Aristo Tile As A Poet

ARISTOTLE
AS POET
This page intentionally left blank
ARISTOTLE
AS POET
The Song for Hermias and
Its Contexts
Andrew Ford
3
2011
3
Oxford University Press, Inc., publishes works that further
Oxford Universitys objective of excellence
in research, scholarship, and education.
Oxford New York
Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto
With ofces in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam
Copyright 2011 by Oxford University Press
Published by Oxford University Press, Inc.
198 Madison Avenue, New York, New York 10016
www.oup.com
Oxford is a registered trademark of Oxford University Press
All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise,
without the prior permission of Oxford University Press.
Library of Congress Cataloging-in-Publication Data
Ford, Andrew Laughlin.
Aristotle as poet : the song for Hermias and its contexts / Andrew L. Ford
p. cm.
ISBN 978-0-19-973329-3
1. Aristotle. 2. PoetryEarly works to 1800. I. Title.
PN1040.A53F67 2010
881
.01dc22 2010000006
1 3 5 7 9 8 6 4 2
Printed in the United States of America
on acid-free paper
For Annabelle and Viviane
quand jai t pre . . . Balzac
C O N T E N T S
Abbreviations xix
1. The Text 1
Aristotle: The Song for Hermias 1
Sources and First Reading 3
2. History and Context 9
Deconstructing Atarneus: Questions of Method 10
Constructing Hermias: The Erythraean Inscription 17
The End of Hermias: Theopompuss Letter to Philip 21
3. Performance and Occasion 27
Commemorative Epigrams: Aristotle and Simonides 29
Book Epigrams: Theocritus of Chios 35
Texts and Things: Herodotus on Hermotimus 41
4. Performance and Context 45
Witnesses: Callisthenes Hermias 48
vii
viii Contents
Sources: Hermippuss On Aristotle 54
Authenticity: Aristotles Apology 60
5. Genres of Poetry 69
Lyric Genres from Plato to Alexandria 71
Impious Song: The Paean to Lysander 80
Paean, Hymn, Skolion? 86
6. Kinds of Hymn 91
Hymnic Form: Ariphrons Paean to Health 91
Hymnic Flexibility: Pindars Fourteenth Olympic Ode 97
Hymns in Hexameters: Homer and Aristotle 105
7. Ethos 113
Ethos in Debate: An Attic Skolion and a Poem
by Sappho 114
Ethos in Protreptic: Aristotles Hymn to Hermias,
vv. 18 121
Ethos in Epiphany: Immortal Virtue in Sophocles
Philoctetes 127
8. Reading 137
Troping: o. uo0o in Euripides and Bacchylides 138
Mythologizing: Hymn to Hermias, vv. 916 144
Immortalizing: Hymn to Hermias, vv. 1721 147
9. Endurance 157
Memorial: Aristotles Elegiacs to Eudemus 160
Survival: A Letter from Plato 166
NOTES 173
BI BLI OGRAPHY 217
GENERAL I NDEX 233
I NDEX OF PASSAGES DI SCUSSED 239
P R E F AC E
People are often surprised to hear that Aristotle wrote poetry,
naturally thinking of him in the rst instance as a philosopher
and indeed as one of the greatest thinkers inthe ancient world. In
fact, Aristotle composed enough poetry to ll two papyrus rolls
in the ancient collections of his works, for it was not unusual
that a well-educated gentleman of his day should be able to come
up with a verse or song to grace special occasions. What is very
surprising is the story toldabout one of his poems, for the sources
that preserve the text also tell us that it came near to costing
the philosopher his life. This lyric, one of only two to survive
complete, will be the central thread in the study that follows,
which combines a close reading of that work with an attempt
to understand its remarkable reception. Though very little of
Aristotles poetic output survives, I hope thereby to cast further
light onhis relationto the Greek lyric traditionandto the musical
culture of the later fourth century.
The poemstrictly speaking, the lyric to a brief song
commemorates Hermias of Atarneus, ruler of a small principality
in the northeast corner of the Aegean. In the late 340s BCE,
ix
x Pref ace
Hermias, who had been Aristotles student, patron, and father-
in-law, became entangled in the tensions between the Persian
Empire to his east and a rising Macedon to his west. When
Hermias was captured by the Persian king and put to death
around 341, Aristotle composed an ode in praise of his friends
character. My original aim had been to call attention to this text,
whichis relatively little discussed today, and to place it withinthe
Greek literary tradition. This I have done mainly in the latter part
of the book. But my project expanded as I found myself drawn
into a very old debate about which genre the poem belonged to:
strictly speaking, was this a dirge for Hermias, a eulogy, a hymn,
a drinking song, or some combination of these? The question
may sound academic, except that it seems to have meant enough
to some of Aristotles contemporaries that they were willing to
threaten him with trial and execution on account of it. This
episode is usually thought to have occurred after the death of
Alexander the Great in 323 BCE when a wave of anti-Macedonian
sentiment swept through Greece. In Athens, where Aristotle was
teaching, his long-standing relations with the regime (his father
had beenthe physicianof Alexanders grandfather and he himself
had been the young kings tutor) would have been a liability.
Political agitators, we are told, began to accuse the foreign-
born philosopher (from Stagira in northern Greece) of being too
sympathetic to tyrants, and the song for Hermias was brought
forth as a prime piece of evidence. According to the version
preserved in Athenaeus (who wrote around ve centuries after
the event), a religious ofcial of Demeters Eleusinian mysteries
teamed up with an Athenian politician to charge that Aristotles
song, ostensibly a lament for his friend, was actually a kind of
hymn implying that Hermias had become a god. This was impiety
in itself, the priest might urge, and the politician could add that
such a song revealed a person unsympathetic to Athenian ideals
of democratic equality. Both could buttress their case by recalling
Pref ace xi
the trial of Socrates, who had been put to death on charges
that included impiety in 399. We are not told whether the case
against Aristotle ever came to trial, but it seems that the threat
of legal action was real enough, for he left Athens for good in 323,
reportedly explaining that he was leaving, lest the Athenians sin
twice against Philosophy.
1
A modern reader may well ask if such a story, like Aristotles
bon mot, is too good to be true. Due scrutiny of the sources will
follow, but it seemed to me that a reading of Aristotles poem
that put this information aside as unreliable or irrelevant could
hardly be called complete. Indeed, it seemed to me impossible
even to construe the text without considering these matters,
for they bear directly on our view of what the poem is trying
to do. The story of the trial also raises questions about Greek
literary culture in Aristotles time. Is it credible that an Athenian
jurywhich typically included hundreds of people selected at
randomshould have cared whether the song was a hymn or
not? What was the prosecution thinking in launching such an
attack, and how did they understand the text? We may also
wonder about the relation between Aristotle the literary theorist
and the wider public: Howis it possible that he of all people could
have opened himself up to a charge of misapplying generic rules?
The Prince of Philosophers was, after all, alsothe Prince of Critics,
and his lectures on Poetics had set out with exemplary clarity
the system of literary genres on which most ancient and much
modern literary criticism is based. He least of anyone should
have erred in the question of what was a hymn and what wasnt.
Finally, if we doubt the story of the trial, ought we also to reject
the poem as a later fabrication?
Such reections led me to include in my literary analysis a
wider view that took in the songs contexts, including its early
reception and transmission. Coming to terms with the words
seemed to require understanding the circumstances under which
xii Pref ace
they were composed, presumably shortly after 341; and some
idea of how such songs might circulate seemed necessary to
understand how a personal lyric could have become a public
scandal, as it seems to have in 323. In addition, once the question
of the songs authenticity arises, we must give some thought toits
later transmission, in particular asking about the circumstances
under which a genuine song of Aristotles might have been
recorded and preserved. The result has been a rather extended
piece of exegesis, but one that I hope is justied both by the
intrinsic interest of the song and by the interpretative issues it
raises; one of the pleasures in reading old poems is that the basic
process of making sense of the words can provide heightened
examples of the choices that arise in literary reading generally.
Aristotles song for Hermias ought in fact to be recognized
as a landmark in the history of Greek literature, because it is
one of the very rst lyric poems for which we have substantial
evidencein some cases going back to contemporariesfor how
and where it was composed, performed, and received. We usually
read early Greek poems knowing next to nothing about their
authors andnothing about the people towhomthey refer (except,
of course, what the poems themselves tell us). But Aristotles
song comes down to us along with considerable information
about its author, subject, and the responses of early audiences;
we thus have an opportunity to supplement a reading we might
give of it as an isolated, authorless fragment with one that can
place it rather precisely withinthe political, religious, andmusical
cultures of the late classical age. Acknowledging that this agenda
will draw me into areas beyond my expertise, and that literary
theory has made the old tactic of putting texts in historical
context a less than straightforward affair, I nonetheless hope
that this attempt to see a lyric in the round, as it were, may be a
useful case study for the more frequent occasions when evidence
is lacking to trace a poems background in detail.
Pref ace xiii
In my literary interpretations I have been guided by the rst
question that comes up when people hear that Aristotle wrote
poetry: Really? Whats it like? I know of no other way to say
what the song is like than to set it beside other poems in the
tradition, both those that closely resemble itAriphrons lyric
in praise of health is the best-known exampleand those that
bring out its distinctive qualities by contrast. I end up comparing
a far wider range of texts than earlier scholars have citedfrom
Sapphic stanzas through Sophoclean trimeters, taking in both
high andpopular verse andprose genres as wellbut I submit
eachas illuminating specic aspects of Aristotles text while being
worth a fresh look in its own right.
The book is organized to place Aristotles poem rst, so that
readers can come back to it repeatedly, as the text does. It is
followed by my translation and a brief run-through of its con-
tents, a rst reading designed to register its principal themes and
tropes as they would have unfolded before an ancient audience.
I then turn to the evidence for Hermias and his relations with
Aristotle and consider howit may affect our interpretation of the
poem. Chapter 2 walks the story back to its sources very carefully,
for the historical texts we use to understand a literary text
are rarely straightforward and have contexts themselves to be
considered. Chapter 3 takes up Aristotles only other poem to
survive complete, an epigram he composed about Hermiass
death, which is reported to have been inscribed on a monument
in Delphi. This and other related epigrams will make us confront
the worrying gap that may arise between textual accounts of
an event or object and the posited event or object itself. Despite
these complexities, and despite some denitely spurious sources,
I conclude in chapter 4 that we should accept Aristotles song for
Hermias as authentic, even though attempts to specify a single
original performative context remain speculative. At this point
we will turn from the songs contexts to the song, which I hold
xiv Pref ace
is best approached by following Aristotles accusers and asking:
What is its genre? The rst of the next two chapters sketches the
traditional Greek system for recognizing forms of lyric (a topic
that deserves more attention than it has received); the second
argues for the exibility and negotiability of a songs genre in
actual practice. My interest ingenre is not judicialto determine
the literary category to which the work properly belongs
but historical, looking at genres as epitomes of cultural norms
and observing how they inuence the meanings of songs and
govern their circulation through society. Only if we appreciate
the close connection between Greek conceptions of genre and
the occasions of social life can we understand why Aristotles
accusers could have expected to arouse a jurys indignation at
this alleged hymn. Chapter 7 begins a re-reading of the Hermias
song, now seen against the panorama of Greek song types that
constituted Aristotles literary horizon. Here and in chapter 8
we will be in a position to see this poem and others like it in a
new way. Even if the sources for the story turn out not to be
trustworthy in all details, they can bring the text into focus by
calling attention to aspects of it that provoked divergent and
apparently heated interpretations. And even if some ancient
readers seem to have misconstrued the text deliberately, this
bizarre episode inthe history of its interpretationis animportant
reminder that we cannot wish away our historical distance and
see the work stripped of all partisan construal and temporal
obfuscation. Indeed, we cannot draw a sharp line separating
modern understandings of Aristotles song from the chain of
its ancient receptions, for the lyric has only reached us by being
recordedagainandagain, eachtime under a particular conception
of its meaning and value. That ongoing process of reception and
interpretation is considered in the nal chapter.
I should say here that I regard the perspectives I bring to
bear as complementary without pretending that they combine
Pref ace xv
to reveal the poems nal and denitive meaning. It would be
nave to claim that putting a poem in its historical context
is sufcient to determine its correct interpretation, however
this be dened. Nor is my aim to understand the poem as
an expression of Aristotles psychology or to discover in it his
personal response to events in his life. For me, the value of
exploring the poems history and the responses it drew from
its audiences is that they enrich our perception of it as a
specic cultural artifact, as a work of art produced in a unique
time and place.
2
Historicizing also allows us to read the songs
language against the language of the time and so to catch its
contemporary accents, for all poems begin as contemporary
poetry. The payoff for reading the lyric in light of its contexts is
a more ne-tuned appreciation for its verbal dynamics. It will be
seen that the song modulates through a variety of lyric styles
and that it shifts the picture it gives of itself as it unfolds.
For such reasons I will decline to pin it down in the end to a
single historical context or a single genre; what seemed more
important was to follow its changing meanings throughout its
dynamic career, from the time it arose among the circle of
Hermiass intimates until it passed, after a contentious entry into
the public sphere, into antiquarian compilations such as that of
Athenaeus, where we can read it today. What follows, then, is less
an exhaustive historical analysis or nal literary interpretation
than notes toward a biography of a song.
A poem by the master of those who know can hardly be
expected to have passed unnoticed among scholars of Greek
literature, and my debts to earlier treatments of this poem
ought to be acknowledged. It will be clear from my discussion
that I have found especially useful Wilamowitz (1893), Bowra
(1938), Jaeger (1948), and Renehan (1982), though I have not
agreed with them on all points. What I have tried to add to
these indispensable studies is a more constant awareness of
xvi Pref ace
Aristotles poem as a piece for performance, as a song (a .o
or e otc y), which is what the Greeks would have called it. We
have Aristotles words because they were written down and then
read, re-read, and re-copied; but their form shows that they
were made as part of a song, a melodic work designed to be
performed, re-performed, and remembered. (Accordingly, I use
the term poem in what follows when considering Aristotles lyric
in its function as a written text, and song when thinking of it as a
performance piece.) Keeping this fact in mind adds important
dimensions to our understanding of the words that remain
and of the meanings they took on through history. In sorting
through the traditions about Aristotle, I have learned much from
Drings superb collection, in particular what value there is in
source criticism properly done. This old approach was out of
fashion when I was in school, in part because it could be seem
naively positivistic to seek to track down the sources of great
books. But the learning and intelligence displayed in such works
as Wormell (1935) on the tradition about Hermias, Bollanse
(2001) on Hermippus, and Harding (2006) on Didymus, along
with, of course, the fundamental workof Wilamowitz andJacoby,
command respect; if we cannot aspire to recover all or even the
key sources that lie behind a given work, such scholarship can be
of great help in recognizing the ways in which texts and songs
were used, passed around, and preserved in the ancient world.
Finally, I would not have ventured so far into elds of
scholarship in which I was little more than a novice if I had
not known I could rely on learned friends for help. Of those
who read and commented on this book in manuscript, I thank
rst Douglas Lane Patey, my long-standing ideal reader on these
matters. Nearby is M. B. E. Smith, whose bracing criticisms
led me to omit many weak arguments and strengthen what
I could not omit. My colleague Michael Attyah Flower lavished
on me his deep and subtle knowledge of Greek history and
Pref ace xvii
its sources. It was a pleasure to be able to impose on the
kindness of Marco Fantuzzi, whose imaginative and learned
criticismI have long admired. Vayos Liapis was similarly generous
with his detailed knowledge and tactful sense of poetry, as was
the versatile and thoughtful Marek Wecowski. Last but not
least, Pauline LeVens valuable suggestions were marked by the
same originality and independence she displayed in the ne
dissertation she wrote with me on fourth-century Greek lyric.
My manuscript was supported at a crucial stage by Stefan Vranka
of Oxford University Press; he secured helpful readers reports
and contributed many wise suggestions on repeated readings of
the manuscript. Thanks to them all, my text has been purged
of errors, inaccuracies, and infelicities of expression, while being
enriched with references to primary and secondary literature.
I thank themmost warmly and avowwith equal warmth that the
defects that remain are mine alone.
A B B R E V I AT I O N S
CA J. U. Powell, ed. Collectanea Alexandrina. Oxford:
Clarendon Press, 1925.
CEG P. A. Hansen, ed. Carmina epigraphica Graeca
saeculorum vii-v a. Chr. n. (Texte und Kommentare xii).
Berlin and New York: De Gruyter, 1983.
DK H. Diels and W. Kranz, eds. Die Fragmente der
Vorsokratiker, edition. 3 vols. Berlin: Weidmann, 1952.
DL Diogenes Laertius, Lives of Eminent Philosophers (Vitae
philosophorum), ed. M. Marcovich and H. Grtner.
Stuttgart and Leipzig: Teubner, 1999.
FGrH F. Jacoby, Fragmente der Griechischen Historiker.
Leiden: Brill, 1923.
Fr(r). fragment(s) in the edition specied
IEG M. L. West, ed. Iambi et elegi Graeci, 2nd edition. 2 vols.
Oxford: Clarendon Press, 19891992.
LSJ H. G. Liddell , R. Scott, and H. S. Jones, A Greek English
Lexicon. 9th ed. Oxford: Clarendon Press, 19251940.
PMG D. Page, ed. Poetae melici Graeci. Oxford: Clarendon
Press, 1962.
xix
xx Abbrevi ati ons
PMGF M. Davies, ed. Poetarum melicorum Graecorum frag-
menta, vol. 1. Oxford: Clarendon Press, 1991.
RE A. Pauly, G. Wissowa, W. Kroll, eds. Real-Encyclopdie
der classischen Altertumswissencschaft. Stuttgart: J. B.
Metzler, 1893.
Rose V. Rose, ed. Aristotelis qui ferebantur librorum
fragmenta, 3rd ed. Leipzig: Teubner, 1886.
SH H. Lloyd-Jones and P. Parsons, eds. Supplementum
Hellenisticum. Berlin: W. de Gruyter, 1983.
Test(t). testimonium (-ia) in the edition specied
TrGF B. Snell, R. Kannicht, and S. Radt, eds. Tragicorum
GraecorumFragmenta, 5 vols. Gttingen: Vandenhoeck
& Ruprecht, 19712004.
Wehrli Fritz Wehrli, ed. Die Schule des Aristoteles, 2nd ed.
10 vols. and 2 suppl. Basel: Schwabe, 19671974.
Chapter 1
The Text
Aristotle: The Song for Hermias
e ,x` e o. uo0 y vt p,ox t
.
,
0 y,ee x e..toxov p t
.
,
o e ,t, e,0 v, o, e
xe`t 0ev tv e.x
v
\
E.. ect xo
xe`t vou x. yvet e.,o` u e x eevxe 5
xo tov
`t , ve p e..t
xe,v t oe0 evexov ,uoo u x x, too
xe`t yov v e.exeuy yxot 0' u
\
vou.
o u c'
\
vxv xe`t o
\
c to
\
H,ex. y A yce x xo u,ot 10
..' e v x.eoev
v
\
,yot
o` ev [e y, u]ovx c uvetv.
oo t x 0ot At. ` u At
e x' A tceo cou

y .0ov
o e c'
\
vxv t. tou o, e Axe,v o 15
vx,oo e . tou y,ov eu y e.

