explained Sjöström. “Hold up

your open hand; your palm represents the body of the capsule flavor and your fingers

represent the odor and flavor notes that emerge. One of your fingers is the fishy note, but

you have reduced it by adding an essential oil. Fold that finger down into the palm of

79
Cairncross and Sjöström 1950: 311.
310
your hand. You have changed the flavor profile."80 The flavor profile could thus make the

meaning of difficult to describe sensations immediately comprehensible to non-

technicians, reifying the total “flavor concept” the method proposed.

Over time, ADL’s visual renderings of flavor profiles became increasingly

stylized and almost decorative, full-color images that depicted perceptible notes as

dynamic, brilliant-hued triangles overlaid upon luminous hemispheres. At this point, they

ceased to be explanatory resources, and instead became icons of the method’s power, its

ability to control and define ephemeral sensations, to shape experience itself.

II. Achieving Flavor Leadership: The Flavor

Profile as a Tool of Sensory Design

The flavor profile method’s widespread adoption in the food and beverage

industry, its fittingness for industrial applications, came to shape the way that things were

made to taste, profoundly affecting the sensory qualities of manufactured foods and

beverages in postwar America. The flavor profile was not a neutral tool of measurement,

but also represented a historically specific “concept of flavor” — as such, it was a

phenomenotechnique that reified a model of flavor that privileged certain aspects of

80
Sjöström 1972: [3].
311
sensory experience and established particular kinds of relations between sensation and

behavior.81

The flavor profile’s particular utility as a design tool derived from this: it

provided a way of representing flavor that was not grounded in the molecular specificities

of individual aromas and tastes. A flavor profile could represent flavor as a temporalized,

experiential entity — a pattern or a sequence of intensities — that came into being during

the act of consumption. This phenomenological model allowed trained flavor panel

members to exclude the personal, social, cultural, and historical particularities of foods

and beverages, and inductively deduce “certain generalizations about flavor that serve as

guides to product assessment and product improvement.”82 That is, the flavor profile

claimed to be able to provide a general model for good flavor in foods, one that could be

productively applied to the flavor choices made in the manufacture of all types of

comestibles, from soup to nuts — as well as the sensory design of inedible consumer

goods.

This capacity was not happenstance, but was built into the very foundations of the

method. In their 1949 paper, Cairncross and Sjöström described the flavor profile as the

81
Gaston Bachelard, The New Scientific Spirit; Hans-Jorg Rheinberger, “Gaston
Bachelard and the Notion of Phenomenotechnique,” Perspectives on Science 13.3 (Fall
2005): 313-8.
82
L.B. Sjöström, “The Descriptive Analysis of Flavor,” in David Peryam, Francis J.
Pilgrim, and Martin S. Peterson, eds. Food Acceptance Testing Methodology: A
Symposium Sponsored by the Quartermaster Food and Container Institute for the Armed
Forces, (Washington, DC: National Academy of Sciences National Research Council,
October 1954): 27. Sjöström and others at ADL categorized the flavor profile as a
“phenomenological” method, in distinction from psychometric (difference tests) and
psychophysical (measurements of reactions and preferences) methods. Sjöström 1972:
[8].
312
basis of “a philosophy of seasoning,” a theory of successful flavor in food. Their IFT

paper concluded with the following paragraph:

Inherent in any successful system of seasoning and flavoring is the

building of an interesting complex of flavor. This is accomplished by the
increase of blending, the building of greater amplitude, and the addition of
interest factors…. This concept supplies a working scheme and philosophy
to be followed in all problems of flavoring and seasoning. The Flavor
Profile method is a means of indicating degrees of success in the
development and control of optimum flavor.83

This was a way of comprehending and designing for consumer desirability while

minimizing the risky, expensive, and uncertain business of consulting consumers. Or, as

explained in a booklet promoting the work of ADL’s Food and Flavor division, the flavor

profile is “a disciplined and codified understanding of the elements of flavor,” a kind of

knowledge “instinctive to the master chef and to others who have been in the business of

creating excellence in food. Using the Flavor Profile method… the process is no longer

instinctive, it is under control, and can be used to achieve specific solutions to specific

problems.”84 Tacit knowledge was thus brought under the scope of technical

determination and control. This is also, incidentally, a way of making the case for the

value of outside consultants, who often lack local, hands-on experience with a company’s

products or culture, and so situate the source of their authority elsewhere, in the ability to

descry optimal operational structures and systems.

83
Cairncross and Sjöström 1950: 311.
84
[Arthur D. Little, Inc.] “Food and Flavor,” [n.d., but likely 1950s] ADL Collection,
MIT, Series 6, Box 11, Folder 5.
313
 Cairncross and Sjöström’s influential 1953 study of ‘flavor leadership,’

mentioned at the outset of this essay, followed through on a suggestion hinted at in their

paper to the IFT, and laid out a “working scheme” to build an “interesting complex of

flavor.”85 In order to determine the qualities that characterized commercial success,

flavor profiles were prepared for each flavor leader and its nearest competitors across

eight different product classes: catsup, mustard, salad dressing, canned luncheon meat,

cola, chocolate bars, peanut butter, and gelatin dessert. The goal was not to compare, say,

salad dressings with each other to identify the specific ingredients, sensory notes, or other

features that distinguished the top-seller from the also-rans in its category, but to compare

sales leaders across categories in order to seek out markers of success. In other words, the

study set out to inductively determine a set of general principles about good flavor — to

discover a “common denominator of quality” shared by all foods — as well as to gain

insight into the “unique factors of flavor and quality responsible for the outstanding

success” of particular products.86

Sjöström and Cairncross presented their findings as a set of design

recommendations. Taken as a whole, these principles offered a choreography of

optimized sensory experience, a temporal sequence of perceptions that unspooled during

a consumer’s intimate experience of ingestion. A successful flavor began by making an

interesting “first impression”: delivering an early-impact note. This was followed by a

sequence of other “interest notes” accompanying the rapid development of pleasurable

85
L.B. Sjöström and S.E. Cairncross, “What Makes Flavor Leadership?” Food
Technology 7 (1953): 56-8.
86
Ibid: 56.
314
mouthfeel and full flavor amplitude.87 This orchestral swell of sensation came to a clean

finale; all “flavor leaders” showed minimal aftertaste, while the taste of many of the

second-place finishers lingered on. “The first place cola drink exhibited a quick, clean

disappearance of taste which encourages the drinker to take a second sip,” Sjöström and

Cairncross explained in one example. “If the sips follow one another steadily, more colas

are sold.”88 Flavor sensations could be choreographed to encourage continued (and

continuing) consumption, the ultimate proof of “good flavor” in a system whose models

were chosen based on commercial dominance and sales volume.

This sequence of sensations and intensities was imagined to operate on a

consumer’s desires, and guide her or his behavior, beneath the level of conscious

awareness. Sjöström dilated on this point in a subsequent paper. The leading brand of

catsup, “which incidentally is seldom advertised… had a profile entirely different from

the lesser lights in the catsup field, though all are nationally advertised,” he observed.

The number-one catsup kicked off with an unexpected baked bean-thiol note, and showed

only a hint of sweetness in a complex blend; lesser catsups started with sweetness and

sustained it, with little sensory variety. If a typical consumer were asked to choose

between sulfury thiol or sweetness, she would almost certainly express preference for the

sweet, and describe the thiol as unpleasant. Yet, assured Sjöström, she likely buys the

leading brand “and unconsciously values its interest factors.” 89 The captivating power of

these “interest factors” could overcome, with implicit action, the explicit exhortations of

87
Ibid: 58.
88
Ibid: 58.
89
Sjöström, “The Descriptive Analysis of Flavor,” in Peryam et al. 1954: 28.
315
advertising. The consumer would continue buying her customary brand, without quite

being able to explain why she liked it. Sjöström offered no discussion of the chemical

origins of the thiol note, nor of the historic, cultural, or material paths by which it came to

be found in the top-selling catsup but not in its competitors. It was not so much the

definable sensory quality of the interest note, but its timing, intensity, and distinctness,

that enchanted the consumer, whose thrill-seeking senses sought out the thiol’s timpani in

the arrangement, preferring it over the monotonous drone of sweetness.

One factor stood apart as the crucial determinant of whether a product was

destined for flavor leadership: high amplitude, “a full body of highly blended flavor.”90

Amplitude was directly related to “blend,” the integral, unanalyzable portion of flavor

that could not be perceived as distinct “notes.” It was also associated with the sort of

flavor balance and “smooth flavor” that was typical of the sensory effect produced by

unprocessed foods. As Cairncross and Sjöström noted: “Flavors occurring in nature are

often blends, but man must work to achieve a satisfying blend of flavor in any processed

food.” However, this effort was always worthwhile, as “blending shouts ‘quality.’”91

Blending was a longstanding term of art among people concerned with the flavor

of commercial goods, used since at least the nineteenth century by highly skilled workers

engaged in the production of consumer luxuries that owe a significant portion of their

value to sensory quality — goods such as whiskey, coffee, tea, or tobacco. Blending may

be used to diminish the perceptibility of sensory faults, increase standardization across

90
Sjöström and Cairncross 1953: 57.
91
Sjöström and Cairncross 1953: 57.
316
batches, or to impose a distinct “house” flavor on a branded commodity.92 Cairncross and

Sjöström’s invocation of blendedness as a sensory characteristic of “natural” flavors was

likewise not new, but had deep roots among those who worked with synthetic flavor

chemicals. One of the earliest American monographs on the subject of flavor additives,

Charles Herman Sulz’s 1888 Compendium of Flavorings, advised soda bottlers that the

success of their beverages “is to a great extent dependent upon the correct blending of

various flavors, which should unite in a harmonious whole,” avoiding “a pronounced

flavor” or any “roughness to the taste” that might suggest artificiality.93 Flavor and

beverage industry trade and technical literature of the 1920s, 1930s, and onwards

frequently testifies to the importance of blend, often in language that echoes that of the

flavor profile’s description of amplitude.94

92
As far as I know, the networks of experts, technologies, and practices associated with
“blending” have not been studied by historians, and have largely been considered craft
practices associated with methods of artisanal production — rather than integral to the
large-scale production of foods and other sensible goods. Although beyond the scope of
this paper, I believe that attending to these bodies of tacit knowledge could illuminate
murky historical questions related to the production of consumer goods and the changing
meanings of quality, and reveal ways of working that have remained more or less
invisible in most accounts of industrialization. For a set of primary documents that reveal
some aspects of this, see: “The English Whiskey Decision,” American Food Journal vol
4. (September 15, 1909): 53-56, see esp. 55 on “process”; John H. Blake, Tea Hints for
Retailers, (Denver: Williamson-Haffner Engraving Company, 1903): Chapter X (“Tea
Blending”); and Tobacco Whiffs for the Smoking Carriage (Mann Nephews: Cornhill,
1874): 3-17. For a related account of the intersection of historical, technological, and
social factors involved in forming an agricultural commodity into a consumer good with
particular sensory qualities, see: Barbara Hahn, Making Tobacco Bright: Creating an
American Commodity, 1617-1937, (Baltimore: JHU Press, 2011).
93
Charles Herman Sulz 1888: 10-11.
94
See, for instance, Melvin De Groote, “The Selection of Extracts for Carbonated
Beverages,” The Beverage Journal 58.3 (March 1922): 52-3; Bernice Challenger, "The
Art of Blending and Its Application in the Bottling of Carbonated Beverages," Beverage
Journal 58.2 (Feb 1922): 91-3. De Groote portrays a high-quality flavor as one that
317
 A blended flavor, then, was a sign of both genuine “naturalness” and exquisite

human skill. In the latter case, a corollary consequence of blendedness was the erasure of

the traces of skilled labor involved in its production. Blendedness gave the impression

that flavor was immanent rather than externally applied. As a prescription for flavor

leadership in processed foods, it put emphasis on control over processes, and on the use

of skillfully crafted flavor additives that reproduced the effect of naturalness.

But, as it turns out, there was a chemical shortcut to achieving blended, high-amplitude flavors:
MSG.

MSG and Flavor Leadership

What was the flavor of MSG? In the first decade of the twentieth century, Ikeda

Kikune, the Japanese chemist who successfully synthesized and commercialized the

chemical, had argued that MSG produced a taste sensation distinct from the recognized

four “basic” tastes of sweet, salty, sour, and bitter.95 He dubbed that sensation “umami,”

deliciousness, although it would not be until the 1990s that Western scientists came to

demonstrates not only a distinct and memorable character, but also ‘roundness’ and
‘smoothness,’ where ‘roundness’ is a balance between different elements within the
flavor compound, and smoothness is a balance between the flavor compound and other
sensory aspects of the drink, such as sweetness, acidity, and carbonation. Challenger
advised that the goal of successful blending is “attaining a harmonious whole. In a blend,
no one flavor is dominant, yet all unite to form a distinct different flavor” (92). For a
discussion of the importance of blend in the production of synthetic flavors, see Bernard
H. Smith, “Modern Trends in Flavors,” Food Research 2.3 (1937): 251-253. Smith, a
chemist, headed the Brooklyn flavor company Virginia Dare.
95
Jordan Sand, “A Short History of MSG: Good Science, Bad Science, and Taste
Cultures,” Gastronomica (Fall 2005): 38-50.
318
accept umami as a basic taste modality.96 Indeed, until the early 1940s, most US food

researchers, manufacturers, and consumers spared little attention for MSG, dismissing it

as an “oriental” seasoning unsuitable for Western cuisines.97

This began to change in the early 1940s, when several US factories came on line

producing MSG from agricultural waste products.98 One of the first major American

markets for domestically produced MSG was in powdered, dehydrated soups shipped

abroad as food aid during the wartime Lend Lease program. In other words, MSG was

initially used for the purpose of making minimally acceptable, highly processed foods

somewhat more palatable. The Second World War fueled interest in exploring MSG’s

potential as a food additive. Much of the research on the chemical was conducted in

connection with the US Army Quartermaster Food and Container Institute, the center of

military food technology research, which was deeply preoccupied with the problem of

enhancing the “acceptability” of foods.99

96
Sarah E. Tracy, “Delicious: A History of Monosodium Glutamate and Umami, the
Fifth Taste Sensation,” (PhD diss., University of Toronto, 2016).
97
Although some wide-awake US gourmandisers sought out MSG in Asian specialty
markets in the 1930s, it remained obscure and little-known; when mentioned in media, it
was often described as an “imitation meat” flavor (and indeed was considered as such by
USDA food labeling standards until the early 1940s). This may in part have been related
to the presence of other amino acids in less-than-pure MSG, lending a meaty taste, but it
also was related to ideas associated with the chemical’s origins. East Asian diets had long
been regarded by some Western observers as monotonous and impoverished, meat-
deficient poverty cuisine. For more on the history of MSG in the US before “Chinese
Restaurant Syndrome,” see Berenstein, “How MSG Became American,” forthcoming.
For Western views of Asian diets, see Belasco, Meals to Come.
98
Marshall, “History of Glutamate Manufacture,” in Quartermaster Food and Container
Institute, Flavor and Acceptability of Monosodium Glutamate, 1948.
99
See Chapter 3. The US Army Quartermaster organized two symposia on the subject of
MSG, the first in 1948, the second in 1955. The ADL group’s work was prominently
319
 The work that Sjöström, Cairncross, and the rest of their group at ADL had done

on behalf of International Minerals featured prominently in the new field of MSG

research. Rather than trying to define the “glutamic taste” by comparing it with other

basic tastes, as previous workers had done, Sjöström, Cairncross, and their colleagues

used the profile method to determine and describe the parameters of the “glutamic

effect,” a multisensory effect including taste, aroma, and tactile sensations, in MSG’s

applications to various different foods.100 Although there were categories of products it

did not improve —namely, sweet foods and beverages and dairy — the chemical was

found to enhance the flavor appeal of a broad set of foods, including canned, frozen,

dehydrated, and other highly processed meat and vegetable products.101 In general, “the

principal effect on food flavor was a balancing, blending and rounding out of total

flavor.”102 It diminished the “steam-table flavor” of vegetables that had been left to

malinger dismally on self-service buffets; it blunted the unpleasant, earthy flavor

sometimes found in potatoes, and the sharpness of onion, in canned soups; eliminated the

“fishy” note sometimes found in canned lima beans. It also boosted desirable flavors:

intensifying carrot and cauliflower; “it makes meat taste more meaty and potatoes taste

featured at both. Quartermaster Food and Container Institute for the Armed Forces,
Flavor and Acceptability of Monosodium Glutamate: Proceedings of the Symposium,
March 4, 1948; Armed Forces Food and Container Institute, Monosodium Glutamate: A
Second Symposium, Held at Palmer House, Chicago, June 7, 1955 (Chicago: Research
and Development Associates, Food and Container Institute, 1955.)
100
S.E. Cairncross and L.B. Sjöström, “What Glutamate Does in Food,” Food Industries
20 (July 1948): 76-7, 200-1.
101
Loren B. Sjöström, Stanley E. Cairncross, and Jean F. Caul, “Effect of Glutamate on
the Flavor and Odor of Foods,” in Armed Forces Food and Container Institute 1955: 31-
8.
102
Cairncross and Sjöström 1948: 33.
320
more potato-y.”103 It provided a highly pleasurable tactile stimulation to the mouth. As

one researcher put it, “it is difficult to describe this sensation other than to call it a

‘feeling of satisfaction.’”104 After swallowing an MSG-containing morsel, a tactile after-

effect remained in the mouth that left the eater “in anticipation of the next mouthful.”105

Further, MSG seemed to perform its effects at subthreshold levels (ie, at

concentrations below conscious perception.)106 Consumers might not taste the MSG, but

they did taste the difference. Especially in processed and canned foods, it suppressed

undesirable flavor notes, while boosting desirable flavors, and produced a mouth-filling

sensation that was highly pleasurable, without raising the awareness of its presence in the

mix. “Monosodium glutamate has a very definite effect on consumer preference for

foods,” concluded one Quartermaster study, which found that MSG “decidedly

improved” the appeal of many of the highly processed foods they added it to.107

As noted, the flavor profile method has its direct origins in the effort to name and

record the effect that MSG had on many foods. The flavor profile was designed to

efficiently capture and communicate MSG’s unique consequences: it blended flavor in

many savory foods, and increased overall flavor amplitude. In other words, the blueprint

for successful flavor, as described by the flavor profile, was modeled on the flavor-

103
Sjöström “Descriptive Analysis of Flavor” in Peryam et al. 1954: 28.
104
E.C. Crocker, “Meat Flavor and Observations of the Taste of Glutamate and Other
Amino Acids,” in Quartermaster Food and Container Institute 1948: 28.
105
Sjostrom et al., in Armed Forces Food and Container Institute 1955: 33.
106
Rosaltha Sanders, “The Significance of Thresholds of Taste Acuity in Seasoning with
Glutamate,” in Quartermaster Food and Container Institute 1948: 70-2.
107
Norman F. Girardot and David R. Peryam, “MSG’s Power to Perk Up Foods,” Food
Engineering 26.12 (December 1954): 71-2, 181-5.
321
boosting, mouth-filling richness of MSG-enhanced foods. MSG was a shortcut to flavor

leadership.

This is crucial, as Sjöström, Cairncross, Caul, and the rest of the Flavor and Food

group at ADL were at the center not only of basic research into MSG, but also of the

chemical’s promotion and commercialization — on behalf of Ac’cent, the trade name for

International Minerals’ MSG product. Rather than a chemical salve for the flavor-

diminished foods of wartime scarcity, MSG had to be demonstrated as an appropriate and

desirable complement to the foods of prosperity, one that could improve the apparent

quality, shelf-life, and appeal of many different kinds of products.

MSG was heavily marketed to manufacturers in the late 1940s and 1950s as a

chemical that had the effect of reversing “flavor loss” and preserving, restoring, or

boosting flavor in canned, frozen, and other processed foods. One advertisement for

Ac’cent, published in the trade journal Food Technology in 1952, explained that the

chemical was:

…the amazing new seasoning that catches — and holds — flavor during
processing, while flavor’s at its peak. Yet it adds no color, aroma, or
flavor of its own. There are wonderful natural flavors already in the foods
you process… with Ac’cent, you intensify these flavors. And, what’s
really important, the flavor-edge your products have over competition
gives your salesmen something to talk about! When they cut a can for a
customer, they’ll have the assurance of not only fine products to back

322
 them up but products with a flavor-edge that means repeat sales for those
products.108

MSG was presented as a chemical solution for a longstanding problem (flavor

loss during processing), and one that acted not by masking problems, but by fully

capitalizing on the latent value (the “natural” flavors) somehow already present in foods,

with the effect of securing repeat business. Ac’cent was not the only MSG product on the

market, but the language of ADL reports about the chemical, and the terminology of the

flavor profile, permeated advertisements for its competitors, including Zest

(manufactured by the Staley Corporation of Iowa, from corn gluten) and Great Western.

In particular, MSG was touted as a way of bringing out the “naturalness” of highly

processed foods. “Magnify natural food flavor as more and more leading food processors

do… with Zest.”109 “MSG blends, strengthens, and preserves the natural fresh flavors of

your product,” ran an advertisement for Great Western’s MSG. “It creates a uniformity of

taste, a flavor identity which is the first step in establishing lasting consumer brand

preference.”110 The slogan “Ac’cent makes food flavors sing” graced advertisements and

MSG tins for much of the latter half of the 1950s. These taglines assume greater meaning

when one understands them as lay explanations of the “amplification” effect described by

the flavor profile.

108
Ac’cent, “One of These will be a ‘Best Seller,’” [advertisement], Food Technology 6
(April 1952): 7.
109
Staley Co., “When the Flavor Is There the Customers Are Too!” [advertisement for
Zest], Food Engineering 26 (February 1954): 108.
110
Great Western, “Is Flavor Control Your Problem?” [advertisement], Food
Engineering 30 (June 1958).
323
 Efforts to promote the use and consumption of MSG were tremendously

successful. Nearly two decades before so-called “Chinese Restaurant Syndrome” was

first reported in the letters section of the 1968 New England Journal of Medicine, MSG

was already becoming a common chemical presence in US processed foods.111 Between

1943 and 1955, domestic production of MSG increased from just over three million

pounds a year to more than thirteen million.112 By 1962, US production topped thirty

million pounds, and would continue to rise; nearly all of this was consumed

domestically.113 Despite concerted advertising efforts to persuade consumers to think of

MSG as the “third shaker,” joining the venerable seasoning duo of salt and pepper in the

home kitchen, the vast majority of MSG in the American food system was used in the

production of “convenience foods” – canned soups, frozen foods, baby foods,

condiments, and other processed foods, especially those containing protein.114 Precisely

the kind of value-added foods that could be sold at a premium, and carried the largest

profit margins. Its pervasive importance in these products was such that in 1964,

researchers from Monsanto announced, “MSG dwarfs in dollar importance any other

flavoring chemical known to man, with the possible exception of salt.”115

111
Ian Mosby, “That Won-Ton Soup Headache’: Chinese Restaurant Syndrome, MSG,
and the Making of American Food, 1968-1980,” Social History of Medicine 22.1 (2009):
133-51.
112
Donald K. Tressler, “History of the Glutamate Industry,” in Armed Forces Food and
Container Institute 1955: 13.
113
This was, however, a fraction of the global production, estimated at 148 million
pounds a year, with Japan accounting for more than half. S.A. Heininger and D.J.
Jorgensen, “Flavor Potentiators: Economic Considerations,” in Arthur D. Little, Inc.
Symposium on Flavor Potentiation (Cambridge: ADL, 1964): 14-15.
114
Heininger and Jorgensen, in Arthur D. Little 1964.
115
Heininger and Jorgensen, In Arthur D. Little 1964: 15.
324
Reconfiguring the Receptive Human Sensorium

But how, exactly, did MSG work? Its effects were described not only in terms of

its “boosting” of food’s latent flavors — but also in terms of the chemical’s mode of

action upon the body. MSG is the sodium salt of glutamic acid, an amino acid known to

be neurologically active. Some of the earliest American studies on the chemical involved

its central nervous system effects and neuropharmaceutical potential, as a possible

treatment for epilepsy and mental retardation in children.116 Indeed, one study in the early

1940s found that treating mentally retarded children with glutamic acid increased IQ and

reduced personality and behavioral problems.117 Although there was no evidence of

MSG’s benefits upon the cognitive capacities of neurologically ‘normal’ individuals,

there was speculation that the chemical could be effective for increasing intelligence in

the population more generally; in any case, one pharmacologist wrote, “considering the

dosage used in food flavoring” MSG’s presence in the food system “could only be

beneficial.”118

One leading hypothesis about how MSG produced its effects drew on its

demonstrated neurophysiological activity, and attributed to the chemical the property of

116
Carl C. Pfeiffer, “Pharmacology of Glutamic Acid,” in Quartermaster Food and
Container Institute 1948: 73-8; Rohland A. Isker, “Notes on the Use and Effects of
Monosodium Glutamate,” Journal of the American Dietetic Association 25 (September
1949): 760-3.
117
Zimmerman, Bergemeister, and Putnam, “Group Study of Effect of Glutamic Acid on
Mental Functioning in Children and Adolescents,” Arch. Neurol. Psychiat. 56 (1946):
489-506.
118
Pfeiffer in Quartermaster Food and Container Institute 1948: 77.
325
“increasing the sensitivity of the taste receptors.”119 In other words, MSG operated by

increasing the human body’s responsiveness to certain compounds in foods, enhancing its

receptiveness to certain forms of sensation. Although there was little direct experimental

evidence in support of this hypothesis, researchers continued to pursue it — especially in

the absence of a working model that accepted Ikeda’s proposition that MSG triggered a

distinct, umami taste modality.120

In 1964, ADL organized a symposium on a concept that had emerged from its

work with flavor profiles and MSG: flavor potentiation.121 “Potentiation” was a term

borrowed from pharmacology, where it signified compounds that had no direct effect “on

a biological system, but which exaggerate[ed] the effect(s) of other agents on that

system.”122 Rather than producing noticeable sensory effects, flavor potentiators were

thought to act directly on the body’s mechanisms of sensation. These were imperceptible

agents which reconditioned the human body’s response to other compounds in the

environment, magnifying, exaggerating, and synergistically enhancing certain perceptual

effects. (Or as one newspaper advertisement for Ac’cent explained it to consumers,

119
Francis J. Pilgrim, Howard G. Schutz, and David R. Peryam, “Influence of
Monosodium Glutamate on Taste Perception,” Food Research 20 (1955): 310-16.
120
Pilgrim et al. (1955) review the experimental record, and find “no evidence… that
MSG has any unusual effect on sensory acuity” (316). Sjöström, writing in 1963, cites
continuing experiments in search of this effect, and disagrees with Pilgrim et al.’s
conclusions. Loren B. Sjöström, “The MSG Story,” in Arthur D. Little 1964: 1-3.
121
Arthur D. Little 1964.
122
“Flavor Potentiation: An Introduction” in Arthur D. Little 1964: [np].
326
“Scientists…have established that, unlike any seasoning known, Ac’cent urges the taste

buds to a quick, intense, and sustained appreciation of food flavors.”)123

MSG, the “first known flavor potentiator of major significance,” drove the search

for similar compounds, including nucleotides such as 5’-IMP and 5’-GMP, which were

already in use in Japan, and which received FDA approval for use in foods in 1962.124

There were two arguments to be made for the value of these substances. First,

potentiators were seen as a means of using less of another ingredient. In dehydrated

soups, the addition of a proprietary nucleotide mixture allowed formulators to reduce the

amount of (presumably more costly) beef extract used, while maintaining (or increasing)

consumer acceptance.125 Were one to be found, the presumed economic value of a

sweetness potentiator, a compound that enhanced the perception of sugars, was

staggering, especially in a marketplace at the beginning of a sustained boom in low-

calorie soft drinks.126

The second argument for the value of flavor potentiators was entwined with their

imperceptibility, their spooky action at subthreshold. Despite the pace of change in the

food industry, food manufacturers were, as a whole, conservative. Having built a market

for a certain product, they were very reluctant to make any changes to it; indeed,

manufacturers hired companies such as ADL to guarantee flavor consistency in cases

123
Ac’cent, “How Much Good Flavor Has a Hamburger Got?” [advertisement],
Washington Post (November 8, 1950): 4.
124
Loren B. Sjöström, “The MSG Story,” in Arthur D. Little 1964: 1; Dudley S. Titus,
“The Nucleotide Story: Applications,” in Arthur D. Little 1964: 11.
125
Titus in Arthur D. Little 1964: 11.
126
Heininger and Jorgensen in Arthur D. Little 1964: 17-8.
327
where ingredients had to be altered. “The reluctance to change the flavor of an existing

product is both natural and logical, since the consumer is very often antagonistic toward

any change in a product to which she has grown accustomed” conceded a representative

from Merck, who spoke at the flavor potentiation symposium.127 Potentiators promised to

enhance existing flavors and improve acceptability without detection, influencing

consumers’ behavior without alienating or alerting them.

Research into potentiators relied on flavor profiles, which made it possible both to

determine the system of synergistic effects produced by the chemicals, and also to predict

whether they would enhance food acceptability.128Indeed, the flavor profile made aspects

of sensory experience perceptible, and measurable, to experts with the power to shape the

qualities of things — precisely those aspects of sensory experience that resisted analytic

quantification or conscious awareness. Commanding these molecular relations, by means

of compounds increasingly designed to be imperceptible, could yield powerful effects.129

127
Titus in Arthur D. Little 1964: 12.
128
Titus in Arthur D. Little 1964: 10.
129
In a certain sense, this may be a special application of what Michelle Murphy has
described as the formation of “domains of imperceptibility,” the designation of the
material limits of knowledge that is integral to the establishment of scientific authority.
“Seeing necessitates the designation of the unseeable, knowing the unknowable, and so
on,” she writes. She also observes that: “over the course of the twentieth century
imperceptibility itself became a quality that could be produced through the design of
experiments or monitoring equipment.” In Murphy’s case study, which concern liabilities
for chemical exposures in built environments, instruments and scientific practices were
used to deny or cast doubt upon the sensed existence of latent chemical presences. In the
flavor and food industries, the imperceptibility of chemical additives was produced
alongside (and indeed, as part of the process of) the enhancement and improvement of
flavors themselves. The goal was not to cast doubt on the sensations provoked by these
chemical presences, but to naturalize them, render them proper to the foods that
contained them and the bodies that responded. Michelle Murphy, Sick Building Syndrome
328
“Because of their remarkable behavior, we believe that these and as yet undefined

potentiators will soon open up new paths to consumer flavor satisfaction,” enthused the

representative from Merck who participated in the symposium.130

Conclusion: The Flavor Boom

“Substance X” dramatized the fine distinction between designing foods to

perfectly satisfy human desires — the utopian ideal of a place where no needs, even

those of pleasure and imagination, go unmet — and configuring consumers to accept the

gratifications that were given to them without resistance. According to the flavor profile,

flavor was not just a sensible quality of certain forms of matter, a set of definite

perceptual effects, or an object of scientific knowledge or connoisseurship, but a

persuasive and influencing agent that acted on human bodies, between physiological

receptivity, psychic effects, and behavioral response. In the dominant contexts of research

into food science and technology — namely, contexts concerned with technics of

industrial food production and processing — flavor became a sensory feature that could

be designed to directly influence consumer motivations and actions en masse by

operating on the intimate level of ingestion. The flavor profile would thus light the way

toward the search for and identification of other neurophysiological effects, which would

act upon and influence human bodies regardless of their particular constitution or specific

situation.

and the Problem of Uncertainty: Environmental Politics, Technoscience, and Women

Workers, (Durham and London: Duke UP, 2006.)
130
Titus in Arthur D. Little 1964: 10.
329
 The flavor profile method was a technoscientific tool that allowed investigators to

compare the flavor of a catsup with that of a cola, or of a canned meat with that of a

gelatin dessert, despite the fact that the foods or beverages in question may not share any

similarities when it came to particular “flavors,” mealtime roles, or other consumer

associations. It was a model uniquely suited to the market for processed foods that came

into being in the postwar era, a market whose defining features were hypertrophic

abundance and competition, where a canned soup must compete for its stomach share not

only with other products of its ilk, but with anything else that a body might care to eat.

In successfully providing a general model for mass-consumption flavor, the flavor

profile also offered the possibility of targeting niche markets. The obverse of flavor

leadership were foods that stoked the passions of a select group of fans. “If a food

product does have a dominant flavor that is strong and distinctive, it will usually appeal

to cultural backgrounds, or to gourmets,” summarized one article that reviewed the

findings of organoleptic studies of flavor. “While these limited groups form faithful

markets for such products, the great mass markets can only be successfully exploited if

the manufacturer designs his flavors to meet mass approval, and changes them to follow

trends in popular taste.”131

This too, then, was a commercial opportunity. As Raymond Stevens, the Vice

President of ADL, said to the Institute of Food Technologists: “We are breeding a nation

of gourmets. A critical consumer means opportunity.” As profit margins continued to

131
“Tailored to Taste,” American Perfumer & Essential Oil Review 60 (August 1952):
99.
330
narrow with competition, finding the right flavor meant the difference between success

and failure.132 The contemporary hyper-refined palate which disdains the mass produced

American food industry could in some ways be said to be the invention of that very

industry — sensitized and called into being by the products designed in laboratories of its

flavor scientists.

132
“Full-Day IFT Symposium Stresses Key Role of Food Engineering,” Food
Engineering 26 (August 1954): 83.
331
 Chapter 6
The Sniffing Machine: Flavor Research
and the “Instrumental revolution” in
Chemistry

When Colonel John D. Peterman, the Commandant of the Quartermaster Food

and Container Institute for the Armed Forces, welcomed researchers from the food and

flavor industry, the military, government, and academic food science departments to the

May 1957 symposium on the “Chemistry of Natural Food Flavors,” he was addressing

scientists in a field that was in the midst of radical change.750 “With the availability and

application of our most advanced chemical and physical techniques and processing

knowledge,” Col. Peterman predicted in 1957, “it seems reasonable to expect that

progress in the next 10 to 20 years can be expected to be much more rapid than in the last

decade.”751 At the Quartermaster, a new program was underway to understand the

chemical changes that occurred to food during processing, in order to improve the

750
Jack H. Mitchell, Norbert J. Leinen, Emil Mrak, and S. David Bailey, eds. Chemistry
of Natural Food Flavors: A Symposium Sponsored by the National Academy of Sciences
National Research Council for Quartermaster Food and Container Institute for the
Armed Services and Pioneering Research Division, Quartermaster Research and
Engineering Center, (Washington, DC: Armed Forces Food and Container Institute, May
1957).
751
Col. John D. Peterman, “Introduction: The Emerging Science of Flavor,” in Mitchell
et al., eds., Chemistry of Natural Food Flavors 1957: 2.
332
“acceptability” of military rations and other manufactured foods, a process aided by an

infusion of funding and adoption of new technologies.752

Food research, like many other areas of basic research, was benefiting from the

surge of Cold War government science funding.753 Flavor research, and indeed all of

chemistry, was in the midst of being radically transformed by an array of new machines.

The development and technical refinement of gas chromatogaphy, mass spectrometry,

infrared and ultraviolet spectroscopy, and nuclear magnetic resonance was one aspect of

a broader transformation in chemistry in postwar America, when wartime advances in

electronics and precision machinery converged with the needs of booming chemical and

petroleum industries, crystallizing into what has been called the “instrumental

revolution.”754 Those technologies seemed to offer the key to tremendous advances in an

area of chemical knowledge that had hitherto remained recondite and obscure. Both army

and industry would benefit from advances in flavor chemistry, Peterman emphasized, as

the bevy of new physicochemical instruments produced more exact knowledge about the

compounds responsible for flavor in foods.

752
Donald K. Tressler, “Interest of the Quartermaster Corps in Flavor,” in Mitchell et al.,
eds., Chemistry of Natural Food Flavors 1957: 3. Tressler was the scientific director of
the Quartermaster Food and Container Institute.
753
Anastasia Marx de Salcedo, Combat-Ready Kitchen: How the U.S. Military Shapes
the Way You Eat, (New York: Current/Penguin Random House 2015): 77-8.
754
See Peter J.T. Morris, ed. From Classical to Modern Chemistry: The Instrumental
Revolution, (Cambridge: Royal Society of Chemistry in Association with the Science
Museum London and the Chemical Heritage Foundation, 2002); Davis Baird, “Analytical
Chemistry and the ‘Big’ Scientific Instrumentation Revolution,” Annals of Science 50.3
(1993): 267-90; Carsten Reinhardt, Shifting and Rearranging: Physical Methods and the
Transformation of Modern Chemistry, (Sagamore Beach: Science History Publications,
2006).
333
 What would progress look like? Perhaps drawing an analogy with the nutritional

fortification of foods with vitamins, Col. Peterman ventured that “there appears to be the

distinct possibility that the acceptability of certain food items might be significantly

improved by fortification with flavor substances.” For instance, “once we know the

chemical nature of desirable meat flavor compounds,” those chemical compounds could

then be synthesized and added to “meat items, soups, and gravies” in the military

canteen, enhancing their sensory appeal to soldiers’ appetites. The loss of flavor in

dehydrated foods could possibly be prevented, or fugitive flavors captured and restored.

Enzymes might be used to break down flavor precursors in foods, capitalizing on latent

flavor to maximize sensory experience. One day, instant coffee might even offer the

satisfaction of fresh-brewed.755

The emerging science of flavor, then, was located at the intersection of chemistry

and desire in the context of food production for the purposes of feeding large groups of

people, whether soldiers or civilians. Flavor science, as articulated by the architects of the

Quartermaster Institute symposium and as practiced by investigators in a range of

institutional settings, aided by powerful new technologies of chemical analysis, explicitly

connected physicochemical research with the improvement of the “acceptability” of

foods — especially processed foods. Aiming for more than just the restoration of flavor

lost during processing, this was a science organized around the optimization of the

sensory possibilities of food. The means of accomplishing this optimization would be

through the manipulation of the chemical components of flavor.

755
Peterman, in Mitchell et al., eds., Chemistry of Natural Food Flavors 1957: 2.
334
 Later commentators would fall into the habit of illustrating the progress in flavor

research in terms of the growing number of known flavor chemicals. “In the 1950s, only

about 500 flavor compounds were known,” wrote USDA research chemist Roy Teranishi

in 1989, in his introduction to an American Chemical Society volume reviewing recent

advances in flavor chemistry. “Since then, with the advent of modern instrumentation,

thousands of compounds have been characterized in hundreds of different foods.”756

Although research into the compounds responsible for flavor in food had been proceeding

for decades using classical chemical techniques, for Teranishi and others who worked

with flavor, the “advent of modern instrumentation” marked an inflection point, an

acceleration in the rate of growth of scientific knowledge about flavor that transformed

the very foundations of their field.