1
2 A R I S T O T L E A S P O E T
xoty` e, e o tcto
,yot,
e 0 evexv x tv eu yoouot Mo uoet,
Mveoo uve 0 uyex,, Lt
v tou o pe eu
ou 20
oet t. te x y ,e ppe tou.
O Virtue of great toil for humankind,
the fairest quarry in life,
for your shape, maiden,
even to die is an enviable fate in Greece
and to endure pains, consuming, unrelenting; 5
such is the fruit you cast into hearts,
immortal-like, better than gold,
than breeding, than sleep with its soft beams.
For your sake even that godly
Heracles and the sons of Leda 10
endured much in their exploits
on the track [?] of your power;
in longing for you Achilles and
Ajax entered the house of Hades;
for the sake of your dear shape, Atarneus 15
nursling left the rays of the sun bereft.
Hence he will be a subject of song on account of his exploits,
and the Muses will grow him into immortality,
those daughters of Memory,
making grow reverence for Zeus, 20
god of guest-friends, and the rewards of steadfast
friendship.
The Text 3
Sources and First Reading
The Greek text above is taken from the standard modern edition
of Greek lyric poetry by Denys Page (with one supplement
at v. 12).
1
It is based on three ancient sources: the oldest is
a commentary from the second half of the rst century BCE
on a speech attributed to Demosthenes; the commentary was
composed by Didymus, an extremely productive and well-read
scholar who worked in Alexandria.
2
In elucidating a speech (most
probably the Fourth Philippic, 10.32) that alluded to Hermias,
Didymus recalls certainpoints inhis career and quotes Aristotles
song. The fullest account of the incidents surrounding the poem
is given by Athenaeus, writing around the beginning of the third
century CE. He quotes it near the end of his Learned Banqueters
(Douglas Olsens translation of Deipnosophistai), a long ctional
account of an impossibly brilliant dinner conversation that
Athenaeus composed by pillaging earlier works of antiquarian
scholarship (Book 15, 696A-697B). Sometime later in the third
century, Diogenes Laertius also quoted the poem in the account
of Aristotle that he composed for his Lives of the Philosophers.
These sources differ in small ways, and the question of what
sources they used will be taken up in due course.
Page is responsible for the colometry of the text above,
the ragged right- and left-hand margins meant to demarcate the
musical phrases of the original; in the Didymus papyrus, the
poem is written en bloc, as lyric poetry was often transcribed in
the Hellenistic age.
3
The patterning of short and long syllables
shows that the song consists of a single stanza, composed in a
fairly common kind of rhythm for which modern scholars have
devised the term dactylo-epitrite. We do not know enough to say
whether dactylo-epitrites were associated with a specic range of
emotions or themes, but a leading expert in Greek metrics has
noted that in the fourth century the rhythm was characteristic
of what he terms educated bourgeois lyric.
4
As a performance
piece, it shouldbe conceptualizedas a short, single-stanza song
dactylo-epitrites were always sungwitha melody unique tothat
song. It is dogma that goes back to the ancients that this meter
implies that the song was for choral performance; the dialect suits
this possibility, since it has the light Doric coloration that was
conventional in choral odes. But I shall argue below that there
is no reason the song could not have been performed or at least
re-performed as a solo piece.
My rather literal translation follows the line numbers and
punctuation of Pages text, though for convenience I have
inserted spaces in the translation to signal the songs three main
conceptual periods. As an introduction to the poem, let us follow
Aristotles words and themes as they would have unfolded before
an ancient audience, bearing in mind that Greek audiences were
familiar with a vast body of songs, many known by heart, and
were capable of delighting in new variations on old themes. This
is to postpone a synoptic examination of the works structure,
giving it a rst hearing, so to speak, as an event that would have
played out in time.
5
No hearing occurs without expectations, and so the question
of howthe song soundedtoanancient audience raises for the rst
time the question of genre. Prima facie, Aristotles song begins
very muchlike a hymn, whichis tosay that its formal components
can be paralleled in innumerable Greek songs composed to praise
a divinity.
6
Hymnic style begins with the very rst word, areta,
a vocative that at once invokes and personies virtue. (I take
over, with some misgiving, this traditional translation of areta
because it is less awkward than a more precise rendering would
be, such as human excellence, which is the way the word is
often rendered in Aristotles ethical treatises. What is crucial
to bear in mind is that Greek virtue has not the moral or
sexual connotations the word later acquired from its use in
The Text 5
Christian literature. In the context of a praise song, the best
denition of areta may be that of Russell and Wilson, who
base it on Aristotles rhetorical and ethical works: aret is the
power to provide and protect good things, and to confer great
benets.)
7
In form, areta cues us to expect the Doric dialect
(slightly different from the pronunciation aret in Athens); that
is, the following speech will not be everyday, unmarked talk.
Songs with touches of Doric were characteristic of Greek cult
hymns, though all we can infer fromthis formal detail is that the
song presents itself as suitable to be sung by a choir and at a cult
site, not that it was ever actually put to use in that way.
The hymnic rhetoric continues as the vocative is followed, as
regularly in hymns, by an epithet: Aristotle calls upon not any
form of Virtue but the one characterized by great struggling,
literally of many toils. He then gets down to the main business
of the hymn, praise of the divinity, for as in many Greek hymns,
the principle here is do ut des: the poet gives praise so that the
god may be gracious in return.
8
Aristotle lls the rst sixteen
verses with praise of Areta as the most desirable object of human
aspiration. This praise is articulated into two parts, with the
end of the rst movement marked by a priamel in vv. 69,
a gure of speech common in praise poetry that lists items
in ranked order to set off the merits of the object of praise.
9
Virtue is thus presented as more desirable than wealth or noble
ancestry, more alluring thanphysical pleasure. Aristotles version
of this gure is made eminently apprehensible by being shaped
as a tricolon crescendo in which the third element is given
capping force by being expanded. (As in, our Lives, our Fortunes,
and our sacred Honor.) Here the word translated with soft
beams, a sesquipedalian and archaic-sounding epithet, rounds
out the rst conceptual period with sleep and at the same
time brings the performer to a metrical pause in which to take
a breath.
The second period (vv. 916) proves the claims of the rst
by adducing a series of admirable gures who spent their lives
in the service of Virtue. These exemplary heroes are organized
chronologicallyHeracles and the Disokouroi did their deeds
before the Trojan War in which Achilles and Ajax won gloryand
metaphysically: Heracles is Zeuss son, as were Castor and Pollux,
the male offspring of his intercourse withLeda; of the Trojanpair,
Achilles had the goddess Thetis for a mother, while Ajax was fully
mortal. We are thus thinking in terms of time and noble ancestry
when a third example comes up, which we may suspect will be
the last, since the number three proved to mean closure in the
priamel. In all respects the third item (v. 15) surprises: Aristotle
names only one gure, and he leaps from heroes of old to a
contemporary and friend, an abrupt move from muthos to logos,
from more than mortal gures to the nursling of Atarneus.
Things conform a little less strictly to the hymnic program
as our song begins its third period after v. 16. In the poems
argument, Hermiass devotion to Areta is the culminating proof
of her worth and so belongs to the hymnic agenda of praise.
At the same timeand here one begins to see an opening for
Aristotles criticsAristotle is also praising Hermias, implying
that his travails, which are put ona par with the mythical exploits
of heroes, make himworthy to be remembered like them. And so
the third movement inaugurates a shift in focus, as what began
sounding like a hymn to a divine principle modulates to sound
like a song of praise directed at a fellow mortal. This shift in
focus is arguably also a shift in genre, since in Aristotles day
there was a long-standing and widely respected tradition that
outstanding human achievement deserved to be celebrated in
song, but that songs for mortals should keep their praise at a level
below that which is offered to the gods. The distinction between
hymns in praise of gods and songs for mortals was preserved in
the popular terminology of Aristotles day, which called a song
The Text 7
celebrating human achievement an enkmion, literally a revel
song sung by a band of cheering young men on a carouse. The
Greeks knewwell that revels can become rowdyFor hes a jolly
good fellow can quickly turn into as nobody will deny!and
so encomia were expected to temper praise with pious warnings
against excess. An illustration is Pindars victory songs, which in
their day were usually called enkmia (as will be seeninchapter 5):
almost any Pindaric epinicianwill showhowinsistently he blends
glorication of athletic triumph with warnings that the happy
victor should not be misled by success to think himself more
than mortal.
In turning to praise Hermias, therefore, Aristotle entered
into the unwritten encomiastic contract. And indeed in v. 17,
he resorts to a common encomiastic topos when he declares
that Hermiass extraordinary qualities make him worthy to be
celebrated in song, adding that the Muses, the daughters of
Memory, will make Hermias immortal (v. 18). Once again,
Aristotles enemies may prick up their ears, but there is nothing
impious here. If Aristotle offers his friend the highest form of
praise a mortal could hope for, it is also quite a conventional
compliment, even stereotypical by this time; there is nothing
novel in the idea of achieving immortality through song, which
already had a long tradition behind it when Homer spoke of
Achilles winning unwithering fame in the Iliad (9.189). The
rhythmof the poemunderscores Aristotles appeal to traditional
ideas of poetic immortalization, for verses 13-14, describing
Hermippus epic predecessors, come near to a dactylic hexameter,
the canonical meter of epic.
Aristotles song contains a minor surprise, since many (though
not all) hymns would close with a petition of some kind. This
ostensible hymn, however, includes no prayer, unless its closing
expectation that Hermiass excellence will be remembered is
taken as a request that he meet his due reward. At the end of
the song Aristotles rhetoric becomes a little involuted and may
seem overwrought, and when the Muses celebration of Hermias
turns out to be a song increasing the honor of Zeus we may lose
sight of whether we are praising Virtue, Hermias, or Zeus. Later
I will argue that Aristotle was not losing control here but was
deliberately and conscientiously pushing poetic forms to a limit.
For now, it will be agreed that at no point does he explicitly
say that Hermias has or will become a god. Though, as we will
see, some of the heroes mentioned managed in some stories to
overcome death, Aristotle is clear that the immortality Hermias
will enjoy will be inpoetry, not onOlympus. Wilamowitz saidthat
only a calumniator could call this a paean to Hermias, and almost
all modern commentators agree that the charge of impiety was
baseless.
10
To the extent that the poemis a hymn, it is a paean to
human excellence personied, which we will see was a perfectly
conventional conceit. Yet a closer reading will show that things
are not altogether simple and that the song does contain a sort of
hymn to Hermias hovering just beneath the surface. The poem
is more artful in its stance than has been shown so far, but this
must be made to appear, paradoxically, by turning away fromthe
text for a moment and thinking about the culture from which
it came.
Chapter 2
History and Context
Two prose texts about Hermias from this period remind us that
Aristotles neat little piece had a public: one is an inscription
recording a treaty concluded in Hermiass name, the other a
contemptuous dismissal of him by a contemporary observer
of the political scene. These diametrically opposed sketches of
Hermias provide a background that can illuminate Aristotles
words, but show as well that reconstructing the historical
context of a song is not simple: it is clear that the texts we
use to reconstruct the context of a poem need themselves
to be contextualized. Even when we can put our hands on
tangible rsthand traces of the historical actorsas in Hermiass
inscriptionwe are not in touch with rock-solid reality but
with partisan representations that need situating and often
decoding. Certain postmodern strains in current criticism would
accordingly suggest that we regard Hermias as, so to speak,
a gure of speech, a verbal sign that we decode by referring
to other signs without ever getting to the truth of things.
The force of such arguments must be acknowledged: the people
referred to by our texts are indeed known to us only through a
9
concatenation of other texts, and it is hard to leap the gap from
words to reality. But I will argue in the following section that to
concede that context is not something easily reconstructed from
texts does not require reducing the people in the story to the
status of signieds or regarding their doings and sufferings as
merely linguistic interactions in the texts recalling them.
1
Deconstructing Atarneus: Questions
of Method
Methodological difculties in contextualizing works of literature
are particularly acute for readers of early Greek poetry because
our historical knowledge is so thin. A Latinist colleague asks me
with only partly feigned exasperationwhy, whenyou place before
Hellenists one of their subtle lyric texts, they start spinning out a
historical novel about how it was composed and performed
matters that are usually impossible to know for certain and
arguably irrelevant to an appreciation of the work as a poem.
One might reply that we are only following the lead of our Greek
texts, which typically refer to the contexts and occasions of
their performance, and of the Greek scholarly tradition, which
typically dened kinds of lyric according to the occasions at
which they might be performed. But it remains true that after
the poststructuralist criticism of the 1970s and 1980s we can
no longer speak of historical context as if it were something
straightforwardly out there behind the texts and easy to reach.
The problems for readers of ancient literature are really two:
one is pragmatic, the result of the fact that historical evidence is
usually rare and is very often contaminated by the texts it would
be used to explain; the other is theoretical, having to do with
postmodern objections to the way the relation between text and
context is conceived. I think readers of Greek poetry need not
Hi story and Context 11
be defeated by the second set of objections and that we can, and
should, cope with the rst.
A simple, I hope not too simple, way to bring out the
theoretical problems with adducing the historical context of a
poem is to pose the question: What in Aristotles text directs us
to look away fromit for information? There is just one place that
will give pause to readers who nd contextualization problematic
or irrelevantthe phrase Atarneus nursling in v. 15. The
expression is a cipher without some reference to a historically
specic placea town that was situated on the northwestern
coast of Turkey, near the modern town of Dikili. Only if we know
this can we know, having consulted other texts, that Aristotle
refers to Hermias, who ruled Atarneus from around the middle
of the fourth century. The little puzzle is easily enough solved,
but solved it must be because this verse is the only time that
Hermias is named in the poem, a poem whose purpose seems to
be to celebrate him.
If it is agreed that the phrase nursling of Atarneus sends
us outside the text, it must be conceded that appealing from
text to context is easier said than done. Over recent decades,
the traditional practice of reading literary works in relation to
their historical contexts has been regarded as highly problematic.
A main basis for such reservations derives from principles of
structural linguistics: reference works not by words pointing to
things in the world but by one linguistic signier directing us
to other signiers, which send us to other signiers in turn.
Atarneus sends us to Hermias, but we only know who Hermias
was from historical documents, that is, from more texts. The
impasse that ensues is epitomized in the slogan, taken over from
Jacques Derrida, that there is no outside the text, by which it
is meant that there is never a leaping of the gap between sign
and referent.
2
Note that what drives Aristotles reader to think
of fourth-century history is not simply the fact that Atarneus
and Hermias are proper nounsfor so are Heracles, Achilles,
and Zeus. These names also set the reader a challenge insofar as
they are not words to be found in dictionaries; but mythological
gures do have a denition of a sort, that is, a more or less
stable mythology that determines whether we regard Aristotles
use of them as correct or solecistic, as plausible or paradoxical.
Names like Zeus and Achilles thus show that proper names
can sometimes function like ordinary words and need not call us
to a specic person or point in history. From a poststructuralist
perspective, it is hard to see that Hermias and Atarneus are much
different; just as we know Heracles only from Homer, Euripides,
Ovid, and so on, the historical Hermias for a modern reader
may be no more than the nexus of a number of more or less
convergent stories inDidymus, Athenaeus, Diogenes, and so on.
3
Mythological personages thus provide an apt illustration of the
view that the person behind a proper name is to be found not in
historical fact but intextinthe sumtotal of the things that have
been said under that name and found acceptable and memorable.
Aristotle was free, then, to bring Atarneus into his poem, even as
he was free to bring in mythological gures. But history, as Mae
West might have said, had nothing to do with it.
Postmodern reading holds further that reference in texts is
not only inescapably intralinguistic but also endless, so that we
create meaning only by arbitrarily breaking into this sign-to-
sign relay. It happens that our phrase can illustrate this point
as well. We hear from Himerius of Bithynia, a public orator
of Roman Imperial times, that Atarneus was the name of a
king of Mysia who founded the city (Oration 40.40-1 Colonna).
A deconstructionist would be happy to accept this testimony
that the one historical name in our poem, Atarneus, properly
belongs not to the city but to its founder, and that it is only
the linguistic process of metonymy, a process of transferring
meaning, that xes the word onto the physical place as its true
and proper name. Nor would such a reading be discomted
by the fact that the story could be told the other way round: a
historian experienced in foundation legends will read Himerius
suspecting that the toponymcame rst and suggested a name for
the founding king when such a gure was wanted. As often, we
are not in a position to know. Behind one name lies another, and
behind that name lies another still. The little riddle in Atarneus
nursling turns out to be hard to solve after all, and the would-be
historicist reader ends up reading like a postmodern, going from
text to text.
The argument can become metaphysical, but pragmatically
speaking there is no denying that the word Atarneus now has
purely linguistic contentthe city no longer survives, having
been abandoned a few centuries after Hermias (or so at least we
are informed by another text, Pausaniass Description of Greece
7.2.11). Deconstruction compensates for this rather bleak (and,
after a while, repetitive) insight by usefully highlighting the ways
by which texts create an illusion of extra-textual realities out
of purely linguistic resources. In this, it can help the reader
of poetry, often raising beguiling paradoxes in the process.
For example, on a deconstructive reading, the problem with
Aristotles poemis not, as his accusers charged, that he collapsed
the distinction between man and god by composing a hymn to
his friend, but that he naively assumed there was a difference
between these signiers when, linguistically, Zeus, Achilles, and
Hermias are all on the same level.
I nd a deconstructive approach inadequate to give a full
account of these texts, though I recognize that it can be effective
at unmasking tricks of logos. Its value derives, I think, from the
fact that deconstruction is a form of rhetorical criticism, albeit
an extreme one that insists that all languagenot excluding
the and ais gurative, allusive, not referential. But for
the classical rhetorician, logos is only one of three aspects of
speech that require analysis; the task of reading should also pay
attention to the ethos, the speakers character suggested by the
words, as well as to pathos, the impression meant to be created in
the audience. We will see that crucial aspects of Aristotles poem
emerge only when we consider these latter two aspects, taking it
as a speech that projects a certain sort of characternot only in
Hermias but also in the poet who ventures to praise himand as
an address that has particular designs on its audiences state of
mind. Bringing tolight the ethos and pathos of the poemwill prove
to be the main reason for not adopting the postmodern view,
because it puts us too much out of sympathy with what Aristotle
is trying to say and do in the poem. At this point, however, we
need only argue that a rhetorical reading of the song militates
against obliterating Hermiass historical status. For it is only as
the name of a historical individual, of a gure who is not a myth,
that Hermias works in the rhetoric of the piece: his name can
only cap the series of devotees to Virtue by being different in
kind from them. Only by referring to a historical Hermias can
Aristotle succeed in the evident ambition of his poem to add a
name to the canon of heroes. It is left to deconstruction to claim
that Aristotles attempt to separate myth from history was in
vain, but it is not an option to say he did not try.
If we have defended the claim that the word Atarneus is
not fully legible without a historical gloss, it still remains to
say what role such knowledge should have in appreciating the
poem as a work of literature. Much, indeed most Greek lyric
poetry is studded with proper names about which we know
nothing from other sources; in some cases we cannot decide
whether these names referred to actual historical persons or were
archetypal names for stock gures celebrated in song. (Pity the
future historicizing critic who tries to track down the original
Yankee Doodle Dandy.) Why should it matter, then, that in the
case of Aristotles song we happen to have external evidence
about its honoree? It is not obvious that this adventitious
knowledge obliges us to interpret this poemin its context. These
details might interest historians of the northeast Aegean but not
necessarily readers of poetry.
Knowledge of a poems historical references might be regarded
as gratuitous information, an extra nuance to be savored by
those so inclined, except that historical glossing is sometimes
inseparable from the basic task of construing the text.
4
Again,
the phrase nursling of Atarneus affords a compact example:
we will see that some contemporaries disputed Hermiass lineage
and legitimacy, claiming he was a slave and a barbarian from
Bithynia.
5
In this light, the metaphor nursling could be read
as a deliberate counterassertion, rooting Hermias in the soil
of Atarneus as a native. Support for the assumption that
legitimacy was at issue can be taken from the similar way
in which Himerius introduces him, Hermias of Atarneus by
birth (
\
E, te Axe,v ` u y vo, Or. 40.40).
6
Whether we take
Aristotles nursling as a pregnant locution can again be thought
optional, but once admitted, considerations of context tend to
extend beyond glossing names and affect the meaning of other
parts of the text. For example, the mention in v. 4 that dying for
virtue is accounted an enviable fate in Greece gains an extra
reason to be in the text if read as an assertion that Hermias,
far from being a barbarian, shared Greek values. The curiously
expressed thought in vv. 45for the sake of virtue, even to die
is an enviable fate in Greece / and to endure painsalso seems
less curious if contextualized. It sounds anticlimactic, having
mentioned that people will die for Virtue, to add that they are
willing tosuffer for her as well. One immanentist explanationof
this arrangement is to cite the trope of husteron proteron (second
things rst), a feature of many Greek poems and narratives by
which what is logically the second of two connected elements
is mentioned before the rst. Identifying a trope is more than
labeling a textual curiosity, for it adduces a host of parallel cases
that reassure us that we are not misconstruing the words. But
naming a trope is never a complete interpretation, for we still
can ask why Aristotle would use husteron proteron here. One
possible explanation might rely on reports that Hermias died
after being tortured by the Persian king. If we take v. 3 as a
tactful paraphrase of this painful (and, in Greek eyes, shameful)
fact, then the husteron proteron emphasizes Hermiass bearing up
under torture as on at least an equal footing with accepting death
as signs of Virtues appeal.
7
These readings might be debated, but we should only be
debating the extent and not the fact that the text of Aristotles
song had particular verbal resonance in the world for which it
was composed, and that though this world is gone, the words
bear ineradicable marks of the historical context from which
they came. I would like to close this methodological discussion
by saying that I know of no more signicant and far-reaching
advance in humanistic studies of the twentieth century than
structuralisms elaboration of the systematic processes (the
langue) that underlie and make meaningful each individual act of
speech (parole) or sign-making. James Redeld has well drawn
some of the implications of this development for criticism:
Social analysis is the necessary precondition of literary criticism
because poetry implies culture just as parole implies langue.
8
Much current criticism in the vein called cultural studies would
agree when Redeld adds, Poetic language is wonderfully dual,
conicted, rich and self-extending language, but it is language
an elaboration of a human collectiveso reading a poem is
learning the language and learning the culture that shaped the
language to be reshaped. But the reading of poetry, the making of
meanings from its words, cannot conne itself to the structural
or even poststructural level. On this basis I aim to bring out
the linguistic processes and cultural ideologies at work in the
song without denying the agency of individual actorsespecially
Aristotle the poet, but also his readers and misreaders, along with
those who transmitted his text.
No doubt many historical meanings are now buried beyond
our power to excavate, but that is of course no reason not to
register the ones we can see. Moreover, even when the historical
motivations underlying certain phrases have been dropped from
the tradition, they can make their presence felt in the text to
the extent that, in their time, they inuenced the way that the
poem was made or affected the way that it was received and
remembered. In this indirect way, even the hidden past can exert
a formal inuence, and such inuences can only be legible if one
is looking for them. I do not claim that Hermias was the onlie
begetter of this song or that a complete knowledge of his life is
essential to understanding it; but I urge that we proceed on the
assumption that the poem would not exist, or would not exist
in the form we know, were it not for his life and friendship with
Aristotle.
Constructing Hermias: The Erythraean
Inscription
The ancient sources for Hermiass life differ radically. As noted,
the paragon of virtue that appears in Aristotles portrait is
countered in other early accounts that revile Hermias not only
as a eunuch, a slave, and a barbarian, but also as a ruthless
and bloodstained tyrant. This is unsurprising; an unelected,
unconstitutional strongman, Hermias had to maintain power
between a restive Persian Empire to his east and competing
Greek interests to the west and south.
9
Indeed, Didymus was
moved to insert an excursus on Hermias into a Demosthenic
commentary precisely because accounts about the autocrat had
become so contradictory (Didymus col. 4.6065). He seems to
have taken control in Atarneus around the middle of the fourth
century when he succeeded, some suggested by assassination,
Eubulus, the previous ruler and his erstwhile patron.
10
Eubulus
began in Bithynia and may have owed his rise in station to
a wave of revolts in the late 360s by the satraps of Anatolia
against the Persian king Artaxerxes II (reigned 404359/358).
In these changing times a number of minor gures succeeded
in installing themselves as independent dynasts.
11
Eubulus had
been a banker, and Hermiass name is from Hermes, a good
patron god for exchange and commerce.
For a long time in its earlier history, the area of Atarneus had
beencontrolled by Chios, and it is notable that some of Hermiass
most determined early detractors come from the island.
12
One
prominent antagonist was the historian Theopompus of Chios,
who, in an open letter addressed to Philip of Macedon, smeared
Hermias as a repulsive person and a dangerously aggressive
ruler. On the other side, Hermiass supporters include several
Peripatetics in addition to Aristotle; they were indebted to him
for his hospitality and so had reason to portray him as an ideal
student of philosophy.
13
Aristotles song for Hermias, then, must
be regarded in some respect at least as a piece of propaganda. The
fact that Theopompus appears to concede that Hermias has a
certain reputation (doxa at Didymus col. 5.23) that has reached
Philip indicates that the tyrants story was of more than local
interest, and when we come to consider the monument Aristotle
dedicated to Hermias at Delphi, it will be easy to see that at least
part of his agenda in his commemorative acts was to project an
image of Hermias and his character that did not reect badly
on his school. As much as acquaintance with Aristotles other
writings may impress us with his integrity, he was of course
no impartial witness, and in defending an associate he was also
defending himself and his profession. The philosopher himself
was hardly immune to slander.
14
We know this best from one
of Aristotles defenders of the early Roman period, Aristocles
of Messene in Sicily, a Peripatetic historian of philosophy who
was generously excerpted by Eusebius in his Preparation for the
Gospels. Aristocles found it convenient to deal with the many
texts slandering Aristotle by breaking their authors down into
those who knew him, those who read those who knew him, and
the rest of the crowd.
15
An indication of some themes raised in
these writings can be seen in Aristocles remark that Aristotle
incited this resentment because of his friendships with kings
and the sheer superiority of his writings (Praep. Ev. 15.2.11).
The written record nowhere provides a straight historical
account to which we can turn to detect historical evasions or
distortions in our poem.
16
One of our oldest pieces of evidence
for Hermiass career is an inscription discovered in Smyrna
(modern Izmir), whose tangible reality is not to be confused
with objectivity. The inscription records in 33 lines a treaty
between that city and Hermiass government at Atarneus.
17
At
the time of the treaty, Hermias has already established himself
along the coast of Asia Minor around 60 miles north of Chios
and seems to be trying to extend his inuence to the south
by forming an alliance with Erythrae, an Ionian mainland city
just opposite Chios. The treaty is thus to be dated sometime
when he was in control, that is, 350342 BCE. The most striking
aspect of the inscription is that it is between the Erythraeans
on one side and Hermias and his companions on the other
(hetairoi, repeated in lines 23, 1011, 1415, 2021, 25). Some
have inferred from the phrase that Hermias had adopted some
form of constitutional rule, in line with the favorable tradition,
found in Philodemus and Didymus, that Hermias softened his
tyranny under the inuence of Aristotle and other students
of Plato.
18
This idealizing view, powerfully advanced in Werner
Jaegers account, would harmonize with the representation of
Hermias found in a prose eulogy for him that was composed
by Aristotles nephew, Callisthenes of Olynthus (to be discussed
below). According to Callisthenes, Hermias died wishing to have
word sent to his friends and companions that he had done
nothing unworthy of philosophy (Didymus col. 6.1516). We
know nothing specic about the form of Hermiass government,
but it may be naive to infer a sort of enlightened power-sharing
from the inscriptions companions. The term could easily have
been a euphemismfor the inner council or bodyguard with which
ancient autocrats were wont to surround themselves. The poetry
of Pindar, for example, shows that the tyrants and monarchs
of the archaic period allowed their paid wise men to describe
themselves as the friends (philoi) or guest-friends (xenoi, see
below) or evenhetairoi of the king.
19
All we cansay for sure is that
Hermiass companions must be seen as a piece of acceptable
self-representation by the ruling power. The inscription itself
(lines 3233) directs that a copy be put on display in both cities;
Hermias will set up his copy inthe sanctuary (hieron) of Atarneus,
the founding hero of the city (Himerius, op. cit.). The placement
of the stone in itself asserts the legitimacy of the autocrat.
The texts about Hermias that we have, then, can be divided
into two main camps. One side presents him as a cut-throat and
tyrant who tried to cover up his crimes and ignoble origins by
cultivating famous wise men as toadies; the Peripatetics suggest
that he died as a martyr to philosophy and so set the stage
for his portrayal by moderns as a edgling philosopher-king.
20
Historians who are familiar with the full range of evidence can
best balance these traditions and give reasons for preferring one
side or the other. My ownviewis that neither extreme will do: the
former portrait is clichd and the latter unconvincingly idealized.
In the long tradition of Greek monarchs who included wise
men in their entourage, the actions of Hermias strike me as
particularly intelligible if we regard him as a harbinger of the
Hellenistic age, one of those rulers who were willingpartly out
of personal taste andconvictionandpartly for political reasons
to establish a place for men of culture to gather and so let the
world know that civilization ourished in their city.
Although there are basic questions about Hermias that we
cannot answer with certainty, to write him off as a retrievable
historical individual puts us too out of sympathy with the
poems ambition to recall and praise him. It would certainly be a
convenient policy in dealing with archaic Greek poems to bracket
historical questions about the proper names they mention; and it
is true that it little matters tothe force of the poemwhichportrait
of Hermias is historically accurate. Yet once the full import of the
poem is in view, it will be clear that a pervasive skepticism takes
a metaphysical view fundamentally at odds with Aristotles and
makes us unable to grasp his point. I turn, then, to the evidence
for Hermiass fall and his relations with Aristotle, promising only
that the reader will understand Aristotles song better by reading
other texts among which it raised its voice and against which it
claimed a place.
The End of Hermias: Theopompuss
Letter to Philip
In348/7Plato died, andAristotle left Athens for a sojournabroad
that would last a dozen years. The two events have often been
connected, but we simply do not know why Aristotle left or even
whether it was before or after his mentor died. Many have seen
signicance in the fact that Speusippus and not he was chosen to
take over as headof the Academy. The implicationthat Aristotles
departure signaled some kind of falling out with the Academy
has appealed to ironists and critics of the philosopher, but may
be a fabrication post hoc ergo propter hoc.
21
Speusippus was,
after all, Platos nephew, and we may be wrong to expect that
there wouldhave beenevident academic standards for choosing
a successor: the Academy was in many ways unprecedented as
an institution of higher learning that was not identied with
its lead teacher and so had to nd its own way in developing a
planfor succession.
22
Aquite different explanationfor Aristotles
leaving the city is that in 348 Athens became uncomfortable for
people withMacedonianconnections: that year Philip sacked and
razed Olynthus, a one-time Athenian ally, and anti-Macedonian
riots broke out in the city. Demosthenes rode such passions
to prominence as the leader of a nationalist, anti-Macedonian
party and his Olynthiac Orations stressed Macedonian iniquity
repeatedly to the public.
23
The issue was still a sore point in
306, well after Aristotles death, when Demochares delivered a
speech Against the Philosophers, charging that Aristotle had
collaborated with Philip by indicating which Olynthian families
were the richest and ripest for plunder.
24
Things are a little clearer if we change the question and ask
why, when he left, Aristotle rst headed for Asia Minor and
Hermias. It seems likely that Hermias and Aristotle had prior
connections to judge fromthe fact that Aristotle was, as Didymus
puts it, treated like family (col. 5.63: ot xtxexe ct xtxo). The
Augustan-age scholar Strabo reports that Hermias had studied
in Athens, whether at Platos Academy or with Aristotle; this is
contradicted by a letter attributed to Plato (the Sixth) in which
the writer says he has never met Hermias; Platonic epistles are
always open to being doubted, but it seems to me that the
position of tyrant was not one that could easily be attained or
maintained through long absences.
25
Aristotle is more likely
to have known Hermias through family connections he had in
Atarneus. At the risk of seeming a romantic novelist, I point
out some suggestive connections that the tradition records.
Some ancient biographies say it was a man from Atarneus called
Proxenus who took charge of Aristotle when his father died;
this Proxenus would have been the one who shipped Aristotle
off when he was seventeen to study under Plato at Athens.
It may be worrisome how aptly named this helping gure is
(something like, ambassador or consul), and we shall later
have to consider the possibility of his being a creature of the
Aristotle legend. But there can be no doubt that Aristotle was
especially close to Hermias; at some point he married Pythias,
recorded as the tyrants niece or adopted daughter. Aristotle and
Pythias had a daughter also called Pythias, who is mentioned in
the philosophers will as recorded in Diogenes Laertius. Indeed,
that document assigns a certain Nicanor to be guardian to
the younger Pythias (whom he eventually married), and an
attractive conjecture by C. M. Mulvaney identies Nicanor as
the son of Proxenus of Atarneus and Aristotles older sister
Arimneste.
26
Apart from these shadowy but possibly close familial ties,
the fact that when he went east Aristotle found himself in
the company of several sympathetic students of philosophy
inclines me to think that he anticipated something like the
substantial hospitality Hermias could afford. Hermias made
Assos, Atarneuss port city to the west (modern Behramkale
in Turkey), available to him and his fellows for research.
27
We
do not know who else exactly was in Assos, but Theophrastus,
Aristotles student, collaborator, and eventual successor in the
Lyceum, was certainly not farif not in Assos, then just opposite
on Lesbos. Nearby to the northeast, Erastos and Koriskos, two of
Aristotles fellows at the Academy, had gone to teach philosophy
in their home town of Sepsis on the slopes of Mount Ida.
The Platonic letter mentioned above seems to be the teachers
recommendation of these young scholars to the care of the
prince. Aristotle might have also have been accompanied by
his nephew Callisthenes. Callisthenes, to whom we will return,
is said to have been the grandson of Proxenus of Atarneus and
Arimneste.
Whatever it was that rst brought philosopher and tyrant
together, there is no doubt that Hermias was interested in
philosophy and specically in the philosophy of Platos school.
This shows clearly through a hostile source, Theopompus of
Chios, who circulated, apparently toward the end of the 340s,
an open letter to King Philip designed to blacken Hermiass
reputation. The following extract, quoted in a relatively tattered
part of Didymuss papyrus, is spoken with scornful irony:
[Eu vo uo (` v) y( ` e,)

y v,
\
oe u]x
.
c` e, tt x(e`t)
t.[xe.]o y[yov] , x(e`t) [p e,p]e,o (` v) ` v (x` e)
x v H[.ex]v tv []t.o[oo] t, co u.o c(` ) yv(v)o
e [c]y eyot uyotv
v xe t evyy u,otv e yv txet

[although a eunuch (?)], he has become a charming man of
taste; and though he is a barbarian he philosophizes with
the followers of Plato; born a slave, he competes at the
Panhellenic games with chariots that cost so much.
28
Theopompuss language need not imply that Hermias visited the
Academy but could refer to his relations with Aristotle and such
fellow philosophers who were near Assos, these being for all
practical purposes still Platonists.
29
His point in any case is
that Hermiass philosophizing is pretentious, an attempt to pose
as a man of culture when he was in reality a debased barbarian
slave. It seems to be true that, whatever Hermiass devotion to
ideas may have been, having wise men in ones court was a source
of what Pierre Bourdieu called cultural capital. The fact that
Theopompus puts Hermiass philosophizing on a par with his
adopting rened manners and competing in the Greek games
suggests that philosophizing could be, in the fourth century,
what athletics had been in the sixtha way for autocrats on
the margins of the Greek world to present themselves to that
world as civilized, enlightened, and legitimate rulers.
30
Such as it may have been, the school of Atarneus proved
short-lived. In 342 Aristotle was summoned by Philip to Pella
to take charge of the education of the prince Alexander. He
spent seven years at the Macedonian court and nally returned
to Athens (after a visit to his hometown) in 335. It was while
Aristotle was in Pella in the late 340s that Hermiass always
precarious political situation turned perilous. The great king
of Persia, now Artaxerxes III Ochus (reigned 358338), became
concerned that Hermias was intriguing with Philip of Macedon;
Philip could use Atarneus as a foothold if he decided to attack
Asia, as some Greeks were urging him to do (and as Alexander in
fact did in 334). It has even been suspected that when Aristotle
showed up in Atarneus in 347 he was acting as a secret agent for
Philip.
31
In his Fourth Philippic, Demosthenes speaks as if there
is no doubt that Hermias had entered into a pact with Philip,
but all we can say is that such negotiations would have been
prudent on Hermiass part in the 340s.
32
His position depended
on Persian acquiescence, for Atarneus lay within Dascyleum,
the Persian satrapy in northwest Anatolia. Persia had been
reasserting itself, notably in the reconquest of Egypt in 343/342
for which expedition the general was Hermiass friend, Mentor
of Rhodes. This minor player in our story had been made satrap
of Asia Minor and was the brother-in-law of Artabazos, who
had been installed as satrap of Dascyleum by Artaxerxes II.
33
Thus it was that Artaxerxes III was able to reach down and
use Mentor to trick Hermias and deliver him into his hands in
Susa, probably in 341. This event is what attracted Demosthenes
notice, apparently in the time between Hermiass capture and his
death.
34
As a rabid opponent of Philip, he was happy to think that
Hermias would betray their mad conspiracy; the Fourth Philippic
calls Hermias the agent and accomplice of everything which
Philip plans against the King [of Persia] and hopes that the
King will hear the whole business . . . from the very person who
carried it out and was responsible for it (10.32). In the event,
Hermias seems to have betrayed nothing, despite being tortured
and, on some accounts, mutilated.
Chapter 3
Performance and Occasion
Having noticed some early documents from Hermiass career,
we turn to the poetry his death elicited. I begin not with
Aristotles song but with a short epigram he composed in
Hermiass memory and an epigram responding to this by a
contemporary poet, Theocritus of Chios. To take up these
poems, however, is not to move at last from talking about
historical context to texts, for these texts propose contexts
and occasions for their own performance that we must take
into account in interpreting them. This returns me to the
argument with my colleague mentioned above, and my position
that, although our reconstructions of performative contexts for
Greek poems must be tentative and sometimes speculative, we
cannot fail to attempt such reconstructions because the words
of the poems often call them forth by alluding to their own
situations of performance. Unlike most modern Western lyric,
Greek lyric poems typically present themselves not as private
meditations but as speeches made in social circumstances.
They create meaning by projecting a persona by whom such
words might be spoken and also occasions at which they might
27
appropriately be delivered. Though Aristotle and his peers were
obviously highly literate and literary, poetry was still in his day
predominantly an art to be heard, and the occasions at which
a song was presented could exert a powerful inuence on how
it was received and interpreted. For this reason, many Greek
songs nd ways to construct for themselves preemptively ideal
occasions inwhichthey wouldbe the perfect thing tosay. Context
thus must be considered not only as a hard reality out in the
world against which songs echoedactual places, occasions, and
audiences before whom they were performedbut also as a
projection of the song by which it hoped to secure a favorable
reception. Our approach to occasions and contexts, then, must
be partly historicalconsidering the social realities that gave
the words of a poem particular relevance and resonanceand
partly rhetoricalallowing that the poetic language could do a
great deal to situate the audience and shape their perception of
the event.
Context in this sense is integral to interpreting the song for
Hermias because it is not too much to say that the genre of a
Greek poem was effectively the same as the occasions for which
it was appropriate. The question of contexts of performance
is thus tied to that of genre, and when we come to consider
Aristotles trial we will have to ask howfar his songs meaning and
genre changed when the lyric was brought into the courtroom.
For the present, however, I want to pursue the idea of context
and occasion as something projected by the poem, partly a
reection of the importance of social conventions in a still
vital performative culture and partly a rhetorical construct, an
imagined world and set of expectations that the performance
itself brought about and fullled. Epigrams would seem to be
easy to discuss in these terms: the obvious context is the physical
site where the inscribed stone is set; the occasion nothing
Perf ormance and Occasi on 29
particular at all, just the chance of a passerby happening upon
the words. It will be seen that things are not so simple on
both counts, and I will propose that with Theocrituss counter-
epigram and possibly with Aristotles original we have to allow
for the possibility that we are dealing with book-epigrams,
ctional representations of stones, locales, andtheir inscriptions.
Considering the verses in these terms will put us in a better
position to ask in the following chapter what was the original
context or contexts for Aristotles song for Hermias.
Commemorative Epigrams: Aristotle
and Simonides
When Hermias was killed, Aristotle was still in Pella supervising
Alexanders education; it is likely that news of his death came
quickly to the capital and that it was fromMacedon that Aristotle
decided to make a public response, indeed to let his view of the
matter be heard throughout the Greek world. Diogenes Laertiuss
Life tells us that Aristotle had a statue (andrias) erected for
Hermias at the sanctuary of Delphi, and quotes the pair of elegiac
couplets he had inscribed beneath it:
1
xvc ox' ou o
\
o t e,ep` e ex e,v 0 tv e
\
yv` yv
xxtvv H,o v xoo,v peot. u,

ou ev, e .yy ov tot
v e y ot x,ex yoe,
e ..' e vc, toxt ,yo evo co. tou.
This man was once upon a time impiously and sinfully
slain by the king of the bow-bearing Persians,
winning his victory not in an open test of deadly strength
but by trust placed in a treacherous man.
A biographical question, naive and unanswerable though it may
be, may provide the quickest way into the specicity of this
verse: In view of the fact that Hermias and Aristotle had been
so close, do we nd the tone of the couplets too cool? Verse
epitaphs were certainly able to strike a plangent, personal note
that is often the more forceful for being compressed; but here
Aristotle deliberately focuses not on the loss of his close friend
and benefactor but on his despicable enemies. The text stresses
the kings impiety and conates his execution of Hermias with
the treachery of Mentor. Context counts for much in Greek
poetry, even for epigrams meant to last for generations, and
it seems that this poem was situated in a Panhellenic sanctuary
primarily to broadcast the perdy of Persia.
Aristotles public tone, more indignant than grieving, should
not mask the deep bond that this dedication represents. We
can see this if we understand that the relationship between
Aristotle and Hermias was of a very special kind that the Greeks
called xenia, a form of ritualized friendship greatly illuminated
in Gabriel Hermans 1987 study. Traditionally translated as
sacred hospitality or the guest-host relationship, xenia named
a broad set of mutual obligations between any benefactor and
beneciary, obligations that were underwritten by no less a
god than Zeus xenios, Zeus the god of guest-friends. Herman
points out that the services typically rendered by xenoi could
include setting up funerary stones for ones xenos and composing
poems to express ones affection. Although it was impossible
for Hermias to get a proper burial among family and friends,
Aristotle could do the next best thing and erect a kind of
cenotaph for him. In addition, a monument afforded a blank
slate for verse, and so Aristotle took the opportunity to execute
a xenos duty to set the record straight and spell out just who
the villain was in this episode. Pindar, a supreme singer of
the ties of hospitality, could have been describing Aristotles
agenda: I am a xenos: warding off blame that obscures, I come
to my dear friend as with streams of water to proclaim his true
glory ( tv tt oxoxtvv e v yyov, / u
\
cexo
x
,
\
o` e t.ov
vc,' e
yv / x. o
x yxuov etv o).

2
Delphi
was an obvious choice for these purposes: seat of one of Greeces
most famous oracles and host of the prestigious Pythian games,
it attracted visitors from the entire Greek world and a little
beyond. There they couldreador have readto themby guides
inscriptions on the monuments that adorned the sanctuary and
lined the road leading to Apollos temple. One of the examples
Hermanmentions, a votive plaque put upby Xenophonof Athens
in a treasury built by the Athenians along Delphis sacred way,
suggests in the name of its dedicatee how deeply the ethic lay:
For Proxenos, for he was a xenos. In addition, Delphi had taken
on an extra symbolic role as a symbol of Greek unity in the face
of Persian invasions in the early fth century. Hence Aristotles
condemnationof Easterntreachery was appropriate to the locale.
The context in which this poem was exposed is reected in its
metrical form, and here I think of form not only from the point
of view of artistic shaping but also as a device facilitating the
circulation and survival of poetry. Of course, Aristotles choice of
elegiac couplets was dictated by a centuries-old traditionof Greek
epitaphs; but tradition settled on elegiacs for these purposes
with reason. First of all, elegiacs are easy to memorize: their
dactylic rhythms have much in common with the great warhorse
of Greek verse, the hexameter, but elegiacs are even more regular
in closing each couplet with an identical heptasyllabic rhythmic
unit. For the same reason, they are easy to read, and because they
require no melody to be recited, they can be re-performed after
being read or heard once or twice. The result is that, though the
stone may stay in place, its message can be widely dispersed by
passersby. In these respects, Aristotles inscribed verse for Delphi
is fundamentally different from the song he composed to Virtue,
even though we call both poems. The words of a song might
be read off a page, as in the Didymus papyrus, but one will not
be able to perform it without learning the melody; without this
melody, the rhythms of the songwhich were unique to that
songare far less easily memorized or even perceived. I shall
return to this basic difference in the two types of rhythmical
composition when I come to Aristotles song, but for the present
observe that not only is a poets choice of form made in light of
what convention and context demand but that convention and
context have favored certain forms because of the way they are
able to move among people. To maintain a sense of song and
poetry as machines for social communication, we ought not to
lump together every text that can be metrically scanned in some
broad and homogeneous notion of poetry.
3
It was as readable,
recitable verse that Aristotles mini-message was memorized by
readers or those within earshot of readers and carried away from
this international meeting place.
Re-performance is also built into the rhetoric of the piece:
as phrased, the inscription can be taken in two ways, as a label
attached to an object on the spot or as a script for a more portable
speech act. As a label, as information carved on a statue base, the
stone speaks the verse in the sense that it imparts to the viewer
the story behind the monument; such speaking monuments
are very common in the rhetoric of Greek tombstones. But the
addition of a little particle, once upon a time . . . (ox), which
is found in other commemorative inscriptions, makes the words
performable in another way.
4
The adverb transforms the verse
from a label afxed to a stone into something that can be said
by an onlooker at some distant time, indeed something that can
be said on an indenite number of occasions in the future. The
reader no longer speaks in the name of the stone, vocalizing its
message, but adopts the persona of a local exegete or tourist
who may stand before the monumentthe deictic force of this
in v. 1 is strong in Greekand declare what it is. This aspect of
the text makes every reading of it a re-play, a re-dedication of the
original act of inscription that can be re-enacted with no end in
sight. As performed, the words are moreover not only a speechact
but something of a curse. It is worth recalling that epigrams were
designed to be recited aloud rather than silently read, and the
Greeks were always aware that their wordfor glory (x. o) meant
something heard; renown meant re-naming and re-hearing.
Hence every reader/performer of this text ensured that the name
of Persia lived in infamy, her impiety literally resounding again.
Aristotle thus contrived for his Delphic monument a message
that was at once authoritatively rooted to the spot and re-
playable, even portable. Our last observation is that it may have
been ctional as well. Consider, for example, the epithet bow-
bearing: this may be taken to hint at Persian cowardice, if
the bow be given its traditional force as a symbol of unheroic
ghtingthe antithesis of Homeric warfare at close quarters
with swords and spears. On the other hand, gods can be bow-
bearing as well, and the epithet is found of divinities in earlier
poetry, including Delphis own Apollo. The choice between these
two avors of the epithet is between reading Aristotles epigram
as an attempt to transcribe a historical event in conventional
form and language or as an attempt to extend and innovate
within the tradition of monumental epigrams. In favor of the
latter possibility is the fact that the same form of the epithet
is found in the same metrical position in an epigram attributed
to Simonides, Greeces most famous memorializing lyric poet.
Bow-bearing also modies the Persians (called Medes in fth-
century parlance) in a pair of elegiacs that purport to celebrate
their defeat by the Athenians at the river Eurymedon in 468 BCE:
Ot
\
c e,' Eu ,u covx e ox' e y.ev
.oev y
\
pyv
e,v evot M ycv xoo,v ,o eot
etyxe t, o t x xe`t
xu,v
`t vy v
x e..toxov c' e ,x y v y'
.tov 0 tvot.
These men once upon a time lost their youth by gleaming
Eurymedon
contending with the champions of the bow-bearing
Medes,
warriors, both foot soldiers and on swift-faring ships;
they left behind the nest memorial of aret when
they died.
The unresolved debate about whether these lines, which are
preserved only in a late anthology, were actually inscribed on
a monument or were what Page calls a later literary exercise
only makes clearer that the memorial (v e) left behind at
the end is gurative; more than any stone that may have stood
near Eurymedon, it is the verse itself that works as a reminder
(vy tov) of these excellent dead.
5
Aristotles epigram too, although the deictic pronoun leading
off the poem would seem to tie it to a stone, has undoubt-
edly had life apart from this monument, circulating in oral
re-performance and making Persian perdy resound among
hearers who would never set eyes on the stone (as it continues
to do this day). Once the this is allowed to be ctional, one
might question whether there was a real stone at all, for no
trace of this monument has ever been found. In other words,
we may underestimate the power of Aristotles rhetoric when
we assume as realists that some actual monument, now lost,
lies behind Diogenes quotation rather than crediting the poetic
this of the verse with generating our belief in a stone on which
the text was rst inscribed. After all, in the fourth century,
Greek poetry had begun to explore the genre of articial and
consciously ironic book epigrams that feature so prominently
in Hellenistic literature.
6
In a book epigram the Here lies . . . ,
which functions as true a deictic in inscribed epitaphs, becomes a
conventional marker of the genre, its meaning ironically reversed
to not here, not on this page. I raise the issue not because
there is a particular reason to doubt Diogenes word, but to call
attention to the fact that Aristotles language in the poem is at
once thoroughly conventional and at the same time capable of
shaping our conceptions of the world that it ostensibly points
to.
7
If occasion and context can powerfully affect the meaning
given a poem, poetic language can do a good deal to determine
the contexts and occasions we imagine for it as wellalmost to
the point of producing a stone out of thin air.
Book Epigrams: Theocritus of Chios
Whether it was ever inscribed or not, Aristotles epigram would
have spread at least as widely by passing from mouth to mouth
as by being read in situ. By either route, the brief and pointed
poem made its way across the Aegean and came to the attention
of a Chian politician and poet called Theocritus. No friend
of Hermias or of the philosopher, he issued a sharp reply
in kind:
\
E, tou u vo uou x xe`t Eu po u.ou xc co u.ou
o ye xvv xv,v x uv A,toxox .y,
o
\
` ct ` e x` yv e x,ex y yeox, uotv t
\
.xo ve ttv
e vx' Axecy te po,p,ou
v ,ooe t.
For Hermias, the eunuch and slave of Eubulus
empty-headed Aristotle made this empty tomb;
doing honor to his unrestrained belly, he chose to dwell
in outpourings of slime rather than in the Academy.
The elegiacs, preserved for us by Didymus and Diogenes, among
others, are a faux-epitaph, an alternate inscription for Hermiass
monument.
8
Posing as an inscription on the same (this)
cenotaph on which Aristotle wrote, Theocrituss epigram is a
kind of grafto scrawled over the original. At the same time,
the verses are crafted so as to be rewarding in oral performance.
The stunning rst line has nine of fourteen syllables composed
of ou- sounds, and the second is neatly divided between ks
and nasals in the rst half and ts and sibilants in the second.
One assumes such a ourish of assonance could sound savagely
taunting. It is also noteworthy that Theocrituss opening allitera-
tion, withthe line-ending doulou, picks up the doliou (deceptive)
that ended Aristotles epigram, suggesting that his own text
would play very neatly if it followed a recitation of that verse.
Theocrituss is a book epigram, a ctional epitaph never
inscribed; he was known as a satirist and his witticisms were
collected in a book by an otherwise unknown gure called
Bryon or Ambryon.
9
Literariness can be contagious, for when
Theocritus took it up he ensured that Aristotles verse
whether or not an original was actually carved anywhere
entered the world of performative epigrams as well. As such,
both texts became subject to minor variation or adaptation in
re-performance. For example, it is a notable detail that Theocritus
says Aristotles poem was on a tomb (a sma), whereas Diogenes
speaks of a statue (andrias).
10
We should not put much weight
on either as historical evidence: there is no reason to assume that
Theocrituss sma represents eye-witness testimony; conversely,
Diogenes statue may be an inference from the initial xvc,
this man, he found in his version of the text. Complicating
any attempt to settle the question is the fact that we are
examining the operations of a highly traditional language of
poetic commemoration in which words and phrases of a given
metrical shape tended to be easily substitutable for each other in
a line. For an audience of such poetry of variety-within-limits, it
made little difference whether Theocritus was misremembering
or modifying his precursor text in order to set up a pun. Sma
suits Theocrituss second verse better than would the roughly
synonymous and isometric mnma (memorial) because one
did speak in Greek metaphor of an empty sma, a cenotaph,
but less readily of an empty memorial or statue. There are
in fact visible traces of the oral circulation that these epitaphs
enjoyed in the transmission of Theocrituss text. Consider, for
example, the minor variations in the way its rst two lines are
recorded. Diogenes Laertius, citing Ambryons On Theocritus
gives (5.11):
\
E, tou u vo uou y
c' Eu po u.ou e
\
e co u.ou
o ye xvv xv,v x uv A,toxox .y
Of Hermias, eunuch and slave to Eubulus at once,
empty-headed Aristotle wrought an empty tomb [sma].
Didymus, citing the same On Theocritus but attributing it to
Bryon, gives:
\
E, to[u] u [vo uou x] x(e`t) E[u po u.ou xc] co u.ou
o ye x[vv] xv[,v 0 yxv A,toxo]x .y
Of Hermias, eunuch and slave to Eubulus, this
is the empty tomb [sma] empty-headed Aristotle
erected.
Eusebius, quoting Aristocles, gives:
\
E, tou u vo uou x xe`t E` upo u.ou xc co u.ou
v ye xvv xv,v 0 yxv A,toxox .y
Of Hermeias [sic], eunuch and slave to Eubulus, this
is the empty monument [mnma] empty-headedAristotle
erected.
Among several small differences, the sources do not settle
whether Theocritus spoke of Hermiass memorial (mnma,
Aristocles) or tomb (sma, Didymus and Diogenes).
11
Didymus
and Diogenes may have found the latter reading in the treatise
On Theocritus. But their versions of v. 2 diverge on whether
Aristotle wrought or provided the tomb (x uv) or erected
it (0 yxv). As is quite understandable in a text so much in the
air, there is also instability in the deictic language: Aristotles
this man becomes this one (xc, i.e., this tomb) in Aristocles
and probably in Didymus, whereas in Diogenes the deictic
disappears altogether under the relatively colorless adverb e
\
e
(at once). Where there is no object in view to which deixis
points, substitutions arise easily in this formula-heavy genre,
especially when sources quote poetry from memory rather than
seeking out and nding the spot in a papyrus roll. The kinds
of variation noted here are less likely to result from a scribe
miscopying letters than from a performer adapting or slightly
misremembering phrases. Such oral variants characteristically
preserve the meter and general sense of the original in more
or less the same words without being resolvable into a clear
hierarchy of archetype and copies. The textual transmission of
Theocrituss poem thus indicates that it enjoyed a degree of oral
performance.
The ctionality of these epitaphs, that is to say their lack of
connection to any original source (whether a stone at Delphi
or a page of On Theocritus), lets them change slightly as they
are reused. Partly because his poem has been removed from
the historical context that would have limited its reference,
Theocrituss meaning is underdetermined at points, notably
in his nal insult and in the word po,p,ou just before the
poems end. The difculty turns, indeed, on whether to take
po,p,ou as a proper or a common noun, a decision that in
turn determines how we date and understand the epigram as a
whole. The eminently well-read Plutarch tells us, when he quotes
the end of the poem, that Borborus was the name of a river in
Macedonia (On Exile 603C); in that case, the poemwould refer to
Aristotles leaving Hermias and the Platonists at Assos to take
upwithPhilipandAlexander in342. But as Michael Flower brings
out in an excellent discussion, we are not bound to accept this
unconrmed geographical tidbit. Writing around four centuries
after Aristotle, Plutarch may well have been repeating an ad
hoc ction designed to help make sense of the epigram as it
aged.
12
Printing lower-case po,p,ou, as I have done, suggests
a different context: Aristotle is being mocked for leaving Athens
to go to Atarneus in 347, the outpourings of slime being the
debased patronage of Hermias. One point in favor of this reading
is that Theocrituss poem becomes a more direct response to
Aristotles if it closes by insulting Hermias and the philosopher
together. The connotations of the common noun are consonant
with this, since in the vocabulary of the sacred mysteries, to
lie in slime or mud was a traditional image for the woeful fate
of a non-initiate.
13
The image would be a direct riposte to the
pretentiousness of Hermiass philosophizing as Theopompus
represented it in his Letter to Philip.
14
Page takes a similar view
of po,p,ou, but for chronological reasons I do not agree when
he adds that the religious connotations of the word hinted at
Aristotles alleged impiety. Theocrituss parodic epitaph would
have had most point if it were composed, like its target poem,
while the events of the late 340s BCE were still fresh in peoples
minds; but it appears that the charges against Aristotle surfaced
several years later, whether after Alexanders accession in 335
or, more likely, his death in 323.
15
Lower-case po,p,ou puts
the parody closer to the death of Hermias in 341 and Aristotles
original. On this interpretation, ,ooe t, (outpourings) has
a possible double entendre: rather than referring to a particular
river, it might suggest the hospitality a wandering philosopher
requires in images of pouring out of toasts and libations.
16
Aristotle abandoned the civilized, Hellenic precincts of Plato
(as Theocritus is willing to characterize the Academy for the
purposes of this epigram) to revel in the polluted streams of
Hermiass hospitality; it is further implied that this hospitality is
uncivilized, for po,p,ou chimes with pe,p e,ou, barbarian.
Some interpreters accept outpourings of slime here but
interpret the insult differently, as intimating that there was
sexual impropriety in the relationship between Aristotle and
the eunuch.
17
That charge was certainly oated in the ancient
sources (e.g., Diogenes Laertius 5.3), but here I think it would,
so to speak, muddy the waters; the basic moral failing that
Theocritus imputes to Aristotle in this poem is gluttony, along
with the hypocrisy to disguise it with ne language. The word
for belly in v. 3, slightly indecorous in Greek verse as in
English, is a traditional poetic symbol for the bodys basic
needs that can drive even a hero to undignied acts of self-
preservation.
18
In the version of verse 3 recorded in Aristocles
and printed by Lloyd-Jones and Parsons, the word helps make
a noteworthy phrase: Aristotle left Athens on account of
his unrestrained belly (ct ` e x` yv e x,ex y yeox, uotv). The
language for lack of restraint (e x,ex y . . . uotv) comes out of
fourth-century philosophical ethics, which debated the nature
and causes of being unable to control oneself (e x,eo te).
19
(It may be signicant that the Peripatetic Aristocles is the
one who preserves this philosophically inected reading.)
20
Pairing this high-culture term with the common belly makes a
phrase that paints Aristotle as philosophe and vagabond at once;
the deliberately incongruous language exposes the euphemistic
rhetoric by which itinerant intellectuals were wont to cloak their
unmentionable needs in high ethical terms.
Theocrituss little poem thus ts squarely in a satirical
tradition that mocked sages who abandoned the cultural centers
of Greece for distant courts and hospitality of a not altogether
spiritual kind. The theme was struck at the end of the fth
century, when Aristophanes teased the tragic poet Agathon for
retiring to the golden isles of Macedon and the patronage of
Archelaus (Frogs 50).
21
Earlier still, the archetypal intellectual
for sale had been Simonides of Ceos: an avowed consort of
tyrants in the late sixth centuryfor whom he composed songs
celebrating the victories that they, like Hermias, won in the
gamesSimonides was remembered as the rst poet to charge
money for songs. The life of a wandering intellectual that
Aristotle led in the years following Platos death might appear
to critics to have much in common with the ambiguous career of
Simonides, who was both counted among the Seven Sages and
given the nickname x tpt (greedy) for the way he capitalized
on his wisdom.
22
The charge of excessive appetite can be found
laid against the Stagirite already in his lifetime by Cephisodorus:
Aristocles reports that Cephisodorus, a pupil of Isocrates,
criticized Aristotle as self-indulgent and a glutton, as did other
of Aristotles detractors.
23
Texts and Things: Herodotus on
Hermotimus
Theocrituss verse activates a tradition of blame epigrams to
paint Aristotle as a hypocrite philosopher enslaved to lowdesires.
The way he inects this tradition in fourth-century ethical terms
suggests that slime is not primarily a sexual metaphor, and
here we may as well address the delicate question raised by
some of our sources: Was Hermias a eunuch or wasnt he? The
question is not altogether facetious; trying to give a denite yes-
or-no answer is a fair epitome of the historicists quest for the
truth, even as the desire to gaze upon the Ding wie es eigentlich
gewesen ist can put the historian in the ridiculous posture of
peeking under Hermiass chiton. Theocrituss statement that
Hermias was a eunuch is, technically speaking, contemporary
evidence, as is the description by Theopompus of Chios.
24
I have
noted that Chians were no friends of Hermias, and so even
early and corroborated evidence can be disputed. But the fact
that Pythias is variously described as Hermiass niece or adopted
daughter suggests that her existence had to be squared withsome
acknowledged reproductive impairment. A different direction is
suggested by Mulvaney, who brought up a fascinating parallel
for these charges in a story Herodotus tells about Hermotimus
of Pedasa, a city a little further down the Ionian coast from
Atarneus (Histories 8.1046).
25
Captured in war, Hermotimus
was sold to a Chian named Panionius who had him castrated
and sold as a slave. Eventually, Hermotimus rose to become the
chief of Xerxes eunuchs and from this position tracked down
Panionius in Atarneus and got his revenge. Mulvaney suggests
that the nexus slave/eunuch/Atarneus is the germ of later
slanders against Hermias, whose name is close to Hermotimus
and whose life, we may add, was also a rags-to-riches story. We
do not have enough evidence to disentangle what, if any, are the
connections between the tales, but the phenomenon it points to
inspires caution: it is obvious that real facts may become legend
in the course of their being selectively recounted and shaped; but
it is also true conversely that legends may give rise to historical
facts by inspiring legend-like behavior in real historical agents.
Because these epigrams proliferated beyond their immediate
contexts, we can no more x the truth behind them than afx
themto an actual stone. The elusiveness of the referents pointed
at by these deictic verses is like the undiscoverability of what we
can call, taking a euphemism from Freudian linguists, Hermiass
master signied. This elusiveness is partly due to the texts
rhetorical posturing; but we should not ignore the historical
forces that are at work to remove us from contact with the
past. This can be seen if we consider one nal stone from
Delphi. Even apart from its prominence as a cultural center
of the eastern Mediterranean, Delphi made sense as a place
for Aristotle to site his monument to Hermias because of the
good connections he enjoyed there. Aristotle had conducted on-
site archival research into the victors at Delphis centuries-old
Pythian games. Working with his nephew Callisthenes, he sifted
through local records and compiled a Hu0tovtx v e vey,e y,
a List of Pythian Victors.
26
Probably sometime in the 330s, the
ruling body at Delphi ofcially recognized the pair for their labors
and erected a stele which still survives bearing an inscription
thanking them. We hear from Aelian (14.1) that the honor was
later rescinded, and scholars have connected this information
with the fact that the stone was discovered in Delphi broken in
pieces and at the bottom of a well.
27
If a change in the political
winds was the reason that Aristotles monument was broken, we
may see why it fell to the book tradition to keep his Delphic
epigram alive. In any case, the shattered v ye for Aristotle
and Callisthenes provides a tangible and eloquent image of the
real danger and violence at the root of these tales, as well as of
the processes of propaganda and counter-propaganda that have
served as a conduit to bring these realities down to us, even if
only in fragments.
The next chapter turns to Aristotles song for Hermias and the
question of its possible performative contexts. It begins with one
other extraordinary contemporary document, a prose encomium
to Hermias written by Callisthenes of Olynthus. Callistheness
text, which can be read as if it were designed to complement his
uncles song, has led some scholars to posit that both were rst
performed in a private ceremony held by Aristotle and a few of
Hermiass close friends. Not all the details of this philosophers
rite of remembrance can be accepted with condence. But even
if we cannot be sure of retrieving the original context and
function of Aristotles song, comparison with Callisthenes brings
out the different meanings that it seems to take on when set in
different contexts. Such a perspective will then enable us to say
whether the story of the song being put on trial is plausible.
Chapter 4
Performance and Context
We can call Aristotles hymn an occasional poemin the sense that
it was composedto respondto a particular set of circumstances to
which its text makes reference. But it may have been occasional
in a stronger sense, in having been written to be performed
at a specic time and place, for a particular event involving
a specic group of people. The idea has been hard to dismiss
since Wilamowitzs 1893 work, Aristoteles und Athen, in which
the great Hellenist mooted the idea that Aristotle wrote the song
for a private commemorative ceremony among the Peripatetics.
Wilamowitz was struck by the fact that Aristotle barely named
the honoree of his song (proper names again) and concluded that
the hymn must have been accompanied by a prose encomium
giving out more information about Hermias.
1
The suggestion
seemed to be conrmed some ten years later when the Didymus
papyrus was discovered, for Didymus had included an extract
from a prose work in praise of Hermias written by Callisthenes,
Aristotles nephew. D. E. W. Wormell, in a fundamental study of
the traditions about Hermias, drewthe pieces together: Aristotle
instituted a memorial ceremony in honor of his dead friend,
45
at which the Hymn was sung. It was this service to Hermiass
memory whichwas later usedas evidence of impiety by Aristotles
enemies . . . . There can be no doubt that Callisthenes Encomium
was written for this occasion.
2
The inference is attractive, but it
is well not to pass over the soft spots in the case. One small
detail, though worth pointing out because it touches on the
question of genre, is that when Callisthenes composition is
identied as an encomium, the basis for this is essentially a
gap in the papyrus where the work is introduced (in the rst
line of the quotation below). Scholars quickly proposed that the
word
yx tov would t the size of the gap and make sense,