But even as new technologies facilitated the accumulation of lengthening lists of

the chemical components responsible for the flavors of different foods, these machines

did not answer questions about perception or about desire — about the sensory effects of

these compounds, what each contributed to the total experience of a food’s flavor, and the

role each played in determining the “acceptability” of the food. The researchers whose

professional lives were devoted to the study of flavor had to find ways of accounting for

the sensory meaning of the increasing number of compounds they isolated and identified.

This chapter considers the consequences of the instrumental revolution in

chemistry for flavor research. I discuss the introduction and adoption of powerful analytic

756
Roy Teranishi, “New Trends and Developments in Flavor Chemistry: An Overview,”
in Teranishi, Ron Buttery, and Fereidoon Shahidi, eds., Flavor Chemistry: Trends and
Developments, ACS Symposium Series 388, (Washington, D.C.: ACS, 1989): 1.
335
machines — most significantly, the gas chromatograph and the mass spectrometer —

which transformed the layout and labor of the flavor laboratory, entwined the study of

flavor with other scientific disciplines and chemical industries, and elevated the

professional status of the scientists who researched flavor. Although their immense

potential was evident from the outset, these analytic machines did not automatically find

a place in the flavor laboratory, and flavor chemists did not simply adopt a standard set of

techniques developed in other research contexts. The challenges that had long bedeviled

research into the chemistry of food flavors — the structural variety of flavor molecules,

their minute concentrations in complex mixtures of other substances, their instability —

persisted, even as the machines vastly increased the efficiency of chemical analysis and

identification. Indeed, flavor chemists often found themselves working at the operational

limits of these technologies. Instruments such as the gas chromatograph and the mass

spectrometer had to be shaped to the particular problems of flavor research, and had to be

proven effective and reliable among those working with this specialized category of

materials. Consequently, the adaptations and innovations that originated in flavor

laboratories would come to influence how machines would be used in other contexts.

Meanwhile, in flavor research, the subjective, sensing body of the investigator would

come to be seen as a necessary complement to the powerful analytic machines.

My story here dwells largely on one of the most productive sites of basic flavor

research in the postwar decades, the USDA’s Western Regional Research Laboratory in

Albany, California, one of the four regional hubs of agricultural research created by the

336
1938 Agricultural Adjustment Act.757 These research centers had been commissioned by

Congress to develop new markets, products, and purposes for farm commodities and

byproducts, and contributed significantly to the industrialization of agriculture.758 During

the Second World War, the Albany laboratory became a center for research into

dehydrating, freezing, and freeze-drying, food preservation technologies that were of

great interest in the development of military rations.759 The Albany laboratory’s special

technical capabilities and dedicated engineering facilities, its application of instrumental

technologies to the flavor problems related to intensive processing, and its public-facing

orientation all contributed significantly to shaping the way these tools would be used.

The Pure and the Mixed: Gas Chromatography at

the Boundary of Research and Industry
Pure compounds are scarce in this world, and often artificially produced; the

material world is made up of complex mixtures. The separation of complex mixtures into

component fractions — the isolation and purification of matter — has been a primary

process for producing knowledge about substance since antiquity, and is a constitutive

practice of the science now known as chemistry. The gas chromatograph was, first and

foremost, a powerful tool for separating volatile mixtures into individual compounds,

757
Since 1953, the regional research laboratories have been known as regional research
centers. For the political context, and chemurgical ideologies, that shaped the creation of
the Regional Research Laboratories, see Mark Finlay, “The Industrial Utilization of Farm
Products and By-Products: The USDA Regional Research Laboratories,” Agricultural
History 64.2 (Spring 1990): 41-52.
758
Ruth Coy, “Regional Research Centers: The First Half Century, 1940-1990,” in 1990
Yearbook of Agriculture, (Washington DC: USDA): 121-4.
759
H.T. Herrick, “New Uses for Farm Crops,” Yearbook of the USDA 1942 (1943): 696-
98.
337
which could then be definitively identified by other instruments and techniques.

Compared with prior chemical and physical methods of separation, gas chromatography

(GC) offered numerous clear advantages: it was faster, more precise, and required less

specialized labor.

Although historians have documented numerous interwar precursors of gas

chromatography, it was not until after the Second World War that theory and technical

know-how met opportunity and need, resulting in the commercial development and

dissemination of these analytic machines.760 Scholars identify a 1952 paper by British

biochemists A.T. James and A.J.P. Martin, of the National Institute for Medical Research

in London, as the crucial publication for both gas chromatographic technology and the

theory behind it.761 A colleague at the Institute, who was working on a tricky problem

involving fatty acid metabolism, asked for Martin and James’ help in separating these

organic compounds. Building on prior work in partition chromatography, the biochemists

devised and demonstrated an instrument capable of executing “very refined separations

of volatile substances.”762

760
Keith D. Bartle and Peter Myers, “History of Gas Chromatography,” Trends in
Analytic Chemistry 21.9-10 (2002): 547-57; Leslie S. Ettre, “Fifty Years of Gas
Chromatography — The Pioneers I Knew, Part I,” LCFC North America 20.2 (February
2002): 128-40; Leslie S. Ettre, “The Early Development and Rapid Growth of Gas
Chromatographic Instrumentation in the United States,” Journal of Chromatographic
Science 40 (September 2002): 458-72.
761
A.T. James and A.J.P. Martin, “Gas-Liquid Partition Chromatography: A Technique
for the Analysis of Volatile Materials,” Analyst 77 (1952): 915-932; and James and
Martin, “Gas-Liquid Partition Chromatography: The Separation and Micro-estimation of
Volatile Fatty Acids from Formic Acid to Dodecanoic Acid,” Biochemical Journal 50
(1952): 679; Bartle and Myers 2002. 

762
Quoted in Bartle and Myers 2002: 548.
338
 All chromatographic methods involve a stationary phase and a mobile phase. As

the sample (the mixed substance to be separated) is carried by the mobile phase, its

components are selectively adsorbed by the stationary phase. In gas chromatography, an

inert gas serves as the mobile phase, carrying the sample along the length of a “column,”

a tube which has been either packed or coated with a liquid adsorptive media. Each

component compound flows at a different rate, depending on its physical and chemical

properties. As the mixture travels through the column, its components separate. Ideally,

by the time the sample has run the course of the column, its components have

disaggregated into pure fractions. A detector at the column’s exit registers both the

“retention time” of each compound — the time it took to travel the length of the tube —

and its relative quantity. The machine’s recorded output, the chromatogram, unspools as

a graphical record of this process of separation, concurrent with the effluent vapor — the

fractionated sample — which can be collected for further analysis and sensory

examination.

GC found its first widespread application not in biochemistry, but in the postwar

petroleum industry, as that substance replaced coal as the primary source of fuels and

chemical raw materials.763 The distinguished chemist Carl Djerassi recalled the dramatic

renovation of the chemical laboratory by analytic machines during this period:

“Laboratory glassware and reagents have been replaced by ‘black boxes’ — and

763
Bartle and Myers: 548; Leslie S. Ettre, “Milestones in Chromatography: Fifty Years of
GC Instrumentation,” Liquid Chromatography Gas Chromatography North America 23.2
(February 2005): 142.
339
expensive ones at that!”764 As Davis Baird and Carsten Reinhardt have argued, the

recession of laboratory wetware into these “black boxes” — metal chassis housing

sensors, circuits, control mechanisms, and other components that owed more to electrical

engineering and physics than to classical chemistry — signaled and helped to perpetrate a

shift in the material culture of the chemical laboratory, the epistemological scale and

scope of chemical work, and the professional identity of the research chemist.765 Where

organic chemists had once sought to identify unknown substances by observing reactions

with known compounds, by the 1960s, professionalized analytical chemists largely

focused their attention on the physical properties of matter disclosed by instruments.

According to the Leslie Ettre, a chemical engineer who helped develop GC at

instrument maker Perkin-Elmer and later became the machine’s chief chronicler, gas

chromatographs “represented the first truly automated, complex analytical instruments

that did not need specially skilled scientists for their operation and could be used by

practically every laboratory.”766 What had once taken exquisite skill and care could now

be simplified and mechanized. Accordingly, some scholars have placed the changing

relationship between chemical worker and analytic instrument within the framework of

the industrial labor phenomenon of de-skilling. In the words of one historian of

chemistry, what had been a “craft, with manual skills learned during an apprenticeship,”

became a series of “standardized procedures,” steps that could be summarized in a

764
Carl Djerassi, “Foreword,” in Morris, ed. From Classical to Modern Chemistry: vi-vii.
765
Davis Baird, “Analytical Chemistry and the ‘Big’ Scientific Instrumentation
Revolution,” Annals of Science 50.3 (1993): 267-90; Carsten Reinhardt, Shifting and
Rearranging: Physical Methods and the Transformation of Modern Chemistry,
(Sagamore Beach: Science History Publications, 2006).
766
Ettre 2005: 142.
340
manual and performed by a technician, with a concomitant decline of autonomy and a

threat to professional status.767

Historians of the “instrumental revolution” have amply documented the central

role that private industry played in the story of the development and dissemination of

these technologies. Tools such as ultracentrifuges, spectrometers, and gas

chromatographs were often designed and developed for industrial applications, such as

process control and the production of synthetic materials, before being adopted for basic

research in the academy.768 Certain industries in particular loom large here: chemical and

petroleum companies, such as Dow, Shell Oil, and DuPont, whose investments in

research, and decades of spectacular growth, fueled the postwar economic boom and

767
Pierre Laszlo, “On the Self-Image of Chemists: 1950-2000,” Hyle 12.1 (2006): 99-
130. See also, Davis Baird, “Encapsulating Knowledge: The Direct Reading
Spectrometer,” Foundations of Chemistry 2 (2000): 5-46; and Stuart Bennett,
“Production Control Instruments in the Chemical and Process Industries,” in Morris, ed.
From Classical to Modern Chemistry. Baird, discussing the consequences of the direct
reading emission spectrometer (developed in the 1940s), on the nature of chemical work,
argues that machines in this period are increasingly presented as “thinking instruments,”
embodying the skills that analytical chemists were once required to possess. “To de-skill
the analyst,” he writes, “the instrument must be skilled with a material form of
knowledge.” (6) Bennett, meanwhile, locates the adoption of chemical control
technologies within the framework of Taylorist management strategies.
768
See Yakov M. Rabkin, “Technological Innovation in Science: The Adoption of
Infrared Spectroscopy by Chemists,” Terry Shinn, “Research-technology
instrumentation: The Place of Chemistry,” Charlotte Bigg, “Adam Hilger, Ltd. and the
Development of Spectrochemical Analysis,” and Stuart Bennett, “Production Control
Instruments in the Chemical and Process Industries,” all in Morris, ed. From Classical to
Modern Chemistry. For a sophisticated analysis of knowledge transfer between industry
and academia, that casts instruments as boundary objects, and technologists/technicians
as go-betweens, see Apostolos Gerontas, “Creating New Technologists of Research in the
1960s: The Case of the Reproduction of Automated Chromatography Specialists and
Practitioners,” Science and Education 23 (2014): 1681-1700.
341
supplied many of the materials of modern living.769 This context is crucial to

understanding some scholars’ interpretation of these instruments as fundamentally de-

skilling, oriented towards purposes of monitoring and control, and implicated in the

replacement of the labor of experienced scientific professionals with machine-operator

technicians.770

However, in flavor research, the introduction of gas chromatography and

associated technologies came to be seen as marking a rupture between flavor chemistry’s

past, and its postwar present and increasingly bright future, a rupture that was observed

not only by workers in the flavor industry but also by those studying flavor in academic,

government, and military contexts.771 Rather than diminishing the importance, or

supplanting the expertise, of flavor chemists, GC was essential in establishing their

professional status as well as the scientific credibility of their discipline.

769
Ralph Landau and Nathan Rosenberg, “Successful Commercialization in the Chemical
Process Industries,” in Rosenberg, Landau, David C. Mowery, eds. Technology and the
Wealth of Nations, (Stanford: Stanford UP, 1992): 73-119.
770
Although variations of this de-skilling hypothesis are proposed by Baird, Bennett, and
others, this is by no means the only reading of the impact of these machines on chemical
work. In Shifting and Rearranging, Carsten Reinhardt meticulously examines the work
practices and experimental techniques adopted in different chemical subdisciplines using
analytic instruments, concluding that scientists’ use of these machines was anything but
routine. What he observes, instead, is that “a novel kind of method-oriented chemist came
into existence. Their main focus was the development of instrument-based problem-
solving methods for the chemical community at large. In doing so, their work contributed
to the creative interplay of physical and chemical techniques, concepts, and theories as
well as it relied on scientific cooperation and academic-industrial collaboration.” (27)
771
E.g., Roy Teranishi demarcates the “past” of flavor chemistry as “the era before
infrared (IR), nuclear magnetic resonance (NMR), gas chromatography (GC), and mass
spectrometry (MS) were widely used.” In Teranishi, “Development of Methodology of
Flavor Chemistry Past, Present and Future,” in David B. Min and Thomas H. Smouse,
eds., Flavor Chemistry of Lipid Foods, (Champaign, IL: American Oil Chemists’ Society,
1989).
342
Making Gas Chromatography a Tool for Flavor Research

Given the centrality that GC would assume in later accounts of the development

of postwar flavor science, its usefulness to flavor research would come to seem almost

self-evident. Yet despite the recognized analytic power of the machine, establishing a

correspondence among the volatile compounds separated by the instrument, and the

sensory qualities of foods, was complex, labor-intensive, and fraught with uncertainty.

Long after GC had become standard machinery in the flavor research lab, “the human

nose” continued and, indeed, continues to be recognized as “the ultimate instrument in

flavor chemistry.”772

Making GC useful, trustworthy, and meaningful for flavor research involved the

deliberate coordination of researchers, technicians, and machinists, continuous empirical

tinkering, and sensory corroborations, as well as the development of new techniques for

the preparation of samples, the adaptation and modification of commercial machine

components, and the interpretation of results. At the outset, the use of these machines in

the study of flavor demanded from the flavor researcher an intimacy with both the

chemical constituents of foods, and the mechanical and instrumental components of the

devices that purported to reveal them. It is no accident, then, that some of the most

significant work applying analytic instruments to fundamental research into flavor

chemistry occurred at a site that was equipped with a machine shop in addition to

772
Roy Teranishi, Phillip Issenberg, Irwin Hornstein, and Emily L. Wick, Flavor
Research: Principles and Techniques, (New York: Marcel Dekker, 1971): 137.
343
conventional chemical laboratories: the USDA Western Regional Research Laboratory in

Albany, California.

During the Second World War, the Albany lab became a center for research into

dehydrating, freezing, and freeze-drying, food preservation technologies that were of

great interest to the developers of military rations.773 Research in these areas continued

after the war. Regrettably, these processing methods often resulted in foods that were

flavorless, off-tasting, or otherwise unappealing. In order to understand and counteract

the factors behind the decline in quality, the laboratory undertook chemical studies of

flavor.

Keene Dimick, a chemist working at the Albany laboratory, was part of a group

investigating how freezing altered the flavor of fruits. A major impediment to flavor

research had always been separating the volatile flavorful essences from the gross

material of the fruit — the water, fiber, waxes, and other stuff which comprised the bulk

of the fruit’s matter, but contributed little of its flavor. Dimick and his colleagues devised

and modified instruments that could strip and recover volatile compounds from fruit

juices and purees, reliably preserving “the naturally occurring volatiles with as little

alteration in composition as possible.774” Once these volatiles were recovered, the

773
Herrick 1943: 696-98.
774
K.P. Dimick and Benjamin Makower, “A Laboratory-Scale Continuous Vacuum Flash
Evaporator,” Food Technology (December 1951): 517-520; K.P. Dimick and Marion J.
Simone, “A Laboratory Continuous Distillation Column for Concentration of Aqueous
Solutions of Volatile Flavors,” Industrial and Engineering Chemistry 44.10 (October
1952): 2487-90; D.G. Guadagni and K.P. Dimick, “Fruit Flavors: Apparatus and
Procedure for Separation and Estimation of Volatile Components,” Journal of
344
analysis of their contents proceeded using microchemical techniques very similar to those

that had been employed in flavor research at the USDA since the 1920s.

Dimick’s special subject was strawberries. Initially, his goal had been to develop

“an objective chemical test… for assaying flavor potency” — a means of chemically (and

perhaps automatically) determining the intensity and quality of strawberry flavor.775 For

more than six years, he and colleagues Joseph Corse and Benjamin Makower labored to

separate, concentrate, and identify the volatile chemicals responsible for the flavor of

Marshall strawberries, processing 30 tons of the fruit to obtain approximately 35 mL of a

highly concentrated aqueous solution of volatile compounds.776 This concentrate was

then divided into two unequal parts. The larger part, accounting for about ninety percent

of the volume of the concentrate, comprised low-boiling compounds. This solution was

“virtually without a characteristic aroma [of strawberry] and of very low flavor

intensity.” In contrast, the much smaller volume of high-boiling compounds had an

intense aroma “bearing the characteristic fresh-strawberry flavor.”777 Although the high-

boiling compounds were present at levels no higher than 7.5ppm in strawberries, these

intensely aromatic and scarce chemicals seemed to hold the key to what made

strawberries smell and taste specifically like strawberries, and unlike any other fruit.

However, this oily mixture stubbornly resisted the chemists’ attempts at analysis, proving

Agricultural and Food Chemistry 1.19 (December 9, 1953): 1169-70. Quote from
Guadagni Dimick 1953: 1169.
775
KP Dimick and Benjamin Makower, “Volatile Flavor of Strawberry Essence. I.
Identification of the Carbonyls and Certain Low Boiling Substances,” Food Technology
10.2 (February 1956): 73.
776
Ettre “The Pioneers I Knew” 2002: 137.
777
K.P. Dimick and Joseph Corse, “The Volatile Flavors of Strawberry,” American
Perfumer & Aromatics (February 1958): 45.
345
itself “almost intractable.”778 Dimick and his colleagues had no good way isolate and

identify compounds present in this laboriously produced extraction.

Then, in 1953, the researchers got wind of James and Martin’s seminal paper

describing the theory, assembly, and utilization of a gas liquid partition

chromatograph.779 News of the technology may have reached them through the

petrochemical industry. Shell Oil maintained an important research center in nearby

Emeryville, and Dimick and Corse thanked them for extending technical and material

assistance.780 Before GC became commercially available, custom-built units were already

in operation in the laboratories of petrochemical companies such as Shell and British

Petroleum.781 The Albany lab had a dedicated machine shop, where about half a dozen

“very good machinists” assembled scientific instruments, pilot food processing

equipment such as dehydrators, and other tools for the center’s researchers.782 Working

with technicians in the Albany machine shop, using commercially available and custom-

778
Dimick and Corse 1958: 45.
779
James and Martin 1952.
780
In an acknowledgement at the close of their initial publication on GC, Dimick and
Corse thank Shell Development in Emeryville “for their friendly and helpful advice in
constructing the GLPC [ie, gas liquid partition chromatography] apparatus and, in
particular, for the use of the coiled column.” They also thank various WRRC coworkers:
Dr. Lloyd Ingraham for the circuit design, Victor Ortegren for engineering suggestions,
and EF Jansen, a lab chief, for his encouragement in the entire flavor-analysis program.
K.P. Dimick and J. Corse, “Gas Chromatography — A New Method for the Separation
and Identification of Volatile Materials in Foods,” Food Technology 10.8 (August 1956):
364.
781
Ettre 2005: 143.
782
Personal conversation with Ron Buttery, September 2016.
346
built machine components, and with assistance from colleagues at Shell Oil, Dimick and

Corse had an operational GC unit assembled.783

“The development of gas chromatography,” Dimick and Corse wrote, “opens the

door to the flavor chemist to problems which were heretofore essentially unsolvable.”784

Their first publication about the device in Food Technology included a pair of schematic

drawings, a detailed components list including suppliers, and a working drawing of the

thermal conductivity cell detector, as well as operational information — how to pack a

column, how to calibrate the instrument, how to prepare samples for the instrument. All

of this material was intended to guide other researchers in devising and using their own

GC units for flavor research.

Dimick and Corse presented the machine not as a radical departure from prior

methods, but as a faster and more powerful way to perpetrate a familiar chemical

operation: the separation of complex mixtures. “Gas chromatography,” they suggested,

“really represents a superlative fractionating column.” They repeated James and Martin’s

estimate of the GC column’s “phenomenal” efficiency: “2,000 theoretical plates.”785 This

was a metaphorical construct, a calculation of the machine’s separating power expressed

in terms of the familiar distillation plates of glass fractionating columns, although the GC

contained no such components. At the same time, Dimick and Corse insisted that the

machine offered functional possibilities unattainable by classical chemical methods.

783
A materials list for the assembly of a GC unit is included in Dimick and Corse 1956:
361.
784
Dimick and Corse 1956: 360.
785
Dimick and Corse 1956: 361.
347
Compounds with similar boiling points could be cleanly separated by GC without the

frustrating formation of azeotropes, mixtures inseparable by distillation. The machine

also allowed for operational versatility. By changing the stationary phase used in the

column — replacing a silicone oil manufactured by GE with Union Carbide’s Carbowax

coating, for instance — the researcher could use the different polar and chemical

properties of these substances to more effectively separate closely related compounds.786

Finally, the device created a “permanent record” for each analysis — the chromatogram

— whose peaks charted the emergence of each substance, and provided some clues to its

identity and quantity.

When Dimick and Corse began publishing the results of their research in 1956,

interest in the instrument was growing among chemists working in a range of different

fields and industries. “No beautiful movie actress could have drawn a more appreciative

and attentive audience… than did the day-and-a-half Symposium on Vapor Phase

Chromatography” at the Dallas meeting of the ACS in April 1956, reported the Journal

of Analytical Chemistry, using an alternate name for GC. More than 600 people crowded

into the standing-room-only sessions, as academics, government researchers, and

scientists from companies such as Dow, Monsanto, Shell Oil, General Foods, and Phillip

Morris discussed theory, design, use, and applications. Corse presented the Albany lab’s

work on strawberry volatiles.787

786
Dimick and Corse 1956: 362-3.
787
“Standing Room Only at Dallas Analytical Meeting,” Analytical Chemistry 28.5 (May
1956): 13A-14A; “Five-Day Program for Analysts at Dallas,” Analytical Chemistry 28.3
348
 The adoption of gas chromatography by chemical researchers was facilitated by

the commercial introduction of GC units, which made it possible for laboratories and

institutions that did not have a sophisticated machine shop to acquire and use the devices.

In 1955, three US companies began selling GC devices; by 1962, twenty-five

manufacturers were building the instruments.788 (Dimick himself would leave the USDA

at the end of 1956 to start a scientific instrument company with his brother-in-law,

initially assembling GC units in a former bicycle shop.789 That company, Wilkens

Aerograph, would be sold to Varian Associations for $12 million in 1965.790)

Manufacturers — such as Perkin-Elmer, whose Model 154 was one of the earliest and

most widely-used GC units — also promoted the adoption and acceptance of the

machines by publishing detailed operational guides, providing technical assistance, and

working directly with chemists in different fields to optimize instrumental conditions for

different analytic problems. Perkin-Elmer was also the first instrument manufacturer to

also provide and manufacture standard packed columns, with different stationary

phases. 791

Nonetheless, the reliability and utility of GC for flavor research was not cut and

dry. In 1958, Max Winter, a chemist at the important flavor and fragrance company

Firmenich in Geneva, published his laboratory’s analysis of strawberry flavor, using

classical chemical separation techniques including paper and column chromatography.

(March 1956): 25A-26A. The Symposium on Vapor Phase Chromatography begins

Wednesday morning, April 11, and runs through the following morning.
788
Ettre 2005: 148.
789
Ettre 2002: 465.
790
Ettre 2002: 466.
791
Ettre 2005: 146-7.
349
He pointedly contrasted his laboratory’s chemical methods with those obtained by the

Albany lab with the GC apparatus. “There is no doubt that gas chromatography is an

efficient separation method which is employed in all modern analytical research work,”

he conceded, before enumerating two objections to using the machine as a basis for

determining the chemical contents of strawberry flavor. First, in order to obtain a

strawberry concentrate which could be used in gas chromatography, “important

preliminary treatment is required, during which flavor alterations and losses often occur.”

That is to say, GC did not resolve many of the recognized challenges to flavor research,

because it continued to require the manipulation of a food into an analyzable flavor

sample.

Of particular concern was what happened within the black box of the machine,

where the separating sample was inaccessible to the active manipulation and sensory

evaluation of the research chemist. “The special physicochemical conditions of gas

chromatography (temperature, column activity) may involve alterations of especially

unstable substances,” Winter fretted, and there was no simple way of knowing whether

the chromatogram was registering artifacts produced within the machine, or the unaltered

constituents of nature. Although gas chromatography may provide complementary

evidence to support results produced by classical chemical methods, Winter argued that it

could not be relied upon as the primary method of flavor analysis.792

792
Max Winter, “Chromatography on Columns and Paper in the Study of the Volatile
Fruit Flavor of Strawberries,” in Arthur D. Little, Inc., Flavor Research and Food
Acceptance, (New York: Reinhold, 1958): 291.
350
 Winter was certainly not the only researcher to worry about the possibility of

artifacts and other chemical changes. This remained a concern for flavor chemists using

the machines, and they adopted protocols to minimize and forestall these risks. Indeed,

one of the things that made GC a persuasive and useful tool for flavor research was that

the material under study remained available for sensory and chemical analysis after it had

eluted from the machine. For Dimick and Corse, this constituted strong proof of the

device’s reliability. “After we had developed a satisfactory apparatus,” they wrote, “we

were able to test the hope that there would be no appreciable change in odor of a sample

run through it.” To their relief, they found no perceptible difference between the aroma of

the “total collection of the effluent fractions” and the starting material.793

A possible explanation for Winter’s reluctance to place primary in trust the GC

may be found in the distinction between his purposes at Firmenich, and the aims of

Dimick and Corse at the USDA. That is to say, the values that guided Winter’s chemical

research were not equivalent to those that shaped Dimick’s program. Winter was

exquisitely aware of the fleetingness of fresh strawberry flavor. When strawberries were

crushed, he lamented, the finest flavor lingered only for a brief minute. Within five

minutes, “a change in the characteristic components is noted.” After ten minutes, “this

alteration is marked.”794 His goal at Firmenich was to chemically identify, and reproduce,

the evanescent components of that freshness. Classic paper and column chromatographic

techniques, which he preferred to gas chromatography, required that these volatile

793
Joseph Corse and Keene P. Dimick, “The Volatile Flavors of Strawberry,” in Arthur
D. Little 1958: 306.
794
Winter 1958: 290.
351
compounds be converted to nonvolatile derivatives by means of carefully selected

chemical reagents. Thus, he explained, “volatile or unstable substances are rapidly fixed

and thus protected from changes,”795 and remained within the chemists’ attentive and

careful control. His concern was not an efficient analysis of the total volatile contents of

strawberries — a comprehensive list of all strawberry flavor chemicals, in order to rectify

the faults or standardize the quality of frozen strawberry slurries — but the identification

of those compounds responsible for certain unmistakable and remarkable sensory

impressions produced by certain strawberries. For Winter, pinpointing the material

correlates of precise sensory effects took priority over the broader compilation of

chemical presences.

Winter’s recalcitrance toward analytic instruments, expressed in his 1958 paper,

was one of the last damp squibs of resistance to the machines. The vast analytic

capabilities of GC, its accessibility, experimental versatility, and wide professional

acceptance across chemical fields, proved persuasive to flavor researchers in government,

academic, and flavor industry laboratories. However, Winter’s insistence on the primary

importance of sensory evaluation in the chemical analysis of foods reflected concerns that

would increasingly come to shape how devices such as GC were utilized to study flavor.

795
Winter 1958: 291.
352
Feeding the Machine: Making Flavor an Object Of Gas

Chromatography

Even as GC became a standard instrument across chemical sub-disciplines, flavor

researchers had to develop techniques to make the machine work for their particular

purposes. Part of the challenge lay in the relationship between the sample input and the

machine. The GC had to be readied to properly accept the sample, to permit its clear

transit, accurate separation, and the sharp detection of each of its constituent compounds.

Ideally, “the chromatogram would consist of a series of sharp spikes, a dream of all

chromatographers,” with each “peak” representing a single compound.796 This was rarely

the case, especially for flavor chemists, who were attempting to identify molecules

present in extremely small quantities, with a wide range of boiling points, and from a

variety of structural groups. Quite often, peaks on a chromatogram included multiple

compounds, which had to be separated analytically in order to make identifications.

Researchers also had to grapple with column “bleed” and other sources of “noise,” from

which the “signal” of pure, isolated compounds had to be disentangled.

In the late 1950s, three features were introduced to GC units, expanding the utility

of the devices for all users, but with particular consequences for flavor chemistry

research.797 First, ionization detectors replaced thermal conductivity detectors, vastly

increasing the sensitivity of the machine — an extremely useful development for flavor

chemists. Second, linear temperature programming, which allowed the researcher to

796
Robert L. Pecsok, “Gas Chromatography: Basic Principles and New Developments,”
Journal of Chemical Education 38.4 (April 1961): 212.
797
Ettre “The Pioneers I Knew” 2002: 135.
353
gradually change the column temperature during analysis, made it possible to separate

and isolate compounds over a wider range of boiling points. Third, capillary (‘open

tubular’) columns were developed and introduced by the Perkin-Elmer corporation.

These columns expanded separation power by several orders of magnitude. However, due

to the material constraints of their samples, flavor chemists often had to utilize open

tubular columns of larger diameter than those used in other analytic contexts.798

Rather than routinizing the ways the GC was used, these features provided further

opportunities for hands-on tinkering and modification. Samples often went through the

GC multiple times, running separated fractions through the machines again, with columns

packed with different liquid media to enhance their separations.799 Even after standard

packed columns became available, flavor researchers continued to prepare columns

themselves, hanging them in stairwells to ensure the even distribution of the liquid

798
This was particularly the case when combining headspace analysis (to be discussed
below) and capillary columns. The necessary sample size for adequate headspace analysis
often overwhelmed the capacity of 0.01-in i.d. standard capillary columns, and so
experimenters used larger-diameter, longer capillary columns, sometimes fusing them
themselves to obtain the desired dimensions. See Jennings, “The Neanderthal Age of Gas
Chromatography.” For a discussion of the headspace/capillary problem in an
experimental context, see Ron G. Buttery and Roy Teranishi, “Measurement of Fat
Autoxidation and Browning Aldehydes in Food Vapors by Direct Vapor Injection Gas-
Liquid Chromatography,” Agricultural and Food Chemistry 11.6 (Nov-Dec 1963): 505.
799
This is known as “preparatory chromatography,” and was a particularly common
practice in flavor research because of the chemical heterogeneity of the substances of
interest in foods. By using different liquid phases, and skillful temperature programming,
a chemist could resolve what had appeared as a single peak into multiple component
fractions in subsequent runs through the GC. Dimick’s GC company, Wilkens Instrument
& Research, specialized in devices that simplified preparatory GC. Ettre “Early
Development and Rapid Growth of Gas Chromatographic Instrumentation” 2002: 465-6.
354
phase.800 Researchers freely experimented with various liquid phases, searching for

substances that helped separate stubborn peaks into constituent compounds. Among

commercially available waxes and silicon oils, Tide, the laundry detergent, was found to

be particularly effective. “It will not be long before everything in the stockroom has been

tried as a stationary phase,” one chromatographer commented in the early 1960s.801 Glass

melting point tubes, standard equipment of any chemical laboratory, were inserted and

attached into columns, to create a “sniff port” by which eluted fractions could be

organoleptically examined.802

It was not only the machine that had to be tinkered with to make it suitable for

flavor research. The sample that was introduced into the machine also raised questions

for researchers. A researcher couldn’t just feed a strawberry into the GC, or a slice of

roast beef, or a wafer of toasted bread, and await the automated results of the machine’s

analysis. Investigators had to produce, from the food, a sample, one that was legible to

the GC and conformed to the machine’s technical requirements, while also accurately

representing the complex of flavor chemicals as they existed in the food.

One approach to converting opaque food into a sample that the GC could analyze

was already very familiar to flavor chemists: careful distillation and extraction, generally

with organic solvents such as isopentane and ether, to concentrate the volatile substances

of interest and separate them from water, waxes, and other materials that could disrupt

800
Walter Jennings, “The Neanderthal Age of Gas Chromatography”; Personal
communication, Alfred Goossens, 2015.
801
Pecsok 1961: 215.
802
Jennings, “The Neanderthal Age of Gas Chromatography.”
355
the sensitivity of the GC. Dimick and his colleagues at Albany had done just this when

applying the GC’s separating powers to the recalcitrant remnant of strawberry volatiles

that they had produced using specially designed flash evaporation and distillation

equipment.803

These processes were laborious and material-intensive, introduced the risk of

producing chemical artifacts (as Winter had warned), and inevitably resulted in the loss

of some compounds and the disproportionate collection of others. Further, even in the

best case scenario — one where procedures were followed impeccably, and contained

unaltered all the volatile compounds that were present in the food — this sample was

only an approximation of the human sensory experience of a food. This is because the

volatility of chemical compounds, and thus their apparent sensory qualities, is affected by

their immediate material environment — by the food system that they are contained

within. This is why adding salt to a broth enhances its aroma (by lowering the vapor

pressure), and why a perfume oil smells different than the same compounds in an

alcohol-based eau de toilette. When volatiles were isolated from food, “the original

quantitative interrelationships of aroma components is destroyed,” as one textbook on

flavor chemistry explained.804 Although the sensory differences may be subtle, they were

inarguable. The chemical composition of the aromatic vapor over a distillate of apple

volatiles, for instance, was not equivalent to the compounds that launched themselves

803
Dimick and Corse 1956: 363-4.
804
Teranishi et al. 1971: 39-40.
356
into the atmosphere, entering the olfactory region either through the nostrils with the

intake of breath, or retronasally, as the apple itself was consumed.805

As flavor chemistry became increasingly successful at extracting information

about the chemical composition of food, the desideratum of analysis — the meaning of

flavor as a scientific object — also changed. Rather than aiming for the identification of

the total content of volatiles present within a certain type of food, flavor researchers

increasingly attended to the relationship between the analyzed sample and the human

sensory experience of the food.

Flavor chemists at the Albany laboratory played a leading role in the development

of a method of flavor sampling that sidestepped many of the more tedious aspects of

preparation and also seemed to offer a more meaningful sample — not of the volatiles

present within the food, but of the aroma perceptible above it. In the early 1960s, Roy

Teranishi, the chemist who had replaced Dimick after he had left to start Wilkens

Aerograph, and his colleague Ron Buttery published a paper describing a technique for

direct vapor sampling and analysis — what would later come to be referred to as

headspace analysis.806 Essentially, the method involved placing a portion of food in a

closed container — often, a 250mL glass Erlenmeyer flask — and allowing it to stand for

several minutes, so that volatile compounds reached equilibrium in the flask’s

atmosphere. A syringe was then plunged into the container, extracting five to ten cubic

805
Teranishi et al. 1971: 19.
806
Ron G. Buttery and Roy Teranishi, “Gas-Liquid Chromatography of Aroma of
Vegetables and Fruit: Direct Injection of Aqueous Vapors,” Analytic Chemistry 33.10
(September 1961): 1439-41.
357
centimeters of vapor, which was immediately injected into the GC column via the input

port.807

Headspace analysis would not have been possible without the coordination of

multiple new technological components within the GC machine complex. Of primary

importance was the detector. The first GC units used thermal conductivity detectors

(TCDs), which registered differences between the pure carrier gas and the carrier gas

mixed with the sample vapor.808 The low concentration of volatile molecules in

headspace vapor would have been well below the threshold of these devices; moreover,

they were sensitive to water vapor and air molecules, which meant that analysis of

headspace samples would be distorted by considerable “noise.” In the late 1950s, flame

ionization detectors (FIDs) were introduced, which were orders of magnitude more

sensitive than TCDs, while also being insensitive to water vapor.809

807
A comprehensive description of the method can be found in: Roy Teranishi, Ron G.
Buttery, T.R. Mon, “Direct Vapor Analyses with Gas Chromatography,” Annals of the
New York Academy of Sciences 116 (“Recent Advances in Odor: Theory, Measurement,
and Control,”) (July 1964): 583-9.
808
“In essence, a differential thermal conductivity cell is a Wheatstone bridge in which
carrier gas flows over or near two of the resistors (thermistors are sometimes used) and
the column effluent over or near the other two…. The cell resistors are heated by passage
of current, and the bridge is adjusted so that it is in balance when only carrier gas is
coming through the column. Emergence of sample from the column causes a change in
the thermal conductivity of the effluent as compared with the carrier gas. This results in a
change in temperature, and consequently a change in resistance, of the resistors exposed
to the column effluent. The resulting unbalance of the bridge is then applied to the
recorder.” This article noted that thermal conductivity cells were commercially available
from multiple manufacturers, but also that “their construction is not beyond the ability of
a good technician,” particularly as wiring diagrams and designs had been published. John
R. Lotz and Charles B. Willingham, “Gas-Phase Chromatography,” Journal of Chemical
Education 33.10 (October 1956): 487.
809
Ettre 2005: 148.
358
 Headspace analysis radically reduced the preparatory labor required to introduce a

flavor complex to the GC unit. Teranishi and Buttery called it “zero time analysis — i.e.,

no time lapse for extraction or concentration.”810 It was also non-destructive. The

sampled food remained intact. This opened up the possibility of studying flavor as a

dynamic phenomenon, including systematically studying flavor changes due to

processing or storage. For instance, Teranishi and Buttery studied the development of off-

flavors in stored dehydrated Idaho Russet potatoes and freeze-dried carrots, and also the

chemistry of browning reactions in these foods.811 Other researchers studied changes in

the composition of onion volatiles over time, tracking via chromatogram of headspace

volatiles how the aroma varied between a just-sliced onion and the same onion seventeen

hours later.812 The method also showed the “release” of a spray-dried banana flavor, after

being mixed with water.813

It also seemed to offer a more meaningful sample — one that directly

corresponded with a food’s flavor as it was experienced. This meant being able to make

ready comparisons between, for instance, the composition of peppermint oils considered

810
Roy Teranishi, Ron G. Buttery, and Robert E. Lundin, “Gas Chromatography. Direct
Vapor Analyses of Food Products with Programmed Temperature Control of Dual
Columns with Dual Flame Ionization Detectors,” Analytical Chemistry 34.8 (July 1962):
1034-5.
811
Ron G. Buttery, “Autoxidation of Potato Granules II: Formation of Carbonyls and
Hydrocarbons,” Agricultural and Food Chemistry 9 (1961): 248-52; Buttery and
Teranishi 1963.
812
Donald A.M. Mackay, David Lang, and Murray Berdick, “Objective Measurement of
Odor: Ionization Detection of Food Volatiles,” Analytical Chemistry 33.10 (September
1961): 1373.
813
Mackay et al. 1961: 1371.
359
“high quality” and “low quality” by sensory evaluation panels.814 Or multifactoral

comparisons between different varieties of frozen strawberries judged to be of varying

quality by taste panels, in order to pinpoint the chemosensory factors responsible for

quality.815 For this reason, Teranishi and Buttery initially presented headspace analysis as

a replacement for time-consuming, uncertain taste panel methods of quality control. In

this application, quality control would become a matter of reading a chart, and watching

for the visual indicators of sensible trouble. Indeed, their paper did not attempt to identify

any of the substances whose presence or absence accounted for the changes in the GC

curves of the vapors above fresh carrots and those that had been in the freezer for two

years; nor of the fresh dehydrated potato granules and those stored a year under

inauspicious conditions. Instead, the chromatogram served as a graphical index of

sensory quality and sensible change, what they dubbed an “aromagram.”816 For instance,

using headspace analysis, Buttery had identified n-hexanal as a correlate for spoilage in

stored dehydrated Idaho taters.817 It was clear that this compound was not the one

responsible for the off-odor that developed in ‘spoiled’ stored dehydrated potatoes. But,

as its presence reliably indicated the degree of spoilage, it could be used as an index of

quality despite the fact that the chemical compounds responsible for the off-flavors

remained unknown.

Other researchers, however, understood headspace analysis not as a replacement

for sensory evaluation techniques, but as a means of correlating sensory experience

814
Mackay et al. 1961: 1370.
815
Teranishi et al. 1964: 588.
816
Buttery and Teranishi 1963: 504-7.
817
Buttery 1961.
360
directly with a GC sample. In other words, they understood the technique as making

possible a direct comparison between the experience of a food’s aroma and the analytic

account of its components as produced by the machine.818 This allowed for the

application of sensory panel and flavor profiling techniques to captured fractions and GC

effluent, with the goal of identifying compounds in terms of both chemical structure and

sensory effect. This method was used, to give one example, by a collaborative group of

researchers associated with the United Fruit Company (one of the leading Central

American banana concerns) and Arthur D. Little, Inc., in order to assess the

physicochemical and sensory qualities of bananas that might be chosen to replace the

Gros Michel cultivar, given that varietal’s apparent susceptibility to fungal Panama

disease blight.819

Headspace sampling, of course, had its limitations, of which flavor chemists were

well apprised. Low-boiling compounds were often below the threshold of sensitivity of

the machine; artifacts could be introduced in numerous ways; often, columns with less-

than-optimal resolving power had to be selected in order to minimize noise.820 But the

development and refinement of these techniques alongside other methods of sampling

show GC as an instrument that was being honed in on questions of “flavor” rather than

818
See Mackay et al. 1961: 1374. Their paper concludes: “The odor of large numbers of
materials can now be assayed by direct, nondestructive sampling of the vapor above the
material. Thus, for the first time, direct correlation with sensory evaluation is possible.”
819
Alice I. McCarthy, James K. Palmer, Carol P. Shaw, and Edward E. Anderson,
"Correlation of Gas Chromatographic Data with Flavor Profiles of Fresh Banana Fruit."
Journal of Food Science 28.4 (July 1963): 379-384.
820
For a discussion of the hazards of headspace sampling and best practices, see: S.G.
Wyllie, S. Alves, M. Filsoof, and W.G. Jennings, “Headspace Sampling: Use and
Abuse,” in George Charalambous, ed. Analysis of Foods and Beverages: Headspace
Techniques, (New York: Academic Press, 1978): 1-15.
361
“volatile materials,” and flavor chemists working to make the machine’s results

meaningful and sensible.

“Truly Synergistic”: Gas Chromatography-Mass

Spectrometry

While GC was a powerful tool for the separation of complex mixtures, it did not

provide a ready route to confidently identifying compounds after they had been

separated. The chromatogram did offer some clues. Because retention time was

logarithmically related to boiling point, the boiling point of an unknown substance could

be approximated by comparing its retention time with those of known compounds that

had been used to calibrate the machine.821 In the 1960s, a standardized set of retention

indices was compiled, which helped guide these types of identifications.822 In most cases,

however, other steps were necessary to conclusively identify the components GC had

separated. Here again, chemists ran into a series of difficulties. The quantity of volatile

material that comprised each peak was miniscule — generally topping out at only a few

micrograms — frequently unstable, and often unknown. Rather than using classical

chemical methods to identify these compounds, analytic chemists generally turned to

other machines — in particular, to spectrometric instruments, such as infrared, nuclear

magnetic resonance, and mass spectrometers. These devices provided information about

molecular structure, which chemists interpreted to identify molecules.

821
Dimick and Corse 1956: 361.
822
I.e,, the Kovats Retention Index.
362
 The most generative instrumental relationship was between GC and mass

spectrometry (MS). Based on electrophysical principles articulated by J.J. Thomson in

the early twentieth century, mass spectrometers ionize and separate molecules in a

vacuum chamber. Particles are then detected and converted into a signal that plots mass-

to-charge ratio against signal intensity/relative abundance.823 These mass spectra can then

be used to deduce molecular structure, thus making it an extremely useful tool for the

identification of unknown compounds.

The pairing of gas chromatography and mass spectrometry was, in the words of

William Stahl, the chief of the analytical section at the Quartermaster’s Pioneering

Research Division, “truly synergistic.”824 Stahl’s group had been instructed in the use of

GC in flavor research by Joseph Corse, Dimick’s partner in the USDA Albany lab.

However, rather than using GC to analyze and identify, they used the machine “simply as

an elegant means of separation.”825 GC fractured complex mixtures into isolated

individual components, but fell short in allowing for confident identifications of those

compounds. MS excelled at producing structural information that facilitated

identification, but interpreting mass spectra could be nearly impossible when the

823
Simon Maher, Fred P.M. Jjunju, and Stephen Taylor, “Colloquium: 100 Years of
Mass Spectrometry: Perspectives and Future Trends,” Reviews of Modern Physics 87
(January-March 2015): 113-135.
824
William H. Stahl, “Gas Chromatography and Mass Spectrometry in the Study of
Flavor,” in Mitchell et al. 1957: 58.
825
Stahl in Mitchell et al. 1957: 63.
363
substance was a mixture of compounds. What GC could purify, MS could readily

identify.826

Initially, researchers manually transferred samples collected from the GC’s

effluent to the MS for identification. In his 1957 presentation at the Quartermaster

symposium on the Chemistry of Natural Flavors, Stahl described a series of traps

connected to the exit port of the GC. By manipulating valves and stopcocks, these could

be used to collect and isolate fractions in separate containers.827 These could then be

introduced one by one into the MS, which would then display spectra for the

investigator’s interpretation.

As can be imagined, this was a fussy, time-consuming, laborious process that

demanded close attention and considerable skill. One analytic chemist working at Dow

estimated that it took about 20 to 40 minutes to obtain a spectrum for each fraction and to

prepare the instrument for the next sample. As he explained: “This would mean that if

one had a chromatogram containing 10 peaks whose identity was desired, it would

require 3 to 8 hours of mass spectrometer instrument time to obtain the mass spectra of

the fractions, in addition to the time required to collect them.”828 Because of the

826
Teranishi et al. Flavor Research 1971: 27
827
The technique of trapping fractions from the GC effluent for further analysis would
become quite common. Trapped “peaks” that were suspected of containing more than one
compound could be run through the GC again under different conditions, to improve
resolution. However, because most commercially available traps were designed for
preparative work that yielded larger quantities of material, these were generally too big
for the sample sizes that flavor chemists generally dealt with.
828
R.S. Gohlke, “Time-of-Flight Mass Spectrometry and Gas-Liquid Partition
Chromatography,” Analytical Chemistry 31.4 (April 1959): 536.
364
instability of many volatile compounds, the time lapse between collection and scan could

result in chemical changes and introduced the possibility of erroneous identifications.

The commercial development of a dynamic MS instrument capable of producing

a complete spectrum every few microseconds transformed the utility and utilization of

both GC and MS, and shaped a conjoined destiny for those instruments in analytic

chemistry, including flavor research. In the early 1950s, physicists working at the Bendix

Aviation Corporation in Detroit developed a new kind of ion gun which was capable of

providing a very high resolution beam. This stable, high-resolution ion source, along with

other modifications, made it possible for Bendix to build a Mass Spectrometer with an

extremely high scan rate: the Time-of-Flight Mass Spectrometer.829 This machine could

produce a complete spectrum every few microseconds. Because it used electronic circuits

rather than magnetic fields, it was also smaller, simpler to build, and easier to operate.830

Bendix began custom-building these machines shortly after the introduction of the

first commercial GC units in the mid-1950s. The company anticipated two primary

applications for these devices. First, the analysis of very fast chemical reactions, which

could lead to the more efficient production of synthetic chemicals. Second, the

identification of separated components as they emerged from the GC. “Using the Bendix

spectrometer,” one of its creators explained, “the identification of the emerging

components can be made simply by allowing a portion of the effluent gas to pass into the

spectrometer.” As each compound passed into the MS, its spectrum would rise and fall on

829
W.C. Wiley and I.H. McLaren, “Time-of-Flight Mass Spectrometer with Improved
Resolution,” Review of Scientific Instruments 26.12 (December 1955): 1150-7.
830
Wiley and McLaren 1955: 1150.
365
the oscilloscope screen, allowing for identifications to be made. A permanent record of

the spectra could be made with an oscillographic recorder or on analog magnetic tape.831

Bendix believed that these machines were primarily suited for process monitoring

and process control in the context of the industrial production facilities, rather than

chemical research in industrial or academic laboratories.832 The actual conjugation of GC

and MS was initially left to instrument users, especially those working in industrial

laboratories within the chemical and petrochemical industries. Conjoining GC and MS

meant more than just connecting the GC’s effluent stream to the MS’s input port. It

required deliberately adapting both technologies to each other in order to produce as

much reliable (and interpretable) information as possible. Flow rate, pressure,

temperature, and other factors had to be adjusted in order to minimize noise and optimize

resolving power for both instruments.833

Once again, the USDA ARS laboratory in Albany played an important role here,

developing techniques, refining technologies, and publishing papers about using

831
W.C. Wiley, “Bendix Time-of-Flight Mass Spectrometer,” Science 124 (26 October
1956): 818. In the 1960s, spectra began to be recorded onto reels of magnetic tape. This
provided a cost-effective and efficient way of recording continuous spectra produced
during long chromatographic separations, allowed for greater flexibility in data handling
and analysis, and could be integrated with computer-aided systems of data recording and
processing that were just beginning to come into use. For a discussion of the multiple
advantages of magnetic tape in this context, see: Phillip Issenberg, Akio Kobayashi, and
TJ Mysliwy, “Combined Gas Chromatography-Mass Spectrometry in Flavor Research:
Methods and Applications,” Journal of Agricultural and Food Chemistry 17.6 (Nov/Dec
1969): 1380-1.
832
Wiley 1956: 819.
833
For a discussion of possible difficulties in joining GC and MS, see Roy Teranishi,
R.E. Lundin, and J.R. Scherer, “Analytical Technique,” in H.W. Schultz, E.A. Day, and
L.M. Libbey, eds. Symposium on Foods: The Chemistry and Physiology of Flavors,
(Westport, CT: AVI, 1967): 176.
366
combined GC-MS in flavor research.834 In particular, Teranishi, W.H. McFadden, and

other chemists at Albany and at nearby UC Davis refined the use of capillary column GC

with Time-of-Flight MS.835 Even though capillary columns had been commercially

available since the late 1950s, their use had been limited in flavor chemistry because of

the challenges of collecting and delivering the increased number of captured, small-

quantity fractions to the MS.836 With the machines conjoined, that difficulty was

removed. Capillary column GC also facilitated the use of mass spectra for identification

as it was more likely to deliver pure compounds to the machine, resolving what had

already seemed like fine separations. As an example, McFadden describes the analysis of

a tiny sample — between three and four microliters — of volatile oil that had been

obtained, “after a laborious series of chemical and extractive separations,” from five

thousand pounds of fresh peas. The pea oil resolved into twenty-two “clear peaks” on a

packed column — at first glance, a fine separation. But an analysis on a capillary column

GC revealed thirty-nine separate compounds.837

The GC-MS did not just expand the capabilities of the flavor lab, it made the

work more efficient — transforming what had been a batch process into a continuous

834
Dennis O’Brien, “Cited for More than 60 Years of Flavor Research,” Agricultural
Research Magazine (May-June 2013): 15.
835
W.H. McFadden, Roy Teranishi, D.R. Black, and J.C. Day, “Use of Capillary Gas
Chromatography with a Time-of-Flight Mass Spectrometer,” Journal of Food Science
28.3 (1963): 316-19; W.H. McFadden and Roy Teranishi, “Fast-Scan Mass Spectrometry
with Capillary Gas-Liquid Chromatography in Investigation of Fruit Volatiles,” Nature
200 (October 26, 1963): 329-30.
836
Teranishi et al. in Schultz et al., eds 1967: 170
837
McFadden et al. “Use of Capillary Gas Chromatography…” 1963: 318.
367
process.838 This is not to suggest that there was anything ‘automatic’ about the process of

separation and identification. Investigators often used both the chromatogram and the

mass spectra to make identifications, a procedure that was facilitated by the two gates on

the Bendix Time-of-Flight MS Model 12; they also relied on a synthesis of the compound

(or a commercially available sample) to confirm identifications.839 Even so, major

challenges still remained to making identifications with full confidence using

instrumental data alone — isomers, for instance, could be almost impossible to

distinguish from mass spectra, and could have substantial sensory consequences.

“GC-MS is a tremendous tool and you can obtain much information from it,”

remarked Roy Teranishi at a 1966 symposium on flavor chemistry. “You can determine a

large number of compounds with this technique or at least determine which ones are

interesting, and go on from there. I think, however, that it is very dangerous to say that

you are going to see all and tell all.”840

By the late 1960s, the instrumental assemblage of the flavor research laboratory

was more or less in place. GCMS, as well as other instruments, including infrared

spectroscopy and NMR were routinely used for separations and identifications.841 The

next apparatus to be added to the instrumental assemblage, the computer, was already

838
This comparison is made in: Roy Teranishi et al. in Schultz et al., eds 1967: 170.
839
McFadden et al. “Fast Scan Mass Spectrometry” 1963: 329; Teranishi et al. in Schultz
et al. 1967: 172.
840
Teranishi et al. in Schultz et al. 1967: 178.
841
Teranishi et al. in Schultz et al. 1967; Issenberg et al. 1969; Teranishi et al. 1971.
368
visible on the horizon.842 But the machines had their limits. “Even when a computer is

added to the system,” one group off flavor researchers based at MIT reported in 1969,

“GCMS will produce only vast quantities of uninterpretable data unless all other

chemical, instrumental, and sensory methods are considered and applied whenever

appropriate to the solution of flavor problems.”843

In particular, the sensible body of the expert flavor researcher was needed as part

of the instrumental assemblage, making sense of the growing number of chemical

compounds that the efficient analytic machine complex produced.

The Flavor Chemist and the Machine

The analytic instruments of flavor chemistry, GC and MS chief among them,

presumed (and produced) a materialist theory of flavor. According to these machines, the

flavor of foods is produced by complex mixtures of volatile organic chemical substances,

which can be separated into discrete compounds that can be identified by their physical

properties.

However, GC stood in contrast to contemporary devices that were explicitly

designed to respond like the human olfactory system, simulating theorized mechanisms

of olfaction to register the presence, absence, or concentration of certain odors.844 For

842
Craig B. Warren and John P. Walradt, eds. Computers in Flavor and Fragrance
Research, ACS Symposium Series 261, (Washington, D.C.: ACS, 1984).
843
Issenberg et al. 1969: 1385.
844
Julian W. Gardner and Philip N. Bartlett, “A Brief History of Electronic Noses,”
Sensors and Actuators B 18-19 (1994): 211-220; R.W. Moncrieff, “An Instrument for
Measuring and Classifying Odors,” Journal of Applied Physiology 16 (1961): 742-9;
369
instance, the smelling machines devised by John Hartman, a professor in the Department

of Vegetable Crops at Cornell, were grounded in a mechanistic analogy between odor

receptors and sensitized electric circuits; “olfactory receptor hairs essentially act as

polarized microelectrodes,” he wrote. 845 His machines deployed a varied array of

microelectrodes to obtain differential responses to distinct odorants.846 Hartman

aspirationally compared his machine to optical machines such as the Color Difference

meter, and hoped it could be used as “a device that can characterize a flavor, both for

quality and quantity, by the pattern of reactions at a series of sensing elements.”847 Once

calibrated to and standardized for consumer preferences, it could—he hoped—become a

machine for objectively assessing quality.

Dravnieks and Trotter, “Polar Vapor Detection Based on Thermal Modulation of Contact
Potentials,” Journal of Scientific Instruments 42 (1965): 624.
845
Walter F. Wilkens and Hartman, “An Electronic Analog for the Olfactory Processes,”
Journal of Food Science 29.3 (1964): 373.
846
John D. Hartman, “A Possible Objective Method for the Rapid Estimation of Flavor in
Vegetables,” Proceedings of the American Society of Horticultural Sciences 64 (1954):
335-42; Hartman and W.F. Tolle, “An Apparatus Designed for the Rapid
Electrochemical Estimation of Flavor in Vegetables,” Food Technology 11 (1957): 130-
2; Wilkens and Hartman 1964: 372-8. Interestingly, John Hartman would later achieve
another kind of notoriety at Cornell in the early 1970s, when he became the first tenured
professor to be considered for dismissal by the University administration. Hartman had
refused to teach any classes or conduct any research since 1969. “Instead of using his
green thumb,” explained an editorial in the Cornell Daily Sun, “Hartman has been writing
redneck essays of extraordinary length and incredible obtuseness for several years,”
including one condemning the University Senate’s decision to boycott lettuce grown by
non-unionized workers in solidarity with the United Farm Workers Organizing
Committee, and a “memorable apologia” for police after a brutal response to student
demonstrators in 1972. Although the editorial writer clearly had no love for Hartman, he
criticized the University’s secret hearings to adjudicate the professor’s case, and their
decision to allow Hartman to take early retirement and become an emeritus professor, a
status the writer did not think he deserved. Instead, he suggested that Hartman was fit for
a position in the University administration. Gordon Chang, “Hartman for Provost,”
Cornell Daily Sun, September 5, 1973, 4.
847
Hartman and Tolle 1957: 130.
370
 The gas chromatograph certainly had the potential to be used in this way. Some

early accounts of the gas chromatograph described it as a “sniffing machine,” one that

could replace human evaluators in determining food quality, while also producing a

“permanent record,” a transcription of aroma that translated evanescent and subjective

experience into a scannable visual data chart. As one 1957 article touting the applications

of GC in coffee roasting explained, “the ‘picture’ of the various aroma components in

each sample comes out as a wavy line on a tape much as human pulse reactions are traced

on a lie detector.”848 And, just as the lie detector produced the visible “evidence” of

subjective internal states of mind, it was hoped that the GC could likewise be used to

accurately determine sensory characteristics without relying on the unreliable disclosures

of human sensors.849 Those who worked with the machine also speculated on its potential

as a quality control device, one that could replace the routine labor of human evaluators.

For instance, in an early paper describing headspace analysis, Buttery and Teranishi had

suggested that the technique could be used to “objectively” monitor product quality,

supplanting subjective, uncertain organoleptic evaluations.850 In this application, quality

control might become a matter of reading a chart, and watching for the visual indicators

of sensible trouble.

But while it remained imaginable that GC could supplant certain kinds of

routinized sensory labor, it was likewise insisted that the machine could never automate

848
W. Schweisheimer, “Sniffing Machine Savors Coffee Producing a ‘Picture’ of Coffee
Aroma,” Perfumer & Essential Oil Review 48 (September 1957): 443.
849
For a history of the polygraph, and instrumental tests of deception, see Ken Alder, “A
Social History of Untruth: Lie Detection and Trust in Twentieth-Century America,”
Representations 80.1 (Fall 2002): 1-33.
850
Buttery and Teranishi 1961.
371
the expert labor of the flavor chemist — who was needed not only to induce the food

sample and machine to produce reliable chemical results, but also to interpret those

results and make them meaningful. What had long been evident to flavorists working in

the flavor industry became increasingly clear to chemists studying flavor in other

institutional contexts: an exclusively materialist definition of flavor, one that relied

exclusively or even primarily on instrumentally aided separations and identifications, was

insufficient.

“NASAL APPRAISAL”

The centrality that instruments had assumed in flavor chemistry is vividly evident

in the 1971 textbook, Flavor Research: Principles and Techniques.851 The textbook dealt

851
Roy Teranishi, Irwin Hornstein, Phillip Issenberg, and Emily L. Wick, Flavor
Research: Principles and Techniques, (New York: Marcel Dekker, 1971). Authored by
two USDA researchers (Teranishi and Hornstein) and two members of MIT’s Nutrition
and Food Science department (Issenberg and Wick), this was the first textbook to
exclusively focus on methodologies of flavor research. It was also the first title published
in a series of food science monographs edited by Owen Fennema, a highly regarded
professor of food chemistry in the University of Wisconsin College of Agriculture
Department of Food Sciences and Industries. As can be inferred from the institutional
position of its authors and its publishing context, this book was intended for students who
were anticipating careers in basic or applied research in the food industry, in government
laboratories, or the academy, and it reflects the increasing professionalization of food
science fields. This textbook did not provide any instruction in or discussion of
techniques for formulating flavor additives or designing food flavors. Nonetheless, the
anticipated need for the kind of skills the textbook taught were framed in terms of the
manipulation and production of flavors for foods. The preface explained: “As the
population increases, our caloric and protein requirements inevitably will be supplied by
unconventional foods…. Protein foods derived from petroleum, oilseeds, legumes and the
like may be man’s dietary lot. The persistence of food preferences for the flavors of
traditional foods may prove a major roadblock to the utilization of these new foods. To
372
almost entirely with the proper use of analytic machines: GC, MS, and combined GC-

MS, as well as other spectrometric methods. Students were led through the special

considerations and techniques required when using these instruments to study flavor —

from preparing the sample, to setting up and operating the instrument, to interpreting the

results. There is no discussion of classical chemical techniques of identification; no

mention at all of reagents or reactions.

In the midst of its near-exclusive focus on the use of analytic machines, Flavor

Research repeatedly stressed that the experimenter’s attentive, sensible body was

indispensable in carrying out this work successfully. Throughout, the textbook prescribed

the necessity of continual and careful “nasal appraisals.” This was a response to the

familiar difficulties that came working with unstable, promiscuous, and volatile flavor

materials. At every stage, something crucial may have been lost, or something may have

changed. It was critical that the experimenter confirm, for instance, that the headspace

sample or aroma concentrate that was delivered to the GC faithfully demonstrated the

organoleptic qualities of the food being studied.852 It was also important to ensure that the

machine’s output duplicated the sensory qualities of its input, and that no important

component had been adsorbed by the column, or altered during its passage.853

There was also the matter of the nose’s superior sensitivity. Even with the

improved sensitivity of FID detectors, human subjects could often detect the presence of

incorporate into fabricated foods, flavors acceptable to populations differing in their food
likes will add a new urgency to the flavor chemists [sic] efforts.” (vi)
852
Teranishi et al. 1971: 72, 126-7, 130.
853
Teranishi et al. 1971: 126-7, 262.
373
chemicals at far smaller concentrations than the machine was able to register. Thus, a

fraction eluted from the chromatogram could appear “chemically pure,” when a sniff

would reveal that it was far from “organoleptically pure.”

Most crucially, however, the instrument’s response to chemical compounds was

not analogous to an embodied response; these were fundamentally different kinds of

phenomena. Absent a functional theory of olfaction that could connect molecular

structure with sensory quality, the body’s response could not be deduced from the

machine’s results. In other words, the issue was not that the nose was more sensitive than

the machine, but that the GC was not a body. “The flavor chemist’s job is, in some sense,

similar to those of the biochemist, pharmacologist, and toxicologist,” the 1971 textbook

explained. “He is interested in small quantities of organic chemicals present in very

complex mixtures, the components of which exert some physiological effect. In flavor

research, this physiological effect is contribution to flavor.”854 The question was not,

what compounds are present in this food? It was, which of the compounds present

produce detectable effects on a body?

“Flavor is more than a pattern of peaks on a chromatogram,” explained Rose

Marie Pangborn, C.S. Ough, and Herbert Stone, in a seminal article on sensory evaluation

published in Advances in Food Research. “Flavor is an integrated response, the nasal

mucosa and the taste buds being the integrators. The chromatograph, on the other hand, is

a separator, which, while an extremely useful tool, must have its responses compared

854
Teranishi et al. 1971: 136.
374
with human responses to have a bearing on flavor.”855 Pangborn and her colleagues were

situated, in terms of disciplinary positioning, within the sensory laboratory, rather than

the chemical lab. However, this redefinition of flavor as an integrated, embodied

response — rather than an analytically produced set of chemical compounds — was

crucial to the turn in flavor chemistry research that occurred in the mid-1960s, one that

occurred in response not only to the claims made by sensory researchers such as

Pangborn, but also to the results produced by the machine itself.

This was not the nose jockeying for sniff-supremacy over the machine, but a

conjugation of the two — a joining of forces. The difference between GC and human

body was the source of the power of the ultimate laboratory instrument, which utilized

machine and body together in complementary ways. This conjugation was facilitated by a

modification to GC units, one which had become common among flavor researchers by

the 1960s. The GC split the effluent to an olfactometry port, which allowed a “human

sensor” to monitor and characterize and annotate fractions as they exited the machine.856

Various modifications were made to these for the comfort of researchers, including the

introduction of moisture into the effluent to prevent bloody noses from intensive sniff

sessions.857

855
Herbert Stone, Rose Marie Pangborn, and C.S. Ough, “Techniques for Sensory
Evaluation of Food Odors,” Advances in Food Research 14 (1965): 19.
856
An STS-informed account of GC-Olfactometry can be found in Christy Spackman,
forthcoming, Senses & Society 2017.
857
R.H. Potter and J. Daye, “Apparatus to Introduce Moisture into Effluent Gas,”
Givaudan Flavorist (1970): 8.
375
 The role of the sensible researcher, then, was to serve as the susceptible, in vivo

medium that could distinguish flavor compounds from other chemicals, to allow for “a

bioassay of aroma based on the stimulation of the human nose.”858 Results would be

gauged qualitatively, based on human judgment of the intensity and characteristics of

different fractions and their relationship to the odor of the whole. “Somehow, one tends

to feel safer with bioassays that use the gain or loss of weight of test animals or the

physiological state of rat livers as evidence of biological activity than one does at the

prospect of asking impressionable, opinionated, and unreliable human beings to judge

aroma in isolates,” the authors conceded. “This attitude must be overcome and very

serious efforts made to use sensory evaluation to select the gas chromatographic peaks of

importance, thus avoiding the necessity of identifying all components and then

determining their flavor contribution.”859 In other words, the “nasal appraisal” was

important not only for confirming accuracy, but also for managing the (increasingly

overwhelming) labor of flavor research. The efficiency of GC’s separations had the

consequence of registering the presence of hundreds of compounds, many of them

unknown, in coffee, cheese, and other foods. This efficient production of chemical data

could overtax researchers.

William S. Ryder, a flavor researcher at the General Food’s Technical Center in

Tarrytown, NY, demonstrated how this was done on a chromatogram of concentrated

flavor sample (he does not say of what, but possibly concentrated tomato or meat) in a

presentation at the 1965 annual ACS meeting. The sequence of sensations noted on the

858
Teranishi et al. 1971: 260.
859
Teranishi et al. 1971: 266.
376
peaks reveals a “’dinner table’” full of odors, shedding light on the various elements that

comprise the complex, integrated aroma of a food. Some peaks were annotated with

chemical names (butyric acid, furfural); others with sensory observations (“harsh

phenolic,” “rubbery sulfur,” “toasted cheese,” “cucumber,” “macaroon.”)860 These

annotations could guide efforts for identifications, shedding light on the constitutive

elements of desirable flavors and suggesting research and development priorities, in a

food production context.

“Without sensory evaluations,” the 1971 Flavor Chemistry textbook cautions,

“chemists have no guideposts and will almost certainly lose their way among the byways

of flavor research.”861 But just as sensory evaluation became recognized as necessary in

order to confirm the machine’s accuracy and guide researcher’s efforts, the matter of

developing a stable epistemological framework and reliable set of practices for

correlating the information produced by the sniffing body with that created by the

sniffing machine remained far from settled.

Achieving Subjective-Objective Correlation

Two paths diverged in the chemosensory woods. One approach, referred to as

isolation and identification (I&I), prioritized the enumeration of chemical presences.

860
William S. Ryder, “Progress and Limitations in the Identification of Flavor
Components,” in Irwin Hornstein, ed. Flavor Chemistry: A Symposium Sponsored by the
Division of Agricultural and Food Chemistry at the 149th Meeting of the American
Chemical Society, Detroit, April 6-7, 1965, Advances in Chemistry Series 65,
(Washington: ACS, 1966): 82-3.
861
Teranishi et al. 1971: 58-9.
377
“After all the compounds have been isolated and identified quantitatively, the flavor

workers can begin to put the flavor together and thus discover the compounds of greatest

interest in creating the particular flavor.”862 The second path, called postulation and proof

(P&P), began by compiling the sensory effects related to a food’s flavor, and then

directed itself toward “isolating specific compounds responsible for each important flavor

characteristic.”863 I&I and P&P were both strategies for bringing chemical and sensory

information in line with each other, for directing the course of research that had begun to

be overwhelmed by the plenitude and complexity of its output.

Irwin Hornstein, a flavor chemist at the USDA’s Market Quality Research

Division, described this challenge in terms of cheese. Temperature-programmed GC

analysis of cheddar cheese concentrate had separated approximately 130 compounds, of

which MS had, so far, helped to definitively identify fewer than 50 of these. (It was also

likely that additional compounds of sensory importance lurked to be discovered, retained

by the column or lost in extraction). “It is this task — to evaluate the significance of the

data — that is the biggest problem facing the flavor chemist today. Detection and

identification of volatile compounds are essential, but the correlation of chemical findings

with organoleptic quality is equally important, and progress in this direction has been

slow.”864

862
“Report for Analytical Chemists: Physicochemical Research on Flavor,” Analytical
Chemistry 30.2 (February 1958): 17A.
863
“Physicochemical research on Flavor” 1958: 17A.
864
Irwin Hornstein, “Preface,” in Hornstein, ed. 1966: vii-viii.
378
 By the mid-1960s, quickening the slow progress of “subjective-objective

correlation” had become a central concern in flavor research, as scientists working across

disciplines and institutional settings attempted to formalize a set of practices for relating

instrumentally produced data and sensory experience.865 The American Society for

Testing and Materials (ASTM) Committee E-18, which worked to develop global

standards for sensory evaluation procedures, spawned a Subcommittee on Instrumental-

Sensory Correlation to examine the issue. A 1967 symposium on the topic was held

during the annual meeting of the ASTM.866 Although partly an extension of ongoing

efforts to standardize sensory evaluation methods, the focus here was not on the operation

of human taste panels but on the integration of human and machine responses.

Bringing together researchers from air and water quality, cosmetics, and foods,

the symposium’s purpose was to address the divide between taste panel research and

instrumental analysis, which often seemed to run on parallel tracks. Sensory research was

usually conducted in absence of corresponding chemical identifications. Similarly,

instrumental analysis often simply identified constituents, without attempting to

determine the role these constituents played in flavor. When laboratories attempted to

find relationships between GC peaks and flavor acceptability, they were daunted by the

865
For instance, see David A. Kendall and Anne J. Neilson, “Correlation of Subjective
and Objective Odor Responses,” Annals of the New York Academy of Sciences (July
1964): 567-75, and Rose Marie Pangborn, “Flavor Perception: Relation of Sensory to
Instrumental Measurements,” In DJ Tilgner and A. Borys, eds., Proceedings of the 2nd
International Congress of Food Science and Technology, (Warszawa: Stampa, 1967):
303-18.
866
American Society for Testing and Materials, Correlation of Subjective-Objective
Methods in the Study of Odors and Taste, a Symposium Presented at the 70th Annual
Meeting, ASTM, Boston, MA, 25-30, June 1967, ASTM Special Technical Publication
440, (Philadelphia: ASTM, 1968).
379
“embarrassing wealth of data,” explained W.H. Stahl, the research manager at

McCormick, the Baltimore spice and flavoring company and erstwhile Quartermaster

head, who was the symposium’s chairman. The hundreds of components that were

revealed by GC analysis led to the adoption of “elaborate statistical procedures… to

determine which peaks, if any, have significant relationship to quality in general and to

flavor in particular.”867 This indirect method of association was not only difficult, it was

less than effective (especially given the continuing and perhaps perpetual presence of

unidentified compounds, both known and unknown unknowns). Basic techniques that

connected organoleptic and chemical information about flavor were sorely needed in

order to make the machine’s results meaningful, and to reliably associate sensory

responses with chemical presences.

As can be gleaned from the term of art used to describe the pursuit, “subjective-

objective correlation” did not seek to expel subjectivity from the process of knowledge-

making. The purpose was not to discipline the senses by demanding that the body

respond more like the machine, nor to require the machine to authenticate the body’s

responses — nor was it to design machines that responded more like bodies. Instead, the

goal here was to leverage the differences between body and machine, and coordinate

them to produce a “definitive account” of the sensible world, one that encompassed both

stimuli (chemicals) and perceptual effects (flavor).868 “We recognize that an instrument

cannot replace the human senses, but we also recognize that it often can complement

867
American Society for Testing and Materials 1968: 2. Stahl was the research manager
at McCormick, the Baltimore flavor and spice company.
868
Kendall and Neilson 1964: 568.
380
them,” said Stahl.869 But how could this complementarity be structured? How could

humans work with machines? What could both elements of the system contribute to the

understanding of flavor?

Contributors pleaded for the integration of standard taste panel evaluation

methods, including both difference tests and descriptive (flavor profile) methods, with

instrumental analysis. So, for instance, flavor profile methods could be used to

characterize either specific components, or flavor quality overall. Difference tests, such

as the triangle test, could be used when a flavor was reconstructed synthetically, to

determine whether there was a perceptible distinction between the reconstruction and the

original.

While it became increasingly possible to describe the sensory qualities of

different compounds in standard and systematic ways, it became more and more fraught

to attribute the ‘cause’ of a quality to a particular chemical compound. Rather, as

instruments revealed more of the chemical constituents related to flavor experience,

chemosensory phenomena were repeatedly shown to be an emergent property of

combinations of chemicals: the complete sensory effect often did not resemble most of its

component parts. “The typical peach aroma is due not to one or two compounds,” Loren

Sjöström summarized in one of the studies he reviewed for the ASTM volume, “but is

869
American Society for Testing and Materials 1968: 1.
381
probably an integrated response to a wide spectrum of compounds whose individual

aromas are not at all peach-like.”870

Because of this, the practice of sniffing the GC’s separated compounds to

determine significance could be misleading. Many organic compounds have an odor, but

the odor’s relation to the total could not be deduced by experiencing it in isolation.

“Sniffing the effluent of a GC column can lead to a morass of differing descriptive

terms,” warned the 1971 Flavor Chemistry textbook. Instead, the textbook advised

researchers to try to collect portions of the effluent in cold traps. “In such efforts,” it

continued, “it is well to use a relatively ‘poor’ column as well as a ‘high resolution’

column.” The good column could separate compounds “too far apart for their odor-

relatedness to be noted,” but the poor column’s muddier separations “may provide

guidance about which of the well-separated peaks contain the compounds of interest.”871

In other words, a “less” efficient resolution could reveal important sensory information

about the relationship between different compounds.

Further, sensory quality was also related to concentration; higher levels of a

compound produced not just a difference in intensity, but sometimes a difference in kind

— shifting not only associations, but also affective responses, from pleasant to

unpleasant. The importance of the subjective-objective approach was illustrated by

another phenomena that it brought to the fore: the sensory effects of compounds present

870
L.B. Sjöström, “Correlation of Objective, Subjective Methods as Applied in the Food
Field,” in American Society for Testing and Materials 1968: 3-16.
871
Teranishi et al. 1971: 277-8.
382
in subthreshold quantities.872 These compounds were, by definition, imperceptible to the

nose. But their presence could demonstrably affect the organoleptic character of the

mixture in various ways: intensifying, suppressing, or changing its qualities.

Determinations of human odor thresholds began to be included in the study of flavor

chemistry of foods.873 The interest in flavor chemicals in terms of sensory thresholds

reflected what had become the prevailing understanding of flavor: molecular substances

defined by their measurable effects on the human body.

The following chapter considers how this scientific knowledge was put to use in

the generation of new sensations, new perceptual effects.