but other supplements are thinkable. One notable candidate is
o uyy,ee(composition), suggestedby the rst editors of the
papyrus and printed by its most recent ones.
3
Filling the lacuna
with o uyy,ee, a general word for prose compositions, would
leave open the possibility that Callisthenes, who was a historian
as well as a friendof Hermias, composedhis account as something
other than a ceremonial eulogy. It may have been a history or a
biographical essay that was intended to be circulated through
written copies (ouyy, eexe) rather than performed at a
singular event. Moreover, even if we could be sure that Didymus
called Callisthenes text an
yx tov, this is not of course

per se evidence for howCallisthenes contemporaries would have
slotted his work into fourth-century literary categories.
Greek literature of the fourth century BCE is marked by
an explosion of new forms of written prose, notably including
experiments in prose eulogy.
4
The novelty of prose encomia
is made explicit in Isocrates Evagoras, written for a pro-Greek
king of Cyprus who died in 374. In the prologue to this work,
Isocrates proclaims that his text constitutes a new form of
writing that breaks with the tradition of having poets eulogize
the noble dead. The tradition had left writers of prose encomia
with only gures from myth or the legendary past to celebrate,
Perf ormance and Context 47
and so Evagoras broke new ground by encomiazing in prose
the excellence (aret) of a man (8: e vc,` o e ,x` yv ct ` e . oyv
yxt etv) who was not from the distant past but who had
just died. The innovativeness of Evagoras is highlighted by its
articial stance as a faux-oration, complete with a ctional
occasion implied in its opening address to Nicocles, whom I
behold as you honor your fathers grave.
5
Isocrates sometimes
exaggerates his artistic greatness, but even his detractors allow
that his claim about the novelty of prose encomia is tenable. An
example fromaround 360is Xenophons celebrationof Agesilaus,
his friend and a notable Spartan general: the Agesilaus was an
attempt, as Xenophon put it, to execute the difcult task of
writing a worthy account of his excellence and fame (1.1: e ,x y
x xe`t c oy ou ,
\
ectov e
tov
etvov y, eyet). Callisthenes
was thus working in newly opened elds when he wrote a prose
encomium for the recently deceased king of Atarneus. Arnaldo
Momigliano notes that Callisthenes Hermias ts beside not only
Isocrates Evagoras but also Theopompuss encomia of Philip and
Alexander of Macedon. One should perhaps add The Funeral-
feast of Plato written by Speusippus, the head of the Academy,
which has been thought to have had a eulogistic function. The
times were also ripe for cross-fertilization, however, and it may
be that the Funeral-feast was also bound up with the literature
recounting symposia in prose (inspired by Socratic Symposia, of
the sort writtenby PlatoandXenophon), making for anextra-rich
generic mix.
6
In the background of all this literary experimentation is a
persistent ethical norm, which was evidently widely respected by
audiences, that the forms of praising men that are appropriate
in a given case depend on the status of those praised, and status
includes not only greatness of birth or of accomplishment but
also whether they are living or dead and whether they belong
to the present or the past. The conventions governing literary
practices in this revolutionary time for prose remain tied to basic
religious considerations of what human mortality may claim as
its own and what it may aspire to.
Witnesses: Callisthenes Hermias
The extract from Callisthenes that Didymus chose to copy into
his text reads very much as if, as Wormell suggests, it was the
climax of the piece.
7
e ..` e y( ` e,) x(e`t) Ke..to0 v[y . . . . . . . .]xt ouv- Col. 5.64
x ee ,`t eu xo u [o.. e x . yt e
.].e x(e`t) [xeu-] 65

x t Ou ovov xot[o uxo

y v
\
x` e ` v
xt] xtv-
c uvv, e ..` e x(e`t) .yo tov x[exeox` e
0' o
\
]oto
` v ctx .t, x(e`t) y t
.
[oxyv x ox toxt]v t-
cx x y e ,x y
v eu x[ t x t 0ev ex]t. ot
\
(` v) y( ` e,) p e,pe,ot 0,o( uv)x[
0e ueov
eu
\
xo u] x
.
` y
.
v
.
70
e vc, tev. o
\
y[o( uv)] peot.[ ` u ou c` v (e,') eu xo u
uv0]e
.
-
v o(v)o
\
x,ov e ..' y` xo` u eu xo( ` u) . oyo(u)
e xo uv, 6.1
e yeo0`t xyv e vc, tev x(e`t) x( ` yv) ppet oxyxe x v
x, o(v), ctvo y0y (` v) eu x` ov o
\
. e tvet vo tv
[y]v o(v)ov eu
\
x t t.ov evxv
oo0et ,yot-
xexov e vxttx ovxv c(` ) Bey ou x(e`t) 5
M vxo,o, ct ` e x` o 0ov tv x(e`t) op to0et ` y (,-)
x
.
.
u
.
oyt e..ov
\
c' eu x v e 0 t, xe uxyv (` v)
e.t-v>xp e.xo x[ ` y]v yv yv, ctx ev c(` )
x v ytyvo( v)v e,[ eu x] t xexoe0t v e
ot-
,ov eu x` ov
o tyo c[t ` e x` yv] e ,x( yv) y

\
(` v) o(u
v)
xote uxy 10
x,t oxy u
\
y,[ e,] ` e x v
0, v e,e-
co[ox e]x
.
[y x(e`t)]
.
[o.` u e,` e x]` ov x v pe,p e,v
x, o[ov, o
\
c' y
\
cy] x.[ux e]v ..[]v, t.t-
.[ . . . . . . . . ]ov [
txe.]o e(v)o e
..[o] (` v)
[ou c` v
tv,
] ox[yy c' e]u x t ,` o[] xo` u[ ] t- 15

[.ou x(e`t) xo` u
\
]xe t,o(u) [
t]ox ..tv,
\
ou c[` ]v
e
.
[(v e)to]v [t
]y t.ooo te[ ou c e
]oyov c(te),e-
y[( v)]o.
And Callisthenes composed a kind of [. . . . . . . .] about him
in which he said a number of things including this: Not
only did he comport himself this way when far from danger,
but he remained the same when threatened. And he gave
the greatest proof of his aret at the moment of death. Now
the barbarians marveled at his courage as they watched him.
And the king, getting nothing more out of him than the
same responses, was awestruck at his courage and rmness
(ppet oxyxe, 6.2) of character and considered letting him
off completely, thinking that he would be the most useful of
all his friends; but Bagoas and Mentor opposed this course
through envy and fear that if released he would become
superior to them, and the king changed his mind back. But
on account of Hermiass manliness (aret) the king ordered
himto be spared the usual tortures. Nowsuch moderateness
coming from an enemy was most unexpected, and indeed
is contrary to the barbarian character; and when he was
about to die he called t.t[. . . . . . .] and asked nothing
more than that he send word to his friends and companions
(hetairous) that he had done nothing unworthy of philosophy
or unseemly to the very end.
The extract seems equally suitable for performance as a eulogy
and as a piece of artistic prose (a o uyy,ee) designed to be
circulated independently of any ceremony. The alternatives are
of course not exclusive, for the scripted remarks could have been
subsequently published by circulating copies among lettered
friends, in the way that Isocrates orations circulated. Jacoby
suggests that as a prose work Callisthenes composition would
have been aptly titled Hermias or On Aret, noting that this
is the style in which the dialogues of Aristotle were titled.
8
In
favor of Wilamowitzs suggestion that Callisthenes work was
intended for a private ceremony is that it resonates so well with
Aristotles hymn: bothtexts focus onthe theme of aret, andboth
suggest a sharp contrast between Greek and barbarian mores;
both conclude by praising Hermias and steadfastness. While we
may be unable to determine the original genre of Callisthenes
Hermiasthat is to say, its original performative context
posing the question at least makes clear how closely Greek genre
is intertwinedwithcontext. This is not simply a pedantic problem
of classication, for our answer will determine whether we read
Callisthenes and Aristotles works as independent compositions
or as a diptych, one being designed to be interpreted in light of
the other.
We can at least certainly say that the extract fromCallisthenes
represents another early (he died ca. 327) strand in the battle
to dene Hermias. As such, it interacts with other accounts
that contradict Callisthenes on what exactly Hermias suffered
in the end. The basic conict is that Theopompus (115 FGrH
F 291) says Hermias endured brutal outrages to his body and
was impaled, whereas Callisthenes says that he was accorded
some exemption from Persian torture because of the manly
virtue, the aret, he displayed (6.810).
9
I cannot sort out what
lies behind these Greek reports of Persian barbarity, but it
can be revealing to ask on what authority Callisthenes reports
Hermiass last words. In reply to those who wonder how an
authentic report might have been carried from Persia to Greece,
Jeffrey Rusten suggests that a messenger from Philip could have
passed through Artaxerxes court and brought Hermiass words
back. This persons name would then presumably be what is
missing in the papyrus after t.t in 6.14; of course a number
of (proper) names beginning in Ph have been suggested to
ll the lacuna. But it seems at least equally possible that, as
Harding thinks, Callisthenes report was a rhetorical fantasy
of a great mans dying words; after all, famous last words were
already a popular anecdotal form (such as I leave lest Athens
should sin twice against Philosophy).
10
Herodotuss story of
Croesus quoting Solon while on the pyre (Histories 1.8687)
and Cyruss being moved to clemency at this point suggests
that the image of the Persian tyrant touched by the spectacle
of philosophical suffering was an archetype.
11
Considering that
Callisthenes does not source his account of Hermiass end,
we may identify his audience with the companions (6.14) to
whom the report is being conveyed from Susa. Whether or not
Callisthenes wrote for a particular ceremony, he seems to have
written for a particular audience, a restricted group who knew
each other and at least indirectly knew Hermias; before such a
group the historian felt no need to avow his trustworthiness or
to disarm skepticism.
With this scenario in the background, we may further
characterize the audience Aristotle aimed at by considering
again the functional differences between a lyric ( .o) and
a reciteable verse (
o).
12
As said above, a song for Hermias
was something quite different from elegiac verses, inscribed or
not, and entailed a more restricted audience by its very form.
A song is memorizable in a quite different way from stichic or
distichic verse: the words of a Greek song were held together
they scanned metricallyonly by the melody unique to that
song; one does not recite .y.one sings them. Elegiacs by
contrast have a regular, audible rhythm that is perceptible
without melody; as in an English verse like iambic pentameter,
their rhythm depends only on the normal prosody of the words
so that an elegiac couplet can be read off the page satisfactorily.
It follows further that an elegy can be transmitted by writing it
down: it takes only literacy and a little familiarity with verse to
operate the text. Asong, however, has to be heard to be learned.
Because the Greeks had not yet developed an adequate musical
notation, a tune ( .o) was not like an epitaph you could leave
out in the sun for passersby to read; a potential singer of the song
had to hear it from another singer. This means that, simple as
its melody may have been, Aristotles song was bound to a more
restricted audience than an epitaph at a Panhellenic sanctuary. A
song, moreover, passes most easily within a like-minded group,
or among extended groups that are sympathetic enough with
one another to sing to each other (at least at rst, when it
is still a potential performance piece and not yet a precious
historical document that is carefully copied out and saved). If
this is so, it is understandable that Aristotle felt no need to name
Hermias directly in the song, whereas his name was likely to
have featured in the title of a prose o uyy,ee about him.
13
A nal respect in which lyric differs from epigram is in its being
suitable for singing in unison. Elegiacs were normally recited by
individuals, but, as noted, Aristotles dactylo-epitrites in Doric
dialect could be performed by a chorus. This is a main reason
that Wilamowitz took the song as a liturgical work, but let us
note for now that it was also simple enough in language and
rhythm, andpresumably inmelody, as not to require professional
performers to be passedaround, something not true of other lyric
compositions of the day.
Considerations of form are a main reason not to adopt
the antithetical view to such historicizing reconstructions as
Wilamowitzs: this would be that the song for Hermias was never
intended for an actual ceremony and that the picture it gives of
itself as a ritual utterance is a ction, a poetic stance. The song
need not refer to any performative context but works as a self-
enclosed composition, giving its listeners the pleasure of being
present, vicariously, at a high-toned and solemn commemorative
occasion. One couldcompare Callimachuss collectionof so-called
mimetic hymns, hexameter and elegiac songs that sometimes
(esp. Hymns 5 and 6) pose as cult hymns, even though scholars
have been unable to attach their allusions to ongoing ritual activ-
ity to any knownrite. Althoughthese literary compositions of the
early third century may read as if they were transcripts of what
was uttered while the rite unfolded, the purpose of such language
is not reference but to invest a poetry modeled on other poetry,
especially on the late archaic Homeric hymns, with the admired
rhetorical quality of vividness (enargeia).
14
If the song to Her-
mias were regarded as a mimetic cult song, one could see Aris-
totle once again as a harbinger of Hellenistic poetics, in the same
way that his Delphic inscriptionfor Hermias seemedtoanticipate
the later vogue for ctional epigrams. This cannot be excluded
as a possibility. But Aristotle does not seem bent on producing a
vivid effect of performance (there are no deictics in his text, for
example) and it is notable that Callimachus created his ctional
ritual experiences in recitable, and therefore legible meters that
canbe fully enjoyed ina study; Aristotles song, incontrast, needs
more than words on a page to produce its full effects.
If Callisthenes Hermias may be read either as a performance
piece or as a prose text, considerations of form support the
hypothesis that Aristotles song was sung at rst within the circle
of the Peripatetics and their friends. We may add, if we assume
that Aristotles song was intended for Hermiass intimates, that
the group must have beenrather small if the ceremony took place
inPella soonafter Hermias was killed. But that of course is hardly
the end of the story: the song seems to have been reprised by
Aristotle and his friends when he returned to Athens in 335; and
by the time Demophilus and his cronies get wind of it, the song
was allegedly being sung at the common meals (sussitia) of the
Lyceum.
15
Having gone this far inspeculating about the premiere
of Aristotles song, let us turn to the time when it exploded onto
the public scene.
Sources: Hermippuss On Aristotle
Athenaeus got his account of the trial from others, and to tease
out the historical strands that he braids together it is helpful to
analyze his account into its basic building blocks. Our passage
occurs in the fteenth and nal book of the Deipnosophistai as
both the party and the work wind down. Athenaeuss speakers
break off a lengthy discussion of perfumes (an emolument of
the nest symposia) by calling for wine. A hubbub then arises,
as some propose to toast the Good Daimon, others Zeus the
Savior, and others still Hugieia (Health), and so the savants
consult the testimony of the poets (692f) about after-dinner
toasting. This in turn quickly becomes a learned discussion
of the genre of the after-dinner song, the skolion. Aristotles
hymn will be brought up at the end of this discussion, and will
be pronounced a skolion. The term and some of the practices
associated with skolia can be traced back to the classical age, but
it is impossible toreconstruct the full early evolutionof the genre.
I go into the subject only so far as to understand howsome could
have applied this label to Aristotles song.
By the time of Athenaeus, and probably already in the sources
he cites on the matter, skolia were most easily dened in terms
of occasion: they were songs suitable for social gatherings
(ouvouo tet.694a), preeminently drinking parties or symposia.
As performance pieces, skolia amounted to stylized graces
or after-dinner speeches; their inviolable generic requirement
was to project an ethos appropriate to a civilized party. In
form, they could be either short, simple quatrains easy to
perform or longer, complex compositions calling for some
musical skill to carry off. Athenaeuss discussants derive the
genres name from the latter kind of skolion, which they
understand as the crooked (ox o.tov) song, because when
it came time to sing the less simple forms not everyone
was expected to contribute and the song would follow a
crooked or zigzagging path around the table (693f-694c). The
etymology is highly dubious, but no alternative presents itself:
the question was already unsettled when skolia were discussed
by Aristotles students Dichaearchus and Aristoxenus.
16
But we
can trust that Athenaeus voices a widely accepted ideal when
he says that the nest (x e..toxov) among skolia were the
ones containing some advice or wisdom that was helpful in
living.
17
Athenaeuss scholars thenexemplify the simple kind of skolion
by taking turns performing a set of old Athenian drinking
songs (694c695e). (The collection, which is datable on internal
grounds to the second quarter of the fth century BCE, seems to
have been known to Aristotle, since one of these skolia is quoted
as historical evidence in the Constitution of Athens [19] that was
produced in his school.) The next step in the discussion comes
when someone appends a slightly longer but easily singable
Cretan song that people say is a skolion (695f696a). Once
the issue of generically uncertain skolia is raised, Democritus
of Nicomedia brings up Aristotles song, which he declares is a
kind of skolion and not a paean as Aristotles accusers had alleged
(696a697b):
xo uxv .0 vxv o
\
Ly ox,txo
y 'e ..` e ` yv xe`t x` o
u
\
` o xo u o.ue0ox exou y,e` v A,toxox .ou t
\
E, tev x` ov Axe,v e ou et ev
oxtv,
\
o
\
x` yv x y
e op te xex` e xo u t.oo oou y,e` yv e vyx evo
Ly ot.oto cx [Gulick; for tetcx] e,eoxueo0`t
u
\
' Eu ,u covxo,
\
e opo uvxo xe`t e
covxo
v xo t
ouootx tot o
\
oy ,et t x` ov E, tev et eve. o
\
xt c`
et evo ou c tev
eotv e, t x` o e
oe, e ..` e x v
oxo. tv
\
v xt xe`t eu x` o
t c o
oxtv
eu x y x y .
ev,` ov u
\
tv ot yo.
At this point in the discussion, Democritus spoke: But the
one [sc. skolion] written by the extremely learned Aristotle
to Hermias of Atarneus is no paeanas was asserted by
Demophilus, when he was stirred up by Eurymedon and
brought a charge of impiety against the philosopher on the
grounds that he committed impiety by singing the song as
a paean to Hermias at the common meals they had each
day. But I will demonstrate from its very language that the
song has no feature of the paean but is its own kind of
skolion.
After quoting the song (essentially in the version printed at the
head of this book), Democritus goes on to give an analysis of its
language with which one is bound to agree (696e):
y ` ` v ou x o
t ce t
x t xt xextc tv
v xo uxot c uvexet
etevtx` ov tc te, oe o
\
o.oyo uvxo xo u yy,e oxo
xx.uxyx vet x` ov
\
E, tev ct'
\
v t
,yxv o e y` e,
t. tou o, e Axe,v o
vx,oo y
. tou y,ov
eu y e. ou x
t c' ou c` x` o etevtx` ov
t,,ye,
xe0 e, o
\
t A uoevc,ov x` ov Ye,xt exyv y,e`t o
vx
et ev, o
\
v yot Lo u,t
v xo t Ye tv
ty,eo vot
\
O,ote
co0et
v Y e. et ` ev c'
ox`tv xe`t o
\
t K,ex,` ov
x` ov Mexc ove y,e t, o
\
`v
xxx yvexo A. tvo o

\
cte.xxtx o, yo`tv
\
E,to o
\
Ke..t eto
v
x , x ,`t A,toxox .ou. e
cxet c` xe`t o
u
\
xo
v
L.o t, .u, tovx o y xtvo etc o.
For my part, I do not know how anyone can see anything
specically paeanic inthese verses, since the author concedes
that Hermias has died in the expression, For the sake of
your dear shape, Atarneus nursling left the rays of the sun
bereft. And it doesnt even have the paeanic refrain, unlike
the song that was composed for Lysander the Spartan, which
really is a paean and which Duris in the work entitled Samian
Chronicles says was sung onSamos. And the song for Craterus
of Macedoncomposed by Alexinus the dialecticianis a paean,
as Hermippus the Callimachean says in the rst book of his
On Aristotle. It is performed in Delphi with a boy providing
accompaniment on the lyre.
This defense of Aristotle makes two assumptions about paeans
as a genre. First, it concedes to Aristotles accusers that a paean
addressed to a mortal would, if not ipso facto constituting an act
of impiety, at least violate paeanic idiom (etevtx` ov t c te),
that is, would be uncharacteristic of the genre.
18
Second, paeans
should have a refrain, that is, some form of the cry Hail Paean
(i Paian) should be in the text. Aristotle is exculpated on both
grounds: he plainly admits that Hermias is dead (i.e., this is
plainly to be inferred fromvv. 1516) and his song does not have
any paeanic refrain.
Now what Athenaeus calls (697a) the paeanic refrain (x` o
etevtx` ov
t,,yeor x` o etevtx` ov
t0ye) is indeedvery
common in paean texts, and is taken to be a mark of the genre in
handbooksancient and modernthat sum up genres in neat
recipes.
19
But in point of literary historical fact, not all paeans
had a refrain. Athenaeus himself cites a conspicuous exception
in this portion of his work when he quotes Ariphrons paean
to Health entire, which has no refrain (701f); other ancient
paeans preserved on papyrus and stone can be found without the
refrain.
20
What I suspect is going on here is that the paean-cry
could be used like anamen: insome songs it was integrated into
the lyric, while inothers it could have beenadded by the company
as an extra-textual way of afrming and participating in the
prayer, what Athenaeus calls an added utterance (epiphthegma,
697a).
21
If the questionhad come up for Aristotles accusers, they
could have exploited this variability to insist that the song was
one of those paeans without the paeanic refrain.
Perhaps because he is aware that the refrain is not an
infallible litmus test for paeans, Athenaeus goes on to buttress
his argument by contrasting Aristotles song with a number of
songs addressed to mortals that are certiable as paeans because
they had the refrain. Here we should take note of the sources
Athenaeus cites, for they show that the genre of Aristotles song
had been a matter of dispute for centuries. Athenaeuss list of
legitimate paeans to mortals begins with Duris of Samos, who
wrote in the later fourth century about a paean the Samians
had sung to Lysander, the triumphant Spartan general at the
end of the Peloponnesian War.
22
(More on this song below.) The
second item is key: Athenaeus cites a paean written by Alexinus
of Elis (ca. 339-265) in honor of Craterus of Macedon, probably
Craterus the elder, Alexanders general who died in 321.
23
This
piece of information Athenaeus attributes to Hermippus the
Callimachean in the rst book of his On Aristotle. From the fact
that Hermippus mentioned a paean to Craterus in a biography of
Aristotle we can infer that the dispute over the genre of the
Hermias poem goes back to his day, the later third century.
We can further infer that Hermippus was already arguing for
a position like that of Athenaeus, for the only reason for him to
cite Crateruss paean, complete with its refrain, would be to argue
that the absence of a refrain in Aristotles song precludes it from
being a paean. It may be that Hermippus also gave Athenaeus
the idea that Aristotles song was in fact a skolion. In that case,
Athenaeuss contribution would have been to add to the dossier
of actual paeans to mortals examples taken from later works
on music, for his list includes a paean quoted by Polemon of
Athens, known as a writer of geography in the second century
BCE. Athenaeus alludes to other now lost works in which the
paean-cry was discussed (cf. 701C); among these, we may do well
to recall that Didymus also wrote On Lyric Poetry, in which he
discussed various kinds of hymns, including paeans.
24
Hermippus of Smyrna was a scholar and man of letters who
made his way to Alexandria and evidently had some association
with Callimachus of Cyrene (c. 305c. 240), one of the great poet-
scholars to be found at the library there. Working at the epicenter
of ancient literary scholarship, Hermippus contributed to the
edgling genre of biography with portraits of poets and thinkers,
including On Aristotle. This may have been the earliest biography
of the philosopher; it certainly was an inuential one. It is often
supposed that Hermippus is Athenaeuss prime source for the
history of the song,
25
and he is among the authorities cited
by Didymus in his discussion of Hermias.
26
Hermippus is also
prominent in the Life of Aristotle by Diogenes Laertius, though
Diogenes probably knewhimin extracted formand lled out the
story about the trial with material (sometimes contradictory)
from Favorinus.
27
It is worrisome if Hermippus is Athenaeuss sole source, since
he has had a spotty reputation for reliability among scholars.
28
His most recent editor, Jan Bollanse, gives a balanced account:
he underlines Hermippuss erudition but points out that he
appears to have had a weakness for dramatic stories, and that
he was drawn to gures like Anaxagoras (1056 FGrH F 65) and
Socrates (F 67) whose ideas led to their being tried for impiety.
29
Hermippuss appearance on the scene must make us concerned
that the story of the trial may contain elements of ction. That
this is certainly the case appears when Athenaeus goes on to
quote what is alleged to be Aristotles actual defense speech
(e o.oy te) from the trial.
Authenticity: Aristotles Apology
After cataloging actual paeans to mortals, Athenaeus continues
his defense of Aristotle by quotingpossibly depending on
Hermippus for this as wella sentence from a Defense Speech
Against Impiety, supposedly by the philosopher himself (697A =
Aristotle Fr. 645 Rose). Athenaeus adds the proviso if it is
not a forgery (t ` y xex yuoxet o
\
. oyo), and we should
say this speech was most certainly a forgery: neither Athenaeus
nor any source says the charge came to trial, and it seems
that if it had, we should have heard much more about the
episode. In addition, ctional speeches of famous defendants
had been a popular sub-genre of rhetorical prose at least since
the 390s when a number of Socratic Apologies were produced
in the aftermath of that famous trial.
30
Even though Athenaeus
is prudent enough to doubt the Apology of Aristotle, it is
cause for concern that a polemical pseudo-literature may have
arisen around these matters early enough to have inuenced
Hermippus, assuming that he continues to be Athenaeuss guide
to the story here. By the same token, even a fake text can supply
valuable evidence, and this Apology may cast a pre-Hermippan
light on the affair and contain one of our earliest references to
Aristotles song.
The sentence Athenaeus quotes from the Apology is an
elaborate antithesis that one can easily imagine being produced
in fourth-century rhetorical schools, perhaps by a novice since
it needs a little emendation to come out nicely balanced: For
if I wished to sacrice to Hermias as if he were an immortal,
I would never have prepared a memorial (mnma) for him as
if he were a mortal; and if I wanted to attribute an immortal
nature to him, I would not have adorned his body with funeral
honors (Fr. 645 Rose: ou y` e, e
v ox
\
E, te 0 utv
\

e 0ev ex ,oet,o uvo
\
0vyx v ye xexox ueov xe`t
e 0evex ttv x` yv uotv pou. ovo
txe tot e` v xte t
x ooyoe x` o -o e>).
31
The rst clause refers to Aristo-
tles Delphic monument for Hermias (designated a memorial,
mnma). The funeral honors in the second clause could refer to
a commemorative ceremony such as Wilamowitz supposed to be
the setting for the hymn, but in any case must include the hymn
which was a part of those honors, and which indeed was faulted
for wanting to immortalize (e 0evex ttv) its subject and calling
him immortal (e 0 evexov in v. 18). Seeing a reference to the
hymn in the second clause gives needed force to the adjective
epitaphiois: literally, over the tomb, the epithet counters the
charge that the song was a paean by conceding, as Aristotle had
in his song, that Hermias was dead. A further implication of
the word is to characterize whatever commemorative ceremonies
were practiced as no more irregular in substance than Athenian
state funerals, at which a funeral oration, an epitaphios logos,
was customary.
It is notable that this author draws the Delphic monument
into the accusations against Aristotle, even though that artifact,
in a most respectable shrine and speaking a perfectly conven-
tional verse, had occasioned no charge graver than witlessness.
It does, however, furnish the orator with the rst half of
a pair of balanced antitheses, and this kind of overwrought
rhetoric may be the source of Diogenes Laertiuss statement,
not conrmed by Athenaeus, that both of Aristotles poems
on Hermias were the basis for the charge of impiety.
32
It
may be that the pseudo-Apology responds to an early stage
of anti-Aristotelian propaganda, one that joined his paean
to other allegedly impious behavior. The Peripatetic Aristocles
cites an allegation by Lycon the Pythagorean, whom he dates
to Aristotles generation or the next, that when Aristotles wife
died he venerated her with the same rite that the Athenians
offer Demeter. This bizarre accusation makes a sort of sense
Aristotle immortalizing the daughter as he had the fatherand
the slur may even have had some basis in fact if, as Wormell
suggested, Aristotle staged a grand funeral for his wife, as he
is thought to have done for Hermias.
33
Lycons charge also
resonates with the report in Diogenes and Athenaeus that it
was an ofcial of Demeters mysteries, Eurymedon, who was
the prime mover of the charge of impiety.
34
We do not hear
that Lycon also mentioned the paean to Hermias, but Diogenes
quotes a somewhat later source who explicitly links that song to
a similar Aristotelian offense against Demeter: But Aristippus,
in the rst book of his treatise On Ancient Luxury, says that
Aristotle was enamored of the concubine of Hermias, and that
when Hermias consented to it he married her; so overjoyed was
Aristotle that he sacriced to her in the way the Athenians do to
Eleusinian Demeter. And for Hermias he wrote a paean, which is
written out below.
35
Diogenes is wrong to attribute On Ancient
Luxury to Aristippus, an associate of Socrates who would not
have lived past around 360. This is too early for Aristippus to
have gone into Aristotles connections with Hermias, and On
Ancient Luxury, characterized by Dring as one of the worst
products of Hellenistic calumny, is rather to be seen as a
piece of anti-philosophical invective forged around 250 BCE. The
similarity of both bizarre stories of Aristotle treating his wife
like Demeter may indicate a common source, even if pseudo-
Aristippus differs from Lycon in making Aristotle venerate his
wife while she was alive. If he is garbling (or simply exaggerating)
the report in Lycon, his linking of this charge with the paean
to Hermias may also derive from Lycon, and so the hymn
to Hermias will have featured in anti-Aristotelian propaganda
from a very early time, that is, from Aristotles generation or
the next.
36
Let us turnnext tothe words of the song itself, for the question
of whether the poem is by Aristotle or is a fake has yet to be
addressed. Hermippus was, as noted, a source for both Didymus
and Athenaeus, and if it was he who provided the text of the
poem to them, one may ask where he got it. Before Hermippus,
the song was mentioned by the pseudo-Aristotelian Apology (on
the reading given here), and by the pseudo-Aristippan On Ancient
Luxury, and perhaps before both by Lycon of Iasos. The only
modern scholar I know to doubt that the song is Aristotles
is the collector of his fragments, Valentine Rose. Rose judged
the poem frigid and jejune, even if not without elegance, and
attributed it to some mediocre talent, more philosopher than
poet, writing betweenAristotle andHermippus.
37
Roses frigid
echoes the one ancient source that seems to express doubt about
the authorshipof the song, thoughthe passage has seemedinreal
need of emendation. The text also comes from Aristocles (F 2.5
Caesarani = T58f Dring) as he defends Aristotle fromnumerous
scurrilous attacks, including that of a certain Euboulides. This
Euboulides has been identied with a contemporary of Aristotle,
Euboulides of Miletus, a Megarian-school philosopher who wrote
a book against the Stagirite (cf. DL 2.109). As transmitted in
Eusebius, Aristocles wrote:
xe`t Eu pou. tcy c` ,oc y.
v x xex' eu xo u ptp. t
y ucxet. , xov ` v ot yexe yu,` e ,oo, ovo
\
yy,e oxv e
..v ,`t xo u y eou xe`t x y ,` o