872
Elizabeth S. Keith and John J. Powers, “Determination of Flavor Threshold Levels
and Sub-Threshold, Additive, and Concentration Effects,” Journal of Food Science 33.2
(1968): 213-8.
873
D.G. Guadagni, Ron G. Buttery, S. Okano, “Odor Thresholds of some Organic
Compounds Associated with Food Flavors,” Journal of the Science of Food and
Agriculture 14 (October 1963): 761-5; Kendall and Neilson 1964.
383
 CHAPTER 7
The Creative Flavorist at Work

In his introduction to A Short History of the Flavor Industry, a history compiled

by members of the Society of Flavor Chemists (SFC)1 – Earl J. Merwin shares the

following remarkable anecdote:

At a cocktail party in about 1948, I described my job in flavors to an IBM

salesman I had just met. He proceeded to tell me I was wasting my time.
He told me that his company had just installed a system which guaranteed
uniformity of flavor of the sausage products of a nationwide processed
meat manufacturer. ‘Flavor companies will be out of business in five
years’, as all manufacturers can avail themselves of computer technology.
He had included the variables of fat and protein content, color, salt, cost,
and many other factors, but I’m sure that he had not factored in what
effect a hurricane might have. Sure enough, it wasn’t long before a
hurricane came through the West Indies and severely reduced the
availability and cost of one or more of the major flavor ingredients for his
customer.” Computers, gas liquid chromatography, and flavor profiling all
have had an impact on flavors and flavorists. But they did not eliminate
the need for the flavorist. They all made the job a bit easier, and enabled
the flavorist to expand the list of new and improved natural and artificial
flavors.2

1
Society of Flavor Chemists, Inc. and Chemical Sources Association, Inc., The Flavor
Industry from 1945 to 1995: A Short History of the Flavor Industry With Emphasis on the
USA and the Past Fifty Years, (Neptune, NJ: Society of Flavor Chemists, 1995).
2
Society of Flavor Chemists 1995: 5.
384
 Merwin, a charter member of the SFC who had worked as a flavorist at Fritzsche

Brothers, Givaudan, and McCormick, intends this anecdote for younger flavorists who

likely entered the field confident in both its professional legitimacy and technoscientific

bona fides. This mid-century encounter, however, stages a scenario where the flavorist’s

purpose and prestige is not yet secure, with representatives of the two industries –

chemicals and microprocessors – that would come to define the character of American

technological dominance in the twentieth century squaring off at a cocktail party. The

IBM salesman predicts a situation where his professional counterpart will be done in by

technology. In essence, he is offering a version of Harry Braverman’s thesis of deskilling,

where skilled labor is shunted aside by machines that have total control over the

processes of production.3 The man from IBM apparently takes the flavorist to be a sort of

artisanal laborer, one whose empirical tinkering on the assembly line of flavors will be

replaced by the more precise control of the computer manager.

But the flavorist holds a trump card: he knows that his task cannot be reduced to

mere information and information processes. Writing in 1995, Merwin knows that

flavorists do not just compound formulas, they create new ones – crafting flavors suitable

for the expanding variety of processed food products, smoothing the gaps where the

fluctuations intrinsic to natural supply do not adequately meet industry and consumer

demands for taste and convenience, at the right price. Indeed, in Merwin’s ultimately

triumphant account, technology facilitates the tasks and expands the capabilities of the

flavorist, making way for “new and improved” products, while in no way reducing the

3
Harry Braverman, Labor and Monopoly Capital; the Degradation of Work in the
Twentieth Century (New York: Monthly Review Press, 1975), 184–235.
385
demand for the professional skills she or he provides. The SFC mission statement

proclaims that one of the goals of the organization is “to foster and encourage the art and

science of flavor technology.”4 The machine complex in this respect would prove an ally

to the true craftsman.5

By the 1960s, the creative flavorist had become the most valuable asset of flavor

companies. A 1968 Arthur D. Little, Inc. report on the US flavor industry, prepared for an

Ohio cosmetics company that was considering expanding into the flavor business,

provides a detailed overview not only of the booming commercial prospects for the flavor

sector, but also of the importance of creative flavorists to the reputation, status, and profit

margins of successful firms.6 According to the report, sales of flavorings had grown at an

average annual rate of ten percent over the previous decade, totaling $130 million in

1967, and growth was accelerating; sales of “specialty flavors,” unique proprietary

formulations developed for specific customers, accounting for an increasing portion of

the gains in sales.7 This booming business was undergirded by the creative labor of

flavorists. “The flavor houses’ creative skill has become the principal ‘priceless’

4
Society of Flavor Chemists 1995: 32.
5
Braverman, in his otherwise pessimistic account of the negative social effects of
mechanization on the labor process, does admit that another narrative is possible. “There
is no question that from a practical standpoint there is nothing to prevent the machining
process under numerical control” – that is, under the control of externally programmed,
rationalized and “objective” systems of management –“from remaining the province of
the total craftsman. That this almost never happens is due, of course, to the opportunities
the process offers for the destruction of craft and the cheapening of the resulting pieces of
labor into which it is broken” (Braverman, 199).
6
Arthur D. Little, Inc. “The U.S. Flavor Industry: Report to The Andrew Jergens
Company,” ADL Report C-69866 (March 26, 1968). AW Noling Collection, UC Davis.
The Andrew Jergens Company was a Cincinnati-based manufacturer of cosmetics,
lotions, and personal care products.
7
Arthur D. Little 1968: 3.
386
ingredient and allows the flavor company great latitude in setting prices,” the report

disclosed. 8 “The ability to price the flavoring at a premium is directly related to the

apparent creativity of the flavor chemist.” The value of the company’s products was so

intertwined with skill of its (unnamed and all but invisible) creators, that “if a flavor

chemist with a good reputation leaves a flavor house, the overall reputation of the flavor

house suffers. The competing firm which the flavor chemist joins benefits by his move,

not only from his skills, but by his presence, which may command greater premiums.”9 A

skilled creative flavorist could also guarantee ongoing revenue for the company she or he

worked for. As specialty flavorings were customized for particular products, “long-term

success for a flavor house is assured by the development of proprietary flavorings that

become successful consumer franchises.” Because the formula for these products was

kept secret, even from the customer, and because of the reluctance of food companies to

risk any changes to a successful flavor, these accounts led to large numbers of repeat

orders.10

But who, exactly, was the creative flavorist? Despite interviews with

representatives from 85 companies that manufactured and used flavoring additives, the

authors were unable to describe, exactly, who these people were, how to find them, or

even how many of them there were. Estimates of the number of “top-notch” working

flavorists ranged from fifteen to one hundred; the authors of the report believed that there

were likely fewer than thirty. “No academic credentials are probative; past experiences

8
Arthur D. Little 1968: 13.
9
Arthur D. Little 1968: 27.
10
Arthur D. Little 1968: 27.
387
and associations are the most critical factors.” Like other artisans, these individuals

developed their “skill and knowledge… through an apprenticeship system,” but a

fundamental “intuitive grasp for what makes a flavor” was also necessary, as was a

familiarity with manufacturing and marketing details of different consumer products.

Finally, a successful flavorist “must be attuned to the ‘taste’ of his market.” For those

who could pull it off, the financial rewards were “high,” with “top notch flavor chemists”

generally earning between $25,000 and $35,000, and with some even pulling down

$50,000 a year.11

As we have seen, the basic principles, goals, and methods of the creative flavorist

had been articulated long before the 1950s. Flavor and fragrance companies such as

Synfleur, Fritzsche Brothers, Dodge & Olcott, and dozens of others had relied on the

skilled labor of a small number of highly specialized workers, who combined precise

chemical and sensory knowledge with improvisatory skill, to formulate distinctive

flavorings that conformed to the technical requirements of the manufacturers who used

them. Yet until the mid-1940s, when the term “flavorist” was coined just as “flavor

chemistry” was beginning to gain recognition as a distinct scientific field, even the

nominal identity of these workers was indeterminate, as were their educational

credentials, the technical prerequisites for their labor, and their positions and

responsibilities within the companies that employed them.12

11
Arthur D. Little 1968: 26.
12
The earliest instance of the word “flavorist” that I have found is in a March 1945
article in the trade journal Food Industries: [E.C. Crocker, “A Flavorist Views Food
Processing,” Food Industries 17 (March 1945): 69-71, 170-4.] Crocker was a chemist at
388
 This chapter tells the story of how flavorists became professionals in postwar

America. In The System of Professions, his bedrock study of the processes of

professionalization, sociologist Andrew Abbott distinguishes professions from craft

occupations by allocating to the first abstract, and to the second, primarily technical and

tacit knowledge. While Abbott concedes that professional work also depends on tacit

skills and often consists of routinized tasks, “here, practical skill grows out of an abstract

system of knowledge, and control of the profession lies in the control of the abstractions

that generate the practical techniques. The techniques themselves may in fact be

delegated to other workers.” Craft occupations, in contrast, “emphasize technique per se”

and protect their authority and legitimacy by controlling the transmission of technical and

tacit skills.13 In Abbott’s system, flavorists would likely be classified as a technical

occupation, rather than a “fully” professional one, both because of the apprenticeship

model of training that persists in the field to this day and the emphasis on tacit and

Arthur D. Little, Inc. who had, since the 1920s, worked on scientific and technical
problems related to sensory quality and control (particularly those related to odor).
Crocker claims to have coined the word, explaining: “Since the dictionary lacks a word
for one whose profession deals with flavor, there term ‘flavorist’ is hereby offered.” In
Crocker’s usage, however, a flavorist is not specifically a creator of synthetic flavor
additives. Instead, he applies the term to food technologists and other technoscientific
professionals who work to improve the flavor of food products, particularly within the
context of industrial food manufacturing, by studying the chemical causes of flavor
changes during production or storage and developing new processes and packaging that
did less damage to, or improved, flavor quality. The word caught on, at least in trade
circles, by the early 1950s, although it had come to refer almost exclusively to workers
who developed flavor additives. Although flavorists also often referred to themselves as
“flavor chemists,” in this chapter, I do not use these two terms interchangeably, but
instead reserve “flavor chemists” for those workers (in government, academy, and
industry) whose primary focus was the identification of flavor compounds in food, rather
than the creation of flavor additive products in flavor industry laboratories.
13
Andrew Abbott, The System of Professions: An Essay on the Division of Expert Labor,
(Chicago: UChicago Press, 1988): 8-9.
389
technical skills. For Abbott, the economic salience and persistence of a profession derives

from the essentially abstract quality of its defining body of knowledge, from “their

abstracting ability to define old problems in new ways. Abstraction enables survival.”14

Occupations that are too entwined with the particularities of technique, too bounded by

the material jurisdictions of their knowledge, are at risk of being shunted aside by real

professionals or actual robots. Despite the smooth assurance of the IBM representative,

software programmers faced a similar occupational crisis in the 1950s and 1960s. As

Nathan Ensmenger has documented, software programming was, during this period, a

craft occupation that seemed to bear many similarities to the work of the creative

flavorist: based on intuition, tacit knowledge, and idiosyncratic virtuosity.15 Further, like

flavorists in the same period, the occupational identity of the programmer was radically

underdetermined; as Ensmenger puts it, “‘programmer’… was not a career choice but… a

vocational path followed by accident and only retrospectively labeled and understood.”16

In Ensmenger’s account, the software craftsman was a poor fit in the corporate culture of

computer companies, which valued hierarchies of management, standardized skills, and

routine processes. As a result, “in the interest of efficient software manufacturing, the

black art of programming had to make way for the science of software engineering,”

increasingly theoretical, academic, and abstract.17

14
Abbott 1988: 30.
15
Nathan Ensmenger, The Computer Boys Take OVer: Computers, Programmers, and
the Politics of Technical Expertise, (Cambridge: MIT Press, 2010).
16
Ensmenger 2010: 12.
17
Ensmenger 2010: 24.
390
 More recently, sociologists of work have argued that technicians should be

understood not as “junior” or “paraprofessionals,” but as practitioners of both craft and

science, and moreover, as central and coordinating figures in modern organizations,

whose occupational importance derives as much from their ability to mediate between

multiple social, material, and informational realms, as it does from their technical

mastery.18 For instance, Stephen Barley and Julian Orr describe technicians as “managing

the empirical interface,” the point at which a system of production meets the material

world, in part by transforming materiality into signs, symbols, and indices, carriers of

meaning and value.19 This central and coordinating role, this translation between realms

of experience, emerges not from an abstract view from above, but precisely from

“situated practice” — the tacit, embodied, fully sensual “know how” that comes from

direct manipulation of basic materials. At flavor companies, flavorists translate between

materials and representations, at the juncture between the synthetic compounds produced

by fundamental research and the pattern of customer needs and consumer appetites. They

assemble distinctive products out of heterogeneous materials: available chemical raw

materials, chemical knowledge, customer requests (which may need to be heavily

interpreted), information about costs and production processes, and regulatory

requirements.

18
Barley, Stephen R. "Technicians in the Workplace: Ethnographic Evidence for
Bringing Work Into Organizational Studies," Administrative Science Quarterly 41, no. 3
(1996); Catherine McKercher and Vincent Mosco, eds. Knowledge Workers in the
Information Society. Lanham: Lexington Books, 2007; Julian E. Orr, Talking About
Machines: An Ethnography of a Modern Job (Ithaca, NY: ILR Press, 1996).
19
Barley, Stephen R. and Julian E. Orr, eds. Between Craft and Science: Technical Work
in U.S. Settings. Ithaca: IRL Press, 1997.
391
 Like software programmers, flavorists in the postwar had to make a place for

themselves in corporate structures that are not reflexively welcoming to their way of

working — which may have, structurally, favored systematic and routinized knowledge

over the idiosyncratic effusions of creative skill. Postwar flavorists found their place, in

part, by distinguishing themselves from their “old school” counterparts, enthusiastically

embracing instrumental technologies such as gas chromatography, scientific research,

and technology. But they also differentiated themselves from analytic flavor chemists, by

their insistently embodied and sensual response to the chemical compounds that their

powerful machines eluted into the world. In other words, flavorists used instrumental

technologies to define and defend the prestige of their jobs and technical knowledge

while continuing to insist upon the creative essence of their work lives. I will argue that

flavor chemistry constitutes what I will call a “scientific craft profession,” a form of

trained, professional labor that self-consciously joins technical mastery, scientific

knowledge, and creative skill. Rather than an exceptional or marginal case, I believe that

flavorists ultimately demonstrate the importance of tacit, embodied, and situated

knowledge to scientific and technological careers.20

This chapter begins by examining the formation of the Society of Flavor

Chemists. Flavorists organized as a professional group at a time when the conditions of

20
See, for instance, Natasha Myers, “Molecular Embodiments and the Body-work of
Modeling in Protein Crystallography,” Social Studies of Science 38.2 (April 2008): 163-
199; Lucy Suchman, “Embodied Practices of Engineering Work,” Mind, Culture, and
Activity 7 (2000): 4-18; H.M. Collins, “The TEA set: Tacit knowledge and Scientific
Networks,” Social Studies of Science 4.2 (1974): 165-185.
392
their work were changing rapidly. Commercial circumstances were favorable for the

field, and for its claims to recognition as an expert profession. There was a rising demand

for the services of flavorists and a growing number of job opportunities, which were

being filled by younger workers eager for their share of promised postwar prosperity.

Further, the recognition of flavor chemistry as a distinct chemical sub-discipline lent

credibility to claims of the scientific basis for this form of work, at a time when science

and technology fields were rising in social esteem. Meanwhile, what it meant to hold a

“job in flavors” was also in flux. An expanding petrochemical industry, powerful analytic

instruments, and developments in food production technologies required an increasingly

sophisticated set of skills, and workers who could “make sense of” newly available

chemical materials in evolving and diversifying contexts of use. At the same time,

cultural and political forces started strongly regulating and limiting the use of new

chemical materials in foods, in a certain regard, denigrating the (largely invisible) labor

of the specialists who worked with them. The moment was ripe for the workers in this

rising field to define themselves and improve their prospects.

Then, I take a detailed look at how flavorists transformed chemical compounds

into flavors. By the 1960s, there were more than a thousand potential chemical

compounds that were approved for use in flavoring additives. There was also a rapidly

evolving chemical knowledge of the constituents of flavor, which not only revealed

previously unknown compounds but also shed light on the dynamic interaction between

smelly molecules in foods to produce the perceptual effect of “total flavor.” I consider the

uniquely probing ways in which flavorists read the flavor chemical literature, their

393
distinctive use of analytic instrumentation, and their contextual role within the structure

of flavor companies. I then consider how flavorists learned to do their work, examining

the educational regimes and training programs that produced skilled flavorists.

Finally, I conclude with some thoughts on the moral and historical purpose that

flavorists saw in their work. At a time when “chemicals” in foods were attracting

increasing concern and even popular abhorrence, flavorists attached their work to a

broader, progressive mission of feeding the world at a time of resource crisis.

Becoming Flavorists: The Origins of the Society of

Flavor Chemists
In 1996, the Society of Flavor Chemists (SFC) published a spiral-bound booklet

documenting the history of their profession four decades after the founding of their

organization.21 The booklet contains dozens of brief biographical recollections submitted

by flavorists, testimonials of their own careers in the field, or remembrances of the lives

of departed colleagues. Reading through these life stories, one is astonished to encounter

again and again variations on a common refrain: “Like most flavor chemists, I got into

the industry purely through dumb luck.”22 Some began working for flavor manufacturers

in other roles — both blue-collar positions (handyman and “bottle washer,” compounder)

or technical positions that required some chemical education (quality control, analytical

21
Society of Flavor Chemists 1995.
22
Carl H. Holmgren in Society of Flavor Chemists 1995: 77. Similar “dumb luck”
accounts of initiation into the profession can be found in the entries for Baranowski,
Clemente, Colovito, DeRovira, Eskin, Farber, Fischetti, Donnarumma, Goossens,
Graham, Heinze, Mandel, McBurnie, Merwin, and Mosciano.
394
chemist), shifting to creative flavor work due to a combination of happenstance,

demonstrated skill and interest, and acute labor market need.23 Others answered classified

advertisements or were placed by employment agencies, apparently with few

expectations about the nature of the work ahead. After Earl Merwin graduated from NYU

with an undergraduate degree in chemistry in 1947, an employment agency sent him to

Fritzsche Brothers. “When I told the agent that I had never heard of that company,”

Merwin recalled, “he suggested that I not mention that to them.”24 When Harvey Farber

graduated Queens College with a degree in applied science a decade and a half later, he

had a similar experience when an employment agency placed him at General Foods. “I

had no idea about flavor chemists or flavor companies,” he wrote. “By chance I was put

in a flavor group. I liked it and I was good at it. I was a junior flavor chemist.”25

The SFC was spearheaded by James Broderick, an ambitious young chemist who

was looking for opportunities to develop his skills as a creative flavorist. Broderick began

working with flavors in 1939, a year after graduating Brooklyn Technical High School.

His first job was as a laboratory assistant at a dessert manufacturer, where he worked

alongside a chemist who had some experience in flavor work. “There was an excellent

library of flavor samples, several books on flavors, and a number of key flavor

materials,” Broderick recalled. When the chemist was fired less than a year later for

union-promoting activities, Broderick had the chance to dabble in creating flavors

23
See, for instance, Thomas J. Bonica, who began as a handyman and bottle washer at
Polak & Schwartz in the 1930s (Society of Flavor Chemists 1995: 45-6); Anthony
Clemente began as a compounder at Fritzsche Brothers before the war (Society of Flavor
Chemists 1995: 52).
24
Earl Merwin in Society of Flavor Chemists 1995: 105.
25
Harvey Farber in Society of Flavor Chemists 1995: 61.
395
himself. He had good sensory instincts and a keen interest in flavors, and some of his

formulations were accepted and used— although he later suspected that this was due to

their low cost rather than their quality. Within a few years, he began working at a small

Brooklyn flavor company. By the time America entered the Second World War, he was

in charge of flavor development.26

War service interrupted his career, and when he returned to civilian life, he

actively sought out a mentor. “I felt the need to work with some talented senior flavorist

to enhance my flavor knowledge,” he recalled. After seeing a newspaper advertisement

for a flavor chemist position at van Ameringen-Haebler, he applied; he had come to

admire the creativity and integrity of the company’s flavoring products, and hoped to

learn from the flavorist who had created them. After being hired, he found out that the

flavorist whose work he had respected had died some time before; his son, whom he had

trained to replace him, was in poor health and no longer at the company. (Broderick, in

fact, had been hired as his replacement.) Instead, Broderick worked with James

McGlumphy, who had recently joined the company from the Iowa State University,

where he had been a professor of chemistry. McGlumphy was an analytical chemist with

a doctorate in the field; he knew little of the flavor industry when he joined the company.

He also, according to Broderick’s account, felt his position threatened by another

“practical flavorist” and flavor salesman at the company who had believed that

McGlumphy’s job should rightfully have gone to him.27

26
Society of Flavor Chemists 1995: 46-8
27
Society of Flavor Chemists 1995: 103-4.
396
 Restless, Broderick left van Ameringen-Haebler in 1952 for a position at

Givaudan, which was expanding its flavor division at the time.28 There, he met two

younger flavorists, Jerry DiGenova and Earl Merwin; all had been hired within a year of

each other. The working conditions at Givaudan were also not ideal. As Merwin

recollected, “We worked in one flavor lab with one technician (Mary Mogavero). Our

‘offices’, with a desk for each, were also all in one room.” Their boss, Hans Kessler, was

not a flavorist; he was the sales director at the company. Two older, European-trained

flavorists, Carl Jensen and Joseph Merory, divided artificial and “true fruit” flavors

between them; both soon left to begin their own companies. “Management had set up a

competitive situation” among the three new hires, Broderick complained, “and did not

apparently see the longer range potential of keeping all content.”29 Within a few years,

both Broderick and Merwin had left the company — Merwin went to McCormick & Co.

in Baltimore; Broderick eventually ended up at Kohnstamm, in Brooklyn. DiGenova

remained at Givaudan for the duration of his career, eventually ascending to chief

flavorist and vice president of the creative laboratories.30

What Broderick and his colleagues were discovering was this: the status and role

of the creative flavorist was uncertain within US flavor companies in the immediate

postwar. Although many aspects of the flavor industry were changing, in many cases,

companies still functioned in traditional ways, passing flavor formulas down along

paternal lines, and relying on older, European-trained flavorists. But as the market for

28
See Chapter 4.
29
Society of Flavor Chemists 1995: 47.
30
Jerry DiGenova, Society of Flavor Chemists 1995: 55.
397
flavoring additives boomed after the war, and flavor and fragrance companies expanded

their production of specialty flavorings, there was a need for more skilled workers who

could transform the growing number of available flavoring chemicals into distinctive new

flavoring products. But where should these new workers come from? Should companies

look for individuals with academic credentials in chemistry, such as McGlumphy?

Should individuals with sales experience, who knew the needs of clients, be in charge,

directing the activities of flavorist-technicians? What sorts of resources, instruments, and

personnel should flavorists be granted? Where did their work fit into the company’s

bigger picture? In this regard, flavorists faced some of the same status anxiety and

hostility encountered by software programmers during the same, as described by Nathan

Ensmenger. Both software programming and flavor creation were seen as “black arts,”

practiced by adepts with idiosyncratic capabilities and unique gifts.31 Yet just as “the

black art of programming had to make way for the science of software engineering,”

there was a pressing need within flavor companies to put flavor creation on a systematic

basis, and a concomitant desire among flavorists entering the field to develop their

capacities and define the trajectory of their careers.32

The 1953 IFT meeting in Boston presented an optimal chance. As Broderick

recalled, “a group of us used the opportunity to recruit additional flavorists with the hope

of forming the Society of Flavor Chemists.”33 The quorum of flavorists at the IFT

31
Ensmenger 2010: 19. For flavor creation as “black art,” see “The Art in Imitation
Flavors: The Aromatic Constituents of Strawberry,” Givaudan Flavorist 1953 (2): 1.
32
Ensmenger 2010: 24.
33
James J. Broderick, “Reflections of a Retired Flavorist Before He Forgets:
Strawberry,” Perfumer & Flavorist 17.3 (May/June 1992): 33.
398
included Broderick, Merwin, and DiGenova from Givaudan, as well as Thomas Bonica

and Charles Fricke, from Polak and Schwarz, Frederick Schumm from Dodge & Olcott,

and Louis Strasberger, who Broderick knew from his time at Van Ameringen-Haebler.

The location of this convocation of flavorists from different companies, the IFT

meeting, signified the growing cultural divide between younger flavorists and their older

counterparts. “Old school” flavorists, such as Merory, formerly of Givaudan, were

trained at a time when information about the chemistry of flavor was relatively scarce;

they worked empirically, by sense and memory, had little interest in new analytic

technologies, and often were expected to be both salesman and formulists.34 Younger

flavorists, especially those who entered the field after the war, were more inclined to see

their place among scientific and technical workers, contributors to, and beneficiaries of, a

growing body of fundamental flavor research. But the Mertonian norm of “communism,”

designating the goods of scientific knowledge as common intellectual property among a

community of scientists, was a poor fit with the values of flavor companies, which were

traditionally extremely secretive. “There was still a strong feeling at the management

level in some companies that flavorists should not meet together or even be seen talking

34
The biography of Merory included in Society of Flavor Chemists 1995 describes him
as a convincing salesman but a somewhat inept formulator, who “knew a little about a lot
of things” but did not have a very ‘scientific image’ among his peers. (Joseph Merory,
Society of Flavor Chemists 1995: 104-5). See also, for instance, Merory’s 1960
handbook on food flavorings, where he describes the craft of creating a synthetic flavor
that “closely resembles” a natural one in these words: “The development of a close
resemblance is creative work and depends on the photographic memory of the flavor
technologist to recall aroma and taste of every flavor which passed his sensory and
gustatory organs. He has to know which ingredients to select and to be able to harmonize
them in a suitable flavor formula.” Younger flavorists writing at this time would
indubitably refer to ongoing analytic flavor research as a source for assistance in
developing a naturalistic flavor.
399
together,” Broderick recalled.35 McGlumphy, the head of flavor research at van

Ameringen-Haebler, strongly opposed to the creation of the SFC, at least initially.36 In

fact, Strasberger, who had served as the first Vice President of the SFC, declined to

accept any further positions with the organization, “citing the displeasure he felt such a

position would generate with his employer.”37

The group recognized that the legitimacy of their nascent organization (and

perhaps their own continued employment) was in question. Strategically, they wanted to

choose a leader and president for the SFC that could serve as a credible intermediary. On

the one hand, an established “older” flavorist might not represent their values and goals;

but selecting one of their own younger cohort would fail to gain credit with the

companies they worked for. The first president of the organization, John Bouton, was a

transitional figure. Having begun his career in the late 1930s, he was neither “young” nor

“old,” but occupied an intermediate generational position, thus bridging the chasm

between the older cadre of flavor workers and the rising class, who had largely entered

the field after the war.38 Bouton was also widely respected, as he was the recognized

creator of Dodge & Olcott’s Dolco 5210 Imitation Strawberry, a distinctive “trade-

famous” strawberry flavor that was celebrated as the industry leader at the time.39

35
Broderick “Strawberry” 1992: 33.
36
James McGlumphy in Society of Flavor Chemists 1995: 104. McGlumphy would
eventually become an honorary member of the SFC.
37
Louis Strasberger in Society of Flavor Chemists 1996: 127.
38
Broderick “Strawberry” 1992: 33.
39
Broderick “Strawberry” 1992: 33. Broderick later discovered that the flavor’s
distinctiveness came from methyl heptine carbonate, which added a green note, as well as
a very small quantity of maltol; Broderick believed Bouton was the first to use the
400
 The early meetings of the Society of Flavor Chemists merged social with

professional purposes. Starting in February 1954, about a dozen workers in the New York

flavor industry convened in restaurants, usually in Little Italy, every other month. “A

group of those interested in flavor chemistry have formed the Society of Flavor

Chemists,” ran an announcement in the Journal of Agricultural and Food Chemistry.

“Purpose of the organization is primarily social but informal talks on matters of mutual

interest will be scheduled occasionally.”40 But the group also hosted scientific talks and

other technical information of interest to their field. For instance, Keene Dimick, of the

USDA Western Regional Research Laboratory in Albany, California, gave a talk in 1956

to the SFC about his pioneering work using gas chromatography to study the flavor

chemistry of strawberries.41 “This meeting was a turning point not only for flavorists but

also for Dr. Dimick,” Broderick later wrote. “For the flavorist it changed his approach

and increased his efforts to obtain, evaluate and utilize the hexenyl compounds. For Dr.

Dimick it gave the opportunity to travel to the East and line up suppliers to build gas

chromatographs and start the Aerograph Company.”42

material. Dolco Imitation Strawberry 5210 is referred to as “trade famous” in the 1951
Dodge & Olcott catalogue, which claims that the flavor has “won wide acceptance in
every industry where strawberry flavor is used.” Dodge & Olcott, Inc. Essential Oils,
Aromatic Chemicals, Perfume Bases, Vanilla, Flavor Bases, [Catalog], April 1951: 32.
Smithsonian Institute Trade Literature Collection, National Museum of American History
Library.
40
“Flavor Chemists Form Society in New York,” J. Agric. Food Chem. 2.5 (1954): 266.
41
See Chapter 6.
42
Broderick “Strawberry” 1992: 33.
401
 When the SFC was formally incorporated as a chartered corporation in 1959, its

size had doubled to include nearly two dozen “charter members.” The organization soon

adopted a Code of Ethics, which asked members to pledge to observe high standards of

personal conduct and professionalism, and to recognize certain responsibilities. The first

duty was to the self, to maintain a standard of individual integrity and professional honor,

which included keeping “in active contact with the progress in my profession.” The

second duty was to the flavorist’s employer, “to serve him undividedly and

conscientiously on the basis of a clear, mutual understanding of our respective interests,

guarding his concerns, reporting fully on all technical matters,” and ensuring that

coworkers also respected the demands of confidentiality. The third duty was to the

flavorist’s profession: to contribute to its progress and to the mutual exchange of ideas, to

recognize the work of others, to observe the highest standards of truthfulness in technical

reports, “but in doing so to faithfully guard against the willful and wrongful disclosure of

trade secrets of former employers.”43 In other words, the Code of Ethics attempted to

strike a deliberate balance between the communal norms of scientific and technical

societies, and the obligations demanded by private employers, to confidentiality and

secrecy.

43
Society of Flavor Chemists 1995: 32.
402
Flavors are Chemicals: Flavorists and the 1958 Food

Additives Amendment

The chemical-material culture of the postwar flavorist was defined not only by the

rapid proliferation of intriguingly smelly molecules, but also on the regulatory side, by

new laws and requirements that sought to limit and constrain the use of these chemicals

in foods.

Concerns about chemicals in the food supply had driven perennial reform efforts,

but the precise nature of these worries, the chemical effects that the government was

asked to protect consumers from, evolved over time. Although the 1906 Pure Food and

Drug Act had been motivated in part by fears about noxious adulterants in the food

supply, the law’s main regulatory muscle was flexed to prevent economic adulteration,

the use of synthetic chemicals to deceive purchasers about the identity or quality of the

product they were buying.44 But that law’s failure to give regulators the power to create

legal food standards, and other perceived loopholes such as the ‘distinctive name

proviso,’ came to be perceived by regulators and food reformers as serious shortcomings

that limited its effectiveness at protecting consumers. One of the changes introduced by

the Pure Food Act’s successor, the 1938 Food, Drugs, and Cosmetics Act, had been to

formalize the process of developing official food standards of identity, which were seen

as essential to protecting “the pocketbooks of consumers” and ensuring that they received

“the ‘value expected’” from foods.45 As defined by the law, food standards were to take

44
See Chapter 2.
45
Suzanne White Junod, “Food Standards in the United States: The Case of the Peanut
Butter and Jelly Sandwich,” in Food, Science, Policy, and Regulation in the 20th
403
the form of “recipes,” specifying required ingredients and optional ingredients. The

contents of these standards were determined during a series of hearings, which

entertained testimony from representatives from various interested parties, including the

food industry. Standard foods were required to list only any optional ingredients on their

labels. Foods for which no standards existed were required to list all ingredients.

“Ironically,” notes FDA historian Suzanne White Junod, “consumers knew less about the

contents of standardized foods than about foods for which there were no standards” and

which were required to list all ingredients on their labels.46 Although this system allowed

the FDA to prohibit some chemicals in some foods by excluding them from the standards

— for instance, benzoate of soda, a preservative, was excluded from the “optional”

ingredients in canned tomatoes, thus prohibiting its use — the legal framework for

challenging the inclusion of these ingredients was economic adulteration — not safety.

This process had worked relatively smoothly at first, as the food standard setting

process prioritized staple foods, which tended to include fewer ingredients and were

simpler to define.47 But as food technology generated new kinds of foods that departed

further from anything that could be whipped up in a home kitchen, and with

modifications to the hearings process that often turned the proceedings into a forum for

Century, 2000: 179. For more on the 1938 law and the contentious process of developing
standards of identity for food products, see: Angie M. Boyce, “‘When Does it Stop Being
Peanut Butter?’: FDA Food Standards of Identity, Ruth Desmond, and the Shifting
Politics of Consumer Activism,” Technology and Culture 57.1 (2016): 54-79; James
Harvey Young, “The Government and the Consumer: Evolution of Food and Drug Laws
and the 1938 Food, Drug, and Cosmetic Act,” Journal of Public Law 13 (1964): 197-203;
Gwen Kay, “Healthy Public Relations: The FDA’s 1930s Legislative Campaign,”
Bulletin of the History of Medicine 75.3 (Fall 2001): 446-487.
46
White Junod 2000: 183.
47
White Junod 2000: 181.
404
litigating internecine trade disputes, fractures began to appear in the regulatory system.48

What should be the standards for new types of processed foods, such as freeze-dried

coffee, instant pudding mixes, frozen dinners? Further, investments in food science and

technology research were yielding an increasing number of new ingredients and

functional chemical additives, which food manufacturers were eager to integrate into

their products, but which regulators were increasingly wary about.49 This was dramatized

during hearings about the standard of identity for bread, in the early 1950s. In particular,

bread companies were eager to have the FDA recognize as optional ingredients a new

category of chemical additives: polyoxyethylene monostearates (POEMS). POEMS were

shelf-life boosting emulsifiers, that kept loaves softer longer on supermarket shelves. “It

was painfully clear to everyone at the hearings,” writes White Junod, “that all 27

emulsifiers had not been subjected to the same level of scientific scrutiny for either safety

or suitability in bread.”50 But the FDA challenged their inclusion in the standards not

because of possible health effects, but on the grounds that their use misled consumers

about the freshness of bread. During the prolonged and frustrating hearings debating the

standard, swarms of psychologists and social scientists were called to testify, and asked

to weigh in on “the task of dissociating softness and freshness.”51 For many, the spectacle

of the hearings proved the wrong-headedness of contesting food additives on the grounds

48
In particular, manufacturers and others criticized the food standards system for stifling
innovation and technological improvement. See, for instance, H. Thomas Austern, “Food
Standards: The Balance Between Certainty and Innovation,” Food, Drug, and Cosmetic
Law Journal 24 (1969): 440-55.
49
Suzanne White, “The Chemogastric Revolution and the Regulation of Food
Chemicals,” in Seymour H. Mauskopf, ed. Chemical Sciences in the Modern World,
(Philadelphia: University of Pennsylvania Press, 1993): 322-55.
50
White Junod 2000: 182.
51
White Junod 2000: 182.
405
of consumer deception. It became clear to Congress that a new process was needed, one

that allowed the FDA to directly address the question of safety.52

In order to better grapple with the growing number of untested chemical additives

in the food supply, in the early 1950s, Congress created a Select Committee to Investigate

the Use of Chemicals in Food Products, which would come to be led by Representative

James Delaney of New York. This investigation culminated in the 1958 Food Additives

Amendment (FAA), which became effective in March 1960. While implicitly

recognizing the utility (and inevitability) of chemical additives in an industrialized food

system, the FAA was an attempt to address the increasing concerns from the public and

the scientific community about the possible long-term health effects of these

substances.53 It did this by implementing a review process for new food chemicals

comparable to the one in place for new drugs.54 The law placed the burden of testing onto

private industry, requiring manufacturers to submit detailed toxicological test results and

other data to the FDA, and obtain pre-market approval from the agency for any new

substance before introducing it into the food supply. In particular, the law reflected an

emergent consensus in the medical and scientific community about the relationship

between dietary habits and the incidence of cancer. A section of the law known as the

52
Xaq Frohlich has convincingly argued that the decline of official “standards of
identity” constituted the beginning of an “informational turn” in food labeling, as
consumers were increasingly made responsible for managing their own health risks by
diligently reading labels (and making responsible choices), while the government’s role
in ensuring food quality diminishes. See: Xaq Frohlich, “The Informational Turn in Food
Politics: The US FDA’s Nutritional Label as Information Infrastructure,” Social Studies
of Science (2016): 1-27; and “Accounting for Taste: Regulating Food Labeling in the
‘Affluent Society,’ 1945-1995,” Enterprise & Society 13.4 (December 2012): 744-61.
53
White 1993: 342.
54
White Junod 2000: 183.
406
“Delaney Clause” withheld approval for any substance that had been shown, in animal

experiments, to potentially cause cancer in humans or that was a known human

carcinogen.55 Crucially, the law allowed this pre-market review process to be bypassed

for any substance that was “generally recognized, among experts qualified by scientific

training and experience to evaluate its safety, as having been adequately shown to be safe

under the conditions of its intended use.” This so-called GRAS (ie, “generally recognized

as safe”) provision would be key to the flavoring industry’s response to the law.