\
E, tev ot xt oxyxo eu x yyovu te,
txe +t. t
eoxv eu x` ov ,oox oyet xe`t x.ux vxt H. exvt ` y
e,eyv o0et x e x ptp. te eu xo u cte0 t,et.
38
Euboulides clearly lies in his book against [Aristotle], rst in
adducing frigid poems about his marriage and his intimacy
with Hermias as though others had written them, and then
by saying that he was offensive to Philip, that he was not
with Plato when he died, and that he corrupted the latters
writings.
This sentence becomes quite signicant if we assume that the
song for Hermias was among the poems Euboulides cited having
to do with the intimacy (ot xt oxyxo; cf. ot xt oxexe in Didymus
col. 5.63) between Hermias and Aristotle. Citing it as an example
of Aristotles toadying to tyrants would be quite apt, since
the song conspicuously mentions their friendship (t. te, v.
21). But in that case, what is Aristocles point about their
authenticity? He seems to say that Euboulides lying or deception
(y ucxet) consisted rst in forging poems, presumably poems
that reected badly on Aristotles marriage and relationship with
Hermias, and theninpretending that they had beencomposed by
others. But it is hard to see that anything turned on the question
of whether the damning verses were composed by Euboulides or
by some other of Aristotles detractors. Accordingly, Mulvaney
has persuaded many that we should insert a word into the
Greek to make it say Euboulides lied in adducing frigid poems
about his marriage and intimacy with Hermias as if Aristotle
had written them when they were by others (ot yexe yu,` e
,oo, ovo
\
A,toxox .ou. yy,e oxv e
..v).
39
On this reading, Aristocles makes a more relevant point in
claiming that the hymnwas one of numerous forgeries by various
hands that were palmedoff as Aristotles by Euboulides. Although
a supporter of Aristotle, Aristocles strategy here is to write off
the poem, which he apparently dislikes, as a malicious ction
so that it cannot be used by the opposition as incriminating
evidence.
However construed, of course, Aristocles text is not evidence
against Aristotelian authorship of the hymn. In favor of poem
being authentically Aristotelian is the fact that it is not the kind
of thing a forger would make up. If the poemwas forged, it was by
someone between Euboulides (who, as Aristotles contemporary,
thus emerges as one of our earliest witnesses to the role of songs
in the charge) and Hermippus (for it seems that Hermippus
reported Euboulides to Aristocles).
40
If the forger were writing
soon after 323, one would expect he would have been more
explicitly impious. Moreover, the song contains no reference
to any major theme associated with Aristotles thought or any
noted episode from his life. One can contrast in this regard the
famous Seventh Letter attributed to Plato, which provides just
the sort of inside information on Platos life and methods that
would appeal to his ancient readers. If this text, like the equally
alluring but more dubious Second Letter, has seemed to some
to be too good to be true, Aristotles hymn by contrast is too
idiosyncratic to be fake. Finally, it is far more common to nd
the ancients devising just-so biographical ctions to explain
authentic, if difcult lyric texts than making up peculiar and
problematic poems around which to spin a ction.
I therefore take the lyric as Aristotles, which is not to say that
the story of the trial is completely reliable. On the assumption
that a real song by Aristotle made its way to later scholars, we
must still be concerned by the fact that ancient critics often made
up special historical contexts that could explain peculiarities in
oldtexts. Wehrli, for example, thinks that the story of the impiety
trial and the defense speech were an invention superposed on a
historically more plausible tradition to account for Aristotles
removal to Chalcis in 323.
41
But precedent supports belief in the
dramatic story, evenif it be somewhat embellishedinour sources.
After all, Socrates was actually tried (and executed) on charges
that included impiety. And in his case, too, a pseudo-literature
arose early: Platos reference inhis Apology (19C) toAristophanes
Clouds shows that in such literature, intertextual squabbling
could begin quite soon after the event. The historicity of that
trial is made harder to know, but is not put in doubt by the fact
that a ctional literature grew up around it.
Nor were Socrates andAristotle the only philosophers harassed
at Athens with charges of impiety. A number of high-prole
impiety trials are attested in the later fourth century that make
sense as attempts to bring down inuential gures, including
philosophers, who were thought to be too pro-Macedonian.
Aristotles associate Theophrastus was charged with impiety in
319, and though he was acquitted, a speech alleging to be his
Apology emerged from the case, possibly in time to provide the
model for the pseudo-Aristotelian Apology.
42
Perhaps the only
secure time for philosophers of Aristotles stripe in Athens was
between 317 and 307, when the Macedonian regency imposed
Demetrius of Phaleron on the city, a man of philosophical culture
and in fact a student of both Aristotle and Theophrastus. We are
told that when Demetrius was later forced to leave the city, he
found refuge with a successor of Alexanders, Ptolemy I, in Egypt;
the result would have been that this student of Aristotle played a
key role in conceiving the great Alexandrian library that would be
realized under Ptolemy II.
43
Demetriuss comparatively benign
rule for philosophers came to an end in 307 when he was deposed
and the city liberated by a new Demetrius, a disgruntled
Macedonian warlord surnamed the Besieger (Poliorcetes). There
ensued another upsurge of anti-Macedonian sentiment from
which another document in the war on philosophers emerged,
this one from the prosecutors side. Almost immediately upon
the Besiegers arrival, philosophers and other teachers found
themselves attacked in a law proposed by a certain Sophocles of
Sounion. Sophocles lawmade it a capital crime for philosophers
or sophists to open a school without a license from the citys
council; the pressure was great enough that Theophrastus and a
number of other philosophers left Athens.
44
One year later, the
lawwas overthrownwhenPhilon, identiedas a pupil of Aristotle
(Athenaeus 610f), charged that Sophocles law was illegal. He
was opposed by Demochares, a nephew of Demosthenes who
had taken part in the prosecution of Theophrastus for impiety.
Demochares Against the Philosophers, according to Aristocles,
reviled Aristotle particularly among philosophers.
45
Like later,
literary slanderers, Demochares pretended to have incriminating
documents from Aristotles own hands, in this case letters from
which he charged that the philosopher had betrayed not only
Olynthus and Athens to Macedon but his own hometown of
Stagira as well. The episode suggests that if an animus against
Aristotle persistedamong some Athenians as late as 306, whenhe
had been dead for more than 15 years, it is entirely possible that
he and his songs could have drawn hostile scrutiny in 323, when
Alexander had just died and resistance to Macedonian hegemony
seemed feasible. This little part of the story at least had a happy
ending: Demochares side was defeated; Sophocles had to pay a
ne; Theophrastus and the other philosophers returned to the
city, fromwhich they would not be expelled againuntil Justinian.
Chapter 5
Genres of Poetry
The previous chapter has argued that Aristotles song to
Hermias is not an ancient forgery and that it was mentioned
early in the denunciations of the philosopher that sprang up
in 323. If these source-critical arguments make the story of
the trial more plausible, they also raise questions about the
literary culture of the time. For the allegation that Aristotles
song was impious turned on a question of genre: whether it
was a paean, a species of hymn, or a kind of song appropriate
to mortals.
1
But when one considers how simply the idea that
the song was a paean is dismissed in Athenaeus, one wonders
what Aristotles antagonists had in mind. Is it credible that
they expected a jury of several hundred Athenians, chosen at
random from the citizen body, to follow, as if they were savants
in Athenaeus, arguments about whether the song was a paean or
a skolion? Would a popular jury feel so much outrage at a generic
misstep as to expose the defendant to capital punishment? The
story of the song at trial thus requires that we consider questions
of a literary historical nature, specically what sort of denitions
69
of paeans were on offer and how far public interest in such
matters reached. I propose to showin this chapter that it is quite
credible that a charge of criminal paean-singing could have been
circulated against Aristotle at the time and that such behavior
could be felt to deserve severe penalties. But to see this we have
to nuance our understanding of literary genres in fourth-century
BCE Greece with religious and political considerations.
Ultimately, I pursue genre to identify features of Aristotles
text that audiences found salient and signicant. The goal is not
to x, once and for all, the true generic identity of the poem, but
to specify how Aristotle evoked and played with contemporary
ideas about different kinds of song. In this way, genre will bring
to light a central dynamic of the lyric, which gives every sign
of beginning as a hymn but seems as it proceeds to lose sight
of its initial hymnic object and to veer off into other business.
A detailed consideration of the hymns formal properties must
await the next chapter. The rst step in characterizing the genre
of Aristotles song will be to consider howthe Greeks broke down
the category we call lyric poetry (.u,tx y, a Hellenistic term),
but which they called songs ( .y, e otce t), into kinds or genres.
The history of lyric classication is a complex area of ancient
criticism that could use more study, but it cannot be left out of
account; it is obvious that we cannot rely on Athenaeus, who
wrote almost half a millenniumafter the fact, to understand how
a paean was recognized when Aristotle was accused. But neither,
as will be seen, can we settle on a single right way to dene
genres in the 320s, for the question was usually negotiable and
the concepts were still evolving. I will focus on how such terms
as hymn (u
\
vo), paean (e tev), skolion, and enkmion were
dened, contrasting early classical usage with the Hellenistic
age when the vocabulary Athenaeus uses was developed. In this
history, political and religious notions will not be absent, which
will help us see that more than literary issues could be involved in
Genres of Poetry 71
taxonomies of song, and why those issues mattered to the people
at large.
Lyric Genres from Plato to Alexandria
To ask what genre a lyric poem belongs to can seem fussy and
academic, especially from a Romantic perspective that tends
to equate lyric poetry with poetry itself. But in Greek literary
culture it is fair to say that a song without a genre, without
a putative context and social function, was inconceivable. In
the performative culture of the classical period, song was for
social life, and different kinds of songs were recognized as
suitable for different events. Because such rules as there were
owed from social contexts, it is important to think of genres
not as recipes that had to be followed to the letter but as
sets of expectations that might be adapted and re-negotiated
for particular occasions. Again, Romantic criticism tends to
envision genre as a constraint on poetic originality, but a classical
perspective would feel that an exclusive focus on the poets
creativity overlooks the audiences legitimate expectation that
performers address appropriate topics and express felicitous
sentiments in suitable language. A good 0, yvo, or funeral
lament, for example, was essentially something that everyone
attending the ceremony would agree had been a suitable thing
to say; a very good one might be remembered and referred to
or re-performed on later occasions. If you were to assemble
enough of the latter kind, you would have materials from which
to deduce general rules for what a dirge-writer ought to aim
at to succeed; this kind of genre-theorizing was exactly what
was going on in institutions of higher study in fourth century
Greece.
In the decades between Aristotles lectures on the art of
poetry, usually dated to around mid-century, and the founding
of the Alexandrian Museum in the early third century, Greek
scholars settled on a vocabulary for dening poetry and its kinds
that is still widely used inWesterncriticism.
2
This obscure period
of literary history seems to have continued a tendency that canbe
seeninfth-century criticismto favor formal denitions of genre
over the more socially based terminology and practice of the early
classical and archaic periods. An example is the word paean itself,
a very old name for a kind of song. The genre term paean derives
froma cult song in honor of the pre-Homeric healing god Pain.
A song of supplication or thanksgiving to this god was called a
paean, a metonymy doubtless supported by the songs use of the
gods name as a refrain. Paean kept this sense even as Pain
came to be displaced by Apollo. A social conception of a paean
would consider the god to which the song is addressed and the
circumstances in which Pain/Apollo might be invoked; a more
formal approach might focus on the refrain Hail Paean!, or the
absence thereof. Conceptions of genre that entailed speciable
properties were especially useful when one wanted to categorize,
and therefore be able to retrieve from library shelves, songs
that had come down independently, without being embedded
in a dramatic text, for example, or in an anthology of verse.
The utility of formalist reductions is evident in the success they
have enjoyed, but we must be wary about attributing them to
Aristotles accusers. To sketch how the main lyric kinds were
recognized in his day, I will begin with songs in praise of mortals,
starting with the early classical period and the epinicians of
Pindar.
A seminal study of the evolution of Greek genre terms by
A. E. Harvey showed that in the fth and fourth centuries the
genre that we and Hellenistic scholars call epinician or victory-
song (epinikion) would have normally been included under the
Genres of Poetry 73
broad term enkmion or revel-song, with its implication of
praise of mortal achievement.
3
Only twice in nearly sixty
epinicans do Pindar and Bacchylides refer to their compositions
as songs upon a victory (epinikia); a far more common self-
description is as a revel-song (engkmios [sc. melos], or simply
kmos), picturing them as sung by a band of young men on
a carouse.
4
This terminology is far from proof that Pindars
songs were actually performed by young male choristers: any
poems references to its own performance may be metaphorical
or ctitious; we cannot tell becauseunlike the case with
Aristotlewe lack reliable early evidence for the how epinician
was performed. indeed, we are solacking inevidence that scholars
differ about such a fundamental question as whether Pindars
odes were choral or solo pieces. Nonetheless, the prominence
of kmos is signicant because it shows Pindar submitting his
songhowever it was actually performedfor social acceptance
under the same terms accorded to a kmos: this form of public
celebration was long sanctioned for a man of outstanding
achievement; one who had, for example, won Panhellenic victory
was entitled, in return for his efforts, to noisy public acclaim led
by those near and dear to him, and so epinicians often pose as
exclamations of the victors boisterous male age-mates. It was
presumably unobjectionable to refer to a Pindaric epinician as a
victory song, as he himself does once, but not very informative,
and indeed almost tautologous in context. (The redundancy is
like calling an anthem sung on July 14th a 14th of July song.)
By contrast, to call an epinician an enkmion implied that it was a
measured and socially appropriate recognition of mortal success
andsince the kmos was more public than a dinner party
contrasted this form of praise with that found, for example, in
sympotic skolia, which could also make praise of a mortal their
theme. This sense of enkmion, not altogether rigid to begin with,
is loosening already in the fourth century when we nd the word
used for any kind of formal eulogy for a great man, whether in
verse or prose.
5
In his Rhetoric, Aristotle tried to impose some
precision in terminology when he recognized a broad category of
speech in praise of humans, whether by poets or prose writers
(2.11,1388b21), and subdivided it: he would use enkmion for
praise of an individuals deeds in particular circumstances, and
epainos (approval) for praise of general personal qualities as
manifested in deeds (1.9, 1367b26-29).
6
But the Alexandrian
editors of Pindars considerable poetic output found themselves
with enough victory songs to ll four books, and so it seems
they decided to put songs having to do (mostly) with victory
celebration into their own (newly titled) class, epinicia; they
resupplied the depleted corpus of Pindars enkmia by taking
in certain convivial songs that would have been called skolia in
their day.
7
Harveys study shows that, for similar reasons, most of the
terminology for lyric genres used in the age of Alexandrian
scholarship either did not exist or had different meanings at
the time when the poets wrote. Plato stands out in the transition
between the two worlds because he rejected both the looseness
of archaic classications and the pragmatism of later literary
systems. His ideal of generic purity, derived fromhis metaphysics
and theology, was the rst to demand that every Greek song have
a single, uncompromisable generic identity. The idea that songs
must belong to one genre or another and that no song should
be mixed is rst explicitly articulated in his attack on modern
musicians (i.e., poets of the last decades of the fth century and
the rst of the fourth) for mixing genres. Ina well-knownpassage
in Laws (700A-E), Plato fantasizes that in the good old days
people regarded prayers to the gods as a distinct kind (eidos)
of song and kept them apart from anything having to do with
songs for mortals. He illustrates by citing traditional hymnic
forms: songs for the gods include the dithyramb, Dionysuss cult
Genres of Poetry 75
song, and paeans, dened as songs for Apollo. (Platos clarity
is purchased at the cost of some reductivism, since, as noted,
paeans are found addressed to Apollos sister Artemis and to
other gods in the fth century.)
8
Plato assigns such songs to
the larger category he calls hymns (u
\
vot), which is the rst
attestation of this word for song in the restricted sense song
for a god. Though humnos was an old word for any kind of song,
in his day hymns were often enough so designated as to make
his restriction of the words meaning plausible. To the class of
hymns Plato opposes all songs for mortals, which he exemplies
by a genre inextricably connected to mortality, the dirge. When
Plato goes on to complain that modern poets go so far as to
mix dirges (0, yvot) with hymns (u
\
vot), he is anticipating, in
theoretical andhighly ideological terms, the essence of the charge
against Aristotles song, which is that it was at once a lament
and a paean for Hermias: in their enthusiasm and excessively
given over to pleasure, they blend dirges with hymns, and paeans
with dithyrambs . . . utterly confounding everything.
9
Platos
viewpoints up the formalismand its theological component that
underlies Eurymedons charge.
Aristotle basically agreed with Plato on these matters, and
his own work in poetics and rhetoric was in line with the
general trend toward making genre terms more formalistic
and prescriptive. Generic classication was fundamental to his
attempt to put the study of poetry on a properly technical
basis in the Poetics or art of poetry (poitik). Students who
attended his lectures learned to dene a tragedy, for example,
without reference to its contextfor example, as the kind of play
one could expect at the Dionysian festivalsbut as a distinctive
form of musico-poetic composition exhibiting a set of formal
and thematic qualities that were laid out in chapter 6 of the
published notes. Aristotle seems not to have gone very deeply
into lyric genres in his Poetics, which in its surviving formfocuses
ontragedy andepic. But we caninfer howhe wouldhave classied
his song from Hermias from the general analysis of kinds of
poetry that begins the work: in formal terms (in terms of its
media), a lyric or a song is a kind of artistic representation
( tyot) that is distinctive fromother forms of mimsis inbeing
composed of melody ( .o), language (.yo), and rhythm all
blended together. Such songs (literally, tunes, .y, a part-
for-whole designation for words set to music) are exemplied in
the Poetics through two popular kinds, the dithyramb and the
kitharodic nome (1447a14 ff.); the pair is probably meant to
suggest that the class of lyric song could be further sub-divided
according to the criterion of objects of imitation that Aristotle
will set out in chapter 2: while dithyrambs were for Dionysus,
nomes were usually for Apollo.
This basic division of song-types following cult recalls Platos
in Laws, and the history of poetry Aristotle gives in the fourth
chapter of Poetics shows that he also subscribed to the idea
that songs to the gods belonged to a separate category from
songs for mortals.
10
Having raised the question of how poetry
and its main genres arose, Aristotle reasons that poetry must
derive from natural human appetites and aptitudes. Where his
evolutionary account will differ from Platos story of a fall is in
referring generic distinctions to inherent human capacities, not
to transcendent religious laws. Aristotle then speculatesfor we
are talking about the deep past herethat two basic genres arose
when serious kinds of people naturally represented the actions
of worthy types, while innately vulgar poets represented people
of a baser sort, the latter composing blame (songs), the former
u
\
vot and
yx te.
11
Subdividing serious poetry into hymns
and encomia compendiously accepts Platos distinction between
songs for gods and mortals. Beyond this, Aristotle has little to say
about specic lyric genres such as the paean.
12
Such questions
he seems to have left to his students, like Dichaearchus, who
Genres of Poetry 77
discussed the nature of paeans and sympotic song, and another
associate, the musical expert Aristoxenus, who considered the
etymology of skolia and debated which songs deserved to be so
designated.
13
The evolving Greek conceptions of lyric genre also underwent
pressure from institutions, notably the Alexandrian library,
founded by a successor to Alexander possibly on the advice
of one of Aristotles students. This was the worlds largest
collection of books (i.e., papyrus rolls), and its sheer size must
have encouraged multiplying literary categories and making
them more precise. Large personal libraries had existed earlier:
Aristotle kept his voluminous lecture notes and published works,
and bought works from others; Athenaeus says his collection of
books was remarkable, even if he is not likely to have been, as
Strabo said, the rst man we know to have had a library.
14
But at Alexandria, scholars were confronted with literary texts
on an unprecedented scale and so needed more ramied and
objective taxonomies. For master-genres like epic and tragedy,
it was usually no problem to t a given text in its appropriate
pigeonhole, but sorting old occasional songs into formal classes
could be tricky, and the schemes for classifying lyric appear
to have developed in a rather ad hoc manner. How librarians
classied songs depended a great deal on how much of a given
authors work was available and what information came down
along with it, as a few classic lyric poets may show.
The evidence for how Sapphos poetry was organized in
Alexandria contains gaps, but it seems that only one of the
nine (or eight) book rolls preserving her songs was unied by
a social occasion, the wedding songs (epithalamia). The rest
were grouped primarily according to meter, an expedient that
was made possible by the fact that very many of the songs
Sappho composed for her circle favored a limited number of
short stanzaic or distichic forms that were repeated from song
to song.
15
The editors formalism was encouraged by the fact
that the social functions her songs had lled in archaic times
had become little understood. In contrast, Bacchylides songs in
various meters were organized according to occasion: one book
of epinikia, then of dithyrambs, paeans, hymns, processional
songs (prosodia), songs for girls choirs (partheneia), dance songs
(hyporchemes), erotica, and enkmia. We know how the lyric
corpus of Pindar was organized in more detail: scholars used
the same occasion-based system as for Bacchylides, with the
difference noted above that Pindars erotic poems were included
among the class of enkmia. Further classicatory renements
were wanted for Pindars epinikia, which survived in sufcient
quantity to require four books. Each of these was dedicated
to a single festival where the victory was won (a method
possibly devised by Callimachus); within each book, songs were
arranged according to a complex logic that took into account
the prestige of the contest and the number of victories of
the winner.
16
Still, some texts remained hard to pin down: an
ancient introduction to a song now cataloged as Pindars second
Pythian ode records that, Some say it is not an epinician;
Timaeus calls it a sacrice-song (0uoteoxtx y); Callimachus a
Nemean, Ammonius and Callistratus anOlympian, some (suchas
Apollonius the eidographer) a Pythian, and others a Panathenaic
(II 31.1014 Drachmann). Classes of lyric poems continued to
multiply thereafter, for there is no end to the taxonomic drive,
as Barthes describes it, trying to hold within a necessarily more
and more discriminating network the manners of speaking, i.e.,
trying to master the unmasterable.
17
By the fth century CE,
Procluss Chrestomathy could list 28 genres (eid) of lyric poetry,
broken down into four main categories: songs to gods, songs for
humans, songs to both, and occasional songs addressing neither
gods nor mortals particularly.
18
Here, at last, epinikia recover
their old place beside enkmia, and together with skolia make up
Genres of Poetry 79
the super-category of songs for human beings. It is noteworthy
that, almost a millennium after Plato, the basic terminological
distinction between songs for gods and songs for mortals is
preserved, even as the system has become very elaborate.
19
It was at Alexandria that Hermippus researched his biogra-
phies in the later third century. He is called the Callimachean
by Athenaeus, thoughthis need not meanthat he worked directly
with the scholar-poet. He could nonetheless have imbibed a great
deal of Callimachuss literary system by perusing his Tables or
Pinakes, the 120 book-rolls he had composed as a guide to those
distinguished in all branches of learning and to what they have
written in the library.
20
Hermippus at all events seems to have
dened paeans in the same way as Callimachus, to judge from
a scrap of papyrus from Oxyrhynchus in Egypt that records a
dispute over genre at Alexandria. The papyrus concerns a poem
called Cassandra, probably by Bacchylides.
21
According to the
note, Callimachus had classed the poem as a paean, whereas
Aristarchus of Samothrace, head of the Library inthe mid-second
century BCE, argued that the poemwas a dithyramb. The papyrus
gives the grounds for each scholars choice, but there is, once
again, a gapthe gods of papyri seem to enjoy teasing genre-
criticsjust where the decisive word occurred: on the reading
proposed by Edgar Lobel, the papyruss rst editor, Callimachus
called Cassandra a paean because it had a version of the paeanic
refrain (reading epithegma). whereas Aristarchus argued that the
prominence of myth in the poem made Cassandra a dithyramb;
he held against Callimachus that the paeanic refrain could be
found in other kinds of poems besides paeans.
22
Athenaeus, who
it will be recalled was relying on Hermippus the Callimachean,
pointed to the lack of refrain in Aristotles song as proof that it
was not a paean.
This, then, was the trend in academic reection into which
Demophilus intruded in 323. Although Plato could give voice to
the reactionarys fantasy that literary kinds should be restored
in line with ancient religious traditions, most of those dealing
with literature took a more pragmatic and empirical approach to
lyric classication. All this time, of course, songs continued to
grace cults, social events, and musical competitions throughout
the Greek world, and so the connection between generic forms
and social occasion was not forgotten; it was philosophers
and librarians who had most reason to prefer neat, formalistic
rather than performance-based denitions of genres. Such an
approach would be happy to have a shibboleth that identied all
paeans (and only paeans) as paeans, despite a certain amount of
messiness in actual practice. But in the 320s all these terms were
being revised: Plato had offered a severely rigorous denition
of humnoi as songs for gods; the category of enkmia was still
accepting new entrants; skolia were easier to dene by their
context than by form, and how to recognize a paean was to
remain for centuries a debatable question. The fact that lyric
genres were being reconceived in the later fourth century as new
approaches encroached on traditional conceptions allows us to
see howthe genre of Aristotles song could be a matter of dispute
among sophisticated readers; but we still have to consider why
a larger public should have cared.
Impious Song: The Paean to Lysander
Werner Jaegers justly inuential study of Aristotle presented
his song as a deant afrmation of loyalty to Hermias in the face
of a hostile world: While the nationalist party at Athens, led
by Demosthenes, was blackening the character of the deceased,
while public opinion was dubious about him in Hellas and
feeling ran very high throughout the land against Philip and his
partisans, Aristotle sent out into the world this poem, in which
Genres of Poetry 81
he declared himself passionately on the side of the dead man.
23
To say Aristotle sent his song out into the world suggests
publication of some sort, but the story of the trial implies that
the song was not widely known outside the Lyceum: if it were
circulating widely, it would have been widely used (for why
concern oneself with it except to sing it?); if it were widely
used, it would have been hard to smear as pernicious. The
only poem Aristotle published on the Hermias affair was the
Delphic epigram. A short song like the Hermias hymn was likely
to have circulated especially by word of mouth, with the result
that it remained mainly within the circles of Aristotles friends
until Demophilus and Eurymedon got wind of it. Its relatively
restricted circulation would have made it all the more shocking
if it were presented as sensational evidence leaked from the
inner sanctum of a dangerous cabal.
In quoting the song, Aristotles accusers would doubtless
have xed on verse 18, which says that the Muses will make
Hermias immortal. Here the ostensibly hymnic form of the
song (detailed more fully in the next chapter) is joined with an
explicit declarationthat its subject is beyond mortality. But to say
that ones commemorative act has made the deceased immortal
was one of the oldest topoi of consolatory eloquence. We can
nd in such classic exemplars as Simonides thrnoi and Pericles
funeral oration the idea that the dead will never die so long as we
remember them.
24
In themselves, the words of this verse are not
excessive, and the charge against Aristotle therefore necessarily
turned on the question of genre. Jan Bolanse (2001, 81) has
judged that it is not hard to see howanAthenianjury could quite
easily have been (mis)led to believe that Aristotles poem was
a paean. One question that scholars seem not to have addressed
is why anyone should have cared.
Is it credible that the anti-Macedonian agitators could have
counted on the public to get riled up over such a point?
Generic expectations are in practice rather uid at any time, and
it is hard to imagine even the litigation-loving Athenians being
eager to enforce literary denitions that were being rethought
and in any case were difcult to apply. However, in the political
context of Aristotles later years, it is plausible that a broad public
would have been concerned over such allegations as Athenaeus
reports. For the issue of whether one is praising a god or a
mortal in song resonated with powerful political currents that
had been set in motion by the rise of Macedon. Alexander and
his father Philip before him showed the Greeks new forms of
absolute power in which, in the words of Momigliano, the
traditional notion of hero lost in importance in comparison with
that of divine man. Greek cities had long offered signal honors
(sometimes in the form of hero cult) to eminent benefactors,
but the self-presentation of Macedonian monarchs augmented a
trend toward blurring the distinction between man and god.
25
Philip II was accused of wanting to be taken for a god, and
when Greek cities later offered Alexander honor as a god, this
made sense of an otherwise incomprehensible intrusion of
authority into their world.
26
From the time of his crossing into
Asia in 334 (possibly passing Atarneus on his way), Alexanders
representations of himself underwent a marked elevation. His
family had traced its lineage through Heracles, but this gave him
a Heracles-like mixed paternity that was heroic, not divine. It
was thus a major step up when he began to represent himself
as a direct son of Zeus and to accept divine worship as Ammon
Ra in Egypt in 331.
27
There also seems to have been an attempt
to introduce into Alexanders court the practice of proskunesis,
the ritual obeisance (or debasement, as the Greeks thought of it)
that was de rigueur at the Persian court. Among those resisting
it was Callisthenes, and his quarrel with Alexander over this
practice is said to have played a part in his downfall and death.
28
Genres of Poetry 83
After his return from the east in 325, a number of Greek
cities inferred that Alexander would be pleased to be offered
divine cult.
29
Demades, an Athenian politician, was prosecuted
and ned for having proposed to make Alexander the thirteenth
Olympian.
30
Alexanders death in 323 made any such request
moot, but its shock was such as not to be forgotten. In the end,
though, Athens was not to hold out against the newtrend toward
treating the eminent living like gods: in291BCEwhenDemetrius
the Besieger had replaced Macedonian tyranny with his own, the
Athenians greeted his arrival in the city with incense, garlands,
and choruses singing paeans and other cult songs, including one
hymn that hailed him as a god we see present before us.
31
Back at the end of the fth century, to sing a paean to a mortal
was a major innovation and a shocking act of obsequiousness.
The traditional way that Greeks honored the very great was
by offering them cult as heroes after they died. In exceptional
cases, hero cult was offered to living benefactors, such as
victorious athletes in the great games.
32
But the paean was
already associated in Homer with Apollo and the healing god
Pain: its purpose was to thank the god for a victory or to pray
for release from ills. The rst mortal to have a paean written to
him that Greek historians knew about was the Spartan general
Lysander at the end of the Peloponnesian war. In the wake of
his stunning defeat of Athens at Aegospotami in 405for which
Lysander had a commemorative monument erected at Delphi
the rulers of Samos (who had been put in place by Lysander) gave
him extraordinary honors: they rechristened one of their old
festivals the Lysandreia, erected an altar in his honor as to a
god, and sang paeans to him. This is reported to us by Duris, the
fourth-century historian from Samos who is cited by Athenaeus
for this unusual paean. In an extract cited by Plutarch, Duris
names Lysander as the rst Greek to receive divine worship while
alive, and records the beginning of one paean (867 PMG = Pai. 35
Kppel):
33
Tv
\
E.. eco e ye0 e
ox,exeyv e ' u ,u,ou
Y e,xe u
\
v yoov,
,
t ` y Het ev.
Let us hymn the Leader
of sacred Hellas,
from wide-wayd Sparta,
Hail Paian!
The verse is little more than a chanta triumphal cry at the
defeat of an enemy harking back to one of the oldest attested
uses of paeans as thanksgiving hymns for military victory. The
use of the verb hymn need have no generic implication beyond
its old, general meaning let us sing.
34
Nor is there anything
suspicious in the epithets: both are traditional, with an ancestry
in epic. They are almost banal, except that the placing of e ye0 e
(sacred in the sense of under divine protection) may invite an
auditor to consider if it is a transferred epithet, in which case the
idea of a sacred general presents itself before being rejected as
violating common usage, not to mention traditional piety.
If paeans to mortals could acknowledge new forms of political
power, we can see why an alleged hymn to a friend of the
self-aggrandizing Macedonian monarchy could raise the hackles
of the Athenian public. We cannot be sure what exactly may
have been on the indictment drawn up by Eurymedon and
Demophilus, but there is nothing implausible inAthenaeuss gen-
eral scenario for the 320s.
35
Insucha political context, Aristotles
old song could have been dredged up as incriminating evidence,
even if only to fan prejudice against Macedonian sympathizers.
Genres of Poetry 85
(The defense speeches of Socrates by Xenophon and Plato show
that his idiosyncratic habit of referring to his daimn, something
like an inner voice urging virtue, was brought up against himand
exaggerated into a sign that he was manufacturing new gods.)
A political gures lyric poetry had gured in another proxy
trial about Macedonian politics around 346, when Aeschines
saw some of his personal love poems read out in court by
opponents who claimed his little verses revealed him as lewd
and hypocritical.
36
Ian Rutherford points to a slightly later
speech by Aeschines to show that a persons way with a paean
couldbe exploitedinpolitical ghts. Demosthenes hadattempted
to smear Aeschines by telling a jury that he hadhadthe effrontery
to sing a paean when being entertained at an ambassadorial
dinner by Philip of Macedon, which was a detestable piece of
obsequiousness givenPhilips recent aggressionagainst Phocis.
37
Aeschines seems somewhat anxious in his defense, not exactly
admitting that he sang the paean but insisting that even if
he had, it was directed to a god and that he had, as an
ambassador, only done the ritually correct thing, and that
(multiplying excuses) Athens had not in any case been directly
harmed by the fall of Phocis: Ifwith my country safe and
my fellow citizens come to no harmif I did join in with the
other ambassadors in singing the paean, so that the god was
honored while the Athenians lost no face, I was being pious
rather than unjust, and it is just that the charge against me
should fail (163: t c` o,0 y y
\
tv x y ex, tco ou
oy, xe`t
x v o.tx v xotv yyc` v e xuo uvxv, ouv ycov x` e x v
e
..v , opv xv et eve, y