The law applied to multiple categories of chemical additives, including

preservatives, stabilizers, emulsifiers, surfactants, nonnutritive sweeteners, and food

colors. Rather than joining in a general effort with other additive manufacturers and

users, the flavor industry’s trade organization, the Flavor and Extract Manufacturer’s

Association (FEMA) labored to distinguish its members’ products from other food

additives, arguing that “special criteria” should be applied to the evaluation of flavoring

materials.56 First and foremost, there were simply more chemicals to consider: a far

greater number and variety of compounds were used in the production of food flavorings

than in all the other types of additives combined.57 Conducting a toxicological review of

55
For a detailed discussion of the 1958 Law, the Delaney Clause, the challenges of
implementing it, and its implications for the relationship between science and
policymaking, see Richard A. Merrill, “FDA’s Implementation of the Delaney Clause:
Repudiation of Congressional Choice or Reasoned Adaptation to Scientific Progress?”
Yale Journal on Regulation 5.1 (1988): 1-88.
56
Richard L. Hall and Bernard L. Oser, “The Safety of Flavoring Substances,” in Residue
Reviews (New York: Springer, 1968): 1-17. Collection of the Society of Flavor Chemists,
Chemical Heritage Foundation, Philadelphia.
57
Richard Hall, the McCormick Research Chemist who served as the first head of the
FEMA Food Additive Committee and was instrumental in organizing the GRAS list
effort, estimated in 1959 that about 1,100 flavor additives and adjuncts were in use in
407
all of these compounds would be prohibitively difficult and impossibly expensive. FEMA

sought GRAS status for these materials. In order to build its case, FEMA formed a Food

Additives Committee, which reached out to its membership of manufacturers, asking

them to disclose information that they had long held very tightly: what chemicals they

used in their flavorings, the concentrations they were used in, the types of products they

were associated with, and the quantities annually produced and sold. FEMA also

assembled a panel of recognized experts to review this data, including medical doctors,

toxicologists, chemists, and others largely from academia, as well as some employed by

chemical companies, including Dow and Eastman Kodak.58

On the basis of the results of their survey to manufacturers, FEMA made a case

that the risk posed by flavoring additive chemicals was negligible, and the costs of

ensuring absolute safety prohibitive. The 1960 survey of flavor manufacturers found that

half of the substances on the GRAS list were used in quantities of less than 100 pounds a

year in the national food system. Only eight percent were used in quantities greater than

10,000 pounds a year, and these were generally spices and other botanical flavoring

materials.59 Very few were used at levels exceeding 500ppm. In contrast, sweeteners,

emulsifiers, and other types of additives were often used at levels between 1,000

to100,000 ppm. Their sensory qualities also made them self-limiting. “In all other

food, about twice the number of additives in use for all other purposes. Richard L. Hall,
“Flavoring Agents as Food Additives,” Food Technology 8.7 (1959): 14. This paper was
presented at the Symposium on Food Additives at the annual meeting of the Institute of
Food Technologists, held on May 20, 1959.
58
Bernard L. Oser and Richard A. Ford, “FEMA Expert Panel: 30 Years of Safety
Evaluation for the Flavor Industry,” Food Technology (Nov 1991): 86-8, 93-7. A list of
members of the expert panel can be found on 87.
59
Hall and Oser 1968: 3.
408
categories of food additives, blandness or absence of flavor or odor is highly desirable

and often essential. Thus there is no organoleptic safeguard against an overdose caused

by accident or ignorance.” But “with rare exceptions” the use of flavoring materials “at

levels substantially in excess of any normal maximum renders food unpalatable.”60

Further, while many categories of additives “involve chemical structures not thus far

found in the natural foods of man, and with which he has little toxicological experience,”

with few exceptions, most flavor additives were either synthetically produced compounds

found in nature, or structurally and thus metabolically related. “Unless there is genuine

reason, based on chemical or pharmacological considerations, to question its safety under

conditions of use, it is neither reasonable nor practicable to place on any substance, used

to the extent of only a few hundred pounds annually, the cost burden of chronic toxicity

studies,” concluded Richard Hall, a research chemist at McCormick who was the head of

the Food Additives Committee and spearheaded the GRAS effort, and Bernard Oser, a

consulting chemist who was the non-voting chair of FEMA’s expert panel.61

Members of the FEMA expert panel were required to be disinterested, with “no

connection with the food or flavor industry that might instill any bias.” The experts were

tasked with developing criteria for evaluating toxicity of flavoring substances, which

came to include the history of the substance’s use (or presence) in foods; predictive

assessments based on studies of the metabolism or toxicity of homologous or chemically

related compounds, where such studies existed; and the levels of typical use in foods.

These experts were given access to “all available information related to safety-in-use of

60
Hall and Oser 1968:7-8.
61
Hall and Oser 1968: 5.
409
each flavoring substance,” and asked to apply their criteria. Experts were asked to certify

not only that they considered a substance safe, but that they expected that their view was

shared by other qualified experts.62 “From the beginning,” explained Oser and a fellow

expert panel member, Richard Ford, in an article about the history of the FEMA panel,

“our policy required that all GRAS decisions of the Panel be unanimous, not merely

consensual, and published in the open literature for comment by the scientific community

at large.”63

The first comprehensive list was published in 1965, in Food Technology.64 It

included 1,124 flavoring chemicals, and also enumerated 267 substances which were to

be dropped from use because they did not meet criteria for GRAS substances. The FEMA

Expert Panel green-lit about 80 additional materials in 1970; many of these were

pyrazines, thiols, and other nitrogen- and sulfur-containing molecules that had recently

been discovered through fundamental flavor research, and which were to form important

components of meaty, chocolatey, and vegetal flavors.65 Subsequent lists are published

on a regular basis, and the FEMA GRAS committee reviews new substances submitted

62
Oser and Ford 1991: 87-8.
63
Oser and Ford 1991: 88
64
This was known as GRAS III, because it followed two earlier preliminary lists.
65
Robert J. Eiserle and William J. Downey, “A Review of the Literature Concerned with
Flavor Research as it Applies to the Problems of the Flavor Industry,” CRC Critical
Reviews in Food Technology 2.2 (July 1971): 159-169. Downey was the head flavor
chemist at Fritzsche Dodge & Olcott.
410
by flavor companies.66 The FDA accepted the agency’s findings almost in their entirety,

and continues to do so to this day.

FEMA’s GRAS committee has been described as a case of “regulatory capture,”

as a compromised group of industry-funded experts coopting the regulatory mechanisms

of the state, thus subverting the public interest in service of private profits.67 The

committee’s scientific procedures, the disinterestedness and legitimacy of their findings,

and the effects of synthetic flavor chemicals on the human body, are beyond the scope of

this dissertation. What I would like to examine here instead are the implications of the

FEMA GRAS list for the materiality of flavor additives and the forms of expert labor

involved in their production — that is, for the work of creative flavorists.

From the outset, FEMA was committed to ensuring that an expansive and

expanding list of chemical compounds were permitted and available for use in flavoring

products. The focus of their effort was concerned not with commodity flavoring

chemicals in mass-use — vanillin, for instance, or MSG — but the compounds that were

produced (and used) in minute amounts. Hall, of FEMA’s Flavor Additive Committee,

offered various examples of such compounds. Furfuryl mercaptan, for instance, was a

“critically important ingredient of imitation coffee flavor,” but it was used in such tiny

concentrations that the total national consumption of the chemical likely did not top 50

66
The most recent list, GRAS 27, was published in 2015, and included approximately
2,500 substances.
67
Marion Nestle, Food Politics: How the Food Industry Influences Nutrition and Health,
(Oakland: University of California Press, 2013); Patrick van Zwanenberg and Erik
Millstone, “Taste and Power: The Flavouring Industry and Flavour Additive Regulation,”
Science as Culture (2014): 1-28.
411
pounds a year. Alpha ionone was a chemical used in concentrations of about 1ppm in

imitation raspberry flavors; no more than 250 pounds of the stuff was used per year, in all

foods and beverages produced. For a typical consumer, Hall estimated that it accounted

for one billionth of their annual diet. With these and other examples, Hall forcefully

argued that, even in the case of individuals who consumed abnormal quantities of foods

flavored with these chemicals, the levels of consumption could not rise to toxicological

significance.68

The cases of these and other chemicals used in minute quantities provided

rhetorical support for two of FEMA’s main arguments: the low toxicological risk posed

by these compounds, and the prohibitive cost of conducting a full toxicological review.

But FEMA’s argument also depended upon the regulatory agency accepting the necessity

of all of these chemicals in the food system in the first place. Hall argued that just

because these concentrations were not of toxicological significance, did not mean that

they were not of sensory — and thus economic — importance. “Who is to determine the

commercial importance of a superior flavor as compared to a merely adequate one?” Hall

demanded. “Flavor compounding is a mixture of intuition, science, experience, and — let

us be candid — hocus-pocus…. The question of whether or not a particular flavor

ingredient is important is not a question on which the opinion of an untrained individual

is entitled to serious consideration.”69 The “hocus-pocus” at the core of flavor creation

was a kind of expert knowledge that could not be duplicated by other scientific experts.

Flavor quality was not easily measured or described; taste panel measurements were at

68
Hall 1959.
69
Hall 1959: 2.
412
“best an approximate science.” For these reasons, the expertise of the people who used

these materials — flavorists — must be taken into account. By insisting on the

significance of molecules used in vanishingly small quantities, FEMA corroborated the

specialized authority of those who did have the authority to weigh in on the importance

of these materials.

This position derived from, and helped to nurture, an increasingly organized

professional culture that recognized a “superior” flavor, and valued the technical and

creative skills involved in its production. It protected the interests not only of the

companies that used these chemicals in their products, but of the laborers (creative

flavorists) who worked directly with these compounds and who, increasingly, derived

their professional identity from their skillful use of these materials. Further, by making

the production (and protection) of a list of allowable flavoring materials a common

project, the FAA, and the GRAS list, also reinforced the bonds between flavorists across

companies, as a community with mutual interests in these chemicals. In the past, some

flavor companies had gained an advantage because of exclusive access to an otherwise

unknown flavoring compound.70 One of the consequences of the new regulatory regime

70
James Broderick has described several such examples in his “Reflections of a Retired
Flavorist Before He Forgets” columns that were published in Perfumer & Flavorist in the
early 1990s. Some of these “proprietary” compounds were created by chance, for
instance the lactone used in Fries’ peach flavor in the late 1930s. Others were the result
of basic analytic research into the flavor chemistry of fruits, such as Firmenich’s
“raspberry ketone” (para hydroxy phenyl butanone), identified by Coppens and
Hoejenbos of Polaks Frutal Works in the Netherlands, and used by both that company
and the Swiss-based Firmenich in their prewar raspberry formulations. James J.
Broderick, “Reflections of a Retired Flavorist Before He Forgets: Raspberry,” Perfumer
& Flavorist 16 (Nov/Dec 1991): 13-14; “Reflections of a Retired Flavorist Before He
Forgets: Peach,” Perfumer & Flavorist 17 (Jan/Feb 1992): 35. In the wake of the FAA
413
was that there were no longer any “secret” ingredients; every permissible chemical was

now openly listed and disclosed. However, just because a chemical was on the GRAS

list, did not mean that it was available in the chemical marketplace in a form suitable for

use in flavors. Many listed chemicals were, in fact, unavailable; many desirable

chemicals were unlisted. Further, because of the sensitivity of the human sensorium to

many odor compounds, flavorists required materials that were exquisitely pure. Often,

commercially available chemicals failed to meet the extreme standards of purity required

for flavor applications. (In practice, this often meant that flavor companies further

processed chemicals that they purchased commercially in order to obtain the required

purity.)71

One of the first major initiatives of the SFC was to compile a common database of

chemical suppliers for the list of GRAS compounds. The members of the SFC Flavor

Chemical Source Committee searched through available chemical catalogues, wrote

letters to suppliers asking for updated information, scoured brochures and advertisements,

and drew upon their own personal knowledge of unlisted supply lines and other sources.

Published in 1968, Food Chemical Sources included an alphabetical listing of flavor

chemicals (each with its corresponding FEMA GRAS number) and confirmed suppliers,

and GRAS list, some larger, research-oriented companies (such as IFF, Firmenich, and
Givaudan) did file for (and obtain) patent protection for chemical compounds they had
synthesized — an example is Furaneol, discovered in the mid-1960s in strawberries in the
laboratories of Firmenich. Firmenich had a patent on the use of furaneol in fruit flavors
— and Unilever had a near-concurrent patent for its use in meat and savory flavors —
until the 1980s. While outside the scope of the current dissertation, patent-protected
flavoring compounds will be a subject of future research and interest.
71
A.V. Saldarini, “The Chemical Sources Association — The Pioneers,” Perfumer and
Flavorist 13 (Aug/Sept 1988):57-8.
414
as well as a directory of 128 flavor chemical supply companies.72 Some flavor chemicals

were available from multiple sources. For instance, ethyl methyl phenyl glycidate

(FEMA 2444), the compound once known as “strawberry aldehyde,” was available from

nine companies, including Dodge & Olcott and F. Ritter in Los Angeles. But quite a few

compounds had no known suppliers. For instance, iso butyl iso butyrate (FEMA 2189), a

chemical with an odor reminiscent of pineapple, had no known commercial sources. The

committee saw this directory as an ongoing project. “We urge all to pass on to us

suppliers of various materials which have been left blank in this directory, as well as

suppliers other than those listed,” urged the introduction from the committee’s chairman,

Frank Fischetti, a flavorist at Fritzsche Brothers.73

With the publication of a second edition of Food Chemical Sources in 1971,

corresponding to the expanded GRAS list published the previous year, the gaps between

“permitted” and “accessible” became even more evident. “For the supplier to justify

production, there had to be a demand, a demand by many companies to increase the

volume and justify research and production costs,” explained Al Saldarini, the first head

of the SFC Flavor Chemical Source Committee, in an article describing its history.

Instead of merely compiling known suppliers, the SFC used its collective power to

demonstrate demand and actively develop new sources. In alliance with like-minded

workers in the fragrance industry, they created a new organization, the Chemical Sources

Association (CSA), in 1972. Chemical suppliers were solicited, and invited to SFC

72
Society of Flavor Chemists, Food Chemical Sources, (New York: Society of Flavor
Chemists, Inc., 1968).
73
Society of Flavor Chemists 1968: [np/2].
415
technical meetings, where they were both educated about existing needs, and invited to

present their products to members. The CSA also funded research into chemical

synthesis, especially for compounds of interest for which no supplier could be found.74

This is not to suggest that flavorists took the toxicological risks of the materials

that they used lightly. But while “chemicals” and “foods” are generally perceived to be

mutually exclusive categories by many ordinary consumers, flavorists’ attitudes were

informed largely by their understanding of their materials as both flavors and chemicals.

This also informed the accepting, but generally skeptical, attitude toward the distinction

between “artificial” and “natural” flavoring materials that the FDA imposed on labels in

the early 1970s. According to flavorists’ ways of working with chemical materials, these

categories were logically inconsistent — regrettable signs of chemophobic attitudes

among certain sectors of the public, enshrined in regulatory law.

Making Chemicals Into Flavors: The Flavorist at

Work

“The problem that exists today, if it can be called a problem, is the rapidity with

which new aromatic chemicals have appeared for flavor use,” wrote Frank Fischetti, a

flavorist at Fritzsche, Dodge, & Olcott, in 1980.75 Although somewhat tongue in cheek,

Fischetti is giving voice to his legitimate sense of the scale and speed of the

74
Saldarini 1988: 58.
75
Frank Fischetti, Jr. “Natural and Artificial Flavors,” in Thomas E. Furia, ed. Handbook
of Food Additives, 2nd Edition, Vol. II, (Boca Raton: CRC Press, 1980): 307.
416
transformation of the material-culture of his trade. Fischetti, who began his career as a

flavorist in the late 1950s, had witnessed massive changes in flavor chemical knowledge,

including the introduction of new families of compounds — pyrazines and thiazoles —

formerly unknown as flavoring ingredients.

Postwar flavorists were the beneficiaries of a scientific, technological and

chemical regime that provided unprecedented access to the chemical secrets of flavors in

foods and the intimate mechanisms of perception, as well as new capacities to obtain and

produce synthetic molecules of olfactory interest. But while fundamental research in

flavor chemistry produced prolific lists of substances, it was up to flavorists to work out

how to apply this knowledge to the creation of distinctive, useful, and compelling

synthetic flavors.

In this section, I consider several published accounts of the work-process of flavor

development, all of which illuminate distinct aspects of the creative labor of making

flavors. The first example I consider dates from the first half of the 1950s, just before the

commercial introduction of analytic instruments and their widespread use. Appearing in

the Givaudan Flavorist, the company’s newsletter, this article vividly depicts the

flavorist’s unique approach to the task of flavor “imitation.” The flavorist smells

analytically, reads the scientific literature not only for facts, but also for suggestions and

clues, and assembles materials to create sensory rhymes rather than produce molecular

replicas. I then consider a set of later texts that directly grapple with the consequences of

instrumental technologies such as GC for the creative labor of the flavorist, and that

sharpen the distinction between flavorists and analytic flavor chemists, while also
417
elucidating their areas of interdependence. Finally, I examine a case which situates the

process of new flavor development within both the institutional structure of the flavor

company, and its broader commercial context.

What these accounts demonstrate is that the flavorist possessed a body of

knowledge about chemical materials that included the analytic findings of research

chemists, but that also exceeded them. Flavorists knew different things about both

chemical materials and sensory experience than flavor chemists, asked different things

from the stuff and the machines they both worked with. Rather than being bound to the

discoveries of analytic chemistry, flavorists put this knowledge to use in ways that

changed the contours of the flavored world, not so much forging entirely new species of

sensation, as intensifying and modifying the familiar. However, this seeming

independence of the creative process, does not mean that flavorists operated as

idiosyncratic “genius” type creators, working in eccentric isolation. On the contrary,

flavorists were creatures of the particular milieu of the flavor company, indeed, perceived

themselves at the very center of it.

“The Art in Artificial Flavors”: Building a Flavor by

Analyzing a Scent

“To the real scientific mind, one thoroughly trained and schooled to think in

precise terms, the creation of flavors has always been looked upon as a mystery, or, kind

of ‘black art’ and the flavor chemist as anything but a scientist,” admits the opening

paragraph of “The Art in Imitation Flavors,” the feature article in the second issue of the

Givaudan Flavorist, the flavor and fragrance company’s newly launched newsletter
418
publicizing the work of its flavor division. “The fear is often subconsciously expressed

that these flavor creators are a long step back to the days of the alchemist.”76

Like many chemical companies serving the needs of industrial manufacturers,

Givaudan put great emphasis on the technological sophistication of its laboratories,

factories, and personnel; the promotion of its flavor division was connected to its

increasing investment in research and development in that sector.77 How could the work

of the flavorist find its place within this intensively, even ostentatiously, scientific

milieu? After all, the labor of flavor creation — especially the creation of artificial

flavors — remained associated with closely guarded “secret formulas,” illusory

resemblances, and even deceptions. With “The Art in Artificial Flavors,” the company

sought to dispel the suspicions around the nature of the flavorists’ work. Taking the

creation of an artificial strawberry flavoring as an example, the article goes on to explain

both the flavorist’s methods, and the necessities (and virtues) of his labor.

While the analytic investigations of research chemists into the chemical

constituents of flavor had only yielded lackluster results, the flavorist’s power came from

the ability to smell analytically. “Although the [flavorist] has trained himself to identify,

by odor, hundreds of aromatic chemicals and essential oils, actually he creates by

building up a series of basic odors that he identifies in the product he wishes to reproduce

synthetically.”78 In other words, when a flavorist sniffs a strawberry, the purpose is not to

76
“The Art in Imitation Flavors: The Aromatic Constituents of Strawberry,” Givaudan
Flavorist 1953 (no 2): 1. This article was likely written by Earl Merwin.
77
See Chapter 4.
78
“The Art in Imitation Flavors” 1953: 1-2.
419
use the nose to discern which chemicals are present, but to determine the sensory

dimensions (“basic odors”) of its aroma. In the case of Givaudan’s strawberry, these

were: fruity/estery, green butter, sweet, balsamic, straw/hay, rose-honey, and sour/citrus.

Having spliced “strawberry” into these seven aromatic shades, none of which were

explicitly strawberry-like, the flavorist then considers the available materials that can

produce the requisite effects.79

How are these materials known and chosen? “The Art in Imitation Flavors”

included ample references to recent findings in the basic chemistry of flavor, presenting

an image of the flavorist as up-to-date on the latest scientific literature. The flavorist

reads the research not only for facts, but also for clues. While a flavor chemist at the

USDA or in a university food science department may see the growing list of identified

compounds as an accomplishment in the pursuit of total chemical knowledge, the

flavorist, attendant instead upon the total sensory effect and its subtleties, is exquisitely

aware of the remaining (innumerable) unknowns. The flavorist uses the research as a

starting point, in order to plunge into the negative space of unknown chemicals and

unknown relationships: to make deductions about implied presences, and draw sensory

analogies with related compounds. For instance, to produce the strawberry’s buttery note,

the flavorist might begin with diacetyl and acetyl methyl carbinol — substances that had

79
These sensory dimensions were, admittedly, arbitrary. “We do not infer that the
breakdown is complete, or that this is the only possible breakdown — such a division will
vary with the individual flavor man,” the article noted. (“The Art in Imitation Flavors”
1953: 2.) For another account of how the green note in strawberry flavors was achieved
by flavorists at different companies, see: James J. Broderick, “Reflections of a Retired
Flavorist Before He Forgets: Strawberry,” Perfumer & Flavorist 17 (May/June 1992):
33-4.
420
been found in both strawberries and butter — and add related compounds to enhance the

effect, such as the higher homologs of diacetyl. To lend a “green” and grassy nuance to

the buttery note, he might then add other materials, including ethyl acetyl acetate (which

had, at that point, only tentatively been identified in strawberries), Siberian Pine Oil (a

natural essential oil unrelated to strawberries), and beta-gamma-hexenol (also known as

3-hexen-1-ol), a chemical compound with an intense, green odor, that had been identified

in “many green plants,” but had never been found in strawberries.80 Thus, the flavorist

could borrow a sensory effect from the broader chemical literature, transposing the vivid

3-hexen-1-ol greenness of a clover-leaf or cucumber into his evocation of the fraicheur of

strawberry.

In some cases, flavorists’ insights led him or her to chemicals that were actually

present in the food in question.81 In an article published the following year, the Flavorist

reported on a “very interesting paper” at the recent annual meeting of the IFT, where

Dimick and Makower, of the USDA Western Regional Research Laboratory in Albany,

presented their pioneering work using gas chromatography to investigate the chemical

constituents of strawberry flavor.82 (This was more than a year before the first official

publication of these findings in Food Technology).83 Among newly identified compounds

was 2-hexenal, an aldehyde that was often found in conjunction with 3-hexen-1-ol, the

chemical previously suggested as a green note in artificial strawberry flavors. (Indeed,

80
“The Art in Imitation Flavors” 1953: 2-3.
81
For instance, in the case of methyl anthranilate and grape flavors. See Chapter 2.
82
“The Green Note in Fruits,” Givaudan Flavorist 1954 (no. 4): 2. See also Chapter 6.
83
KP Dimick and Benjamin Makower, “Volatile Flavor of Strawberry Essence. I.
Identification of the Carbonyls and Certain Low Boiling Substances,” Food Technology
10.2 (February 1956): 73.
421
observing that 3-hexen-1-ol “readily” oxidizes to form 2-hexenal, the Flavorist article

speculated that the ripening of the fruit may correlate with the change from alcohol to

aldehyde.) “Work done on natural products has been an aid in creating new flavors, and

we flavor chemists anticipate even greater assistance in the near future,” the article

concluded. “However, with a deep sense of humility, we would like to state that it is our

observation that the ‘art,’ personified in the flavor chemists’ nose and sense of taste, still

has the edge on the rapidly approaching science.”84

Even as the flavorist’s chemosensory savvy led him to foresee the determinations

of the analytic machine, his interest was not constrained within the limits of confirmable

chemical presences. The 1954 Flavorist article went on to observe that the esters of 3-

hexen-1-ol, which (with one exception) had not been identified in nature, “are even more

interesting, from a flavor standpoint, than the alcohol or aldehyde. They have a pungent

but soft fruitty [sic] green odor which has a greater utility in imitation flavors than the

parent alcohol.”85 That is, the flavorist’s engagement with the material dimension of

flavor is not speculation in search of objective confirmation, but sensory extension,

invention, and imagination.

One necessary context for appreciating the forms of the flavorist’s work was the

vastness of the set of chemical unknowns. Given the lack of certain knowledge about the

chemistry of flavor and mechanisms of sensory perception, the creation of synthetic

flavors required the highly specialized skills of individuals equipped to negotiate those

84
“The Green Note in Fruits” 1954: 2.
85
“The Green Note in Fruits” 1954: 2. The exception was the phenyl acetic ester of 3-
hexen-1-ol, which had been found in Japanese mint oil.
422
unknowns in order to produce chemical mixtures that reliably produced desired effects.

However, the unknowns of flavor chemistry were not, per se, unknowable. Written

before the introduction of commercial GC devices, but after Martin and James’ seminal

paper on the subject, “The Art in Imitation Flavors” clearly anticipates the

transformations of analytic chemistry that appeared on the horizon. “It is to be expected,”

the article foretells, “that at some future date a scientist… will be able to test a given odor

in a man-made piece of laboratory apparatus and break down this odor into its component

and basic parts, which information can then be used to duplicate the odor from its basic

materials or other materials with the same odor characteristics.”86

But even in the scenario of total chemical knowledge, “even if the complete

reproduction of the aromatic constituents of strawberry were possible,” the article insists

that the flavorist’s peculiar capabilities would still be required.87 Chemical knowledge is

not enough. The same configuration of volatile organic molecules will perform

differently in a cellophane-wrapped fruit-creme-filled snack cake on a supermarket shelf,

than in a fruit dangling from its stalk in a farmer’s field. The flavorist’s savvy

substitutions can “give the same flavor effect” in the radically different contexts of

production and consumption.

Nine years later, the Givaudan Flavorist revisited its early articles on flavor

creation, reprinting revised versions of the articles, now attributed to Earl Merwin.

“Flavor creation is based on science and art,” Merwin recapitulated, before continuing.

86
“The Art in Imitation Flavors” 1953: 1.
87
“The Art in Imitation Flavors” 1953: 4.
423
“In addition to science and art, there is a third tool that the flavorist must use. The third

foot in the triangular base on which the development of flavors stands is ‘technology.’

Neglect any one of these three and the results will be poor — insufficient for application

to the present state of our food industry.”88 The gas chromatograph (GC) had become an

essential tool of the flavorist. However, no matter how sensitive the machine becomes, it

“will not replace the flavorist’s nose because it is not hooked up to the flavorist’s brain. It

can and does help the flavorist’s nose — implementing the third leg of the triangle — the

artistic quality of flavor creation.”89

“Fresh-from-the-field flavor!” exclaimed a Givaudan advertisement adorned with

a colorized black-and-white photograph of intensely red fruit; it claimed that “nothing has

duplicated” the flavor of “fresh wild strawberries… as closely as Givaudan’s Imitation

Strawberry.”90 The science and technology of flavor creation was put in service of the

flavorist’s art, which did not so much duplicate the strawberries of the field, as create

unprecedented versions, situational strawberries for specific and proliferating

applications, which themselves (at least, ideally) would soon become familiar.

88
Earl Merwin, “The Art in Flavor Creation I: True Fruit Flavors,” Givaudan Flavorist
no. 3 (1962): 5.
89
Earl Merwin, “The Art in Flavor Creation II: Imitation Flavors,” Givaudan Flavorist
no 4 (1962): 3.
90
[Givaudan] “For Fresh-from-the-field flavor!” [Advertisement], Food Technology 11
(November 1957): 23.
424
The Flavorist at the Machine

A 1959 article in the Givaudan Flavorist by V.D. Johnston, the company’s chief

analytic chemist, offered readers a virtual tour of the analytic laboratory, where

“conventional chemical and physical methods are being replaced or supplemented by

modern instrumental methods.”91 Photographs of white-jacketed male and female

researchers intensely preoccupied with the knobs and registers of various gleaming

machines accompanied his descriptions of the different spectrophotometers and gas

chromatographs that were constantly in use in the large air-conditioned space. He

explained that the machines were used for quality control, process control, and

fundamental research, saving the company and its customers both time and money. “But

what is most valuable,” Johnston said, “they give more information; sometimes too much

information.”92

By the 1960s, most flavor companies — and particularly research-oriented

companies such as Givaudan, IFF, and Fritzsche, Dodge & Olcott — employed both

analytic flavor chemists and flavorists on staff. The technical instruments of the modern

flavor laboratory, particularly the GC, were used by both groups of workers. These

machines helped reveal the chemical complexities that produced the effects of flavors in

foods, and were integral to the increasing material sophistication of flavoring additives.

On the other hand, as Johnston suggests above, the machines often provided “too much

information,” disclosing chemical presences that were irrelevant to the sensory qualities

91
V.D. Johnston, “Instrumental Methods of Analyses Save Time — Give More
Information,” Givaudan Flavorist 1959 (1): 1-4. For more on the instrumental revolution
in chemistry and its effects on flavor research, see Chapter 6.
92
Johnston 1959: 1.
425
of a flavor or that were artifacts. Both analytic research chemists and flavorists had to

grapple with the problem of signal and noise when it came to these machines, but their

priorities were different.

A 1971 article by Robert Eiserle and William J. Downey, flavorists at Fritzsche,

Dodge & Olcott (FD&O), reviewed recent scientific literature on fundamental research in

flavor chemistry, meticulously drawing out the points of interest for flavorists in the latest

studies on meat flavor volatiles, trace compounds contributing to roasted barley flavor,

pyrazines in peppers, potatoes, and more.93 They are insistent throughout that simply

having more information about chemical constituents of foods was not of great value.

“Finding new components does not always help the flavor chemist to prepare better

flavors.”94 First, research findings were often impossible to apply directly, at least

immediately. “Often such information is useless to the creative chemist since the

materials identified as being naturally present are not found on the official lists of

approved flavoring ingredients,” they note ruefully.95 Further, research chemists often did

not include organoleptic evaluations of the compounds they discovered, or attempt to

understand the role that various components played in “the total flavor effect” perceived

by the “ultimate consumer.”96 For this pair of flavorists, research findings in flavor

chemistry often provided too much information that was also insufficient for their

93
Robert J. Eiserle and William J. Downey, “A Review of the Literature Concerned with
Flavor Research as it Applies to the Problems of the Flavor Industry,” CRC Critical
Reviews in Food Technology 2.2 (July 1971): 159-169. Downey was the head flavor
chemist at Fritzsche Dodge & Olcott.
94
Eiserle and Downey: 165.
95
Eiserle and Downey: 160.
96
Eiserle and Downey: 163.
426
purposes. In a near-contemporary article, James Broderick, the SFC founder and

Kohnstamm flavorist, summarized the distinction between the “basic researcher” and the

“practical flavorist”: “the researcher’s goal is to identify all components of a flavor, and

the flavorist is frustrated by the fact that much of the research has little practical value for

him. The flavorist needs to identify the key components, and this is an area in which

flavorists and researchers should work more closely together.”97

The distinctions between the flavorist and the analytic flavor researcher are

highlighted in their different occupational attitudes toward GC and its output. It should be

noted, first, that GC’s output assumed multiple forms. The machine produced a

chromatogram, a permanent graphical record of peaks and valleys that registered

detectable chemical presences in the flow of inert vapor as it passed out of the machine,

indicating both the time at which the “peak” eluted from the machine, which could help

with identification, and also its relative quantity in the mixture. But the GC also produced

the separated chemicals themselves. Each peak on the chromatograph indicating a

fraction of the initial mixture, ideally, an isolated compound. These fractions could be

collected at the GC’s exit with specially designed traps, and then subjected to further

instrumental analysis — either subsequent GC separations, or structural identification

with spectrometric instruments, such as the mass spectrometer (MS). Beginning in the

early 1960s, GC and MS were often directly conjoined in a powerful device that

combined separation and identification in a continuous process. But the versatile GC also

allowed another modification, the diversion of some of the vapor-stream effluent to a

97
James J. Broderick, “Fruit Flavor Research: The Practical Flavorist vs. the Basic
Researcher,” Food Technology 26 (November 1972): 37.
427
“sniffer port,” where each fraction could be olfactually evaluated and savored by the

sniffing researcher in synchrony with the detector’s production of the chromatogram.

The sniffer port was critical to the GC’s usefulness to flavorists. Broderick

described the flavorists’ attraction to the sniffer port in memorable terms. He vividly

described an experiment in entomology, where “male moths were strapped to a board at

the outlet of a gas-liquid chromatograph, and an extract from female moths was injected

into the GLC. The key component was pinpointed by the agitation of the male moths

when that component was emitted from the GLC.” He added, “Although this technique is

not generally applicable to fruits, I’ve seen some happily agitated flavorists when they

sniffed a key component sought in a complex run.”98 In other words, flavorists were

excited, inspired, fascinated by smells — but in the olfactory panorama that unspooled

from the GC, what kind of smell would pique this sort of interest? What, exactly, were

flavorists sniffing for?

Increasingly, of course, flavor chemists were sniffing, too — integrating methods

of sensory evaluation into experimental protocols, integrating “nasal appraisals” into their

work process.99 In their review of recent literature in flavor chemistry, Eiserle and

Downey reserve praise for studies that combined instrumental analysis with organoleptic

panels, or that use specially trained judges to make odor determinations. But even when

flavor chemists attended to the sensory characteristics of the compounds they identified,

and included trained organoleptic panels in their experimental protocol, their methods

98
Broderick 1972: 37.
99
See Chapter 6.
428
and conclusions lacked the insights that flavorists possessed. As an example, Broderick

describes recent research into the chemistry of apple flavor at the USDA Western

Regional Laboratory in Albany.100 The Albany team used high-resolution capillary GC

and mass spectrometry to separate and identify 56 different volatile compounds in

Delicious apple essence. Unlike most analytic chemists, they went further, and attempted

to determine which of these compounds contributed to the apple’s flavor. Each of the

separated compounds was subjected to organoleptic evaluation by a panel specially

selected and trained judges, who were asked to indicate which components possessed

apple-like aromas. Broderick commends the researchers for demonstrating the sensory

significance of several previously unreported volatiles in apples, but also notes that when

all of the components identified as “apple-like” were blended together, the result was

something other than apple. “Something was missing.”101

He elaborates: “the fallacy in this approach is that total apple flavor is far more

than just these ‘apple-like’ components — many important nuances of total apple flavor

are not apple-like.” Clove-like eugenol, he observes, plays a crucial role in cherry flavors;

the honeyed green rose of phenyl-acetaldehyde has an important part to play in

strawberry flavor. “A panel cannot pick out all of the key components of apple flavor,

although they may pinpoint the apple-like components. Other apple flavor nuances can

only be pinpointed by someone with the ability and training to break down the flavor into

100
R.A. Flath, D.R. Black, D.G. Guadagni, W.H. McFadden, and T.H. Schultz,
“Identification and Organoleptic Evaluation of Compounds in Delicious Apple Essence,”
Journal of Agricultural and Food Chemistry 15 (Jan/Feb 1967): 29-35.
101
Broderick 1972: 37.
429
its various nuances and evaluate the individual components in relation to these nuances,

and this person is the flavorist.”102

What did the flavorist know that the flavor chemist did not? According to

Broderick, the flavor chemist lacked a working sensory understanding of “total flavor.”

The flavor chemist thinks in terms of chemical building blocks; the flavorist begins with

sensory ones. “The flavorist mentally breaks down a flavor into various nuances and then

tries to simulate each nunace with the materials at his disposal, blending them to get a

final effect,” he explained. “The quality of the final product is dependent upon the

knowledge and artistry of the flavorist and the materials available to him to simulate the

flavor nuances.”103 This recalls the description offered in “The Art in Imitation Flavors”

— which, indeed, was written shortly after Broderick left Givaudan, so the similarity may

be due in part to a common company style rather than to a broader occupational praxis.

Even more, the flavorist knew that these “sensory building blocks” were often quite

dissimilar from the character of the “total flavor.” This was part of the flavorist’s

attunement to the contribution of dissimilar, unlikely, and perhaps repugnant sensations

to the total perception of a flavor, and it became even more acute as more was known

about the chemistry of flavors in nature. Sniffing at the GC could reveal “a whole ‘dinner

table’ full of odors” in a tomato, to use a phrase from a flavorist from General Foods:

bacon and vanilla, cucumber and macaroon, as well as rubbery sulfur and stale hay.104

102
Broderick 1972: 37.
103
Broderick 1972: 37.
104
William S. Ryder, “Progress and Limitations in the Identification of Flavor
Components,” in Irwin Hornstein, ed. Flavor Chemistry: A Symposium, ACS Advances
in Chemistry Series 56 (Washington: ACS, 1966): 82.
430
The flavorist was sniffing for the rubber and sulfur, the stale hay and the roast meat, in

search of difference and distinction. For the flavorist, sniffing was not so much about

obtaining certainty about any of the components, as about gaining insight into the whole.

Indeed, Merwin explained that each flavorist used the GC “in a slightly different manner,

just as he uses every other piece of laboratory apparatus.”105 While analytic flavor

chemists strove for technical mastery of the machine and the chemical world it separated

and fractured, flavorists used the GC as a tool to cultivate the total sense of flavor, and to

inform their own personal style and approach.

Frank Fischetti, of Fritzsche, Dodge & Olcott, explained how this factored into

the process of creating flavor compositions. A flavor can be thought of as consisting of

distinct parts: “flavor character items,” ingredients whose “aroma and/or taste is clearly

reminiscent of the named flavor”; “flavor contributing items,” compounds which while

“not necessarily (and by itself) reminiscent of the named flavor… when used in

conjunction with flavor character items, tends to bring it closer to the named flavor,” and

finally, “flavor differential items.”106 Unlike the first two, these ingredients or

combinations “have little, if any, character reminiscent of the named flavor. These items

are added to a flavor compound to give it individuality, imagination, and difference.

These are items a flavorist employs to create special effects,” such as lift, nuance,

undertone, and aftertaste.107 Essentially, the “flavor differential” factors were the

flavorist’s signature — the “creator’s mark” — and the signifiers of a house style,

105
Merwin “The Art in Flavor Creation II: Imitation Flavors,” 1962: 3.
106
Fischetti 1980: 313.
107
Fischetti 1980: 316.
431
“distinguish[ing] the products of one flavor house for another.”108 Like most marks of

style, they also served a distinct commercial purpose: they prevented copying. “Used in

extremely small quantities, they render duplication almost impossible,” Fischetti

advised.109 Thus, the flavorists’ way of working was never mere replication; the goal was

not to present a sensibly indistinguishable copy of original nature. Instead, the motive and

the interest was to create distinction, difference, and variety — to author a strawberry,

rather than simply paint its portrait.

Finally, none of the flavorists described present the flavorist’s way of working as

superior to that of the research flavor chemist. Though distinct, these were not rival

bodies of knowledge jockeying for jurisdiction, but collaborative fields; the relationship

between these two professions was not seen as adversarial, but complementary. But the

benefit could also extend in the other direction. Broderick estimated that between seventy

and ninety percent of the key components of most major commercial fruit have been

identified. Rather than simply trying to identify as-yet-unidentified chemicals, the

analytic researcher should work with the flavorist to determine “what important effects

are missing.” This way, chemical research could be directed towards the identification of

these key components, rather than the “trace peaks that have little or no effect” on flavor.

“The benefits,” Broderick assures, “will be lower cost/results ratios, better imitations,

faster results, and sounder conclusions.”110 The potential applications of this

collaboration went beyond the improvement of flavor additives. Broderick notes that

108
Fischetti 1980: 255, 316.
109
Fischetti 1980: 316.
110
Broderick 1972: 48.
432
Oregon and Washington strawberries “have more and better flavor than do California

strawberries.” Identifying the chemical components that gave the berries of the Pacific

Northwest a flavor advantage would lead not only to better imitation flavors, but also to

better strawberries. By understanding the biochemical pathways by which these flavor

chemicals are formed within the fruit, “techniques could be developed that would enable

the farmer to grow more flavorful strawberries.”111

The Flavorist in the Flavor Company

Finally, the flavorist’s labor of flavor creation should be understood within a

commercial and business context. The flavorist’s creative work did not occur in isolation,

but was part of a broader, coordinated effort among multiple corporate divisions —

including management, R&D, toxicology, legal, sales, purchasing, and production.