\
v txe o
\
0 ` v
xt exo,
A0yve tot c` yc` v y
couv, u o pouv, e ..' ou x y
c txouv, xe`t
ctxe t e
v oo tyv). Rutherford is right, I think, to infer

from this text that even at the time Aristotle composed his
song, the way a person engaged in paean-singing could be
taken as an index of his piety and basic decency. Platos Laws
connected the collapse of traditional distinctions among song-
classes with political instability, and Aristotles paean could
well have struck listeners as a sign that this known Macedonian
sympathizer was willing to praise a tyrant as if he were addressing
a god.
Paean, Hymn, Skolion?
We come then to the question of whether Aristotles accusers
were right: Was the song a paean or some other kind of
hymn or something else? Ancient references support all three
possibilities. Hermippus, as we have seen, seems to have taken it
as a skolion; according to a probable restoration, Didymus called
it a paean (col. 6.19: o
\
y,e`t
' eu x [t Het]e v); Athenaeus

introduces it as a song or lyric poem (e
oe), and argues that it

belongs in a category of its own, being a unique kind of skolion.
38
Diogenes Laertius takes it, I think, as a hymn, for this is the title
under which he quotes the song (5.6.10; cf. 5.5.10); he calls it
a paean when he rst refers to it (5.4.1), but the term may be
taken over from pseudo-Aristippus, whose attacks on Aristotle
Diogenes is at that point retelling.
39
Modern scholars on the whole have agreed that it cannot be a
paean, but differ about what it might be. Wilamowitz concluded
that it belonged to the class of ritual religious hymns, but he
found the lack of refrain decisive against taking it as a paean;
he would not take it as a thrnos because the taking up of the
deceased into the circle of heroes was too pronounced.
40
Richard
Reitzenstein also denied that the song could be a paeanin his
case because he held all paeans were choral, whereas Aristotles
song would have been a monody if, as reported, it was sung at
meals in the Lyceum. Reitzensteins conclusion was similar to
that of Athenaeus: Aristotles song was a free development of
Genres of Poetry 87
the skolion that incorporated some features of the contemporary
paean; his book argued that mixing of genres was typical of
mid-fourth century skolia when the older sort (as exemplied by
Athenaeuss Attic collection) were on their way out and there was
a tendency for the skolion to merge with the paean.
41
Conversely,
H. W. Smyth assumed Aristotles song was a choral piece, but he
could not accept a song directed to an abstraction as a genuine
paean. Neither could he see it as a conventional lament, which
he expected (on a view now discarded) to be in elegiacs; he
presented it as religious hymn inuenced by the style of the
fourth-century dithyramb, which was somewhat less extravagant
than the new dithyramb of the late fth century.
42
Other
scholars see a mixture of two genres in the poem: C. M. Bowra
makes it a paean blended with a thrnos, part of a secularizing
trend inthe history of Apollos song; Plato would have agreed and
deplored the development. Jaeger calls it a hymn to Areta but
understands it as primarily a tribute to Hermias, a view shared
by a number of scholars.
43
More recently, some have recognized
in the song elements of epinician, suggesting an attempt by
Aristotle to transmute Hermiass death into a triumph.
44
Such
a range of views may tempt one to throw up ones hands with
A. E. Harvey: If even in those days people could not always tell a
paean when they sawone, we cannot expect to discover a reliable
criterionourselves; of paeans he expostulatedimpatiently there
seem, indeed, to have been practically no rules at all.
45
These scholars do not consider the possibility of a songs
adopting a new genre when introduced into new contexts. This
clearly happened to Aristotles song if there is any basis to the
allegation that it was sung at the common meals of the Lyceum.
The idea that poems should be consistently and unambiguously
one genre or another, or even a mixture of clearly denable
genres, is useful in attaching labels to boxes of song-books but
not necessarily related to the ways they work in performance.
The Platonic assumption that the generic identity of songs could
not be compromised neglects the power of context to reshape
a songs meaning. In reading song texts, one has less need for
inexible generic rules than for a situation generics in which a
song is what a performer can persuade a given audience to call
it on a particular occasion. Such a perspective suggests that the
genre of Aristotles song was up for renegotiation when it was
performed in new places. If it were, as Wilamowitz imagined,
intoned by all at a memorial ceremony for Hermias, it was a sort
of philosophers hymn. But those who learned the song for that
occasion could carry it away and sing it, for example, at drinking
parties. There it could function as a paean if it were sung by
the company among the preliminary libations or at the partys
close, at both of which points a group paean was customary;
libation paeans were conventionally directed to Zeus and the
heroes, both of whom are well celebrated in the lyric.
46
The
fact that paeans were required at the beginnings and ends of
symposia as well as on a great many other occasions meant that
there was always a demand for songs that could do the job. Their
popularity is implicit in Platos reference to the poet Tynichus,
a one-hit wonder who achieved a single musical success, the
paean that is on everyones lips (Ion 534D = 707 PMG). Finally,
the short song was suitable for performance as a solo piece after
the libation paeans, as a contribution to the evenings drinking
songs in which a skolion was wanted. In this case, the allusion to
sleep with its soft beams in the ofng becomes pregnant.
47
Thus, whatever its original genre, Aristotles song could
have functioned as a hymn or a sympotic paean, and, once
it passed outside the circle of those who used it in a ritual
way, could become a skolion simply by being performed after
the inaugural rituals. In entering new contexts, the speech act
scripted in the text would change: what a hymn prays to be
accomplished becomes in a skolion a generalizing statement
Genres of Poetry 89
offering wise advice about living, as Athenaeus characterized
the genre. One can extend this line of analysis and say that the
song changed genre yet again when it entered the pages of
the Deipnosophists. To be sure, Athenaeus, with his essentially
Alexandrian view of genre, holds that the song falls in the
category of skolia; but when he quotes it along with a number
of other paeans in the closing pages of the work, the song reverts
in effect to being a paean as part of the textual ceremonies by
which he brings his long sympotic book to a close.
The best modern approaches to Greek lyric genres avoid
Procrustean classications, and indeed a recent study by Anna
Santoni holds that Aristotle deliberately combined elements
of ritual song, encomium, and sympotic skolion to make the
song adaptable to different contexts.
48
In a similar vein,
Renehan judges it an experimental poem.
49
There is something
unsatisfying, however, about the extreme position that comes
into viewholding that the genre of Aristotles poem was
whatever occasion and context held it to be. One problem is
that such a perspective gives us no answer to the question
of what kind of song he composed. The question might be
thought too intentionalist, a futile attempt to look into the
philosophers mind, but some answer to it is necessary if we
are to have a view on the legitimacy of the condemnation of
the poem by the prosecution.
50
More generally, it is wrong in
my view to conceive genres as so uid in action that they can
never provide any push-back, so to speak, limiting their songs
adaptability. Boundaries between one kind of song and another
were doubtless negotiable in practice, but the situation could not
have been completely up for grabssomeone who attempted
to recite the Iliad as a skolion, for example, would have been
accounted a boorish symposiast (as Aristophanes Peace 1265 ff.
shows), and a poet proffering skolia is likely to have been drubbed
at a competition in epic rhapsody. If any song could be used
for any occasion at all, the concept of genre becomes vacuous.
Finally, giving generic expectations some sort of traction is not
only necessary to stave off interpretative chaos; the putative
stability of generic form is also a resource by which poets can
make songs more resonant, giving themthe opportunity to evoke
and exploit the social, political, and religious values packed into
conceptions of genre, and to extend them by crossing one
genre with another. Only if performers and audiences agree to
endow conceptions of genre with normative force can genres
arouse and satisfy expectations, or toy with expectations after
having evoked them through signicant forms. Robert Frost
once said he did not write free verse because it seemed to him
like playing tennis with the net down. This is a playful but fair
analogy for the constraintsat once arbitrary and enabling
that a conventional genre can provide. I submit that Aristotles
poem was made to be something, not everything, and I do not
see it as a catchall in the way Santoni suggests. If it was an
experiment, as Renehan concludes in his valuable discussion, it
is legitimate to ask just what Aristotle was trying to prove.
Whatever we decide is the best way to categorize Aristotles
song, this study of Greek lyric genres has suggested that they
had a paradoxical role. Genre must have acted as an anchor,
a drag on free motion; and yet it was an anchor that could
be moved and set down in new places. In a culture that
had at once numerous performance venues and deeply seated
expectations about propriety, genre had to be at once changeable
and authoritative. The genre of Aristotles song might shift as
it moved into new situations, but in those situations it had
to continue to provide some restraint, limiting and stabilizing
meaning before releasing it to move again.
Chapter 6
Kinds of Hymn
Let us turn now from reception to internal evidence, that is,
to the explicit formal indications the text gives of afliation
with other recognized kinds of song. On this basis, Aristotles
song for Hermias appears as a hymn, and indeed seems very
like a fourth century paean, as his accusers charged. Of all
Greek songs to have survived, the closest to Aristotles is a song
quoted in the Deipnosophists and called a paian (701f ).
1
It was
composed by Ariphron of Sicyon, whose date is unknown, and
is addressed to Hugieia, Health personied (813 PMG = Pai. 34
Kppel). Like Aristotle, Ariphron opens with an invocation to a
deiedabstraction, pronounces anepithet, andgives anextended
argument complete with priamel for her worth.
Hymnic Form: Ariphrons Paean to Health
\
Yy tte p,oxo tot ,op toxe ex e,v, x` e o u
ve tott x.tvovptox e, o` uc ot ,,v uv ty.
t y e, xt y` .o` uxou e,t y` xx v
91
y` x e t ooce tovo e v0, ot peot.y tco e , e y` 0v
ou
\
` x,u tot A,oc txe
\
,xotv 0y, uov, 5
y` t
xt e
..e 00v e v0, otot x ,yt y` vv

e vo` e evxet,
x` e o to, exet,'
\
Yy tte,
x 0e. xe`t . et Xe, txv o e,ot
o 0v c` ,`t ou
xt u ce tv
u. 10
Health, the rst of blessed gods for mortals, with you
may I dwell for the rest of my lifetime, and may you be
gracious to me;
for if there is any delight in riches, or in children,
or in the royal power that makes mans fortune match the
gods, or in the longings
we hunt down with Aphrodites secret snares, 5
or if any other heaven-sent delight for mortals or respite
from toils has been revealed,
it is because of you, blessed Health,
that it owers and shines in the talk of the Graces;
apart from you is no man happy. 10
Lucian refers to this lyric as that extremely well known song
on everyones lips, and the number of inscribed copies that
survive conrms it was widely popular.
2
For this reason people
have tended to assume that Aristotle is the debtor, but as we do
not know the date of Ariphron, we cannot determine whether
Aristotle is imitating him or the reverse. Some direct contact
seems in either case undeniable: resemblances between the two
songs extend beyond the merely generic and include meter (both
employ dactylo-epitrites with an anapestic lead-in), themes, and
a number of specic phrases (for hunting, being like divinity,
toiling, and longing). The main difference is that Ariphron places
Ki nds of Hymn 93
a petition immediately after the invocation, before proceeding
into his argument (for, v. 3).
3
There is no reason to think that either Aristotle or Ariphron
was doing something shocking in addressing a hymn to an
abstraction. A song in praise of Virtue would have been no more
unconventional than the song composed to Kairos, Opportunity,
by the fth-century litterateur Ion of Chios (742 PMG) or a
short anonymous hymn of the same type to Tukh, Chance (1019
PMG).
4
Greek mythopoetic thought drew no clear line between
divinities and personied abstractions: a passage like Hesiods
extended description of Phm or Rumor in the Works and Days
can be read as praise of a minor deity, as a vivid description of a
social process, or even as an allegory. One distinction historians
of religion sometimes make is that full-edged divinities, unlike
abstractions, were the object of actual cult. This may be useful
if it is borne in mind that the line between them was not
impermeable: in fourth-century Athens there was an altar in the
city center to Rumor.
5
Returning to Ariphrons hymn, the epithet ,op toxe in his
invocation stands out in its form (the usual superlative of this
ancient adjective of unknown etymology is ,op uxexo) and
perhaps in its meanings, which suggest that his poem, though
popular, sounded some modern and even ironic notes. My
rendition, rst of gods combines two notions the word can
project, very venerable and, because age was associated with
honor, eldest.
6
Although Health was not the rst born of
the gods in any standard account, the implication that she is
very old (a possible nuance of the superlative) is appropriate
in two ways. One is obvious and conventional: Health is a
very ancient divinity and so venerable.
7
However, the lack
of support in traditional theogonies for making this (rather
new) divinity so early may encourage us to take the epithet
anthropologically, suggesting she was one of the earliest
divinities to be granted cult. This would be in line with an
idea found among Aristotles enlightened contemporaries that
religion began with elemental goods being deied by primitive
people. The sophist Prodicus, for example, had taught that
Demeter was nothing more than a personication of early mans
appreciationof the importance of bread, as Dionysus was of wine.
Health is obviously a primary good of the same sort, and since
health, like bread, is necessary for the races survival, it can be
reasonably afrmed that she was among the rst divinities to be
venerated. This double implication of the epithet, traditional and
innovative, is signaled in the pre-posed word for mortals (v. 1):
this at once strikes a note of piety (mortals cannot hope to live
forever, but may pray for health at least) and of anthropological
awareness (venerating abstractions is what human beings do). It
did nothing to curtail the popularity of the song that its opening
phrase, rst among the blessed gods as far as mortals are
concerned, expressed a sophisticated, humanizing attitude in
religiously unobjectionable terms. In their study of Greek hymns,
Furley and Bremer associate the songs to Health and Virtue as
Philosophical hymns, anticipating the Hellenistic trend away
from the gruesome myths of the Olympians.
8
The witty inventiveness of Ariphron (which Aristotle shared)
is played out in prose in the after-dinner speeches in Platos
Symposium in honor of Eros, passion personied. Although
Hesiod had assigned Eros an early place in Greek cosmogony
(Theogony 120), he played a minor role as a mythological gure
and seems not to have been the object of cult before Hellenistic
times. Platos rst speaker, Phaedrus, champions Eros as a god,
and complains we have numerous hymns and paeans composed
by poets to the other gods, but for Eros we have not even an
encomium (Symp. 177AB). Phaedruss prose performance has
much in common with Ariphrons hymn: he begins by proving
that Eros is among the oldest gods with mythological evidence
Ki nds of Hymn 95
that includes Theogony 120 to afrm that it is agreed on all
sides that Eros is among the oldest gods (ou
\
x o..e0v
o
\
o.oy txet o
\
E
,
v xo t ,op uxexo
t vet); being very

old, he continues Eros is also the cause of the greatest boons to
us (,op uxexo c` ` v y toxv e ye0 v y
\
tv et
xt
oxtv,
178AC).
9
It is all (Socrates speech excepted) sophisticated fun:
later the exquisite Agathon politely takes the opposite view and
argues for Eros as the youngest of the gods (195C). In this
move Agathon seems to have had a predecessor in Ions hymn to
Kairos mentioned above (742 PMG), which saluted Opportunity
as the youngest of the immortals, presumably because kairos
in the sense of the right moment to strike is always arising
now.
Turning back to Ariphrons song, we observe its body is
mostly priamel, listing aspects of human life that are better
when accompanied by health (vv. 37). The elements are not
ranked in any obvious way, but four ors clearly punctuate the
series before the period closes with a return to Health in v. 8
(the vocative absent since vv. 12). The extended list varies the
hymnic topos of listing the gods many cult locales: Ariphron
begins with (a) wealth, which also comes rst in Aristotles
priamel (gold, v. 7),
10
and then names (b) having children,
(c) royal power, and (d) Aphrodite. This can be closely correlated
with Aristotles tripartite priamel comparing Virtue to (a) gold,
(b + c) noble ancestry,
11
and (d) the sensual pleasure of sleep,
perhaps not unconnected with sex.
12
Ariphronconcludes (vv. 810) ona complex and self-reecting
note (as we will see Aristotle doing). Having shown how
many human activities ourish (thallein) with Health, he adds
that she is also the precondition for their being discussed in
graceful conversation. This nice old, partly obscure word for
conversation (oaroi) is suited to cap the priamel because it
could also be a stylized way of referring to poetry (as in Pindar
Nemean 7.69).
13
The implication is that performing the song
itself is an expression of the kind of happiness Heath makes
possible. The connection between Health and poetry would be
especially vivid if the song were performed as a sympotic paean,
making it at once a prayer of thanks and an instance of the
goods for which the group is thankful. In such a case, the paeanic
function of the song could have been signaled by the companys
adding hail Paian at the end. Given the brevity and simplicity
of Ariphrons song, we should also consider the possibility that
in sympotic contexts, the same song, with no paian added, could
have functioned as a solo piece, one of many skolia praising the
joys of health. Sympotic songs often represent themselves as
a form of gracious speech, and Ariphrons song contains in its
oarois an allusion to such stylized talk high-minded, pious but
enlightened reections on human life and the good. In such a
context, Ariphrons would have been a wise and popular choice,
and the performance would have conrmed the songs assertion
that Every happy thing depends on Health.
14
If Ariphron shows what a paean could look like in the fourth
century, one can see that William D. Furley and Jan Maarten
Bremer had good reason to include Aristotles song in their
collection of Greek hymns. They recognize that the nal purpose
of the song seems to be to honor a mortal, but remark with
justice that Aristotle wanted his poem to sound like a hymn
(1.265). It is hard, however, to go further and try to decide,
as is often done, whether either song was a religious hymn
for actual cult or a literary composition executed in hymnic
style.
15
One can conceive of situations in which Aristotles or
Ariphrons song could be used as part of a ceremony appealing to
a god; but by virtue of being arguments for praise, both texts also
constitute coherent and forceful statements of practical ethics
suited for other occasions. We will consider in the following
Ki nds of Hymn 97
chapter onethos howAristotles song might have playedas a form
of intellectual debate; at present, it is enough to have troubled
the Platonic idea of genres xed at a songs birth by pointing
to the possibility that the hymns of Ariphron and Aristotle
could have had a second career in schools or dining halls.
16
The
idea of hymns being reused as other kinds of songs is not often
entertained by literary scholars, and so may be worth supporting
by an earlier example in which cult song-forms are adapted to
other ends. It comes from the rst half of the fth century and
from an author no one would accuse of impiety.
Hymnic Flexibility: Pindars Fourteenth
Olympic Ode
As extensive as are the resemblances between Ariphrons paean
and Aristotles song, they do not prove that the latter is a paean
as well. For it was a long established lyric practice to borrow the
modes and style of hymns and incorporate theminto other kinds
of song. Pindar often put a hymnic faade, as he put it (Ol. 6.4),
in front of his epinician odes; indeed, a number of these hymnic
proems are, like Aristotles, addressed to personied abstractions
such as Fortune or Peace.
17
One of Pindars shorter epinicians is
especially worth reading in relation to Aristotles song, since its
rst stanza resembles the philosophers hymninseveral respects.
The ode repays study as an example of a songs projecting the
context of its own performance, and also as suggesting how it
too might have been adapted for re-performance.
The ode we call Olympian 14 celebrates the crown won at
Olympia by Asopichos of Orchomenos in 488. Orchomenos was a
major city in the central Greek region of Boeotia and was known
as the site of a very ancient cult to the Graces. Pindar accordingly
seems to anchor his occasional poemby invoking these Graces in
hymnic fashion:
18
Keto tv u
\
c exv
.eo toet, et
\
x ve tx xe.. t.ov
\
c,ev,
.te, e e
o tctot peo t.tet
X e,tx O,ovo u, e.etyvv Mtvu ev
toxoot,
x. ux,'
`t u
oet o` uv y` e, u
tv x` e x,v` e xe`t 5
x` e y.ux ' e
vxet ` evxe p,oxo t,

t oo, t xe., t
\
xt e y.e e v y,.
ou c` y` e, 0o`t ov ev Xe, txv e
x,
xot,ev ototv o,o` u ou
x ce txe e ..` e evxv xe tet
,yv
v ou ,ev , ,uoxoov 0 vet e,` e 10

H u0tov A..ve 0,vou,
e veov o povxt ex, O.u toto xt ev.
xvt' Ay.e te
t.yo to. x' Eu ,oo uve, 0 v x,ex toxou
e tc,
exoo tx v uv, Oe. te x 15
,eo to., t co toe xvc x ov
' u v t x ue
xo ue ptp vxe Auc y` e, Ao tov x,
v x . xet e tcv
o.ov,
ou
\
vx' O.utvtxo e
\
Mtvu te
o u
xext. .evoxt e v uv cov 20
+,ove
.0', Ao t, ex,`t x.ux` ev ,oto' e yy. tev,
K.ceov o
,' tco to,' ut
\
v t
o
\
xt ot
\
v ev
x.ot e,' u cot H toe
ox evo xuc tv e 0.v x,o tot e txev.