Increasingly, this work took place in the midst of large research-based flavor companies,

which employed multiple flavorists specializing in different product applications.112 The

success of a flavor was ultimately determined by its commercial performance: its ability

to find a market among food manufacturers, and the ultimate popularity of products that

contained it with consumers.

In a 1981 article, Manfred Vock, a senior research flavorist at International

Flavors & Fragrances (IFF), located the “creative flavor chemist or flavorist” at the center

of the flavor development process. “All flavor research efforts are channeled through

111
Broderick 1972: 48.
112
Arthur D. Little 1968: 25.
433
him, and the flavor which he creates is the link between R&D and sales.”113 The article,

perhaps the most detailed published account of this internal process, tells the story of the

development of cocoa flavorings at IFF in the late 1960s.114 IFF, a public company

formed in 1961 by the merger of van Ameringen-Haebler and Polak & Schwarz, was

known for its robust research and development program.115 The decision to study cocoa

flavor began with the company’s management. Market research had concluded that an

unmet demand existed for a high-quality synthetic cocoa flavoring designed for use in

new types of convenience foods. Further, low-cost cocoa powders fell far short of

delivering the flavor of high-quality goods. “The objective was to develop a cocoa flavor

which would enhance the cocoa and chocolate aroma and taste of various cocoa products

such as instant powders for milk beverages and instant desserts,” Vock explained — a

synthetic cocoa flavoring product that could replace low-quality “natural” cocoa

powders, both by delivering improved flavor quality and enhanced flavor performance in

processed foods. “This need is obvious, because cocoa powder develops full aroma and

113
Manfred H. Vock, “Development of a Flavor at IFF, Planning, Creation, and
Commercialization,” in George Charalambous and George Inglett, eds., The Quality of
Foods and Beverages: Chemistry and Technology, vol 2. (New York: Academic Press,
1981): 198.
114
Vock 1981: 198-209.
115
According to Dorland and Rogers, when IFF went public in 1961, “the infusion of
new capital made it possible to expand in most any direction management selected, and
also provided ample funds for research, as well as for equipping plants and laboratories
with the sophisticated apparatus required for production and quality control of aroma
chemicals under modern conditions.” In 1976, the company spent $16 million on research
on flavors and fragances, and expected to spend even more in 1977. In 1976, the research
staff numbered more than 500 individuals, at the company’s research center in Union
Beach, NJ and at laboratories in 20 countries. Dorland and Rogers 1977: 197.
434
taste after heating, which is naturally not available for instant cold chocolate flavored

foods and beverages.”116

The product development process started with a literature search: a review of

published analytical studies of cocoa and chocolate flavor, as well as spices and essential

oils which were used as cocoa enhancers or extenders, in order to compile a list of all

known compounds and materials associated with cocoa flavor. Meanwhile, an IFF flavor

profile panel evaluated a variety of cocoa powders, establishing the primary and

secondary sensory qualities characteristic of high-quality cocoa — which were used to

create an ideal flavor profile of the target flavor. The flavor profile panel then

“systematically evaluated” the organoleptic qualities of all of the chemicals identified in

the scientific literature, a crucial step given that flavor chemistry research often did not

provide reliable sensory characterizations. Only compounds that showed qualities related

to the target flavor were considered for further research. This narrowed the field of

chemicals under consideration substantially, and also revealed a sensory gap that no

identified compound currently satisfied: “the delicate cocoa/rose related aroma of high

quality cocoa powders.”117 This guided analytic research at IFF, where chemists

succeeded in identifying a quartet of previously unknown unsaturated aldehydes that

produced desired cocoa-rosey and bittersweet-nutty flavor effects.118 The cocoa flavor

project also received “unexpected help… from completely unrelated flavor work.” The

116
Vock 1981: 200.
117
Vock 1981: 206.
118
Vock 1981: 206-7. IFF protected its research by patenting cocoa flavoring
compositions using these compounds. US Patent 3,582,360, “Cocoa Flavoring
Composition Containing 2-phenyl-2-alkenals and method of using same.” June 1, 1971;
US Patent 3,754,038, “2-Phenyl-2-Alkenals,” August 21, 1973.
435
company’s biosynthesis group had produced a new cocoa-like flavor, which delivered

some of the characteristic bitterness of good chocolate. As this substance was

biosynthetically produced, it was technically a “natural” flavor.119 Identifying significant

compounds was not enough, of course. They also had to be produced synthetically.

Research chemists at IFF developed syntheses for many of the new or unavailable

compounds of interest, including pyrazines, furans, and the unsaturated aldehydes. Then a

search of the patent literature was conducted, to ensure that no components or processes

violated existing patents.

Once all of these steps were completed, the flavorist was ready to begin doing his

(or her) work: the actual labor of flavor creation. Essentially, the preceding steps have

assembled a library of possible materials for the flavorist to use. Similar to the examples

of flavor creation described above, Vock begins with a sensory portrait of cocoa, and

selects chemical compounds that corresponded with those effects. He uses the metaphor

of building to describe the work process. First the “corner stones of cocoa flavor

structure” were laid down; key chemicals were selected to produced desired primary

notes, such as cocoa, floral/rosey, and malt, and used at concentrations that correlated to

their sensory thresholds. As the sensible building was chemically assembled, the flavorist

began to attend to secondary notes and nuances: “‘edges’ were smoothed and ‘holes’ were

filled by decreasing or increasing the concentrations of the flavorings.” It was a gradual

process of constant adjustment, especially as new additions could enhance or affect the

underlying blend in synergistic ways. All throughout, the flavorist continues tasting

119
Vock 1981: 207.
436
components and blends in water or sugar water, “until a harmonious cocoa flavor was

achieved.”120 He subsequently offers another common metaphor for the creative labor of

the flavorist, comparing it to fine art painting — “especially… the color combinations of

an abstract work.” Just as the painter relies on his (or her) eye and “modifies the

available colors and shades until the desired effect is achieved…. Taste and smell are the

creative senses of the flavorist.”121 Quite a few flavorists, Vock remarks, also happen to

be excellent painters.

Quite a few steps remain before the flavorists’ “harmonious cocoa flavor” makes

it to the production and sales stage. The composition is evaluated in various applications,

where its flavor profile is compared with the target flavor profile. Then follow stages of

preference testing, stability evaluation, quality control work, and applications

development, all of which require the flavorist’s adjustments, modifications, and input.

Vock compares his form of sensory craftsmanship with “other approaches” to

flavor creation, both of which, he counsels, are less likely to be successful. The first rival

strategy is to begin with the chemicals: combining all the chemicals identified by the

analytical work, not only the ones selected by organoleptic panels as significant. This

approach is flawed: chemical knowledge of flavor must always be assumed to be

incomplete and provisional, thus it is likely to produce a product whose sensory qualities

leave something to be desired; further, the synthetic reproduction of all identified

chemical components is not economically viable. The second strategy is to work with

120
Vock 1981: 208.
121
Vock 1981: 208.
437
“total flavors,” rather than individual chemical components; for instance, utilizing a

previously produced “malt flavor” rather than chemical compounds associated with that

effect to produce the malty quality of cocoa. This leads to snowballing problems, as

interactions between components in different flavor mixtures can lead to unpredictable

sensory consequences and high production costs.122

Vock’s implicit purpose in including these two rival strategies seems clear. The

craftsmanlike method that he favors, where the artisanal flavorist is guided by his or her

senses, skill, and experience, superficially appears less systematic and more inefficient

than the rival modes he describes — both of which, essentially, ask the flavorist to

execute a composition that replicates the findings of more conventionally “scientific”

research workers, whether that of the analytic flavor chemists who describe flavor in

terms of chemical presences, or of the flavor profile panel, which describes it in terms of

sensory qualities. The model that Vock presents continually links the two bodies of

knowledge, chemical and sensory, and indeed more than that — as the flavorist must also

be knowledgeable about the processes of food production, the requirements of different

applications, consumer preferences and desires, as well as regulatory and legal

requirements. The creative flavorist was the indispensable, irreplaceable expert figure

who mediated among all of these different sources of information within the flavor

company, and between the company, its customers, and their (satisfied) consumers.

“Science and art are combined into an ideal marriage to give birth to good flavor

122
Vock 1981: 208-9.
438
creation,” Vock pronounced, somewhat non-idiomatically.123 The flavorist’s ultimate

obligation is to his or her company, “to make certain that the research dollars have been

rightfully spent.”124 The flavorist’s “success is due to this team effort,” Vock writes, and

in return, his labor is integral to the team’s cohesion and efficiency.125

The creative flavorist as the organizing, mediating force at the center of

the regulated, scientific flavor company. From Vock 1981: 199.

123
Vock 1981: 208.
124
Vock 1981: 198.
125
Vock 1981: 198.
439
Learning to Think Like a Flavorist

A 1957 article in The Givaudan Flavorist addressed the widespread curiosity

about what it takes to make flavors:

We are often asked to comment upon the basic talents, above and beyond
a knowledge of chemistry, which are necessary for success in the field of
flavor chemistry. We usually reply that there are two qualifications that all
good flavorists have in common. These are imagination and a thorough
knowledge of raw materials.126

How exactly did the flavorists' "imagination and thorough knowledge of raw

materials" operate in the real world? The author of the article gives an example. While

the "obnoxious" stink of a dead skunk by the side of the road would disgust "the average

consumer," the flavorist has a different reaction:

"The flavorist... sends an active imagination to work the minute an

aromatic material enters his nostrils and goes through that wonderful
process known as smelling. The aromatic ingredient usually recognized as
‘skunk’ is known to the flavorist to be... n-butyl mercaptan. A dilute
solution of this ingredient may be imagined as one of the missing nuances
in a coffee flavor he is working on.127

The successful flavorist is someone whose sensory capabilities are developed to a

degree beyond mere refinement; he (or she) bypasses disgust in favor of informed

126
"Sulfur and Aromatics," The Givaudan Flavorist 1957 4.4 (1957): 1.
127
“Sulfur and Aromatics” 1957: 1.
440
analysis (the "thorough knowledge" that permits the identification of the n-butyl

mercaptan in the skunk's stink) and productive synthesis (the "imagination" that

associates it with the roasted odor of coffee). But how does one become the kind of

person who thinks of coffee after smelling dead skunk? Or, to use another example from

the article, how do you cultivate the capacity to recognize a resemblance between the

noxious fumes drifting from an oil refinery, and "a roast loin of beef, a steak smothered

with onions, or a special blend of tobacco"?

Night School for Flavorists

Between 1946 and 1952, while working at the New York City Department of

Health, Jacobs taught several evening courses in food and flavor technology at the

Brooklyn Polytechnic Institute as an adjunct professor.128 His course, “Technology of

Food Flavors, Colors, and Synthetic Additives,” was first offered by the Department of

Chemical Engineering in the Spring semester of 1946. It was described as a professional

development course for graduate students in the department of chemistry and chemical

128
The 1948-1949 Polytechnic Institute Course Catalog lists him as lead professor of
several courses, all of which are indicated as being offered in alternate years: A year-long
course on Food Technology; A Fall semester course on the Technology of Dairy
Products; a Spring Semester course, “Technology of Food Flavors, Colors, and Synthetic
Additives”; a fall semester course on the technology and chemistry of “economic
poisons” (ie, insecticides, fumigants, fungicides, and pesticides); and a spring semester
course on the technology of alcoholic beverages. [Poly course catalog 1948-9, pp. 66-7.]
See also: “Food Coloring, Flavor, Part of Technology Course at Poly,” Brooklyn Daily
Eagle (January 6, 1946): 30; “Spring Course in Food Technology,” Brooklyn Daily Eagle
(January 12, 1947): 22; “Food Technology Courses Offered by Polytech,” Brooklyn
Daily Eagle (September 15, 1948); “Poly to Give Graduate Course on Brewing-Distilling
Skills,” Brooklyn Daily Eagle (January 28, 1950): 3.
441
engineering, as well as “well-equipped men from industry.”129 The course met for two

hours on Wednesday evenings, and covered the chemistry and use of food additives,

including natural and synthetic colors and flavors, synthetic sweeteners, emulsifiers,

stabilizers, preservatives, and vitamins. “There will also be taken up in detail,” the course

catalog read, “the compounding of synthetic flavors.”130 Jacobs’ teaching appointment

ended in 1952, and the course does not seem to have been renewed under a different

professor.131

In the early 1950s, NYU expanded the scope of its existing aromatics course —

which covered the industrial applications of aromatic chemicals for the perfume

industries — to include the creation and blending of flavors for the beverage,

confectionery, food, and tobacco industries. The semester-long evening class, offered

through NYU’s Division of General Education (the precursor to its present-day School of

Professional Studies) was “intended for persons engaged in the flavor and perfumery

industries, for users of such materials and for those interested in the art,” and included

hands-on work with aromatic raw materials in order to promote the “development of keen

olfactory perception and recognition” as well as how to use them.132 It also featured

frequent guest lectures from perfumers and flavorists working for regional companies

129
“Spring Course in Food Technology,” Brooklyn Daily Eagle (January 12, 1947): 22.
130
1948-9 Poly Course Catalog, Course number 2780, p.66.
131
The circumstances under which Jacobs lost his teaching position are somewhat
obscure, but a series of letters in the Othmer archives suggests that an ongoing dispute
between Jacobs, one of his graduate students, and Othmer over a method they had
developed to process orange and lemon oils may have contributed to this. [Donald
Othmer Papers, Chemical Heritage Foundation, Philadelphia]. The course was not
continued under a different professor after Jacobs was dismissed.
132
In Special Interest Courses, Division of General Education Bulletin for 1951-2.
“Aromatics: Perfume and Flavor Evaluation and Blending.”
442
including Givaudan, Polak & Schwarz, Norda, and Fries Brothers.133 Versions of many

of these lectures were published in the flavor section of the American Perfumer and

Aromatics, which remained under Jacobs’ editorship.

The New York City metropolitan area had long been, and remained, a center of

the flavor industry, with many companies headquartered in Manhattan, and maintaining

production facilities across the river in New Jersey, in the outer boroughs, and in Long

Island.134 Both of these evening courses seem to have served the career-development

needs of local flavor companies, providing introductory training to prospective or current

workers at a moment when the demand for flavor creation skills were particularly acute,

as the food industry’s use of specialty flavorings boomed. However, the failure of these

courses to flourish and persist, to spawn intermediate- and advanced-level classes or

departmental divisions, should not be taken as an indication of reduced demand for these

skills, but rather of alternate routes to their acquisition.

Rather than learning in the classroom, flavorists learned on the job. “There are no

text books or university courses in which this art and science are taught,” wrote Vock, of

IFF, in 1981. “Flavor creation is learned only in industry laboratories by working with

experts,” senior flavorists at the companies that employed them.135 Until at least the late

1980s, these apprenticeship relationships appear to have remained relatively informal,

133
“Aromatics in Food and Tobacco to be Included in NYU Course,” American Perfumer
56 (September 1950): 229; “Guest Speakers for NYU Course on Aromatics,” American
Perfumer 56 (November 1950): 399; “Guest Speakers for NYU Aromatics Course,”
American Perfumer 58 (November 1951): 383. The conductor of the course was Samuel
Klein, consultant perfumer.
134
Rogers and Dorland 1977: 171-240.
135
Vock 1981: 197.
443
that is, designed and operated under the discretion of senior flavorists, rather than

formalized by management as institutional protocols.136

It was this tacit, embodied, sensual knowledge of materials, after all, that defined

the jurisdiction of the flavorist, and distinguishes it from that of the research chemist.

This required not only mastery of a technical curriculum of chemical knowledge and

olfactory acuity, but also the acquisition of a particular attitude towards chemical

materials: an open-minded interestedness and curiosity.

“What Does it Remind you Of?” The Flavorist-Apprentice

Is the flavorist born or made? Do certain individuals have exceptional sensory

capacities, and excel beyond others in the field? And can standard training methods

produce creative professionals?

136
Information about this aspect of flavor industry operations is rather thin in the
published records. However, the protocols for training flavorists apparently contrast with
programs for training perfumers, who, like flavorists, also learn the work on the job. In
the 1970s, several major European companies operated their own schools for perfumers
— often in Grasse, the long-standing center of the French perfumery and essential oil
business. American companies IFF and Monsanto Flavor/Essence are also reported to
have operated training programs for perfumers, which combined formal laboratory
training with on-the-job experience. (Dorland and Rogers 1977: 397-404.) I have found
no mention of similarly formal, organized programs for training flavorists during this
period. At some point, likely in the late 1980s, the Society of Flavor Chemists formalized
a certification program for flavorists. Currently, to become a certified flavorist, an
individual must apprentice for seven years with a senior certified flavorist, pass an
examination and interview, and finally, be voted upon by SFC members. Certified
members of the SFC can sponsor and train apprentice members.
https://flavorchemists.com/become-a-member
444
 In order to consider these questions, I will examine in detail two brief accounts of

programs to train flavorists presented in 1974 at the Society of Flavor Chemists'

Twentieth Anniversary Symposium, "The Multifaceted Nature of the Flavor Chemist,"

held at Rutgers University in New Brunswick.137 Harris Shore, a flavor consultant for

Fries & Fries, opened the symposium with a description of the "one on one" mentorship

relationship between apprentice and master flavorist — preferable, he claims, to batch-

processing prospective flavorists in groups via more schematized curricula.138 In

contrast, Frank Fischetti followed Shore’s account of the relationship between mentor

and apprentice with a description of the training program that he superintends at FD&O:

a schematized, “organized program,” complete with sample quizzes and creativity-

boosting games.139 Despite this key difference, Shore and Fischetti's training methods

reveal a similar set of needs, concerns, and challenges. I will compare these with two

other accounts – the first by E. Cowley of the British flavor firm Bush Boake Allen, Ltd.,

137
These papers were compiled in a post-symposium publication, which is my source for
them: Society of Flavor Chemists, “The Multifaceted Nature of the Flavorist: Papers
Presented by the Society of Flavor Chemists Symposium Held at Rutgers University,
New Brunswick, New Jersey, March 21, 1974,” Society of Flavor Chemists Collection,
Chemical Heritage Foundation, Philadelphia. The papers were also reprinted in the
July/August, September/October, and November/December 1974 issues of The Flavour
Industry, a British trade journal, and these texts were bound together in a separate
booklet. These printed texts likely varied somewhat from the content of what was
delivered at the Symposium – several, for instance, include footnotes to research articles.
With that caveat, I am taking them as generally reflective of the day’s program. Frank
Fischetti's program was also described in Earl J. Merwin, ed., The Development and
Application of Natural and Artificial Flavor Systems, (Wheaton, IL: Allured, 1988).
138
Harris Shore, “The Training of a Flavorist – One On One,” Society of Flavor
Chemists 1974: 2-3.
139
Frank Fischetti, Jr. “The Training of a Flavor Chemist – An Organized Programme,”
Society of Flavor Chemists 1974: 4-6.
445
published in The Flavour Industry in 1973,140 and the second by Agusti Vidal, of the

Spanish flavor and fragrance house Lucta SA, which ran in Perfumer & Flavorist in

1989.141 Because these are foreign firms, and factors related to personnel and business

structure may vary from the American flavor houses, I will draw on these sources only to

point out areas of similarity with the other accounts – which may indicate common trends

– or particularly suggestive differences.

Although Shore hints at a preference for sensory acuity, both he and Fischetti are

explicit that anyone with “normal” taste and smell can be a potential flavorist. Shore

somewhat apologetically excludes those with chronic sinusitis and allergies from the

candidate pool. Fischetti assumes that the training program will screen out those who

really do not have adequate sensitivity, aptitude, or desire to pursue the field, but

consoles listeners that these unfortunates generally opt out on their own and pursue other

opportunities within the company. Cowley, of Bush Boake Allen, indicates that

candidates are screened for “flavor blindness.”

It is expected that candidates will have some basic knowledge of the sciences, but

no advanced degree is specified. Indeed, Fischetti says, “we like a minimum of two years

of college, preferably in chemistry.” In his suggested training program, the first stage

offers students a basic outline of organic chemistry – with a focus on reactions pertaining

to flavor. Cowley looks for candidates with high ratings in both the natural sciences and

140
E. Cowley, "The Training of a Flavourist," Flavour Industry 4 (January 1973): 18-20.
Reproduced in Wayne E. Dorland, The Fragrance and Flavor Industry (Mendham, N.J:
W. E. Dorland Co, 1977): 417-20.
141
Agusti Vidal, "New Comprehensive Training Method for Perfumery and Flavoring,"
Perfumer & Flavorist 14:2 (March/April 1989): 25-44.
446
the visual arts. Vidal, of the Spanish firm Lucta, deliberately includes not only

prospective flavorists and perfumers in the training program, "but [also] everyone who

interacts with the creative team" – including marketing, sales, and purchasing personnel –

in order to ensure clear communication within the company and with customers.142 The

inclusiveness of Lucta's program highlights an issue that underlies all the programs: the

lack of a standard vocabulary with which to discuss and describe the sensory experiences

of flavor and fragrance.

A primary task of these programs, then, is to introduce prospective flavorists to

categories that they can use to describe the aromatic universe in a comprehensive and

comprehensible way. Although Shore does not mention a specific rubric under which

flavor components are presented or taught to the trainee, Fischetti and Cowley specify

that they begin by introducing families of natural essences – essential oils – and move

from there to synthetic compounds. Both justify this as a way of leading trainees from the

familiar to the unfamiliar. That is, apprentices are first taken through the steps of

understanding chemicals in terms of attributes or associations – fruity, lemony, summery

– before then moving on to chemical structures and families – aldehyde, terpene. Lucta's

method operates in a similar manner, dividing the universe of fragrance and flavoring

materials into twenty-five categories, and then taking students from a comparison of a

natural product with its most significant nature-identical synthetic chemical components,

and then proceeding to the artificial chemicals that complement or replace it.143 These

categories permit prospective flavorists to make comprehensible distinctions among

142
Vidal 1989: 25.
143
Lucta 1989: 26.
447
materials in the natural world, but they must also permit analysis and recombination of

sensory properties of the experience of these materials – something that Fischetti

underscores. He describes training prospective flavorists to break "whole" flavors into

constituent parts, and to bring those parts together to form new wholes. For instance, a

flavorist may be asked to take a strawberry flavor and make it greener, buttery, estery, or

jammy –breaking down the barriers between discrete flavor families.

Recognition of flavor families depends upon the acuity of sensory memory.

Shore, Fischetti, and Cowley discuss the challenge of helping the aspiring flavorist to

build reliable mnemonic techniques. Both Shore and Cowley use the image of a library –

a library of tastes and smells – that the flavorist can access to make precise

identifications.144 Fischetti lists “enhanc[ing] the technician’s flavor memory through the

use of mnemonic devices" as one of the explicit goals of his program.145 He trains the

technician-flavorist to build associative relationships between odor and experience. For

instance, he suggests presenting the trainee with samples of essential oils and asking

questions such as: “What oil reminds you of a dentist’s office, sausage, lemon peel, Vicks

Vapo-Rub, pizza, chili?” These referential experiences are selected because they are

common among the potentially diverse group of trainees. Indeed, “these were the very

descriptions the technicians themselves used to describe the oils. What we are attempting

to do is reinforce these descriptions in their minds.”146 Much of Fischetti’s program

seems to involve learning to think of a flavor component in all of its potential

144
Cowley 1973, in Dorland 1977: 418; Shorr 1974: 2.
145
Fischetti 1974: 5.
146
Fischetti 1974: 5.
448
manifestations – as a component of different kinds of flavoring compounds – and

associations. To give just one other example from a program rich with them, he mentions

a game called “Who do you remind me of?” where the technician is given an aromatic to

sniff that is a constituent of an essential oil. The technician must name the oil as well as

other aromatics that might be confused with the sample.

Significantly, Shore, Fischetti, and Cowley place strong emphasis on the need to

cultivate habits of creativity and imagination, associative reasoning rather than dogmatic

thinking. Shore assigns this responsibility to the teacher, suggesting rather vaguely that

the empirical methods of the past be modulated with structured knowledge: “His teacher

will inculcate in him a blend of logical thinking and the skill of sophisticated artistry.”147

However, Cowley cautions that “too much instruction” in the advanced stages of the

training regimen “can stultify the imagination. We are trying to produce a creative

individual who will produce original concepts, and if the flavorist is directed too forcibly

into another person’s channel of thinking, then we may defeat the objective we are trying

to achieve.”148 In other words, in order for creative play to be possible and successful, a

structured program must give way to a more open-ended framework once the trainee has

reached a critical level of material mastery.

In addition to advocating for ample free experimentation time in the course of the

training program, Fischetti confronts this as a challenge: "How do you teach [the trainee]

to be creative? One doesn’t teach creativity really…you foster it…you set up the

147
Shore 1974: 2
148
Cowley 1973, in Dorland 1977: 419.
449
environment, you give him a minimum knowledge, you suggest ways to remove the

cultural, emotional or perceptual blocks he may have and finally you encourage him to

create. Creativity is not only an ability, but a pattern of behavior. How do we set up this

pattern of behavior? We play games."149

Fischetti then discusses several training games at length. In his program, for

instance, the flavor-chemist-in-training is invited to "go to the shelf and pick up any

bottle he chooses and let his mind wander. He is told to write what flavors he thinks it

could be used in. We do not ask him what it is, but rather how many flavor uses he can

think up for this material.... He is never criticized for his suggestions... We want to

encourage a large number of ideas."150

Free and structured play, habits of daydream, associative thinking, empirical

experimentation with materials, a lengthy, somewhat open-ended training process: all of

these seem ill-suited to the needs of flavor and fragrance manufacturers, which are, after

all, commercial enterprises that need to be able to reliably produce dependable products

on a schedule determined by clients. As an attempt to explain how this discrepancy was

managed, I would like to draw attention to something that is largely absent from these

accounts: instrumental technologies. Although Shore and Fischetti mention

instrumentation in passing as a possible later stage of the training program, operation and

149
Fischetti 1974: 5.
150
Fischetti 1974: 5-6.
450
use of these machines, especially GC, was far from simple or self-evident, and they were

an increasingly fundamental part of the process of flavor creation.151

Indeed, their use of machines was where flavorist’s artistic skill, experienced

judgment, and imagination was most evident – the qualities that are at the core of the

training program. But this emphasis on craft, and the allegiance between the ultimately

commercial work of the flavorist and that of fine artists, is not just a professional

necessity, but also a source of professional pride and identity. In "The Flavorist as an

Artist," his address to the SFC’s twentieth anniversary symposium, Jerry Di Genova, the

head of Givaudan's flavor laboratories, describes the work of the flavorist in light of the

simulations that attempted to reproduce nature on a molecular level:

A flavor simulation of a natural product that is composed with the

qualitative and quantitative ingredients only as found in that product will
be nothing else but a comparatively crude simulation and often unlike the
natural counterpart. Knowing this, the flavorist/artist makes his
modifications to arrive at the desired effect. He must be as exact as a
scientist, but, more importantly, as flexible as an artist.152

151
See, for instance, V.D. Johnston, "Instrumental Methods of Analyses Save Time --
Give More Information," The Givaudan Flavorist 1959 (1), 2; Richard H. Potter, "Vapor
Phase Chromatography as a Tool in Flavor Creation," The Givaudan Flavorist 1963 (1),
5-6; Potter, "Further Thoughts on the Use of Vapor Phase Chromatography in Flavor
Creation," The Givaudan Flavorist 1963 (4), 5-6; and Potter, “Gas Chromatography – A
Flavorist’s Tool,” in Society of Flavor Chemists 1974: 18-19.
152
Jerry Di Genova, "The Flavorist as an Artist," in Society of Flavor Chemists 1974: 10-
11.
451
 He then described the task of flavor chemistry in explicitly synesthetic terms –

emphasizing associations with color and sound, and drawing an extended metaphor with

painting, where flavor chemicals are the colors on a flavorist's palette. Ultimately, the

flavor chemist combines art and science by means of technology, using “the latest

scientific knowledge" to "build... a rough flavor frame," and then "as a true artist, build[s]

around it the desired notes, nuances and effects which the instrument has either failed to

deliver or the researcher has failed to identify.”153

The flavor chemist, then, must be taught to recreate those effects of nature that

cannot be fully objectified and quantified – that resist systematization.

Conclusion: The Virtues of Flavor Creation in a Resource-

Depleted World

The November 23, 1962 edition of Life magazine was a special issue, celebrating

the “Bounty of Food.” The cover featured apples, grapes, broccoli, artichokes, and other

fruits and vegetables heaped against a jet-black background suggestive of a depthless

void, and promised articles detailing “Secrets of Taste… $50 Billion Spectacle… Harvest

Splendor… And other stories on the miracle of our plenty.”154

But in the midst of all this domestic abundance, loomed the epic and impending

fact of future hunger: the Malthusian crisis of resources that seemed to be once again on

153
Di Genova 1974: 10.
154
Life, (November 23, 1962).
452
the horizon, threatening American imperium and global stability.155 Postwar America’s

global perspective had laid bare the scale of the problem; it was estimated that “half of

the world’s population still lives under circumstances where enough food to prevent old-

fashioned hunger is a problem of high priority,” a problem that population growth

threatened to exacerbate.156 Solutions for this world problem were sought in food

technology and synthetic chemistry.

Life mentioned “a variety of bizarre solutions,” currently being studied, “each

with a distinct flavor problem.” Soybeans, petroleum, cottonseed cake, farmed chlorella

algae, and other substances were considered potential raw materials for the manufacture

of macronutrients, especially protein, but these substances often carried off-flavors and

odors that rendered them unpalatable. Other alternative sources of calories and

macronutrients were valued for their absence of qualities. For instance, the US Bureau of

Fisheries was developing a process to manufacture a fish protein concentrate (FPC) from

bycatch (“trash fish” that made up about half of fisheries’ harvest’), collateral life that

had “no commercial food value” and was routinely thrown back into the ocean for the

gulls to feast upon.157 The Bureau of Fisheries’ process transformed “the entire fish,

scales and all, into a powder that is tasteless, odorless, chemically pure and rich in vital

155
The perennial reappearance of Malthusian rhetoric, and its consequences for culture
and policy, are discussed in Warren Belasco, Meals to Come: A History of the Future of
Food, (Berkeley: University of California Press, 2006). See also: Warren Belasco, “Algae
Burgers for a Hungry World? The Rise and Fall of Chlorella Cuisine,” Technology and
Culture 38.4 (July 1997): 608-34.
156
C.G. King, “Nutrition in Relation to Flavor and World Food Acceptance,” in David E.
Schafer, ed. Proceedings: Flavor Chemistry Symposium – 1961, (Camden, NJ: Campbell
Soup Company, 1961): 102. King was the head of The Nutrition Foundation.
157
Warren R. Young, “Cracking the Secret Riddle of Flavor,” Life, (November 23, 1962):
120; “A Miracle of the Fishes,” Life (June 29, 1962): 33-4.
453
animal proteins.”158 At a cost of half a cent per person per day, this was a dirt-cheap

protein source; if processed into FPC, the bycatch from U.S. coastal waters alone could

make up for seventy-five percent of the global protein deficit. “The attractive feature of

this diet-fortifier” was its absence of sensory qualities; “its flavor can be adjusted to suit

local palates,” noted Life.159 The new vanguard of global reformers were increasingly

aware that in developing products for global food aid, “one has to be very sensitive to

their flavor traditions, not our flavor traditions.”160

James McGlumphy of IFF, writing in 1966, also emphasized the increasing need

for flavor. Efforts to increase food production by the use of chemical fertilizers, hybrid

cultivars, and improved farming methods, often came at a cost: flavor. An increase in

agricultural yields was “almost always… paired with a decrease in natural flavor levels”

— a phenomenon McGlumphy referred to as nature’s “contrary streak.”161

158
“A Miracle of the Fishes” 1962: 33. Tellingly, the challenge that stood in the way of
fully developing this food source (as presented by this article) was not technological but
bureaucratic. The FDA’s food standards rejected the inclusion of heads, scales, and
entrails in FPC, on the grounds that consumers “would regard the product… as filthy.”
The agency thus forbade it from being sold as food in the US. But if the fish were
required to be cleaned before processing, FPC would be prohibitively expensive.
Although the FDA ruling only applied domestically, there were concerns that shipping
FPC abroad as food aid under the FDA’s domestic prohibition would invite Soviet
criticism: “The Russians could say, ‘See, the Americans are sending you food they
consider too filthy to eat themselves.’”
159
Young 1962: 120.
160
King 1961: 105.
161
James McGlumphy, “Progress in Flavor Research,” in “Flavor: Reflections and
Directions,” a report from the Flavor Update Symposium (November 16, 1965) at MIT,
sponsored by the Northeast Section of the IFT. Published in Food Technology (December
1966): 48-50.
454
 If food technologies and advances in agricultural science could produce solutions

to the problem of scarcity, could stave off global crisis, they apparently did so at the cost

of flavor. But science and technology provide the solution here too: flavor additives. But

then again, scenarios, scientific and technical knowledge was not enough. “The skill of

the flavorist is in greater need than ever before,” McGlumphy wrote.162

Even as a political and cultural establishment turned against the synthetic,

increasingly favoring the “natural”, flavorists (and the industries they labored for)

attached themselves to the broader purpose of world salvation. An October 1949 editorial

in the American Perfumer & Essential Oil Review explained:

At recent international scientific meetings the growth of world population

has caused considerable speculation about the available food supply. It has
been suggested that there is a vast amount of food, other than fish, [ie,
algae] in the oceans which should be useful for amplifying the world food
supply. This will be a challenge to the flavorist, for it will be his function
to make such potential food sources available food sources by making
them palatable.163

A few months later, David Lakritz, flavorist at Florasynth Laboratories in

Brooklyn, restated this professional goal in an article in Drug and Cosmetic Industry:

“Because of the ever increasing world population with a consequent drain on the

162
McGlumphy 1966: 50.
163
“Flavored Notes,” American Perfumer (October 1949): 304.
455
available food supply, it will undoubtedly be the function of the food flavoring chemist to

make flavorsome large potential amounts of wholesome but unpalatable food.”164

This purpose is restated repeatedly in publications by flavorists, for decades. For

instance, in 1971, FD&O’s Eiserle and Downey, calling for fundamental flavor research

more attuned to the purposes of synthetic flavor production, wrap up their plea: “In the

future, it will help us accomplish our industry’s main purpose, namely, to prepare

synthetic flavors, reproducing the type of flavor found in those food products which

undoubtedly will be in short supply in the future.”165

But the long-anticipated seven lean years never arrived, at least not in the

industrialized nations of the West where the products of flavorists’ creative labor were

most often consumed. Instead, flavor and food science were deployed in a landscape of

continued caloric abundance. The meanings of this scenario, the consequences of the

imagination of future scarcity for the ways that foods were made to taste, will be one of

the subjects of my continuing work in this field.

164
David E. Lakritz, “Development of Flavors,” Drug and Cosmetic Industry 65
(December 1949): 724.
165
Eiserle and Downey 1971: 169.
456
 BIBLIOGRAPHY

Archives and Manuscript Collections

Hagley Museum and Library. Wilmington, DE.

Arthur D. Little Collection. MIT Institute Archives and Special Collections. Cambridge, MA.

A.W. Noling Beverage Literature Collection. Special Collections, University of California-Davis.

Davis, CA.

Othmer Library and Special Collections. Chemical Heritage Foundation. Philadelphia, PA.

Records of the Food & Drug Administration, Record Group 88; U.S. National Archives at
College Park, MD.

Smithsonian Trade Literature Collection. National Museum of American History LIbrary.

Washington D.C.

Society of Flavor Chemists Collection. Monell Chemical Senses Center. [Now at Chemical
Heritage Foundation/Science History Institute]

Books and Journal Articles

“50 Years of Service by Van Dyk & Co.,” American Perfumer and Essential Oil Review
63 (June 1954): 453-5.

Abbott, Andrew. The System of Professions: An Essay on the Division of Expert Labor.
Chicago: University of Chicago Press, 1988.
457
“Advances in Spray Drying Improve Flavor Quality.” Food Engineering 28 (February
1956): 77-8.

“Aerosols and the Food Industry, Part I: Beverages.” The Givaudan Flavorist 5.1 (1957):
1-4.

Aftel, Mandy. Essence and Alchemy: A History of Perfume. New York: North Point
Press, 2001.

Alder, Ken. “A Social History of Untruth: Lie Detection and Trust in Twentieth-Century
America,” Representations 80.1 (Fall 2002): 1-33.

American Society for Testing and Materials. Correlation of Subjective-Objective

Methods in the Study of Odors and Taste, a Symposium Presented at the 70th Annual
Meeting, ASTM, Boston, MA, 25-30, June 1967. ASTM Special Technical Publication
440. Philadelphia: ASTM, 1968.

Amerine, Maynard, Rose Marie Pangborn, and Edward Roessler. Principles of Sensory
Evaluation of Food. New York: Academic Press, 1965.

Anderson, Earl V. “Food Aerosols.” Chemical & Engineering News (May 2, 1966): 99-
101.

Armed Forces Food and Container Institute. Monosodium Glutamate: A Second

Symposium, Held at Palmer House, Chicago, June 7, 1955. Chicago: Research and
Development Associates & Food and Container Institute, 1955.

“Art in Imitation Flavors: The Aromatic Constituents of Strawberry.” Givaudan Flavorist

1.2 (1953): 1-4.

458
“Art vs. Nature.” The American Journal of Pharmacy and the Sciences, (April 1852):
184-5.

Arthur D. Little, Inc. Flavor Research and Food Acceptance. New York: Reinhold, 1958.

———. Symposium on Flavor Potentiation. Cambridge: ADL, 1964.

——–. “The U.S. Flavor Industry: Report to The Andrew Jergens Company,” ADL
Report C-69866 (March 26, 1968). A.W. Noling Collection, UC Davis.

Ashworth, William J. “‘Between the Trader and the Public’: British Alcohol Standards
and the Proof of Good Governance.” Technology and Culture 42 (January 2001): 27-50.

Auerbach, Jeffrey. The Great Exhibition of 1851: A Nation on Display. New Haven: Yale
UP, 1999.

Austern, H. Thomas. “Food Standards: The Balance Between Certainty and Innovation.”
Food, Drug, and Cosmetic Law Journal 24 (1969): 440-55.

Baab, Bill. “TIP Baby’s Father Shares Gary Beverage Co. History,” The Federation of
Historical Bottle Collectors Newsletter (July-Aug 2007).

Bachelard, Gaston. The New Scientific Spirit. Translated by Arthur Goldhammer. Boston:
Beacon Press, 1984 [1934].

Backer, Kellen. World War II and the Triumph of Industrialized Food. PhD Diss,
University of Wisconsin-Madison, 2012.

Baird, Davis. “Analytical Chemistry and the ‘Big’ Scientific Instrumentation

Revolution.” Annals of Science 50.3 (1993): 267-90.
459
———. “Encapsulating Knowledge: The Direct Reading Spectrometer,” Foundations of
Chemistry 2 (2000): 5-46.

Barley, Stephen R. "Technicians in the Workplace: Ethnographic Evidence for Bringing

Work Into Organizational Studies." Administrative Science Quarterly 41.3 (1996).

Barley, Stephen R. and Julian E. Orr, eds. Between Craft and Science: Technical Work in
U.S. Settings. Ithaca: IRL Press, 1997.

Bartle, Keith D. and Peter Myers. “History of Gas Chromatography.” Trends in Analytic
Chemistry 21.9-10 (2002): 547-57.

Bastick, W. “Artificial Essence of Pineapple.” In G.W. Septimus Piesse, The Art of

Perfumery…2nd American ed., 384-5. Philadelphia: Lindsay & Blakiston, 1867.

Bedoukian, Paul Z. “The Perfumery Aromatics Industry in the United States, Part I.”
American Perfumer and Aromatics (November 1957): 33-6., (December 1957): 31-5,
(January 1958): 43-50

———. “The Perfumery Aromatics Industry in the United States, Part II.” American
Perfumer and Aromatics (December 1957): 31-5.

——–. “The Perfumery Aromatics Industry in the United States, Part III.” American
Perfumer and Aromatics (January 1958): 43-50.

Belasco, Warren. “Ethnic Fast Foods: The Corporate Melting Pot.” Food and Foodways
2 (1987): 1-30.

———. “Algae Burgers for a Hungry World? The Rise and Fall of Chlorella Cuisine,”
Technology and Culture 38.4 (July 1997): 608-34.
460
———. Meals to Come: A History of the Future of Food. Berkeley: University of
California Press, 2006.

Belasco, Warren and Roger Horowitz, eds. Food Chains: From Farmyard to Shopping
Cart. Philadelphia: UPenn Press, 2009.

Belasco, Warren and Philip Scranton, eds. Food Nations: Selling Taste in Consumer
Societies. New York: Routledge, 2002.

Bensaude-Vincent, Bernadette and Isabelle Stengers. A History of Chemistry. Translated

by Deborah van Dam. Cambridge: Harvard UP, 1996.

Bentley, Amy. Eating for Victory: Food Rationing and the Politics of Domesticity.
Urbana: University of Illinois Press, 1998.

———. Inventing Baby Food: Taste, Health, and the Industrialization of the American
Diet. Oakland: University of California Press, 2014.

Berenstein, Nadia. “Making a Global Sensation: Vanilla Flavor, Synthetic Chemistry, and
the Meanings of Purity.” History of Science 54.4 (December 2016): 399-424

———. “The History of Banana Flavoring.” Lucky Peach (August 2016).

http://luckypeach.com/the-history-of-banana-flavoring/

Besky, Sarah. “The Future of Price: Communicative Infrastructures and the

Financialization of Indian Tea.” Cultural Anthropology 31.1 (February 2016): 4-29.

Bijker, Wiebe E., Thomas Parke Hughes, and Trevor Pinch. The Social Construction of
Technological Systems: New Directions in the Sociology and History of Technology.
Cambridge: MIT Press, 1987.
461
Bix, Amy Sue. “Equipped for Life: Gendered Technical training and Consumerism in
Home Economics, 1920-1980.” Technology and Culture 43.4 (October 2002): 728-54.

Blaszczyk, Regina Lee. Imagining Consumers: Design and Innovation from Wedgwood
to Corning. Baltimore: JHU Press, 2000.

———. The Color Revolution. Cambridge: MIT Press, 2012.

Blumenthal, Saul. Food Products. Brooklyn: Chemical Publishing Company, 1947.

Boggs, Mildred M. and Helen L. Hanson. "Analysis of Foods by Sensory Difference

Tests." Advances in Food Research 2 (New York: Academic Press, 1949): 219-258.

Borges, Jorge Luis. Labyrinths: Selected Stories and Other Writings. Edited by Donald
A. Yates and James E. Irby. New York: New Directions, 1964.

Borgerson, Janet and Jonathan Schroeder. “The Lure of Paradise: Marketing the Retro-
Escape of Hawaii.” In Stephen Brown and John F. Sherry, Jr., eds. Time, Space, and the
Market: Retroscapes Rising, 219-36. London and NY: Routledge, 2003.

Boring, Edward G. Sensation and Perception in the History of Experimental Psychology.

New York: Appleton-Century-Crofts, 1942.

Bourdieu, Jerome, Martin Breugel, and Peter Atkins. "'That Elusive Feature of Food
Consumption:' Historical Perspectives on Food Quality, a Review and Some Proposals,"
Food & History 5.2 (2007): 247-266.

Bourdeiu, Pierre. Distinction: A Social Critique of the Judgment of Taste. Translated by

Richard Nice. Cambridge: Harvard UP, 1984.

462
Boyd, William. "Making Meat: Science, Technology, and American Poultry Production."
Technology and Culture 42. 4 (October 2001): 631-664.

Boyce, Angie M. “’When Does It Stop Being Peanut Butter?’: FDA Food Standards of
Identity, Ruth Desmond, and the Shifting Politics of Consumer Activism, 1960s-1970s.”
Technology and Culture 57.1 (2016): 54-79.

Braverman, Harry. Labor and Monopoly Capital; the Degradation of Work in the
Twentieth Century. New York: Monthly Review Press, 1975.

Briot, Eugenie. “From Industry to Luxury: French Perfume in the Nineteenth Century,”
Business History Review (Summer 2011): 273-294.

Broderick, James J. “Blazing the Trail to… Superior Powdered Flavors.” Food
Engineering (November 1954): 83-4.

———. “Fruit Flavor Research: The Practical Flavorist vs. the Basic Researcher.” Food
Technology 26 (November 1972): 37, 48.

———. “Reflections of a Retired Flavorist Before He Forgets: Raspberry.” Perfumer &

Flavorist 16 (Nov/Dec 1991): 13-14.

———. “Reflections of a Retired Flavorist Before He Forgets: Peach.” Perfumer &

Flavorist 17.1 (Jan/Feb 1992): 35.

———. “Reflections of a Retired Flavorist Before He Forgets: Strawberry.” Perfumer &

Flavorist 17.3 (May/June 1992): 33-4.

463
Brooks, Benjamin T., Dillon F. Smith, and Harry Essex. “The Manufacture of Amyl
Acetate and Similar Solvents from Petroleum Pentane.” Journal of Industrial and
Engineering Chemistry 10.7 (July 1918): 511-15.

Buck, Elizabeth. Paradise Remade: The Politics of Culture and History in Hawaii.
Philadelphia: Temple UP, 1993.

Bureau of Human Nutrition and Home Economics, Agricultural Research

Administration. Sensory Methods for Measuring Differences in Food Quality: Review of
Literature and Proceedings of Conference. Agriculture Information Bulletin No. 34.
Washington, D.C.: U.S. Department of Agriculture, August 1951.

Buttery, Ron G. “Autoxidation of Potato Granules II: Formation of Carbonyls and

Hydrocarbons.” Agricultural and Food Chemistry 9 (1961): 248-52.

Buttery, Ron G. and Roy Teranishi. “Gas-Liquid Chromatography of Aroma of

Vegetables and Fruit: Direct Injection of Aqueous Vapors.” Analytic Chemistry 33.10
(September 1961): 1439-41.

———. “Measurement of Fat Autoxidation and Browning Aldehydes in Food Vapors by

Direct Vapor Injection Gas-Liquid Chromatography.” Agricultural and Food Chemistry
11.6 (Nov-Dec 1963): 504-7.

Cairncross, S.E. and L.B. Sjöström. “What Glutamate Does in Food.” Food Industries 20
(July 1948): 76-7, 200-1.

———. “Flavor Profiles: A New Approach to Flavor Problems.” Food Technology 4

(1950): 308-311.

Cartwright, L.C. and P.H. Kelley. “Sharper Flavor Ratings with Improved Profile Test.”
Food Engineering 23 (September 1951): 71-3, 215.

464
Cartwright, L.C. and Robert A. Nanz. "Flavors Improved, Sales Boosted Through
Organoleptic Tests." Food Industries 20. 11 (November 1948): 1608-9.

Caul, Jean F. “The Profile Method of Flavor Analysis.” In E.M. Mrak and G.F. Stewart,
eds., Advances in Food Research 7, 1-40. New York: Academic Press, 1957.

Caul, Jean F., S.E. Cairncross, and L.B. Sjöström. “The Flavour Profile in Review.”
Perfumer and Essential Oil Review 49 (March 1958): 130-3.

Challenger, Bernice. "The Art of Blending and Its Application in the Bottling of
Carbonated Beverages." Beverage Journal 58.2 (Feb 1922): 91-3.

Chandler, Alfred D. Scale and Scope: The Dynamics of Industrial Capitalism.

Cambridge: Belknap, 1990.

——–. Shaping the Industrial Century: The Remarkable Story of the Evolution of the
Modern Chemical and Pharmaceutical Industries. Cambridge: Harvard UP, 2005.

Chase, A.W. Dr. Chase’s Recipes, or, Information for Everybody: An Invaluable
Collection of About Six Hundred Practical Recipes… 8th edition. Ann Arbor, MI: A.W.
Chase, 1860.

Clarke, Adele and Joan Fujimura, eds., The Right Tools for the Job: At Work in
Twentieth-Century Life Sciences. Princeton: Princeton UP, 1992.

Classen, Constance, David Howes, and Anthony Synnott, “Artificial Flavors.” In The
Taste Culture Reader: Experiencing Food and Drink, edited by Carolyn Korsmeyer, 337-
42. Oxford and New York: Berg, 2005 [2007].

Cogdell, Christina. Eugenic Design: Streamlining America in the 1930s. Philadelphia:

UPenn Press, 2010.

465
Cohen, Benjamin R. “Analysis as Border Patrol: Chemists Along the Boundary Between
Pure Food and Real Adulteration,” Endeavor 35 (2011): 66-73.

Cohen-Cole, Jamie. The Open Mind: Cold War Politics & The Sciences of Human
Nature. (Chicago: University of Chicago Press, 2016).

Cohen, Lizabeth. A Consumer’s Republic: The Politics of Mass Consumption in Postwar

America. New York: Knopf, 2003.

Cohen, Harry. “Foods, Flavors, and Aromatics.” American Perfumer 56 (July 1950): 47-
49.

Collins, H.M. “The TEA set: Tacit knowledge and Scientific Networks.” Social Studies
of Science 4.2 (1974): 165-185.

[J.W. Colton Co.] Centennial Cookbook: J.W. Colton’s Choice Cooking Recipes,
Preparation, and Calendar for 1876-1877, [pamphlet], (Westfield, MA: J.W. Colton Co.,
1876). Courtesy Alfred Goossens.

Coon, Deborah J. "Standardizing the Subject: Experimental Psychologists, Introspection,

and the Quest for a Technoscientific Ideal." Technology and Culture 34.4 (October
1993): 757-783.

Corbin, Alain. The Foul and the Fragrant: Odor and the French Social Imagination.
Translated by Miriam L. Kochan, Roy Porter, and Christopher Prendergast. Cambridge:
Harvard UP, 1986.

Cover, Sylvia. "A New Subjective Method of Testing Tenderness in Meat -- The Paired-
Eating Method." Food Research 1.3 (May-June 1936): 287-295.

466
———. “Some Modifications of the Paired-Eating Method in Meat Cookery Research.”
Food Research 5 (1940): 379-94.

Cowan, Donald R.G. "Developing and Improving Foods by Consumer Testing," Food
Industries (March 1941): 41-44.

Cowan, Ruth Schwartz. More Work for Mother: The Ironies of Household Technologies
from the Open Hearth to the Microwave. New York: Basic Books, 1983.

———. “The Consumption Junction: A Proposal for Research Strategies in the

Sociology of Technology.” In Wiebe E. Bijker, Thomas P. Hughes, and Trevor J. Pinch,
The Social Construction of Technological Systems: New Directions in the Sociology and
History of Technology, 261-80. Cambridge: MIT Press, 1987.

Cowell, Andrew. “The Apocalypse of Paradise and the Salvation of the West: Nightmare
Visions of the Future in the Pacific Eden.” Journal of Cultural Studies 13.1 (1999): 138-
60.

Coy, Ruth. “Regional Research Centers: The First Half Century, 1940-1990.” In 1990
Yearbook of Agriculture, 121-4. Washington, D.C.: USDA, 1990.

Cravens, Hamilton. “Establishing the Science of Nutrition at the USDA: Ellen Swallow
Richards and Her Allies.” Agricultural History 64.2 (Spring 1990): 122-33.

Crocker, E.C. “Measuring Food Flavors.” Food Research 2.3 (1937): 273-86.

Crocker, E.C. and Washington Platt. "Editorial Review. Food Flavors -- A Critical
Review of Recent Literature." Food Research 2.3 (1937): 183-96.

Cross, Gary S. and Robert N. Proctor. Packaged Pleasures: How Technology and
Marketing Revolutionized Desire. Chicago: University of Chicago Press, 2014.

467
Daston, Lorraine and Peter Galison. Objectivity. Cambridge: Zone Books, 2007.

Daston, Lorraine and Fernando Vidal, eds. The Moral Authority of Nature. Chicago:
University of Chicago Press, 2004.

de Nicolai, Patricia. “A Smelling Trip Into the Past: The Influence of Synthetic Materials
on the History of Perfumery.” Chemistry and Biodiversity 5 (2008): 1137-1146.

De Groote, Melvin. “The Selection of Extracts for Carbonated Beverages.” The Beverage
Journal 58.3 (March 1922): 52-3.

———. “Chemical Research in Beverage Ingredients,” Beverage Journal 58. 5 (July

1922): 50a-h.

Deutsch, Tracy. Building a Housewife’s Paradise: Gender, Politics, and American

Grocery Stores in the Twentieth Century. Chapel Hill: UNC Press, 2010.

Dickau, Joel. “Inventing Texture: Edible Science and the Management of Familiarity,
1963-1975,” Global Food History 3 (2017): 1-23.

Di Genova, Jerry. “The Flavorist as an Artist.” In Society of Flavor Chemists, The

Multifaceted Nature of the Flavorist, 10-11. [1974]

———. “Food Aerosols.” The Givaudan Flavorist 11.2 (1963).

Dimick, K.P. and Joseph Corse. “Gas Chromatography — A New Method for the
Separation and Identification of Volatile Materials in Foods.” Food Technology 10.8
(August 1956): 360-4.

468
———. “The Volatile Flavors of Strawberry,” American Perfumer (February 1958): 45-
53.

Dimick, K.P. and Benjamin Makower, “A Laboratory-Scale Continuous Vacuum Flash

Evaporator.” Food Technology (December 1951): 517-520.

——–. “Volatile Flavor of Strawberry Essence. I. Identification of the Carbonyls and

Certain Low Boiling Substances,” Food Technology 10.2 (February 1956): 73-5.

Dimick, K.P. and Marion J. Simone. “A Laboratory Continuous Distillation Column for
Concentration of Aqueous Solutions of Volatile Flavors.” Industrial and Engineering
Chemistry 44.10 (October 1952): 2487-90.

[Dodge & Olcott]. The Story of An Unique Institution: Dodge & Olcott, Inc. 1798-1948.
New York: Dodge & Olcott, 1948.

Dorland, Wayne and James A. Rogers, Jr. The Fragrance and Flavor Industry.
Mendham, NJ: Wayne E. Dorland Co., 1977.

Dove, W. Franklin. “Developing Food Acceptance Research.” Science 103.2668 (Feb 15,
1946): 187-91.

———. “Food Acceptability: Its Determination and Evaluation.” Food Technology 1.1
(January 1947): 39-50.

Dravnieks, A. and P.J. Trotter. “Polar Vapor Detection Based on Thermal Modulation of
Contact Potentials,” Journal of Scientific Instruments 42 (1965): 624.

Dreyfus, William. “Alois von Isakovics,” Industrial and Engineering Chemistry 9 (July
1917): 716.

469
Dupree, Ruth. “‘If It’s Yellow, It Must Be Butter: Margarine Regulation in North
America Since 1886.” Journal of Economic History 59.2 (June 1999): 353-71.

DuPuis, E. Melanie. Dangerous Digestion: The Politics of American Dietary Advice.

Oakland: University of California Press, 2015.

Eiserle, Robert J. and William J. Downey. “A Review of the Literature Concerned with
Flavor Research as it Applies to the Problems of the Flavor Industry.” CRC Critical
Reviews in Food Technology 2.2 (July 1971): 159-169.

Elmore, Bartow J. Citizen Coke: The Making of Coca-Cola Capitalism. New York:
Norton, 2015.

Endrijonas, Erika. “Processed Foods from Scratch: Cooking for a Family in the 1950s,”
in Sherrie A. Innes, ed. Kitchen Culture in America: Popular Representations of Food,
Gender, and Race, 157-169. Philadelphia: UPenn Press, 2001.

Ensmenger, Nathan. The Computer Boys Take Over: Computers, Programmers, and the
Politics of Technical Expertise. Cambridge: MIT Press, 2010.

Ettre, Leslie S. “Fifty Years of Gas Chromatography — The Pioneers I Knew, Part I.”
Liquid Chromatography Gas Chromatography North America 20.2 (February 2002):
128-40.

———. “The Early Development and Rapid Growth of Gas Chromatographic

Instrumentation in the United States.” Journal of Chromatographic Science 40
(September 2002): 458-72.

———. “Milestones in Chromatography: Fifty Years of GC Instrumentation.” Liquid

Chromatography Gas Chromatography North America 23.2 (February 2005): 142-8.

Fehling, M. “Artificial Fruit Essences.” American Journal of Pharmacy, 1853

470
Finck, Henry T. Food and Flavor: A Gastronomic Guide to Health and Good Living.
New York: Century, 1913.

Finlay, Mark. “The Industrial Utilization of Farm Products and By-Products: The USDA
Regional Research Laboratories.” Agricultural History 64.2 (Spring 1990): 41-52.

Fischetti, Frank Jr. “The Training of a Flavor Chemist – An Organized Programme.” In

Society of Flavor Chemists, The Multifaceted Nature of the Flavorist, 4-6. [1974]

———. “Natural and Artificial Flavors.” In Thomas E. Furia, ed. Handbook of Food
Additives, 2nd Edition, Vol. II, 229-336. Boca Raton: CRC Press, 1980.

Fitzgerald, Deborah. “Deskilling Farmers: Hybrid Corn and Farmers’ Work.” Technology
and Culture 34.2 (April 1993): 324-43.

———. Every Farm a Factory: The Industrial Ideal in American Agriculture. New
Haven: Yale UP, 2003.

Flath, R.A., D.R. Black, D.G. Guadagni, W.H. McFadden, and T.H. Schultz.
“Identification and Organoleptic Evaluation of Compounds in Delicious Apple Essence.”
Journal of Agricultural and Food Chemistry 15 (Jan/Feb 1967): 29-35.

Flavor and Extract Manufacturer’s Association. FEMA 100: A Century of Great Taste.
Washington, DC: FEMA, 2009.

“Flavor & Extract Manufacturers’ Association: The First 75 Years.” Perfumer &
Flavorist 9.3 (June/July 1984).

“Flavor Chemists Form Society in New York,” J. Agric. Food Chem. 2.5 (1954): 266.

471
“The Flavor of Pineapple.” Givaudan Flavorist 5.2 (1957): 1-4.

Frank, Thomas. The Conquest of Cool: Business Culture, Counterculture, and the Rise of
Hip Consumerism. Chicago: UChicago Press, 1997.

Frear, William. “Standards for Flavoring Extracts: Address before Flavor Extract
Manufacturers Convention, Atlantic City, June 29, 1916.” Simmons’ Spice Mill 39
(September 1916): 1035-6.

Freedman, Paul. Out of the East: Spices and the Medieval Imagination. New Haven: Yale
UP, 2008.

Freidberg, Susanne. Fresh: A Perishable History. Cambridge: Harvard UP, 2009.

French, Michael and Jim Phillips. Cheated, not Poisoned? Food Regulation in the UK,
1875-1938. Manchester: Manchester UP, 2000.

Friedel, Robert. Pioneer Plastic: The Making and Selling of Celluloid. (Madison:
UWisconsin Press, 1983.

Frohlich, Xaq. “Accounting for Taste: Regulating Food Labeling in the ‘Affluent
Society,’ 1945-1995.” Enterprise & Society 13.4 (December 2012): 744-61

———. “The Informational Turn in Food Politics: The US FDA’s Nutrition Label as
Information Infrastructure.” Social Studies of Science (2016): 1-27.

“Full-Day IFT Symposium Stresses Key Role of Food Engineering.” Food Engineering
26 (August 1954): 83.

472
Gardner, Julian W. and Philip N. Bartlett. “A Brief History of Electronic Noses.” Sensors
and Actuators B 18-19 (1994): 211-220

Gately, Iain. Drink: A Cultural History of Alcohol. New York: Gotham, 2008.

G.E.W. “Some Soda Water Fountain Statistics.” Scientific American 81.7 (August 12,
1899): 99.

Gerontas, Apostolos. “Creating New Technologists of Research in the 1960s: The Case
of the Reproduction of Automated Chromatography Specialists and Practitioners.”
Science and Education 23 (2014): 1681-1700.

Gershun, Gloria. “Taste Testers,” Progress Thru Research 7.4 (Summer 1953): 5-7.

Gildemeister, Eduard and Friedrich Hoffmann. Die Ätherischen Öle. Berlin: Springer,
1899.

Girardot, Norman F. and David R. Peryam. “MSG’s Power to Perk Up Foods.” Food
Engineering 26.12 (December 1954): 71-2, 181-5.

Gohlke, R.S. “Time-of-Flight Mass Spectrometry and Gas-Liquid Partition

Chromatography.” Analytical Chemistry 31.4 (April 1959): 535-41.

Goldstein, Carolyn. Creating Consumers: Home Economists in Twentieth-Century

America. Chapel Hill: UNC Press, 2012.

Goodrich, Charles Rush. ed. Science and Mechanism: Illustrated by Examples in the New
York Exhibition 1853-1854. New York: Putnam & Company, 1854.

Goodwin, Lorine Swainston. The Pure Food, Drink, and Drug Crusaders: 1879-1914.
Mcfarland, 1999.
473
Green, Christopher D. “Scientific Objectivity and E.B. Titchener’s Experimental
Psychology.” Isis 101 (2010): 697-721.

“The Green Note in Fruits.” Givaudan Flavorist 2.4 (1954): 2.

Guadagni, D.G., Ron G. Buttery, and S. Okano. “Odor Thresholds of some Organic
Compounds Associated with Food Flavors.” Journal of the Science of Food and
Agriculture 14 (October 1963): 761-5.

Guadagni, D.G. and K.P. Dimick. “Fruit Flavors: Apparatus and Procedure for
Separation and Estimation of Volatile Components.” Journal of Agricultural and Food
Chemistry 1.19 (December 9, 1953): 1169-70.

Haagen-Smit, A.J., Justus G. Kirchner, Clara L. Deasy, and Arthur N. Prater. “Chemical
Studies of Pineapple (Ananas sativus Lindl). II. Isolation and Identification of a Sulfur-
Containing Ester in Pineapple,” Journal of The American Chemical Society 67.10 (1945):
1651-2.

———. “The Chemistry of Flavor.” Engineering and Science Monthly 12 (January

1949): 3-7.

Hahn, Barbara. Making Tobacco Bright: Creating an American Commodity, 1617-1937.

Baltimore: JHU Press, 2011.

Haley, Andrew P. "The Nation Before Taste: The Challenges of American Culinary
History." The Public Historian 34.2 (Spring 2012): 53-78.

Hall, Richard L. “Flavoring Agents as Food Additives,” Food Technology 8.7 (1959): 14-
19.

474
Hall, Richard L. and Bernard L. Oser. “The Safety of Flavoring Substances.” in Residue
Reviews (New York: Springer, 1968): 1-17. Collection of the Society of Flavor Chemists,
Chemical Heritage Foundation, Philadelphia.

Haller, John S., Jr., “With a Spoonful of Sugar: The Art of Prescription Writing in the
Late 19th and Early 20th Century.” Pharmacy in History 26.4 (1984): 171-8.

Halttunen, Karen. Confidence Men and Painted Women: A Study of Middle-Class Culture
in America, 1830-1870. New Haven: Yale UP, 1982.

Hamilton, Alissa. “World War II’s Mobilization of the Science of Food Acceptability:
How Ration Palatability Became a Military Research Priority.” Ecology of Food and
Nutrition 42 (2003): 325-56.

Hamilton, Shane. “The Economies and Conveniences of Modern-Day Living: Frozen

Foods and Mass Marketing, 1945-1965,” The Business History Review 77.1 (Spring
2003): 33–60.

———.“Cold Capitalism: The Political Ecology of Frozen Concentrated Orange Juice,”

Agricultural History 77.4 (Autumn 2003): 557–581.

Harrop, Joseph. Monograph on Flavoring Extracts with Essences, Syrups, and Colorings.
Also Formulas for the Preparation with Appendix. Intended for the Use of Druggists.
Columbus, OH: Harrop & Co, 1891.

Hartman, John D. “A Possible Objective Method for the Rapid Estimation of Flavor in
Vegetables.” Proceedings of the American Society of Horticultural Sciences 64 (1954):
335-42.

Hartman, John D. and W.F. Tolle. “An Apparatus Designed for the Rapid
Electrochemical Estimation of Flavor in Vegetables,” Food Technology 11 (1957): 130-2.

475
Haynes, William American Chemical Industry: A History, Volume I: Background and
Beginnings. Toronto, New York, and London: D. Van Nostrand, 1954.

Herrick, H.T. “New Uses for Farm Crops.” Yearbook of the USDA 1942 (1943): 696-98.

Herzka, A. and J. Pickthall. Pressurized Packaging (Aerosols), 2nd ed. New York:
Academic Press, 1961.

Higby, Gregory J. “American Pharmacy’s First Great Transformation: Practice, 1852-

1902.” In American Pharmacy (1852-2002): A Collection of Historical Essays. Edited by
Higby and Elaine C. Stroud, 1-4. Madison: American Institute for the History of
Pharmacy, 2005.

High, Jack and Clayton A. Coppin. “Wiley and the Whiskey Industry: Strategic Behavior
and the Passage of the Pure Food Act.” Business History Review 62.2 (Summer 1988):
286-309.

Hines, Thomas. Populuxe. New York: Knopf, 1987.

Hisano, Ai. “Eye Appeal is Buy Appeal: Business Creates the Color of Foods, 1870-
1970.” PhD diss., University of Delaware, 2016.

Hiss, A. Emil. The Standard Manual of Soda and Other Beverages: A Treatise Especially
Adapted to the Requirements of Druggists and Confectioners, 10th ed. Chicago: G.P
Engelhardt & Co., 1901 [1897].

Hobson, Asher and Marvin A. Schaars. “Consumer Preferences for Cheese.” University
of Wisconsin Agricultural Experiment Station Research Bulletin 128 (October 1935).

Hoffer, Peter Charles. Sensory Worlds in Early America. Baltimore: Johns Hopkins
University Press, 2003.

476
Hofmann, August. “Chemistry Applied to Arts and Manufactures: Application of Organic
Chemistry to Perfumery, from a Letter written by Dr. Hofmann to Prof. Liebig.”
Chemical Gazette 10 (March 1, 1852): 98-99

Holmes, Frederick and T.H. Levere, eds. Instruments and Experimentation in the History
of Chemistry. Cambridge: MIT Press, 2000.

Hornstein, Irwin, ed. Flavor Chemistry: A Symposium Sponsored by the Division of

Agricultural and Food Chemistry at the 149th Meeting of the American Chemical Society.
Advances in Chemistry Series 65. Washington, D.C.: ACS, 1966.

Hounshell, David A. "The Evolution of Industrial research in the United States." In

Engines of Innovation: U.S. Industrial Research at the End of an Era. Edited by R.
Rosenbloom and W. Spencer. Boston: Harvard Business School Press, 1996.

Hounshell, David A. and John Kenly Smith. Science and Corporate Strategy: DuPont
R&D, 1902-1980. Cambridge: Cambridge UP, 1988.

Howes, David. Sensual Relations: Engaging the Senses in Culture and Social Theory.
Ann Arbor: University of Michigan Press, 2003.

———. “Marx’s Skin.” Adapted from Howes, Sensual Relations: Engaging the Senses in
Culture and Social Theory. Ann Arbor, MI: University of Michigan Press, 2003.
http://www.david-howes.com/senses/marxsskin.html (accessed October 2016).

———. “The Science of Sensory Evaluation: An Ethnographic Critique.” In The Social

Life of Materials: Studies in Materials and Society, edited by Adam Drazin and Susanne
Küchler, 81-97. London: Bloomsbury, 2015.

Hui, Alexandra. The Psychophysical Ear: Musical Experiments, Experimental Sounds,

1840-1910. Cambridge: MIT Press, 2014.
477
“Human Analyzers.” Chemical Industries 67 (November 1950): 721-2.

Hutt, Peter Barton and Peter Barton Hutt, II. "A History of Government Regulation of
Adulteration and Misbranding of Food." Food Drug Cosmetic Law Journal 39 (1984): 2-
73.

Hyde, A.D. “How General Mills Develops Its New Food Specialties.” Food Industries 19
(October 1947): 110-12, 212-15.

Igo, Sarah. The Averaged American: Surveys, Citizens, and the Making of a Mass Public.
Cambridge: Harvard UP, 2007.

“Imitation Flavors Must Be Designated as Such,” The Beverage Journal 63.4 (April
1927): 47-8.

Isakovics, Alois von. Synthetic Perfumes and Flavors: A Lecture Delivered at Columbia
University. Monticello, NY: Synfleur, 1908.

———. “Essential Oils, Synthetic Perfumes, and Flavoring Materials.” In Allen Rodgers,
ed., Industrial Chemistry: A Manual for the Student and Manufacturer, 766-802. New
York: Van Nostrand, 1914.

Isker, Rohland A. “Notes on the Use and Effects of Monosodium Glutamate.” Journal of
the American Dietetic Association 25 (September 1949): 760-3.

Issenberg, Phillip, Akio Kobayashi, and TJ Mysliwy. “Combined Gas Chromatography-

Mass Spectrometry in Flavor Research: Methods and Applications.” Journal of
Agricultural and Food Chemistry 17.6 (Nov/Dec 1969): 1377-86.

478
Jackson, Catherine M. “Synthetical Experiments and Alkaloid Analogues: Liebig,
Hoffman, and the Origins of Organic Synthesis.” Historical Studies of the Natural
Sciences 44.4 (2014): 319-63

Jacobs, Morris B., ed. The Chemistry and Technology of Food and Food Products.
Volume 2. New York: Interscience, 1944.

Jacobs, Moris Boris. Synthetic Food Adjuncts. New York: D.Van Nostrand, 1947.

———. “How to Make Flavoring Powders,” American Perfumer and Essential Oil
Review 57 (May 1951): 391-3.

James, A.T. and A.J.P. Martin. “Gas-Liquid Partition Chromatography: A Technique for
the Analysis of Volatile Materials.” Analyst 77 (1952): 915-932.

———. “Gas-Liquid Partition Chromatography: The Separation and Micro-estimation of

Volatile Fatty Acids from Formic Acid to Dodecanoic Acid,” Biochemical Journal 50
(1952): 679.

Johnson, Eli. Drinks from Drugs, or the Magic Box: A Startling Exposure of the Tricks of
the Liquor Traffic. Chicago: Revolution Temperance Publishing House, 1881.

Johnston, James F.W. The Chemistry of Common Life. Vol. 2, 8th ed. New York: D.
Appleton, 1856.

Johnston, V.D. “Instrumental Methods of Analyses Save Time — Give More

Information.” Givaudan Flavorist 7.1 (1959): 1-4.

Jones, F. Nowell. “History of Psychophysics and Judgment.” In Edward C. Carterette and

Morton P. Friedman, eds. Handbook of Perception, Vol. II. Psychophysical Judgment
and Measurement, 2-20. New York: Academic Press, 1974.

479
Jones, Geoffrey. Beauty Imagined: A History of the Global Beauty Industry. Oxford:
Oxford UP, 2010.

Jones, Lyle V., David R. Peryam, and L.L. Thurstone. “Development of a Scale for
Measuring Soldiers’ Food Preferences.” Food Research 20 (1955): 512-20.

Josephson, Paul. "The Ocean's Hot Dog: The Development of the Fish Stick."
Technology and Culture 49.1 (January 2008): 41-61

Jou, Chin. “Controlling Consumption: The Origins of Modern American Ideas about
Food, Eating, and Fat, 1886-1930.” PhD Dissertation, Princeton, 2009.

Kahn, E.J., Jr. The Problem Solvers: A History of Arthur D. Little, Inc. Boston: Little,
Brown, 1986.

Katz, Alexander. “Newly Developed Flavoring Aromatics.” American Perfumer &

Aromatics 68 (September 1956): 66-70.

———. “The Development of Aromatics,” American Perfumer & Aromatics 68 (July

1956): 31-2, 35.

———. “Newly Developed Flavoring Aromatics.” Perfumer & Essential Oil Review 48
(March 1957): 131-4.

Katz, Alexander and Abraham Seldner. “New Aromatics to Enhance Fruit Flavors,”
American Perfumer & Aromatics 55 (January 1950): 45

———. “California Essential Oil Development,” American Perfumer & Aromatics 57

(May 1951): 357-60.

480
Kavaler, Lucy. The Artificial World Around Us. New York: John Day, 1963.

Kawash, Samira. Candy: A Century of Panic and Pleasure. New York: Faber and Faber,
2013.

Kay, Gwen. “Healthy Public Relations: The FDA’s 1930s Legislative Campaign.”
Bulletin of the History of Medicine 75.3 (Fall 2001): 446-487.

Keith, Elizabeth S. and John J. Powers. “Determination of Flavor Threshold Levels and
Sub-Threshold, Additive, and Concentration Effects.” Journal of Food Science 33.2
(1968): 213-8.

Kendall, David A. and Anne J. Neilson. “Correlation of Subjective and Objective Odor
Responses.” Annals of the New York Academy of Sciences (July 1964): 567-75.

Kevles, Daniel. “Testing the Army’s Intelligence: Psychologists and the Military in
World War I.” Journal of American History 55.3 (1968): 565-581.

Khatchadourian, Raffi. “The Taste Makers,” The New Yorker (November 23, 2009): 129-
35.

Khong, Rachel. “The End of the World as We Know It: Interview with Michael Pollan,”
Lucky Peach (September 10, 2014).

Kiechle, Melanie. “’The Air We Breathe’: Nineteenth-Century Americans and the Search
for Fresh Air.” PhD diss., Rutgers, 2012.

King, C.G. “Nutrition in Relation to Flavor and World Food Acceptance.” In David E.
Schafer, ed. Proceedings: Flavor Chemistry Symposium, 1961, 97-108. Camden, NJ:
Campbell Soup Company, 1961.
481
King, Florance B. “Obtaining a Panel for Judging Flavor in Foods,” Food Research 2.3
(1937): 207-19.

Kleber, Clemens. “The Occurence of Amyl Acetate in Bananas.” American Perfumer and
Essential Oil Review 7.10 (December 1912): 235.

Klein, Ursula. Experiments, Models, Paper Tools: Cultures of Organic Chemistry in the
Nineteenth Century. Chicago: Chicago UP, 2003.

———. “Technoscience Avant la Lettre.” Perspectives on Science 13.2 (2005): 226-66.

Kletzinsky, V. Mittheilungen aus dem Gebiete der reinen und angewandten

Chemie. (Vienna: Selbstverlag des Verfassers, 1865): 45.

[Kletzinsky, V.] “Ueber die sogenannten Fruchtessenzen.” Dingler’s Polytechnische

Journal 180 (1866): 77.

Kletzinski, M. [sic]. “On Fruit Essences.” Druggists’ Circular and Chemical Gazette
(April 1, 1867): 82.

———. “On Fruit Essences,” American Journal of Pharmacy (May 1867): 238.

Klymentiev, Maksym. “Creating Spices for the Mind: The Origins of Modern Western
Perfumery.” The Senses & Society 9.2 (2014): 212-31.

Knight, David M. and Helge Kragh, eds. The Making of the Chemist: A Social History of
Chemistry in Europe, 1789-1914. Cambridge: Cambridge UP, 1998.

482
Koehn, Nancy F. “Henry Heinz and Brand Creation in the Late Nineteenth Century:
Making Markets for Processed Food.” The Business History Review 73.3 (Autumn 1999):
349-93.

———. Brand New: How Entrepreneurs Earned Consumers’ Trust from Wedgwood to
Dell. Boston: Harvard Business School Press, 2001.

Koehler, Franz A. Special Rations for the Armed Forces: 1946-1953. QMC Historical
Studies, Series II, no. 6, 1958.

Korsmeyer, Carolyn. Making Sense of Taste: Food and Philosophy. Ithaca: Cornell UP,
2002.

Lacour, Pierre. The Manufacture of Liquors, Wines, and Cordials without the Aid of
Distillation, (New York: Dick & Fitzgerald, 1868) [Originally published by subscription,
New York: R. Craighead, 1853]: 57

Lahne, Jake. “Tasting in Context: Consumer Sensory Perception of Vermont Artisan

Cheese.” PhD diss., University of Vermont, 2014.

———. “Sensory Science, the Food Industry, and the Objectification of Taste,”
Anthropology of Food 2016.

Laird, Pamela Walker. Advertising Progress: American Business and the Rise of
Consumer Marketing. Baltimore: JHU press, 1998.

Lakritz, David E. “Development of Flavors.” Drug and Cosmetic Industry 65 (December

1949): 647-8, 720-4.

Langlois, F.N. “The Formulist,” Ungerer’s Bulletin 2.3 (February 1921): 3.

483
Laszlo, Pierre. “Georges Darzens (1867-1954): Inventor and Iconoclast,” Bulletin for the
History of Chemistry 15/16 (1994): 59-65.

———. “On the Self-Image of Chemists: 1950-2000.” Hyle 12.1 (2006): 99-130.

Latour, Bruno. "How to Talk About the Body? The Normative Dimension of Science
Studies." Body & Society 10.2-3 (2004): 205-229.

Landau, Ralph and Nathan Rosenberg. “Successful Commercialization in the Chemical

Process Industries.” In Rosenberg, Landau, and David C. Mowery, eds. Technology and
the Wealth of Nations, 73-119. Stanford: Stanford UP, 1992.