By Cephisus waters,
Your lot to dwell in a place of ne horses,
O, celebrated in song, royal
Graces of splendid Orchomenos, who oversee the ancient
clan of Minyans,
Ki nds of Hymn 99
hear me as I pray. For it is with you that all that is
pleasant 5
and sweet arises for mortals
whether someone is skilled in art, or is fair, or gleams with
success.
Not even the gods do without the august Graces
in arranging a dance or feast; they are the stewards
of all that is done in Heaven, and setting their thrones
beside 10
the one with the golden bow, Pythian Apollo,
in reverence they make the honor of their Olympian father
ow forever.
Lady Aglaia
and Euphrosuna fond of song, the greatest gods
children, be our witnesses now, Thalia, too, 15
lover of song, as you behold our band of high stepping
revelers
at this happy event: for to sing of Asopichos have we come,
well rehearsed in the Lydian mode,
because the city of the Minyans has won at Olympia
with thanks to you. And now, to the dark hall 20
of Persephone, Echo, bring the resounding message
to his father, and tell Cleodamus when you see him
that his young son, in the famous glens of Pisa,
with wings of glorious conquest crowned his hair.
It is common in Pindaric epinicians that the specics of the
victory being celebrated are not given immediately; here we are
well into the second stanza before we hear details of the victory
the name of the victor, his parentage, home city, and the venue
(vv. 1723). Only after this stanza has sounded, in fact, is it clear
that that what began as a hymn is a victory ode. The two-stanza
song seems to evolve from one genre into another.
Because of the way particulars are withheld, the rst stanza
holds together as a self-containedsong for no particular occasion.
It can play as hymn to the Graces, a general song of praise
without a petition. The opening epithet-phrases invoke and
localize the Graces, calling themto the Cephisus in Orchomenos,
one of two famous rivers by that name. The body of the hymn
then praises them by adducing arguments (for, v. 5) to show
how extensive is their domain: a priamel lists three spheres of
humanexcellence withwhichthe Graces are intimately involved
wisdom (which includes poetry), physical beauty, and success
(which includes athletics) all come about with the Graces.
19
The priamel is capped by turning to their role among the gods:
surely, the Graces are stewards of divine felicity, for it is
unthinkable that the dances and feasts of Olympus should lack
graciousness.
20
As a hymn, the rst stanza offers no myth,
in the sense that no traditional story is alluded to; but there
is mythologizing in the image of the Graces setting up their
thrones beside Apollos. This snapshot, executed in the third
person rather than in the you-style more common in the hymn,
encapsulates a common plot in archaic hexameter hymns in
which one of Zeuss children, such as Hermes or Apollo, is shown
making his way to Olympus and claiming an honorable place
among the great gods. With this image Pindar grants Panhellenic
importance to a proud ancient cult of Thebes.
Taken by itself, the rst stanza of Olympian 14 has numerous
similarities to Aristotles hymn (and to Ariphrons as well).
Formally, both contain an opening invocation, a priamel, and an
argument. When Pindar caps his praise of the Graces by moving
fromoccasions of mortal felicity (v. 6) to those onOlympus (v. 8),
he makes the same swerving move as Aristotle, though in the
opposite direction. The end of Pindars rst stanza also recalls
Aristotle in that both leave us with the picture of a maidenly
Ki nds of Hymn 101
groupof Zeuss daughters perpetuating their fathers honor (with
o povxt of 12, cf. Aristotles o pe eu
ouoet, v. 20). Pindar

manages to impart a sense of closure to the stanza with the
evocative adjective ever-owing (aenaon given emphasis by
its separation from the noun it modies in v. 12): the metaphor
returns us to the river waters in the opening line, while the
religious charge in ever-owing concludes the mini-hymn on
an image of eternal festivity.
21
Through this combination of ring-composition and imagistic
poise, Pindars rst stanza achieves such closure that it is not
hard to imagine it being sung separately as a one-verse song
about the size of Aristotles. By placing in the second stanza
the details about Asopichos and his victory, Pindar has made
the rst more generally relevant, all the more so by including
no specic request. At Orchomenos this mini-song could be
used in a number of contexts. Feasts followed athletic victories,
and such occasions would welcome a pious skolion (cf. p,oxo t,
v. 6), an acknowledgment that festivity and song are divine
gifts, and a reminder that these pleasures derive from properly
conducted prototypes on Olympus (vv. 89). Symposia would
also have afforded an opportunity to replay the song as an
epichoric paean: an address to one of the most ancient cults of
the Graces may perhaps seemnot easily convertible into Apollos
song though he is notably prominent (v. 11). The opening
libations of symposia, however, called for songs addressed to
various divinities, and this stanza could be suitably performed
while wine was poured to the gods and the group prayed. Zeus
in particular was regularly invoked in sympotic prayers, and in
such a context Pindars closing image of honor owing to Zeus
would coincide with the libations poured out in Thebes. The rst
stanza of this epinician, a prayer-song celebrating graciousness,
would be a perfect grace to begin a civilized feast.
22
My speculation about possible re-uses of part of a Pindaric
ode ought to be balanced by restating that in this case, as usual,
we know nothing about the performance of the song that does
not come fromthe song.
23
But even on that limited evidence, we
can say that the song envisions its own repetition outside of its
premiere. Lets put the two stanzas back together and focus on
Pindars most striking literary conceit, the invocation of Echo at
the end. It is worth noting that Echo in some accounts was the
daughter of the river Cephisus, and so may be apostrophized here
as a gure of local cult. Be that as it may, invoking Echo raises
the possibility that the songs message will be heard by other
audiences beside the one awaiting the chorus at a Theban shrine.
In summing up for Echo the information to be relayed to the
victors deadfather (vv. 2224), Pindar incorporates intohis song
its rst echo, so to speak. Inaddition, the rhetoric of these closing
verses conates Asopichoss dedicationof his crownto the Graces
with his original crowning at Olympia and so makes the present
performance of the song a follow-up to an earlier, original rite.
Even at its putative premiere before the Theban Graces, then, the
song posed as a sort of reenactment, and appropriately so since
all human occasions for enjoying wisdom, beauty, and glory are
reections of the divine festivity that the Graces superintend
in the hymn. Pindars song in many respects tied closely to a
particular Theban cult and locale thus closes by making the act
it professes to accompany a ritual repetition of a gesture already
performed, and one whose substance is to be preserved into the
future. The gure of Echo reinforces this conceit, authorizing
future repetitions of the song, even past the barrier of death.
Crossing this nal barrier opens the possibility for the song
to convey its message to an entirely different set of listeners,
including readers who were not present at the Graces shrine on
that happy occasion.
Ki nds of Hymn 103
It remains of course possible that all my suggestions are off or
wrong andare basedona misreading of statements inPindar that
ought to be taken at a symbolic level. But to read what presents
itself as Theban cult song as a context-less fantasy is to turn
it into an artistic phantasmagoria one has difculty imagining
being of interest to, far less being commissioned by anyone, least
of all by Greek aristocrats anxious to advertise the excellence of
their families and cities. My own response to this predicament is
that the danger in attempting to conceive a songs contexts is not
so much that we may write a historical novel as that we may write
a bad one, forgoing vital detail by not pressing the poem for all
the evidence it gives about its situation in the world. To change
the analogy, reading Greek lyric withnothought for performative
context is like regarding Greek sculptures as pure white objects,
forgetting the vivid colors and jewels with which they were once
bedecked but which time and pilferage have stripped away. In
some Greek lyric texts, as sometimes in sculpture, residual ecks
of color remind us of the fuller range of their effects in situ.
Such a reminder is the way that Olympian 14 represents itself as
a processional, sung by a troop of revelers celebrating Asopichos
by Cephisuss waters, for I repeat that the picture a song gives of
its own performance qualies it signicantly, even if the scenario
was never actually played out.
The declared destination of this high-stepping band of singers
(x ov, v. 16) is the shrine of the Graces, where they hope
the hymn turns out to have a formal petition after all (v. 15)
to be received. The deictic in v. 16 referring to these revelers/
performers is the moment when the song most concretely ties
itself to a context and event, which prompts further speculation
about possible cult performance. Crown-bearing processions
were a common form of Greek religious celebration, and if such
were the rst context of Pindars song, the nal image would
take on extra point: the song ostensibly describes the victor
bearing his wreath through the streets of Orchomenos, but it
ends by making the bestowing of the votive at a Boeotian shrine
coincide with the crowning ceremony at Olympia. The result is
that the this in this chorus is a little destabilized, reenacting
a past celebration and beckoning to a re-performance in the
future. This motile stance is supported by the songs metrical
scheme. Matching stanzas rather than triads supports the self-
description as a processional hymn performed on the way to the
shrine. Triads are best performed by dancing so as to lend visual
reinforcement to the songs complex metrical articulation into
turn, counter-turn, and stand; but the isometric, end-stopped
stanzas of this song canbe repeated without breaking the rhythm
of the procession. Ending on an echoing note also allows one
to make a repeating loop of the song, as it must have been
sung again and again while the parade went down the streets of
Orchomenos. Wherever else it was carried, Pindars song let his
patrons bring their Pan-Hellenic glory back home in resounding
fashion.
Pindars hymn to the Graces, then, may support the
hypothesis that Ariphrons and Aristotles hymns were re-
performable pieces, and that they may contain reections on
their possible reuse in the future, for good hymns were reused.
To conclude this considerationof lyric genres, we should contrast
the one other most important form for Greek hymns, though
prosodically very different, the recited dactylic hexameter. The
short example I select will afford a convenient opportunity to
notice two hexameter lines from other hymns Aristotle wrote.
More importantly, comparing sung and recited hymns will show
that, although the Greek poetic tradition developed an array of
distinct musical modes and distributed these among different
contexts and occasions, Greek poets delighted in translating the
themes of one idiominto another, inother words, inadapting the
Ki nds of Hymn 105
same kind of speechto different performative styles. The result is
to see the extraordinary degree of integration among the entire
poetic tradition, which can make individual poems richer and
more intelligible.
Hymns in Hexameters: Homer and
Aristotle
To begin, let us return to Ariphrons epithet , optoxo, which
we remarked was used in a way that did and did not rely on
its possible sense of oldest. Surprisingly for a word that can
be used loosely as very venerable, presbistos is not widespread
as a laudatory epithet before Ariphron, and his use was subtle
enough that later readers had to explain why it was appropriate
to health.
24
One other instance of presbistos in the sense of
eldest is worth a look because it occurs in a hexameter hymn
that shows that certain hymnic topoi we have met in lyric
remain recognizable when executed in the meter and dialect of
heroic epic. There are also, unsurprisingly, differences entailed
by specics of each performative mode, notably in this case a
different strategy to attain closure.
Hymns in recited hexameters are attested from the archaic
age, and are mainly known froman ancient anthology attributed
to Homer but containing works from all periods.
25
One of its
shorter items may have been composed later than Aristotle,
though the language is so traditional it is hard to be sure. It
shows in any case how hymnic form persists across different
metrical modes:
Ie tev e yxt,ev e tooet y
u0 0.ov
,op toxyv, y
\
` ,pt
`t 0ov`t ev0' o
\
o'
ox tv
y
` v o
\
oe 0ve c tev
,xet y
c' o
\
oe vxov
y
c' o
\
oe x vxet, x ec ,pxet
x o 0v o
\
.pou.
x o o c' u
etc x xe`t u
xe,ot x. 0ouot 5
xvte, o u c'
xet co uvet p tov y
c' e . o0et
0vyxo t e v0, ototv o
\
c' o
.pto o
\
v x o` u 0u
,,v xt yoy x x' e
0ove evxe e,oxt.

p, t0t v otv e
,ou,e , opto, y
c` xex' e y,o` u
xx yvotv u 0yv t, o
t xo c'
t.exet
o0. v 10
eu xo`t c' u vo tyot .tv x exe xe..ty uvetxe
xot,ev ouo,' o
.po c` o.` u xe`t .o uxo oyc t
e tc c' u ,oo uvy vo0y. t xuctot,
e,0vtxe t x o,o t ,oev0 otv u
,ovt 0u
e touoet oxe t,ouot xex' e
v0e e.0ex` e o ty, 15

ou
\
x o` u xt yoy ov` y 0` e e
0ov ce tov.
Xe t, 0 v yxy,, e
.o' Ou ,evo u e ox,vxo,

,,v c' e vx'
c y p toxov 0u y,' o
e
eu x` e,
y ` xe`t o to xe`t e
..y v yoo' e otc y.

Earth all-mother I shall sing, she with a rm foundation,
the eldest, who nourishes all things on earth:
whatever passes over the divine land or over the sea,
and whatever ies, these are nourished from your bounty.
You cause people to be blessed with children and
fertility, 5
Mistress, yours is the giving of sustenance and its
taking away
for mortal men: happy is he whom you graciously
honor: for him all stands ready in abundance.
His life-nurturing plough-land is laden, and in his elds
ocks and herds ower, his house is lled with good
things. 10
Men like this are lawful leaders in a city of fair women,
and prosperity and wealth attend them.
Their boys rejoice in youthful happiness,
Ki nds of Hymn 107
in ower-bearing choruses their young girls with glad hearts
play and sport among the grass soft with owers 15
such are those you honor, reverend goddess, boundless
divinity.
Rejoice, Mother of gods, wife of starry Heaven,
and kindly in exchange for my song grant heart-cheering
sustenance;
and I shall call you to mind and another song.
Among the usual hymnic elements present here invocation,
praise focusing onthe scope of the divinitys power, petition the
description of Earths blessings are particularly developed; the
effect is, in Bowras words, a vivid document of the aristocratic
life.
26
For our purposes, more informative may be the close,
which is typical of hexameter hymns but not what we have seen
in Aristotle or Ariphron (or Pindar). This hymn signals that the
end is coming by the stereotyped cry Rejoice (khaire, in v. 17),
a vocable found at the close of many hexameter hymns. This
is followed by a formulaic nal line that is repeated a dozen
times in the corpus.
27
Stereotyped language is emphatic here
because closure needs to be marked more strongly by diction
in hexameters, which, metrically, can be added to ad lib.; in a
song, closure is at least in part signaled by music the melody
gives shape to each stanza, and melodic ends of repeated stanzas
reinforce the end of the whole.
Conventional as this hymn is, each of the opening epithets
has something to arrest attention. Most immediately striking
is y
u0 0.ov, which is found only here in Greek literature.

Apparently coined for this song, having a solid foundation is
appropriate for Earth qua earth, as in Hesiods anthropomorphic
broad-breasted Earth (Ie t' u , uox,vo, Theogony 117). But
0 0.e were also the foundations of temples, and so the epithet
evokes the hymnic topos of praising a gods haunts or places
where his shrines were established. Earth is the seat of all holy
places and so offers a ne point of departure for the poets praise.
More complex is e yxt,ev all-mother.
28
Saying that Earth
was the mother of all is not new (e.g., Hesiod, Works and
Days 563, y y evxv yxy,; cf. Pindar, Ol. 7.38, Aeschylus,
Libation Bearers 128). But the epithet e yxt,e is oddly
formed in its second element, differing from the synonymous
and more regular compound e yx,, which is applied to
earth in the Prometheus (90: e yx, x y y). The second
element of e yxt,ev is possibly reecting a ne point of
Homeric scholarship in Alexandria, which wondered whether in
Iliad 14.259 Night is called the mother ( yxt,e) or subduer
(c yxt,e) of gods and men. It is a complex and not fully
known story, but Earth clearly has the pedigree to claim this
old epithet.
29
So too with ,op toxe: Mother Earth is certainly
very venerable, but there is point in activating its other sense
and naming as oldest of the gods the Earth, who springs up in
Hesiods Theogony immediately after Chaos, the primal void.
30
The epithet shared by Ariphron and the Homeric hymn
is also found in the opening of a hexameter hymn written
by Aristotle. Surveying the books written by the philosopher,
Diogenes mentions one in hexameters (
y) and quotes the
opening line from what must have been the rst poem in
the collection: Pure one among the gods, eldest, far-darter
(
\
Ayv` 0 v , opto0'
\
xexyp., DL 5.27 = Fr. 671 Rose).
The rst epithet suggests, and the third clinches Artemis as
the likely addressee: e
\
yv, pure both as undeled and as
chaste, is an epithet of a variety of gods in early poetry, most
often of the virgin Artemis, though Demeter bears the epithet
in Hesiod;
31
Artemis is clearly pointed to by
\
xexyp. (an
ancient epithet which I render by an old convention as far-
darter because its true etymology was not clear); the epithet
\
xexyp.o is exclusive to Apollo in Homer and Hesiod, and
Ki nds of Hymn 109
preeminently associated with him thereafter. But the form is
potentially feminine and is so applicable to his sister Artemis, to
whom the word is applied in a Homeric Hymn (9.6). Reinforcing
this relation suggests that Aristotles is a hexameter paean
invoking the god of paeans as well as his sister. As there is no
tradition or plausible reason to call Artemis eldest, presbiste
here might be rendered very venerable, unless what followed
was a sophisticated hexameter hymn of the sort Callimachus
wrote, justifying the words other sense in a syncretizing or even
allegorizing fashion.
Of course, if we hadmore of the poemwe might ndthat it was
an ordinary enough hymn, and the same uncertainly would be
involved in thinking about another recitable hymn that Aristotle
composed, this time in elegiacs. The same list in Diogenes also
records a book of elegiac poems (
.y te, no. 146) and quotes

the opening phrase of what sounds like a hymn to Artemis:
O Daughter of a mother with fair children (xe..tx xvou
yx, 0 uyex,, Fr. 672 Rose = 672 IEG). It is hard to say
much about these scraps of recitable hymns because they are
heavily formulaic; without a context to resonate against, the
traditional language conveys little that is particular. The lyric
hymns we have seen, in contrast, seem to devise ways to import
something of their performative contexts into their rhetoric,
either because they were composed for specic festival events
or because they wish to pose as such. One other way to provide
poems with contexts, of course, was for readers to tell stories
about howthey were composed, and the difference this can make
can be illustrated by a paean said to have been composed by
Socrates. Diogenes, allowing that there has been some question
as to its authenticity, quotes the opening line of what he calls
a paean by Socrates (DL 2.42 =[2] IEG): Delian Apollo, rejoice,
and Artemis, glorious children both (L y.t'

Ao..ov e t,,
xe`t

A,xt, e tc x.tv ). A suggestive context for such a
poem can be found in a well-known passage of Platos Phaedo,
in which Socrates says he has been inspired by a dream to
take up composing verse while awaiting execution in prison
(60C61B). Among Socrates poetic activity, Plato mentions
putting Aesops tales into verse and composing a prelude for
Apollo (60D:
vx tve xo` u xo u At o ou .you xe`t x t

xv A.. ,oo ttov xe`t e
..ot xtv ). The poem cited by

Diogenes could have been composed as, or forged to represent,
the Apolline prelude. Its genre cannot be pinned down; Diogenes
calls it a paean, presumably because Apollo and his sister are
featured. The paeans we have seen are in lyric meter for musical
accompaniment, but either because he is in prison or because
his musical education did not advance that far, Socrates does
not avail himself of a lyre but resorts to a recitable meter to
give his song formal integrity. At a festival for Delian Apollo,
however, hexameters on a suitable theme could be set to a lyre
and be performed as a kitharodic prelude or overture to the
gods celebration. As always, occasion can powerfully inuence
meaning, and the context Plato suggests for a text such as
Diogenes cites would make it richly ironic: it is tting that, even
in prison, Socrates should have done his pious best to compose a
paean or proemto greet the god whose festival is underway while
the Phaedo plays out. This is also, of course, the same festival
at the conclusion of which Socrates is scheduled to die (Phaedo
58AC).
32
Such scanty fragments can perhaps best serve to remind us
of the range of forms that hymns could take and yet remain
recognizable as hymns. They also indicate some ways in which
Greek poems were enriched by tying themselves, or by being tied
by readers, to specic and highly charged social events. It seems
that lyric hymns designed to be repeated in the future were more
apt than recitable poetry to import something of their inform-
ing context with them into different performative situations.
Ki nds of Hymn 111
This means that, for readers, Greek lyric genre may be dened as
the frame of expectations that social context ordains, even when
that context is to be found only in the rhetoric of the poem.
With this in mind, we return to Aristotles song, this time in its
rhetorical aspect.
Chapter 7
Ethos
Having examinedthe formal means by whichAristotle positioned
his song as a kind of hymn, I turn to the ethos of the piece. This
aspect of Greek poetry is hard to discern solely through verbal
analysis but would have meant a great deal to ancient audiences.
I have noted that most Greek lyric texts read more like scripts
for giving a speech than records of private meditations. As such,
they project a persona that is ethical in the Greek sense of
the word, that displays the speakers character (thos) and values
by declaring, as Aristotle would put it in his ethical treatises,
what is to be pursued and what avoided in life. As a mode of
rhetorical analysis, ethos considers a song as a performance
of self, a metrically and otherwise stylized version of what a
personof a certainsort might say ina certainsituation. Fromthis
perspective, a strong ethical strand emerges in Aristotles song,
one that blends his hymnic argument with another discursive
mode.
To bring out this dimension of Aristotles text, I will adduce
two earlier lyrics that on rst glance may seem neither much like
it nor like each other, a ve-stanza song by Sappho and a quatrain
113
by ananonymous Atheniansymposiast. Althoughthese songs are
very different from Aristotles on their surface, fundamental to
themall is the way they take up and address a popular and often-
discussed topic: What is the most important thing to pursue in
life? What deserves to be called the nest (kalliston) thing? The
ethical stands of these different kinds of song will be illuminated
less by Aristotles philosophical ethics than by his remarks in the
Rhetoric on how to praise and what to praise. Rhetorical ethos is
expressed most obviously in what speakers choose to praise and
the reasons they give for it, but almost every word of Aristotles
song has ethical force when one appreciates that it is arguing for
a vision of what is kalliston. A lyric by Sappho in the same vein
will bring out most clearly the rules of the game they both were
playing. But rst it will be useful to exemplify the genre through
a short skolion mentioned by Plato.
Ethos in Debate: An Attic Skolion and a
Poem by Sappho
If the opening of Aristotles poem suggests a hymn is under
way, the second line complicates things when he declares Virtue
the nest (kalliston) thing to hunt down in life. Piling up
complimentary epithets is common in hymns, but predicating a
superlative would have been heard by Greek audiences as a moral
proposition that they were invited to accept or dispute. Such a
declaration functioned as an opening gambit, inviting members
of the audience to take the oor in their turn and discoursein
whatever poetic or prose formwas appropriate in that context
upon their view of the best or nest (kalliston, ariston) thing
in life. Speakers can be found striking this note in virtually every
kind of Greek poem, and in prose texts as well, so that we are
not dealing with a literary genre but with a genre of discourse, a
Ethos 115
traditional concatenation of themes, rhetorical forms, and rules
for transforming them that had been developed in different
performative modes to suit different groups andoccasions. It was
a widely adaptable formof civilizedentertainment, of employing
ones leisure time in a noble way as Aristotle would put it.
1
We
nd out from Plato that a favored space for such performances
was the symposium. His Socrates gives a capsule description
of the form in a debate about moral education with a sophist:
I assume that you have heard at symposia men singing this
skolion in which they count off as they sing, Health is best, and
second is being fair, but third, as the poet of the skolion says,
is wealth won without trickery.
2
The reference is to a famous
verse that is quoted by Athenaeus among the Attic skolia, but
Platos contextualization lets us see that it has a slightly more
than playful purpose:
3
\
Yyte tvtv ` v e
,toxov e vc,`t 0vex ,