Laudan, Rachel. Cuisine and Empire: Cooking in World History. Oakland: University of
California Press, 2013.

———. “A Plea for Culinary Modernism: Why We Should Love New, Fast, Processed
Food/” Gastronomica 1.1 (Winter 2001): 36-44.

Law, Marc T. “The Origins of State Pure Food Regulation.” Journal of Economic History
63 (December 2003): 1103-1130.

———. “How do Regulators Regulate? Enforcement of the Pure Food and Drugs Act,
1907-1938.” Journal of Law, Economics, and Organization 22, (2005): 460-86.

Lazarsfeld, Paul and Marjorie Fisk. "The 'Panel' as a New Tool for Measuring Opinion."
Public Opinion Quarterly 2.4 (October 1938): 596-612.

Leach, William. Land of Desire: Merchants, Power, and the Rise of a New American
Culture. New York: Pantheon, 1993.

484
Lemov, Rebecca. Database of Dreams: The Lost Quest to Catalog Humanity. New
Haven: Yale UP, 2015.

Levenstein, Harvey. Revolution at the Table: The Transformation of the American Diet.
New York: Oxford UP, 1986.

———. Paradox of Plenty: A Social History of Eating in Modern America. New York:
Oxford UP, 1993.

Mack, Adam. "'Speaking of Tomatoes': Supermarkets, the Senses, and Sexual Fantasy in
Modern America." Journal of Social History 43 (Summer 2010): 815- 842.

Mackay, Donald A.M., David Lang, and Murray Berdick. “Objective Measurement of
Odor: Ionization Detection of Food Volatiles,” Analytical Chemistry 33.10 (September
1961): 1369-74.

MacKenzie, Donald and Judy Wajcman, eds. The Social Shaping of Technology: How the
Refrigerator Got its Hum. Philadelphia: Open University Press, 1985.

Maher, Simon, Fred P.M. Jjunju, and Stephen Taylor. “Colloquium: 100 Years of Mass
Spectrometry: Perspectives and Future Trends.” Reviews of Modern Physics 87 (January-
March 2015): 113-135.

Marchand, Roland. Advertising the American Dream: Making Way for Modernity, 1920-
1940. Berkeley: University of California Press, 1985.

Marcuse, Sophie. "An Application of the Control Chart Method to the Testing and
Marketing of Foods." Journal of the American Statistical Association 40.230 (June
1945): 214-222.

“Markets: Aerosols Reaching New Highs.” Chemical & Engineering News (December
19, 1955): 5518.
485
“Markets: Aerosols Crack Billion-Dollar Mark.” Chemical & Engineering News (May
21, 1962): 36-7.

Marling, Karal Ann. As Seen on TV: The Visual Culture of Everyday Life in the 1950s.
Cambridge: Harvard UP, 1994.

Marx de Salcedo, Anastasia. Combat-Ready Kitchen: How the U.S. Military Shapes the
Way You Eat. New York: Current/Penguin Random House, 2015.

Mathews, Burgin. “Sinners and Saints (1926-1931)/Document.” In Vladimir Bogdanov,

Chris Woodstra, and Stephen Thomas Erlewine, eds., All Music Guide to The Blues: The
Definitive Guide to the Blues, 3rd ed., 672. San Francisco: Backbeat Books, 2003.

May, Elaine Tyler. Homeward Bound: American Families in the Cold War Era. New
York: Basic Books, 1988.

Mazzola, L. Charles. "Grading Food by a Descriptive Method." Food Industries 2 (May

1930): 214-5.

———. "How a Formula for Descriptive Grading Was Developed." Food Industries 2
(August 1930): 340-44.

McCarthy, Alice I., James K. Palmer, Carol P. Shaw, and Edward E. Anderson.
"Correlation of Gas Chromatographic Data with Flavor Profiles of Fresh Banana Fruit."
Journal of Food Science 28.4 (July 1963): 379-384.

McFadden, W.H. and Roy Teranishi. “Fast-Scan Mass Spectrometry with Capillary Gas-
Liquid Chromatography in Investigation of Fruit Volatiles.” Nature 200 (October 26,
1963): 329-30.

486
McFadden, W.H., Roy Teranishi, D.R. Black, and J.C. Day, “Use of Capillary Gas
Chromatography with a Time-of-Flight Mass Spectrometer,” Journal of Food Science
28.3 (1963): 316-19.

McGee, Harold. On Food and Cooking: The Science and Lore of the Kitchen. Revised ed.
New York: Scribners, 2004.

McGlumphy, James. “Progress in Flavor Research.” Food Technology (December 1966):

48-50.

McKenna, Christopher D. “The World’s Newest Profession: Management Consulting in

the Twentieth Century.” Enterprise and Society 2.4 (December 2001): 673-9.

McKercher, Catherine and Vincent Mosco, eds. Knowledge Workers in the Information
Society. Lanham: Lexington Books, 2007.

Meikle, Jeffrey. American Plastic: A Cultural History. New Brunswick: Rutgers

University Press, 1995.

Meinel, Cristoph. “Molecules and Croquet Balls.” In Models: The Third Dimension of
Science, edited by Soraya de Chadarevian and Nick Hopwood, 242-75. Stanford:
Stanford UP, 2004.

Meiselman, Herbert L. and Howard G. Schutz, “History of Food Acceptance Research in

the US Army.” Appetite 40 (2003): 199-216.

Merrill, Richard A. “FDA’s Implementation of the Delaney Clause: Repudiation of

Congressional Choice or Reasoned Adaptation to Scientific Progress?” Yale Journal on
Regulation 5.1 (1988): 1-88.

Merwin, Earl. “The Art in Flavor Creation I: True Fruit Flavors.” Givaudan Flavorist
10.3 (1962): 5-8.
487
———. “The Art in Flavor Creation II: Imitation Flavors.” Givaudan Flavorist 10.4
(1962): 3-8.

———, ed. The Development and Application of Natural and Artificial Flavor Systems.
Wheaton, IL: Allured, 1988.

Metcalf, Lyne S. “Spray Drying of Flavors.” American Perfumer & Aromatics (April
1955): 64-5.

Mintz, Sidney. Sweetness and Power: The Place of Sugar in Modern History. New York:
Viking, 1985.

Mintz, Sidney W. and Christine M. Du Bois. “The Anthropology of Food and Eating.”
Annual Review of Anthropology 31 (2002): 99–119.

Mitchell, Jack H., Norbert J. Leinen, Emil Mrak, and S. David Bailey, eds. Chemistry of
Natural Food Flavors: A Symposium Sponsored by the National Academy of Sciences
National Research Council for Quartermaster Food and Container Institute for the
Armed Services and Pioneering Research Division, Quartermaster Research and
Engineering Center. Washington, DC: Armed Forces Food and Container Institute, May
1957.

Moir, H.C. "Some Observations on the Appreciation of Flavor in Foodstuffs," Chemistry

and Industry 14 (February 21, 1936): 145-8.

Mol, Annemarie. “I Eat an Apple. On Theorizing Subjectivities," Subjectivity 1 (2008):

28-37.

——–. “Good Taste: The Embodied Normativity of the Consumer-Citizen." Journal of

Cultural Economy 2.3 (November 2009): 269-284.

488
——–. “Tasting Food: Tasting Between the Laboratory and the Clinic." In A Companion
to the Anthropology of the Body and Embodiment, edited by Frances E. Mascia-Lees,
467-80. Wiley-Blackwell, 2011.

Moncrieff, R.W. “An Instrument for Measuring and Classifying Odors.” Journal of
Applied Physiology 16 (1961): 742-9.

Morris, Peter J.T., ed. From Classical to Modern Chemistry: The Instrumental
Revolution. Cambridge: Royal Society of Chemistry, 2000.

Mosby, Ian. “That Won-Ton Soup Headache’: Chinese Restaurant Syndrome, MSG, and
the Making of American Food, 1968-1980.” Social History of Medicine 22.1 (2009): 133-
51.

Moser, Helen A. H.J. Dutton, C.D. Evans, and J.C. Cowan. "Conducting a Taste Panel
for the Evaluation of Edible Oils." Food Technology 4 (1950): 105-109.

Moskowitz, Howard R. “How Psychophysics Changed the Food Business, and How the
Food Business Forever Changed Psychophysics.” International Society for
Psychophysics 25 (2009).

Moss, Michael. Salt, Sugar, Fat: How the Food Giants Hooked Us. New York: Random
House, 2013.

Mowery, David C. “The Relationship Between Intrafirm and Contractual Forms of

Industrial Research in American Manufacturing, 1900-1940.” Explorations in Economic
History 20 (1983): 351-374.

Mudry, Jessica. Measured Meals: Nutrition in America. New York: SUNY UP, 2009.

——–. "The Mindful Measurement of Food: Quantification, the Food Pyramid and
Discourses of Taste." Material Culture Review 70 (Fall 2009): 12-22.
489
Murphy, Michelle. Sick Building Syndrome and the Problem of Uncertainty:
Environmental Politics, Technoscience, and Women Workers. Durham and London:
Duke UP, 2006.

Myers, Natasha. “Molecular Embodiments and the Body-work of Modeling in Protein

Crystallography.” Social Studies of Science 38.2 (April 2008): 163-199.

Ndiaye, Pap. Nylon and Bombs: DuPont and the March of Modern America. Translated
by Elborg Forster. Baltimore: Johns Hopkins University Press, 2007.

Nestle, Marion. Food Politics: How the Food Industry Influences Nutrition and Health.
Oakland: University of California Press, 2013.

Nestle, Marion and W. Alex McIntosh. “Writing the Food Studies Movement.” Food,
Culture and Society: An International Journal of Multidisciplinary Research 13, no. 2
(2010): 159–179.

“New Dry Flavorings Marketed by Britons.” Food Industries 20 (May 1948): 144.

Nickles, Shelley. "'Preserving Women': Refrigerator Design as Social Process in the

1930s." Technology and Culture 43.4 (October 2002): 693-727.

Noling, A.W. History of Hurty-Peck & Company: Its First 50 Years, 1903-1953.
Annotated manuscript. Indianapolis: A.W. Noling, 1968. A.W. Noling Collection, UC
Davis.

Nye, Mary Jo. From Chemical Philosophy to Theoretical Chemistry: Dynamics of Matter
and Dynamics of Disciplines, 1800-1950. Berkeley: University of California Press, 1993.

490
O’Brien, Dennis. “Cited for More than 60 Years of Flavor Research.” Agricultural
Research Magazine (May-June 2013): 14-16.

O’Connor, Kaori. “The Hawaiian Luau: Food as Tradition, Transgression,

Transformation, and Travel.” Food, Culture & Society 11.2 (2008): 149-172.

Okihiro, Gary Y. Pineapple Culture: A History of the Tropical and the Temperate Zones.
Berkeley: University of California Press, 2009.

Oliver, Henry K. “Report on the Character of Substances Used for Flavoring Articles of
Food and Drink.” Annual Report of the State Board of Health of Massachusetts, Vol. 4,
January 1873: 145-74.

“Origin of Powdered Flavors.” American Perfumer (March 1956): 62-3.

Orr, Julian E. Talking About Machines: An Ethnography of a Modern Job. Ithaca, NY:
ILR Press, 1996.

Oser, Bernard L. and Richard A. Ford. “FEMA Expert Panel: 30 Years of Safety
Evaluation for the Flavor Industry,” Food Technology (Nov 1991): 86-8, 93-7.

Owens, Larry. "Engineering the Perfect Cup of Coffee: Samuel Prescott and the Sanitary
Vision at MIT." Technology and Culture 45.4 (October 2004): 795-807.

Parascandola, John. “The Pharmaceutical Sciences in America, 1852-1902.” In American

Pharmacy (1852-2002): A Collection of Historical Essays. Edited by Gregory Higby and
Elaine C. Stroud, 19-23. Madison: American Institute for the History of Pharmacy, 2005.

Pangborn, Rose Marie. “Sensory Evaluation of Foods: A Look Backward and Forward.”
Food Technology 18 (September 1964): 63-7.

491
——–. “Flavor Perception: Relation of Sensory to Instrumental Measurements.” In DJ
Tilgner and A. Borys, eds., Proceedings of the 2nd International Congress of Food
Science and Technology, 303-18. Warszawa: Stampa, 1967.

Parkin, Katherine J. Food is Love: Advertising and Gender Roles in Modern America.
Philadelphia: UPenn Press, 2011.

Parrish, Edward. An Introduction to Practical Pharmacy. 2nd ed. Philadelphia: Blanchard

& Lea, 1859.

Parry, Ernest J. Chemistry of Essential Oils and Artificial Perfumes. Scott, Greenwood &
Co., London and D. Van Nostrand, New York, 1899.

Pecsok, Robert L. “Gas Chromatography: Basic Principles and New Developments.”

Journal of Chemical Education 38.4 (April 1961): 212-6.

Peryam, David R. "Quality Control in the Production of Blended Whiskey," Industrial

Quality Control (November 1950): 17-21.

———. "Sensory Evaluation -- The Early Days," Food Technology 46.1 (January 1990):
86-89.

Peryam, David R. and Norman F. Girardot. “QM Pins Food ‘Likes’ and ‘Dislikes’ with
Advanced Taste-Test Method.” Food Engineering 24.7 (July 1952): 58-62, 194.

Peryam, David R. and Francis J. Pilgrim. “Hedonic Scale Method of Measuring Food
Preferences.” In “The Methodology of Sensory Testing: Contributions to the Institute of
Food Technologists Symposium in Pittsburgh, Sponsored by the Committee on Taste
Testing and Consumer Acceptance for the 17th Annual Meeting of the IFT,” Food
Technology 11 (September 1957): 9-14.

492
Peryam, David, Francis J. Pilgrim, and Martin S. Peterson, eds. Food Acceptance Testing
Methodology: A Symposium Sponsored by the Quartermaster Food and Container
Institute for the Armed Forces. Washington, DC: National Academy of Sciences National
Research Council, 1954.

Petrick, Gabriella M. “The Arbiters of Taste: Producers, Consumers, and the

Industrialization of Taste in America, 1900-1960.” PhD diss., University of Delaware,
2006.

———. "'Purity as Life': H.J. Heinz, Religious Sentiment, and the Beginning of the
Industrial Diet." History and Technology 27 (2001): 37-64.

Phillips, Christopher J. “The Taste Machine: Sense, Subjectivity, and Statistics in the
California Wine World.” Social Studies of Science 46.3 (2016): 461-481.

Pilgrim, Francis J. “The Components of Food Acceptance and Their Measurement.”

American Journal of Clinical Nutrition 5.2 (1957): 171-5.

Pilgrim, Francis J., Howard G. Schutz, and David R. Peryam. “Influence of Monosodium
Glutamate on Taste Perception.” Food Research 20 (1955): 310-16.

Platt, Washington. "Rational Methods of Scoring Food Products," Food Industries 3

(March 1931): 108-11.

———. “Some Fundamental Assumptions Pertaining to the Judgement of Food

Flavors,” Food Research 2.3 (1937): 237-49.

Potter, Richard H. "Vapor Phase Chromatography as a Tool in Flavor Creation." The

Givaudan Flavorist 11.1 (1963): 5-6

———. "Further Thoughts on the Use of Vapor Phase Chromatography in Flavor

Creation." The Givaudan Flavorist 11.4 (1963): 5-6.
493
———. “Gas Chromatography — A Flavorist’s Tool.” In Society of Flavor Chemists,
The Multifaceted Nature of the Flavorist, 18-19. [1974]

Power, Frederick B. “The Detection of Methyl Anthranilate in Fruit Juices.” Journal of

the American Chemical Society 43.2 (1921): 377-81.

Power, Frederick B. and Victor K. Chestnut. "The Odorous Constituents of Apples."

Journal of the American Chemical Society 42.7 (1920): 1509-1526.

———. "The Odorous Constituents of Peaches." Journal of the American Chemical

Society, (1921): 1725-1740.

———. “The Occurrence of Methyl Anthranilate in Grape Juice.” Journal of the

American Chemical Society 43.7 (1921): 1741-2.

———. “Examination of Authentic Grape Juices for Methyl Anthranilate.” Journal of

Agricultural Research 23.1 (1923): 47-53.

Quartermaster Food and Container Institute for the Armed Forces. Committee on Food
Research Conference on Food Acceptance Research. Quartermaster Corps Manual QMC
17-9. Washington, D.C.: War Department, Office of the Quartermaster General, 1946.

———. Flavor and Acceptability of Monosodium Glutamate: Proceedings of the

Symposium. Chicago: Food and Container Institute, 1948.

Reineccius, Gary A. “The Spray Drying of Food Flavors.” Drying Technology 22.6
(2004): 1289-1324.

Reinhardt, Carsten. Shifting and Rearranging: Physical Methods and the Transformation
of Modern Chemistry. Sagamore Beach, MA: Science History Publications/USA, 2006.
494
———. “The Olfactory Object: Toward a History of Smell in the 20th Century.” In
Ursula Klein and Reinhardt, eds. Objects of Chemical Inquiry, 321-40. Sagamore Beach,
MA: Science History Publications, 2014.

“Report for Analytical Chemists: Physicochemical Research on Flavor,” Analytical

Chemistry 30.2 (February 1958): 17A.

“Report of the Proceedings of the Fourth Annual Convention of the Flavor & Extract
Manufacturers Association,” American Perfumer & Essential Oil Review 8.4 (June
1913): 86-93.

Rhees, David Jerome. “The Chemist’s Crusade: The Rise of an Industrial Science in
Modern America, 1907-1922,” PhD Diss., University of Pennsylvania, 1987.

Rheinberger, Hans-Jorg. “Gaston Bachelard and the Notion of Phenomenotechnique.”

Perspectives on Science 13.3 (Fall 2005): 313-8.

Risch, Sara J. and Gary A. Reineccius, eds. Flavor Encapsulation. ACS Symposium
Series 370. Washington, D.C.: American Chemical Society, 1988.

Roberts, Lissa. “Death of the Sensuous Chemist: The ‘New’ Chemistry and the
Transformation of Chemical Technology.” Studies in the History and Philosophy of
Science 26.4 (1995): 503-29.

Roberts, Lissa, Simon Schaffer, and Peter Dear, eds. The Mindful Hand: Inquiry and
Invention from the Late Renaissance to Early Industrialization. Chicago: University of
Chicago Press, 2008.

Rocke, Alan J. Image and Reality: Kekule, Kopp, and the Scientific Imagination.
Chicago: University of Chicago Press, 2010.

495
Roosth, Sophia. "Screaming Yeast: Sonocytology, Cytoplasmic Milieus, and Cellular
Subjectivities." Critical Inquiry 35.4 (2009): 332-350.

Rouse, Morleen Getz. "Daytime Radio Programming for the Homemaker: 1926-1956."
Journal of Popular Culture 12.2 (1978): 315-327.

Saldarini, A.V. “The Chemical Sources Association — The Pioneers.” Perfumer and
Flavorist 13 (Aug/Sept 1988): 57-8.

Sale, J.W. "Labeling of Flavoring Extracts." American Perfumer & Essential Oil Review,
July 1925.

———. “Labeling Beverages and Beverage Materials Under the Federal Food and Drugs
Act.” In R.O. Brooks, Critical Studies in the Legal Chemistry of Foods. New York:
Chemical Catalog Company, 1927: 263-71.

Sale, J.W. and W.W. Skinner. “Food Flavors: Their Source, Composition, and
Adulteration: Conclusion,” Beverage Journal (October 1922): 50a-50c.

Sale, J.W. and J.B. Wilson. “Distribution of Volatile Flavor in Grapes and Grape Juices.”
Journal of Agricultural Research 33.4 (1926): 301-310.

Samuel, Lawrence R. Freud on Madison Avenue: Motivational Research and Subliminal

Advertising in America. Philadelphia: University of Pennsylvania Press, 2011.

Sand, Jordan. “A Short History of MSG: Good Science, Bad Science, and Taste
Cultures.” Gastronomica (Fall 2005): 38-50.

Sandage, Scott. Born Losers: A History of Failure in America. Cambridge: Harvard UP,
2005.

496
Sandelowski, Margarete. Devices and Desires: Gender, Technology, and American
Nursing. Chapel Hill: University of North Carolina Press, 2000.

Sauer, C.F. “Why Flavoring Extracts are Essential Food Products,” Simmons Spice Mill
(August 1918): 1014.

“Save the Volatiles.” Food Industries 6 (November 1934): 485.

Schafer, David E., ed. Proceedings: Flavor Chemistry Symposium, 1961. Camden, NJ:
Campbell Soup Company, 1961.

Schaffer, Simon. “Astronomers Mark Time: Discipline and the Personal Equation.”
Science in Context 2.1 (1988): 115-145.

Schatzker, Mark. The Dorito Effect: The Surprising New Truth About Food and Flavor.
New York: Simon & Schuster, 2015.

Schivelbusch, Wolfgang. Tastes of Paradise: A Social History of Spices, Stimulants, and

Intoxicants. Translated by David Jacobson. New York: Vintage, 1993.

Schleifer, David. "The Perfect Solution: How Trans Fats Became the Health Replacement
for Saturated Fats." Technology and Culture 53.1 (January 2012): 94-119.

Schlosser, Eric. Fast Food Nation: The Dark Side of the All-American Meal. New York:
Perennial, 2002.

Schrank, H.C. “Pure Food Laws” Paper read before the conventions of the Wisconsin and
Illinois State Bottlers’ Associations. American Bottler 29.4 (April 1909): 20.

497
Schüll, Natasha Dow. Addiction by Design: Machine Gambling in Las Vegas. Princeton
and Oxford: Princeton UP, 2012.

Schultz, H.W., E.A. Day, and L.M. Libbey, eds. Symposium on Foods: The Chemistry
and Physiology of Flavors. Westport, CT: AVI, 1967.

Schutz, Howard G. "Evolution of the Sensory Science Discipline." Food Technology 52.8
(August 1998): 42-6.

Schupphaus, R. “On the Alcohols of Fusel Oil.” Journal of the American Chemical
Society 14.3 (1892): 45-60.

Schweisheimer, W. “Sniffing Machine Savors Coffee Producing a ‘Picture’ of Coffee

Aroma.” Perfumer & Essential Oil Review 48 (September 1957): 443.

Scott, R.D. “Methyl Anthranilate in Grape Beverages and Flavors.” Industrial and
Engineering Chemistry: 15.7 (July 1923): 732-3.

Scranton, Philip. Endless Novelty: Specialty Production and American Industrialization,

1865-1925, (Princeton UP: 1997)

Scrinis, Gyorgy. Nuritionism: The Science and Politics of Dietary Advice. New York:
Columbia UP, 2013.

Seldner, Abraham. “New Aromatics for Flavoring and Perfume Industries.” American
Perfumer & Essential Oil Review 54 (October 1949): 295-6.

Shapin, Steven. "The Invisible Technician," American Scientist 77.6 (1989): 554-563.

498
———. The Scientific Life: A Moral History of a Late Modern Vocation. Chicago:
University of Chicago Press, 2010.

——–. “Changing Tastes: How Things Tasted in the Early Modern Period and How They
Taste Now.” Hans Rausing Lecture, Uppsala University, 2011.

———. "Sciences of Subjectivity." Social Studies of Science 42.2 (2012): 170-184.

Shapiro, Laura. Something from the Oven: Reinventing Dinner in 1950s America. New
York: Penguin, 2005.

———. Perfection Salad: Women and Cooking at the turn of the Century. Berkeley:
University of California Press, 2008.

Shepard, James H. “Like Substances,” Pure Products 3.11 (November 1907): 507-13.

Shore, Harris. “The Training of a Flavorist – One On One.” In Society of Flavor

Chemists, The Multifaceted Nature of the Flavorist, 2-3. [1974]

Singerman, David Roth. “Inventing Purity in the Atlantic Sugar World, 1860-1930.” PhD
diss., Massachusetts Institute of Technology, 2014.

Sink, Gabriel. “A Tribute to the Oldest American Flavor and Fragrance House.”
Perfumer and Flavorist 17 (January/February 1992): 37-9.

Sjöström, L.B. and S.E. Cairncross. “What Makes Flavor Leadership?” Food Technology
7 (1953): 56-8.

Sloterdijk, Peter. “The Crystal Palace,” Public 37 (2008): 11-16.

499
Smith, Andrew. Drinking History: Fifteen Turning Points in the Making of American
Beverages. New York: Columbia UP, 2013.

Smith, Bernard H. “Modern Trends in Flavors,” Food Research 2.3 (1937): 251-253.

Smith, Ellen Goodell. A Practical Cook Book and Text Book for General Use: The Fat of
the Land and How to Live On It. Amherst: Carpenter and Morehouse, 1896.

Smith, Mark. How Race is Made: Slavery, Segregation, and the Senses. Chapel Hill:
UNC Press, 2006.

Smith, Pamela H. The Body of the Artisan: Art and Experience in the Scientific
Revolution. Chicago: University of Chicago Press, 2004.

Smith, William Abbots. On Glycerine, and Its Uses in Medicine, Surgery, & Pharmacy.
Being Principally an Abstract of M. Demarquay's Treatise, 'De La Glycerine,' &c.
London: H.K. Lewis, 1863.

Smith-Howard, Kendra. Pure and Modern Milk: An Environmental History Since 1900.
New York: Oxford UP, 2013.

Society of Flavor Chemists. Food Chemical Sources. New York: Society of Flavor
Chemists, Inc., 1968.

———. The Multifaceted Nature of the Flavorist: Papers Presented by the Society of
Flavor Chemists Symposium Held at Rutgers University, New Brunswick, New Jersey,
March 21, 1974. Society of Flavor Chemists Collection, Chemical Heritage Foundation.

500
Society of Flavor Chemists, Inc. and Chemical Sources Association, Inc. The Flavor
Industry from 1945 to 1995: A Short History of the Flavor Industry With Emphasis on the
USA and the Past Fifty Years. Neptune, NJ: Society of Flavor Chemists, 1995.

Spackman, Christy. “Transforming Taste: The Twentieth-Century Aesthetic Remaking of

Water.” PhD diss., New York University, 2015.

Spang, Rebecca L. The Invention of the Restaurant: Paris and Modern Gastronomic
Culture. Cambridge: Harvard UP, 2000.

Spary, E.C. Feeding France: New Sciences of Food, 1760-1815. Cambridge: Cambridge
UP, 2014.

Spiekermann, Uwe. “Redefining Food: The Standardization of Products and Production

in Europe and the United States, 1880-1914.” History and Technology 27 (March 2011):
11-36.

Stanziani, Alessandro. Histoire de la qualité Alimentaire: XIXe-XXe Siécle. Seuil: Paris,

2005.

———. “Negotiating Innovation in a Market Economy: Foodstuffs and Beverage

Adulteration in Nineteenth-Century France.” Enterprise and Society 8 (June 2007): 375-
412.

Star, Susan Leigh and James R. Griesemer. “Institutional Ecology, ‘Translations’ and
Boundary Objects: Amateurs and Professionals in Berkeley’s Museum of Vertebrate
Zoology, 1907-39.” Social Studies of Science 19.3 (1989): 387-420.

Steen, Kathryn. The American Synthetic Organic Chemicals Industry: War and Politics,
1910-1930. Chapel Hill: UNC Press, 2014.

501
Stone, Herbert, Rose Marie Pangborn, and C.S. Ough. “Techniques for Sensory
Evaluation of Food Odors,” Advances in Food Research 14 (1965): 1-32.

Sterne, Jonathan. The Audible Past: Cultural Origins of Sound Reproduction. Durham:
Duke University Press, 2003.

Strasser, Susan. Satisfaction Guaranteed: The Making of the American Mass Market.
New York: Pantheon, 1989.

Strauss, GLM et al. England’s Workshops. London: Groombridge & Sons, 1864.

Suchman, Lucy. “Embodied Practices of Engineering Work.” Mind, Culture, and Activity
7 (2000): 4-18.

"Sulfur and Aromatics." Givaudan Flavorist 4.4 (1957): 1-4.

“Synthetic Perfumes: Lecture by Dr. Alois von Isakovics.” American Druggist and
Pharmaceutical Record (April 13, 1908): 176.

“Tailored to Taste.” American Perfumer & Essential Oil Review 60 (August 1952): 99.

Teranishi, Roy. “Development of Methodology of Flavor Chemistry Past, Present and

Future.” In David B. Min and Thomas H. Smouse, eds., Flavor Chemistry of Lipid
Foods, 13-25. Champaign, IL: American Oil Chemists’ Society, 1989.

Teranishi, Roy, Ron G. Buttery, and Robert E. Lundin. “Gas Chromatography. Direct
Vapor Analyses of Food Products with Programmed Temperature Control of Dual
Columns with Dual Flame Ionization Detectors.” Analytical Chemistry 34.8 (July 1962):
1033-5.

502
Teranishi, Roy, Ron G. Buttery, and T.R. Mon. “Direct Vapor Analyses with Gas
Chromatography.” Annals of the New York Academy of Sciences 116 [“Recent Advances
in Odor: Theory, Measurement, and Control”] (July 1964): 583-9.

Teranishi, Roy, Ron Buttery, and Fereidoon Shahidi, eds. Flavor Chemistry: Trends and
Developments. American Chemical Society Symposium Series 388. Washington, D.C.:
ACS, 1989.

Teranishi, Roy, Phillip Issenberg, Irwin Hornstein, and Emily L. Wick. Flavor Research:
Principles and Techniques. New York: Marcel Dekker, 1971.

Teranishi, Roy, R.E. Lundin, and J.R. Scherer. “Analytical Technique.” In H.W. Schultz,
ed. Symposium on Foods: The Chemistry and Physiology of Flavors, 161-81. Westport,
CT: AVI, 1967.

Thompson, Emily. The Soundscape of Modernity: Architectural Acoustics and the

Culture of Listening in America. Cambridge: MIT Press, 2004.

Teil, Genevieve. "No Such Thing as Terroir? Objectivities and the Regimes of Existence
of Objects." Science, Technology, and Human Values 37.5 (2012): 578-505.

Teil, Genevieve and Antoine Hennion. “Discovering Quality or Performing Taste? A

Sociology of the Amateur.” In Mark Harvey, Andrew McMeekin, and Alan Warde, ed.
Qualities of Food, 19-37. Manchester: Manchester UP, 2004.

Todes, Daniel P. “Pavlov’s Physiology Factory.” Isis 88.2 (June 1997): 205-46.

Tracy, Sarah E. “Delicious: A History of Monosodium Glutamate and Umami, the Fifth
Taste Sensation.” PhD diss., University of Toronto, 2016.

503
Trout, G. Malcolm and Paul F. Sharp. “The Reliability of Flavor Judgments, with Special
reference to the Oxidized Flavor of Milk.” Cornell University Agricultural Experiment
Station Memoir 204, June 1937.

Troesken, Werner. “Exclusive Dealing and the Whiskey Trust,” Journal of Economic
History 58.3 (1998): 755-88.

Trubek, Amy. The Taste of Place: A Cultural Journey into Terroir. Oakland: University
of California Press, 2008.

Turin, Luca. The Secret of Scent. New York: Ecco, 2006.

Ulloa, Ana Maria, Josep Roca, and Heloise Vilaseca. “From Sensory Capacities to
Sensible Skills: Experimenting with El Celler de Can Roca.” Gastronomica: The Journal
of Critical Food Studies 17.2 (2017): 26-38.

U.S. Congress. House. Committee on the Judiciary. Report on the Whisky Trust
Investigation. 52d Cong., 2nd sess., 1893. H. Rep. 2601.

U.S. Congress. Senate. Committee on Manufactures. Adulteration of Food Products. 56th

Cong, 1st sess., 1900. S. Rep. 516.

U.S. Department of Agriculture. Standards of Purity for Food Products (Circular 19).
Washington D.C., June 26, 1906.

U.S. Federal Trade Commission. “Complaint No. 1199: Federal Trade Commission v.
The NuGrape Co. of America. Charge: Unfair Methods of Competition,” Annual Report
of the Federal Trade Commission for the Fiscal Year Ended June 30, 1925, Washington,
DC: Government Printing Office, 1925: 179-80

U.S. Federal Trade Commission. In the Matter of NuGrape Company of America. Docket
1576. Complaint, Feb 27 1929; Decision, May 19, 1931. 15 FTC.
504
U.S. Tariff Commission. Census of Dyes and Other Synthetic Organic Chemicals 1923.
Tariff Information Series 34. Washington, D.C.: Government Printing Office, 1924.

van Zwanenberg, Patrick and Erik Millstone. “Taste and Power: The Flavouring Industry
and Flavour Additive Regulation.” Science as Culture (2014): 1-28.

Veit, Helen Zoe. Modern Food, Moral Food: Self-Control, Science, and the Rise of
Modern American Eating in the Early Twentieth Century. Chapel Hill: UNC Press, 2013.

Vidal, Agusti. "New Comprehensive Training Method for Perfumery and Flavoring."
Perfumer & Flavorist 14:2 (March/April 1989): 25-44.

Vincenti, Walter. What Engineers Know and How They Know It: Analytical Studies From
Aeronautical History. Baltimore: Johns Hopkins University Press, 1990.

Vileisis, Ann. "Are Tomatoes Natural?" In Martin Reuss and Stephen H. Cutliffe, eds.,
The Illusory Boundary: Environment and Technology in History. Charlottesville: UVA
Press, 2010.

Vock, Manfred H. “Development of a Flavor at IFF, Planning, Creation, and

Commercialization.” In George Charalambous and George Inglett, eds., The Quality of
Foods and Beverages: Chemistry and Technology, Vol. 2, 198-209. New York:
Academic Press, 1981.

Walter, Erich. Manual for the Essence Industry. 1st ed. New York: John Wiley and
London: Chapman & Hall, 1916.

Wansink, Brian. “Changing Eating Habits on the Home Front: Lost Lessons from World
War II Research.” Journal of Public Policy and Marketing 21.1 (Spring 2002): 90-99.

505
Ware, Thomas M. So Little Soil… So Little Time: The Story of the International Minerals
& Chemical Corporation. Newcomen Address delivered at a National Dinner of the
Newcomen Society in North America, Chicago, May 25, 1967. New York: Newcomen
Society, 1967.

Warner-Jenkinson Co. Bottlers’ and Ice Cream Makers’ Handy Guide. St. Louis:
Warner-Jenkinson, 1921. A.W. Noling Collection, University of California Davis.

Warren, Craig B. and John P. Walradt, eds. Computers in Flavor and Fragrance
Research. ACS Symposium Series 261. Washington, D.C.: ACS, 1984.

Weir, R.B. “Distilling and Alcohol, 1870-1939.” Agricultural History Review 32.1
(1984): 49-62.

Wenneis, J.M. “Making Dry Flavors by Spraying.” American Perfumer (December

1955): 48-50.

“What Has Happened to Flavor?” Fortune 45 (April 1952): 130-3, 146-52.

White, Suzanne. “The Chemogastric Revolution and the Regulation of Food Chemicals.”
In Seymour H. Mauskopf, ed. Chemical Sciences in the Modern World, 322-55.
Philadelphia: University of Pennsylvania Press, 1993.

White Junod, Suzanne. “Food Standards in the United States: The Case of the Peanut
Butter and Jelly Sandwich.” In David F. Smith and Jim Phillips, eds. Food, Science,
Policy and Regulation in the Twentieth Century: International and Comparative
Perspectives, 167-88. New York: Routledge, 2013.

Wiley, Harvey. “Life in the Husk.” Good Housekeeping 63.3 (September 1916): 65-6.

Wiley, W.C. “Bendix Time-of-Flight Mass Spectrometer,” Science 124 (26 October
1956): 817-20.
506
Wiley, W.C. and I.H. McLaren. “Time-of-Flight Mass Spectrometer with Improved
Resolution.” Review of Scientific Instruments 26.12 (December 1955): 1150-7.

Willkie, H.F. and E.H. Scofield. “Some Factors Influencing Determination of Relative
Preferential Values of Distilled Alcoholic Beverages.” Institute of Food Technologists:
1941 Proceedings, 203-8. Champaign, IL: Garrard Press, 1941.

Wilkens, Walter F. and John D. Hartman. “An Electronic Analog for the Olfactory
Processes.” Journal of Food Science 29.3 (1964): 372-8.

Wilson, Bee. Swindled: The Dark History of Food Fraud, from Poisoned Candy to
Counterfeit Coffee. Princeton: Princeton UP, 2008.

Winter, Max. “Chromatography on Columns and Paper in the Study of the Volatile Fruit
Flavor of Strawberries.” In Arthur D. Little, Inc., Flavor Research and Food Acceptance,
290-301. New York: Reinhold, 1958.

Winton, A.L. and E.H. Berry. “The Chemical Composition of Authentic Vanilla Extracts,
Together with Analytical Methods and Limits of Constants.” In Harvey Wiley and A.L,
Pierce, eds., Proceedings of the 28th Annual Convention of the Association of Official
Agricultural Chemists, USDA Bulletin 152, 146-58. Washington: Government Printing
Office, 1912.

Wood, Donna J. “The Strategic Use of Public Policy: Business Support for the 1906 Food
and Drug Act.” Business History Review 59 (Autumn 1985): 403-32.

Woloson, Wendy. Refined Tastes: Sugar, Confectionery, and Consumers in Nineteenth-

Century America. Baltimore: Johns Hopkins University Press, 2002.

507
Wyllie, S.G., S. Alves, M. Filsoof, and W.G. Jennings. “Headspace Sampling: Use and
Abuse.” In George Charalambous, ed. Analysis of Foods and Beverages: Headspace
Techniques, 1-15. New York: Academic Press, 1978.

Yoder, Robert. “The Man with the Million-Dollar Nose.” Saturday Evening Post 224.13
(September 29, 1951): 27, 110-12.

Young, James Harvey. Pure Food: Securing the Federal Food and Drugs Act of 1906.
Princeton: Princeton UP, 1989.

———. “The Government and the Consumer: Evolution of Food and Drug Laws and the
1938 Food, Drug, and Cosmetic Act.” Journal of Public Law 13 (1964): 197-203.

Young, Warren R. “Cracking the Secret Riddle of Flavor.” Life (November 23, 1962):
110-23.

Zeide, Anna. “In Cans We Trust: Food, Consumers, and Expertise in Twentieth-Century
America.” PhD Diss., University of Wisconsin-Madison, 2014.

Zimmerman, Bergemeister, and Putnam. “Group Study of Effect of Glutamic Acid on

Mental Functioning in Children and Adolescents.” Arch. Neurol. Psychiat. 56 (1946):
489-506.

508