c ux,ov c` u` ev xe.v yv o0et,
x, txov c` .oux tv e c.,
txe x xe,xov y
\
p ev x` e x v t.v.
Health is best for a mortal man,
and second is to be born fair;
third is to grow wealthy without trickery
and fourth is youthful sport among friends.
The simple song is lighthearted, but strikes an ethical posture
from the end of the rst line: the addition of for a mortal
man indicates the singer will not aim at some impossible
perfection but will content himself with those goods a mortal
can hope to attain (what was called at the time the human
good, to anthrpinon agathon). The choice shows a symposiast
whokeeps a sense of humanlimits without losing anappreciation
for pleasure. Health, as we saw, was praised by Ariphron in
religious terms as most venerable for morals. The skolion
shows that choosing to praise health could have a political
dimension as well: the singer gives rst place to a good that
is in principle available to all, not the desirable but less widely
distributed goods of inborn beauty or wealth. The nal verse
can also be seen as achieving closure by a bit of self-reference,
as in Ariphrons paean. In the company of friends, being young
together (hban) is virtually synonymous with sympotic activity.
(The marriage of Heracles upon his ascent to Olympus to Hb,
youth, youthful vigor, symbolizes not only perpetual youth
but also the enjoyment of perpetual felicity.) The symposiast
thus closes with the implication that another candidate for the
nest thing to do in life is to sing this song among friends.
This wittiness and the mild irony may be why in part the
verse was sometimes ascribed to wise Simonides (651 PMG;
cf. 579 PMG).
Aristotle, we recall, also began by declaring that Areta was
kalliston, and ranked her above ne birth, wealth, and pleasure.
Both verses defend an ethical position. This is the reason that
Socrates adduces the Attic skolioninhis debate witha sophist, and
for this reason too Aristotle quotes it in his Rhetoric as the kind
of thing an orator should say to project his values.
4
The game of
striking an ethical attitude in verse was played in various musical
modes, as in this elegiac couplet from Theognis (255256):
K e..toxov x ctxetxexov . toxov c u
\
yte tvtv
, eye c` x,vxexov, xo u xt
, et, x xu tv
What is nest [kalliston] is what is most just; health is most
to be desired;
but the most pleasing thing is to get hold of what one
loves.
Ethos 117
Difference within the requirements of form is prized in this
competitive game, and Theogniss catch offers us a slightly
different, justice-centered ethics along with an off-beat focus
on ranking superlative adjectives as much as goods. Another
aspect of this verse worth remarking is that the elegiac couplet,
as noted, is legible in the sense that it could be learned right
out of a songbook, and indeed the poetry collection that went
under Theogniss name appears to have been such a collection of
easy sympotic pieces. Neither the lyric skolion nor the songs of
Ariphron and Aristotle are legible in the same way. (In Gorgias,
Plato assumes that his interlocutor knows the skolion from
having heard it: 451E: ot
oet y e, o e xyxo vet.) Thus it was an
elegiac version of the sentiment that was chosen to be inscribed
on Letos shrine in Delos for passersby to read. We know this
fact from Aristotle, who quotes and discusses its wisdom in the
beginning of the Eudemian and the Nicomachean Ethics.
5
One may
say that the kalliston game takes on new life when such verse is
quoted in these treatises, as it does in Platos dialogue, though
its rules are reformulated to suit the new mode of discourse that
came to be called philosophy.
This game, really a pretext for descanting on a moralizing
theme in a suitable register, was already rened in Lesbos around
600 BCE when Sappho gave the following elaborate reply. It
has come down to us on a papyrus with four of its ve stanzas
relatively clear:
6
ot ` v t yv ox,xov ot c` ocv
ot c` v ev e to
`t y ev .etvev
vet x e..toxov,
y c` x yv o
x-
x xt
,exet
eyu c u
e, o uvxov yoet 5
evxt xo ux, e y` e, .u ,ox 0otoe
x e..o e v0, v E. ve xv e
vc,e
xv ev e,toxov
xe.. toto
pe T,ot
ev . otoe
xu c` e tco ou c` t.v xox yv 10
eev
v eo0y, e ..` e e, eyey eu
xev
[ . . . ]oev
[2 lines illegible]
. .] v uv Avexxo, te ov vet 15
o ou ] e,o toe,
x e x po..o tev
,exv x p ee
xe e,ue . e,ov t
cyv ,oo
y` x` e A ucv e
,exe xe v o
\
.otot
oco evxe. 20
Some say an array of cavalry, others foot soldiers,
and others say ships are, in this wide world,
the nest [kalliston] sight to behold. But I say it is
whoever one loves.
It is easy to make this clear to all: 5
for that one who surpassed
all other mortals in beauty Helen abandoned
her most excellent husband
and sailed off to Troy,
without a thought for child 10
or her own parents. But she was swept
away [sc. by passion]
[2 lines illegible]
And when I think of Anactoria,
no longer with us here, 16
Ethos 119
I would rather look on her lovely step
and the bright glance of her face
than Lydian chariots
and soldiers in panoply. 20
Sappho opens with a priamel naming three forms of military
splendor as candidates for the nest sight to be found on earth.
She then, like Aristotle, caps the tricolon with a strong shift
that presents us with the spectacle of Anactoria dancing. (So
I think we should understand bama, footstep, in v. 17.) Like
Aristotle too, Sappho goes on to argue her case frommythology.
7
Perhaps most strikingly, both poems turn at the end from myth
to mortal. If Aristotle shifted from muthos to logos in moving
from heroic gures to a friend, Sappho makes an equally broad
jump in turning from war to love. Her poem is a version
elaborate, rened, and with an ethos that is open to oriental
luxury of the kalliston game played by the skolion. Indeed,
the conclusion of Sapphos argument, though worked out with
exquisite control of tone and pace, is not dissimilar from the
Theognidean entry in the game: the most desirable thing is to
get what one loves (256).
Putting Aristotles song against these formally disparate texts
shows that all exhibit the same deep structure, so to speak.
In the background is a grammar that organizes the range of
conceivable ethical postures and provides the elements by which
these can be transposed and presented intelligibly. The form
is nally determined by the kind of speech that is appropriate
to the occasion and the attitudes and abilities of the group.
8
Whether one recited a familiar elegiac sentiment conned from a
songbook, or tossed off a four-line skolion fromthe good old days
in Attica, or exhibited real virtuosity by singing a full ve-stanza
song in Lesbian dialect, one put familiar themes in play within
recognizable transformations, so too, inevitably, whenone raised
ones voice in Doric hymnic style to performAristotles song. It is
noteworthy that this deep structure can emerge in quite concrete
details of the text: these can range fromthe conjunctive for as a
signal that the argument is beginning (Ariphron v. 3, Sappho v. 6,
Aristotle v. 17) to the trick of closing a priamel with a categorical
twist and the predilection for ending the song on a self-reective
note.
In Sappho, however, we are pulled back from general mor-
alizing by Anactoria, which Sapphos scholars have conjectured
must be the name of a beloved former member of her group.
She is the Hermias of her poem: as a proper name, Anactoria
would seem to focus this brilliant exercise in traditional form
on a specic person in a specic place and time. And yet
Anactoria is all but unknown to us outside of Sappho; the
few brief ancient references to her may well be based on
later attempts to understand this passage. Indeed, Page raised
the possibility that the name is a shell and that Anactoria
is a pseudonym for one of Sapphos pupils whose real name
was Anagora.
9
We cannot tell. In favor of Pages suggestion,
however, we may observe that pseudonymity was intrinsic to
later instantiations of Sapphic love poetry, such as the odes
Catullus addressed to Lesbia, the woman from Lesbos. Within
Sapphos text, the function of Anactoria may be regarded as
purely rhetorical, what Elroy Bundy in his classic analysis of
praise poetry termed a concrete name cap to close a priamel.
10
In a similar way, more recent perspectives from structural
anthropology would identify Anactoria as a name for some
archetypal role that many Lesbian girls, whatever their actual
names, took on: she is the girl who has graduated from the
choral group and gone away (e.g., found a good marriage). This
happens to every debutante in every season, and there is no
possibility or need to look for a specic Anactoria outside
the text.
Ethos 121
If there was an Anactoria, even under another name, we
would say that from our perspective she has moved further
thanHermias along a path that leads fromhistorical particularity
toward a kind of literary katasterismos, becoming a purely textual
effect. The comparatively greater obscurity that a historical
Anactoria has suffered, however, is not simply due to her
belonging to an earlier, less well-documented age than Hermias,
for she is already moving in this direction within Sapphos song:
her epiphany in the poem, which is to say the moment when
the image of her dancing is superposed on the visions previously
evoked, arrives with the qualifying phrase, who is not present
(v. 16). For all her elusiveness and playfulness, Sappho does not
pretend her words can conjure the loved girl to the occasion of
the song. Much as with Hermias, absence is not denied.
Ethos in Protreptic: Aristotles Hymn to
Hermias, vv. 18
Considered in this light, the opening words of the song for
Hermias take on extra weight: e ,x e o. uo0 is not only
a hymnic invocation but also Aristotles answer to the question
of what is most choiceworthy in life (kalliston, v. 2). The answer
is not virtue per se, but the kind of excellence that can only
be achieved through toil (0o). Aristotle thereby commits
himself to a well-established set of ethical beliefs. That aret was
worth pursuing despite the toil it entailed is taught in Hesiods
Works and Days: with personication verging on allegory, the
archaic poet declares that Baseness or Misery (Kakia) is easy to
attain, for it dwells nearby and the road thereto is smooth; but
the gods have put sweat before Virtue, who dwells at the end of
a long, steep road that is painful going at rst, though wonderful
upon arrival (287292). This passage was much reprised in Greek
discussions of excellence, including a paraphrase in lyric form by
Simonides: there is a story that Virtue dwells amid rocks difcult
to reach, he sang in the sixth century, and is only beheld by
those whose hearts have felt the sting of sweat (579 PMG).
To choose to praise Virtue of much toil, then, was hardly
breaking new ground; but a song in her honor may have been
a novelty: it seems that no one had previously addressed a hymn
to Virtue personied, and one may suppose this was part of
Aristotles inspiration.
11
If this were not distinctive enough, he
has animated conventional hymnic topics with the discourse of
popular ethical debate. The double game continues as the poet
pursues two discursive agendas at once. In this light, let us go
through the beginning of the song again and observe its shifting
ethical stances.
The end of the rst verse inserts a qualication for Aristotles
assertion by specifying that it is valid for the race of mortals
(y vt p,ox t: the stance is almost too insistently pious, for
the mention of p t in v. 2 is nearly pleonastic). The dative can
be construed with e ,x e o. uo0 in two ways: if we take it
with e ,x e, virtue is the best thing we mortals can strive for,
though if we were gods we might have different priorities; taken
with o. uo0, the expression implies that virtue is absolutely
desirable, though it is our sad lot as mortals that it only comes
with much toil. The key word o. uo0 thus wavers in tone,
depending on whether the speaker adopts the perspective of a
mortal or of a god who might pity mortals.
12
The metaphor in v. 2, 0 y,ee, quarry, is conventional,
almost banal, but suggests something about the target audience:
this is hunting not as a necessity to get the basics of life one
can eat acorns for that but as sport or avocation, a civilizing
art that, notionally at any rate, tames land for agriculture and
prepares young men for leadership and war. An audience so
addressed is very much like the audiences wooed by sophists
Ethos 123
and other teachers of advanced education in the classical age.
They developed a rich prose literature of protreptic in order to
convert young men at the turning point of adulthood to the path
of Excellence as they dened it. In such discourses, the hunting
metaphor appealed to young menwho had the freedomto choose
their occupations and who, we may add, had the talent, training,
and leisure to enjoy songs that deliberated in stylized formabout
the nest object in life.
13
The social signicance of Aristotles
metaphor can be seen in a prose essay written not very much
earlier, Xenophons On Training Hounds (Kunegetika). This
manual on hunting commends the pursuit to noble youths as
a preparation for leadership in civic life, and it concludes with an
impassioned attack on the amorality and uselessness of sophistic
education of the day. In the fth century, the sophist Prodicus
had created a prose discourse in praise of aret, which Xenophon
esteemed enough to copy into his Memorabilia (2.1.2134 =
Prodicus 84 B2 DK). Prodicus anticipated Aristotle by developing
the personication of Virtue in Hesiod and wise Simonides into
a prose drama: he pairs Aret with Kakia (Baseness), as Hesiod
had, but represents them as two female gures accosting young
Heracles at the crossroads of his life. The personied gures,
each arrayed as one might expect, speak in turns for and against
the worth of a life of virtuous toil. These protreptic passages
suggest that Aristotles discourse could have had a similar role if
it were reprised in sympotic contexts. For in the course of being
young together over wine and song, as the anonymous skoliast
put it, old men were wont to give advice about life to the young.
The similarity in ethos between Aristotles song to Areta and
Prodicuss Heracles at the Crossroads does not mean that we
should reduce the lyric to a recruiting anthem for the Lyceum.
But in considering its meanings, it is well to bear in mind
that one of the most active and culturally signicant branches
of fourth-century literature was protreptic (in which I would
include swathes from early Plato).
14
Protreptic discourses are
ethical in a double way, not only in projecting the persona of
a trustworthy and well-intentioned speaker but also in aiming
to transform the character of the addressee. In adopting the
protreptic mode, a poet aspired to induce a particular pathos in
his audience; a protreptic poem offers the possibility of shaping
its audience on repeated hearings, forming in them a habit of
taking pleasure in the proper praise of excellent actions.
15
This
was, in Aristotles view, the noblest function of poetry, including
song, in the education of the young.
16
Such a context gives more force to the toil in Aristotles
keynote epithet. In sense, 0o is not far fromvou, which
occurs just below at 5. Both words are formulaic in the epinician
vocabulary that praises an athlete for expending effort (and
money) to demonstrate aret in the games and win glory for
his city.
17
They name the pain in the American protreptic
adage, no pain, no gain. The underlying ethic is articulated in
Aristotles sectiononpraise inthe Rhetoric (1.9, 1366a2368a37)
in which he says that virtue is particularly praiseworthy when it
comes not from birth or chance but as the result of effort. In an
epinician perspective, toil makes virtue more admirable; what
epinician praises, protreptic commends as the means to the goal.
Some readers would take a further step and adduce Aristotles
philosophic books to argue that his poemexpresses his particular
conception of the good. Jaeger, for example, took Aristotles
Areta with her fair shape (morph, v. 3) as a correction
of Platos metaphysical conception of the form (eidos) of
excellence.
18
But this seems an approach that is easily overdone.
The discourse genre in which Aristotle participates encourages
the expression of attitudes that are intelligible and acceptable
to the group. Convivial song is not a medium for debating
metaphysical systems, and Aristotles song was bound to have
a short afterlife if it rattled sensibilities or overturned ideas that
Ethos 125
had been expressed on the subject for centuries. The after-dinner
performer may be a philosopher, but one does not behave only
as a philosopher.
With his ethical proposition fully if somewhat enigmatically
statedinthe rst twoverses, Aristotle reasserts hymnic style with
a series of second-person pronominal forms in v. 3 and following,
initiating what Nordenidentied as the you-style characteristic
of a great many ancient hymns.
19
With these apostrophes Areta
undergoes, guratively, a reverse metamorphosis: the quarry
turns froma wild animal to an addressable maiden. The grammar
becomes a little tense when the physical form of Virtue is
broached: there is anastrophe of the preposition and hyperbaton
of the vocative in v. 3, o e ,t, e,0 v, o, e. Language
becomes a bit jumpy when Virtue makes her epiphany.
20
As a maiden, Areta is both beautiful and elusive all men
desire her but she is not easily possessed; this Virtue is more
erotized than the prudent, temperate counselor of Prodicus.
When in vv. 67 she casts a fruit before the minds eye, the
imagery may suggest a scenario like the Atalanta story, in which
the speedy maid was outraced by a cunning suitor who distracted
her by throwing apples of Aphrodite in her way.
21
But the epithet
that follows immediately changes the picture: calling the fruit of
Areta immortal-like (xe,v t oe0 evexov) moves us away from
Atalanta and toward the apples of immortality that grew in the
garden of the Hesperides at the edge of the world: seizing these
apples was among the rst of Heracles labors and a prototype
of any heros quest for immortal glory. The compound epithet
that I translate as immortal-like (t oe0 evexov) seems to be
Aristotles deliberately awkward invention, and such compounds
are among the stylistic features that make critics call this
poem dithyrambic. Wilamowitz divined that such a word must
have underlain the banal epithet immortal transmitted in the
sources knownat the time (Athenaeus has x e 0 evexov, Diogenes
t e 0 evexov), and his suggestion was conrmed when the
Didymus papyrus appeared. The key semantic difference is that,
while t oe0 evexov may evoke the idea of literal immortality (as
does e 0 evexov), such equal- compounds imply that Virtues
reward is not actually deathless but only equal to immortality
inits extraordinary worth.
22
Hence it seems that Aristotle coined
this new word precisely not to destroy the sober ethos he was
projecting. The thought-arresting combination immortal-like
fruit creates an oxymoronic conjunction of permanence and
seasonal change that suits the dominant theme of the poem, and
the novel epithet shows that Aristotle was willing to strain poetic
diction so as not to trespass on the mortal/immortal divide that
kept hymns and encomia distinct.
Even so, the Prince of Critics got in trouble because of the
way he phrased his theme. The ashpoints can be identied by
comparing one nal poetic protreptic with a quite extraordinary
ethos. When a deied Heracles descends from Olympus near
the end of Sophocles Philoctetes, he delivers a speech on aret
that is, upon analysis, virtually a synonym of Aristotles song.
Although many texts in the tragic corpus treat the themes
of excellence, its price, and reward, none so closely resembles
Aristotles song as these spoken verses. What makes Sophocles
text particularly revealing to set beside Aristotles is that the
tragic poet, in striving to formulate the promise of posthumous
glory in a vivid new way, exhibits two moments of verbal excess
(see underlined notes on p. 127 last line and p. 128 2nd from
last line) that are similar to the controversial spots in Aristotles
song: Sophocles uses the word athanaton in a disconcerting
way, and he ends with an apothegm that, on its face, sounds
impious. More generally, Sophocles gambit of putting this
traditional theme in the mouth of an Olympian highlights the
strong connection between the message and the ethos of the
speaker.
Ethos 127
Ethos in Epiphany: Immortal Virtue in
Sophocles Philoctetes
Produced in 409, Philoctetes is centrally concerned with heroic
suffering and its reward.
23
The suffering is Philoctetes, cast
ashore by the Greek army on its way to Troy because of a noisome
and excruciating wound; the reward is the prospect of glory to
be won if he rejoins the expedition and takes his destined part in
the fall of Troy. Our passage comes from the plays conclusion,
which suddenly and radically reverses the plot: up to v. 1409,
Philoctetes appears determined to nurse his bitterness against
the Greek army and go back home to his father. As he is about
to depart, abandoning any prospect of heroism for himself and
for his newly won ally, Achilles noble young son Neoptolemos,
a godlike gure, appears over the stage and in a coup de thtre
changes Philoctetes mind. Heracles, nowmade an Olympian and
privy to Zeuss great designs, persuades his protg to go to Troy,
where he will nd a cure for his ills and play a glorious part in
the fall of the city. Omitting a brief aside to Neoptolemos, the
climactic speech (14091471) runs as follows:
24
M y y, ,`tv e
v x v y
\
x ,v
e t
y u0v, e t Ho tevxo 1410

eoxtv c eu c` yv x` yv H,ex. ou
e xo y x x. utv . uootv x o
ytv.
t ` yv o` yv c y
\
x e,tv ou ,ev te
\
c,e ,o.t v, 1415
x` e Lt x , eov pou. uex e oot,
xex,yx uov 0 o
\
cv y
\
`v ox ..y
o` u c
v u0v
exouoov.
ke`t , xe v oot x` e
` e . x ue,
o
\
oou ov yoe xe`t ct.0 ` v vou
e 0 evexov e ,x` yv
oov,
\
e,o0 o
\
, ev. 1420
xe`t oo t, o e t
o0t, xo ux o t.xet e0 tv,
x x v vv x vc u x. e 0 o0et p tov.
.0 ` v c` o` uv x c e vc,`t , x T,txv
.toe, , xov ` v voou e uoy.uy, e,
e ,x yx , xo
xx,t0`t ox,ex uexo, 1425

H e,tv v, o
\
` x vc et
xto xex v
u,
xotot xo t
o tot voot t p tou,

,ot x T,o tev, ox u. e x
.e0,e o` e
yt, e ,tox t
x.ep ` v ox,ex uexo,

Ho tevxt ex,`t , ex,e Ot
xy . exe. 1430
a
\
` c e` v . epy o` u ox u.e xo uc xo u ox,exo u
xv
v vy te , u,` ev
` yv
xt.
. . . . .
Ey ` c Aox.ytv
euox y,e y o y voou ,
I.tov.
x c ux,ov y` e, xo t
o t eu x` yv , ` v
xot e
\
. vet. To uxo c
vvo t0, o
\
xev 1440
o,0 yx ye tev, u op tv x` e , 0o u
\
xe
..e evxe c ux, y

\
y txet ex` y,
Z u y
\
y` e, u o pte ouv0v yoxt p,oxo t, 1443
xe` v ot xe` v 0 evotv, ou x e ..uxet.
Not yet, not before you hear our
speech, son of Poias;
conceive that this is the voice of Heracles
that you hear, and the one you see here is he.
For your sake I have come down from my heavenly seat
in order to reveal Zeus plans to you 1415
and to check you from your intended course.
Hear then what I say:
First I shall tell you of my own fortunes,
How after toiling through many labors
Ethos 129
I won immortal aret, as is plain for you to see. 1420
For you too, be assured, the same is in store,
to make your life right glorious (eu-klea) after these travails.
When you go with this man here to the city
of the Trojans, you will rst off nd relief from this painful
disease,
and then be judged foremost of all the army in aret; 1425
Paris, the cause of all these ills,
with my bow you will absent from life,
25
and you will sack Troy, and booty back to your halls
you will send, taking the top prizes for military valor,
back to Poias your father by Mt Oetas plateau. 1430
And whatever booty you get from these troops,
carry off to my pyre to make a memorial offering
for my bow.
. . . . .
For my part, I will send Asclepius 1437
to Troy to be the healer of your sickness.
For once again by my bow she must be
captured. And know this, when 1440
you come to sack the land, keep a pious respect for the gods
things;
for father Zeus considers everything else of less import.
For piety dies along with mortals 1443
and whether they live or die is not destroyed.
Opening anapests (14091417) inform the audience that the
gure appearing is none other than Heracles, whom Philoctetes
had last seen when agonies drove him to immolate himself on
Mount Oeta. For his tendance on that occasion, Philoctetes has
inherited the bowthat Heracles used in an earlier capture of Troy
and that is destined to be present when she is conquered again.
Sophocles exploits these and other mythological connections
that Heracles has with Philoctetes to compose a unique kind
of exhortation. If Heracles was usually, as in Prodicus, on the
receiving side of advice, Sophocles allows him to play the role
of the one who commends aret to a young comrade. It is
also ironic that this Heracles, like Aristotle, can nd no better
example of the rewards of toilsome virtue than the myth of
Heracles.
The divine Heracles, like the singer of Aristotles hymn, speaks
of aret, its price, and its immortal reward, and ends with a
tribute to piety (eusebeia 1441, cf. sebas in Aristotle v. 20).
But at two points underlined above Sophocles language
is so unusual that scholars have been unsure of what he is
saying. The rst comes early on when Heracles says that as a
reward for his labors he has won immortal excellence, as you
can plainly see (1420). Immortal (athanaton) is an unusual
epithet for aret, perhaps because aret is so closely associated
with mortal excellence; but the phrase is easy enough to make
sense of if we interpret aret as a cause-for-effect metonymy
meaning glory.
26
Heracles phrase thus condenses the common
topos that aret leads to renown and the greatest aret to glory
that never dies.
27
Understanding the phrase as immortal glory
makes the reward Heracles won not much different fromwhat he
promises Philoctetes just below, that the latters toils will make
his life glorious (eu-klea).
28
The problemwiththis construal of athanatonaretn is that it is
not clear howHeracles immortal glory is, as the rest of the line
indicates, something visible. Accordingly, Richard Jebb extended
the phrase still further, taking it as the glory of [Heracles]
becoming immortal, the unique reward he won for his uniquely
toilsome labors. Heracles immortalizationcould easily have been
made visible, either by costume, as Jebb suggests, or from the
fact of his appearing in the way gods usually did, from an
elevated platform (theologeion) or a cranelike suspension device
Ethos 131
over the stage. (It was Aristotles e yev y inPoetics 1454b2
that gave such forced plot-resolutions as that in Philoctetes the
name deus ex machina.) On this view, athanaton aretn is a
riddle; the expression, like the heros fate, is unprecedented.
29
It follows that we must contrast this form of immortalization
with that promised Philoctetes, survival after death through
commemorative speech (u x. e) being a pale reection of the
apotheosis that Heracles obtained.
Both immortal glory and the glory of immortalization
are tenable construals of athanaton aretn, and it may be that
Sophocles has made bothavailable because immortal excellence
means different things to Heracles and to his human auditors:
mortals undertake extraordinary efforts in hopes of a glory
that does not die, but Heracles has found out that Herculean
efforts lead to an even greater reward. Pointing to a gap
between Heracles perspective from Olympus and what mortals
are permitted to hope for is a subtle but unambiguous signal in
his promise to send Asclepius to cure Philoctetes (1437). This
contrasts sharply, though for no immediately apparent reason,
with a promise made by Neoptolemos not much earlier in the
play, that if Philoctetes came to Troy he would be cured by
Machaon, the son of Asclepius (1333). Relying on the legend that
Asclepius was brought back from death and made an Olympian
at the urging of Apollo, Sophocles has Heracles up the ante
on Neoptolemos: whereas the best hope of remedy a merely
heroic person can promise is a descendent of Asclepius, the
immortalized hero enjoys access to the healing god-man himself.
The gap between the epigone and the immortal healer, like that
between the two meanings of athanaton aretn, is that between
human and divine. It is surely relevant that the main reason that
Zeus rst struck down Asclepius was that he had in one way
or another attempted to close this gap by healing death.
30
As
in Aristotles poem, Sophocles wants a strong new statement of
excellences survival, and like himhas used the epithet athanaton
in a way that raises difculty.
Allowing for a special double ethos in this speech may
help with another puzzling expression near its end Heracles
injunction to the young warriors not to lose a pious respect for
the divine in victory. This warning evokes the story told in epic
that after the fall of Troy, Neoptolemos impiously killed old king
Priam as he was huddled at Apollos altar in supplication. Within
the dramatic frame, piety (sebas) and its link to Zeus provide
Heracles with closure on a high note, as they do for Aristotle. The
hero expands on this, however, with a puzzling nal thought:
Zeus regards all other matters as secondary; for the sense of
reverence [eusebeia] dies along with mortals and whether they
live or die is not destroyed (14431444).
The adage seems scandalous on its face, and it has needed a
change only of two or three letters to convert the manuscripts
for . . . surely piety dies into its opposite: for piety does not
die together with mortals, but whether they live or die it is not
destroyed.
31
On this reading, printed in most modern editions,
piety must be taken as the reputation for doing pious acts (a
metonymy analogous to the use of aret at 1420). Heracles thus
adds to the prospect of Philoctetes glorious life (1422) the
additional incentive of being celebrated after death for piety.
32
A problem with this view, which its proponents admit, is that
the emended text is a banality. Nor is it clear why, apart from
the reference to an extra-dramatic event, Heracles should stress
piety among the virtues, rather than courage, or indeed steadfast
friendship. The alternative is to take the manuscript reading as
another strong guration for living on after death, along lines
proposed by Wilamowitz: piety dies with men not in the sense
of perishing away but of accompanying them in the afterlife.
33
Though there is much that must be given up when the body dies,
Ethos 133
piety is the one excellence that, even when we are dead, does not
pass from us.
34
On the manuscript reading, Heracles would be giving a new
perspective on the usual sentiment one hears, for example, in
Euripides that aret, even if one dies, is not destroyed, but
lives when the body is no more (e ,x` y c` xe` v 0 evy xt ou x
e ..uxet, / y c ou x x o
vxo o exo, Fr. 734.12 TrGF).
The mortal Olympian initiates us into the mystery that, among
the virtues, there is something unique about piety and the way
it survives death. (Its uniqueness is reinforced by the almost
outrageous apothegm.) In support of his gnome, Heraceles can
cite no less an authority than Zeus, who cherishes piety above all
other forms of excellence.
35
As an Olympian, Heracles can tell us
howthe greatest god answers the What is best? question. From
the other side, aret and piety are said to bring greater rewards
than we can imagine.
Quite unlike Aristotle, Sophocles enjoyed a reputation in his
time as especially pious, and so in refurbishing the ancient
exhortation to virtue he seems to have been free to utter
an almost impious paradox with impunity.
36
His language in
this speech also differs from Aristotles in being striking at
points but without resorting to the attention-getting epithets
of dithyramb, for Heracles is a solemn speechmaker, not a
singer. The dramatists primary creative move in this speech is
to invigorate the protreptic to virtue by composing one that a
man-god might give, and so the fact that the speech is spoken
in epiphany turns out to be crucial. At the same time, this is an
eminently performable stretch of iambics on a standard theme,
and its general conventionality makes it suitable to be recited as
a stirring discourse on its own; we knowthat in Aristotles era an
ability to recite speeches (rhseis) from tragedy was considered a
social accomplishment.
37
In its context as part of Philoctetes,
however, the unique ethos of the speech is more apparent.
Taking advantage of his medium, of the power of spectacle over
narrative, Sophocles can bring Heracles before our eyes, so that
we see him and not just hear him speak or hear about him, as
we might from an epic rhapsode or mythologist. We register
the strong language that crops up in the speech in light of its
unique speaker, whose status is reinforced, as you can see, by
the dramatists resources.
The interdependence of ethos, genre, and context to be seen
in this speech suggests a nal general formulation for the subtle
ways in which, when we read a poem removed from its original
setting, the context from which its words rst emerged is a
hovering presence behind the words themselves. Epiphanies in
Greek poetry often allow for an ineliminable ambiguity about
what actually occurs: songs bidding a god to come usually request
a literal apparition, but oftensuggest that the arrival of the divine
will be noticeable only as a general enhancement of the scene,
a scene that may remain unchanged in other respects.
38
The
ambiguities of presence and absence in Greek representations
of epiphany seem to me to provide a good way for thinking
about how contexts may appear in texts. In theory and often
pragmatically, words in a text refer only back and forth to each
other, but they can in the aggregate evoke an outside world,
an external presence to which they point and where they will be
received. In this outside where the text is bodily performed,
words nd their things and ethical exhortation nds its end in
the soul of a hearer. But as with a divine epiphany, the presence
of this outside world cannot be pressed too hard. Like a prayer, a
poetic text calls forth an external presence, even if only through
a single proper noun or demonstrative adjective; but the power
that is invited to appear will always remain somewhat less solid
than the audible or visible poem itself. All that is needed for
a successful poetic epiphany is that the scene of its utterance
Ethos 135
be enhanced, even if only by the addition of the hymn itself.
Conceived in magical terms, a prayer aims to constrain a god to
appear on a scene; conceived as song, an effective hymn changes
the scene rst of all by entering it, and then by bringing some
other reality or context, however glancingly, into view.
Chapter 8
Reading
My nal approach to Aristotles text examines its metaphorical
and gurative language in light of similar turns of phrase to be
found in the Greek poetic tradition. I consider this to read the
poetry in the poet, for I hold with Aristotle that the poets native
gifts are to be sought on the level of language. By gurative
language I mean the kinds of things that the Poetics says make
language distinctive: archaic words (glssai), metaphors, unusual
forms, and all deviations from ordinary usage; in particular, an
ability to manage metaphors well is a sign of the born poet, for
the crucial skill of divining resemblances is one that cannot be
taught.
1
I would blend Aristotles essentially rhetorical approach
to poetic language with the practice of Harold Bloom, who shows
that these devices must be considered not inertly, as merely
ornamental variations, but as purposeful deviations, swerves
as Bloom would say, from earlier poetic expressions; moreover,
these earlier texts were themselves swerves, so that the tradition
passes down not only memorable turns of phrase but patterns
of evasion. Bloom teaches us to analyze gurative language not
simply as the lexical difference between two statements but as
137
an active interaction between two energetic systems. It allows
us to read not only Aristotle the poet but Aristotle the reader
of poetry.
An attempt to read Greek lyric in this light should begin
by acknowledging that we have lost so many poetic textsto
say nothing of the innumerable songs that went unrecorded
that we cannot be sure we possess what Bloom would call the
most important precursor texts for Aristotle. Nevertheless,
looking at Aristotles tropes in this dynamic way, as far as the
evidence allows, is the best way to bring the song into focus as a
particular poetic utterance, and to bring out the unexpected and
illuminating connections it has with earlier texts that appear to
be concerned with other matters.
Troping: o oo in Euripides and
Bacchylides
In poetry at once traditional and competitive, engaged with a
range of generic expectations and aiming to make a unique
statement on a common theme, the identity of the poem as
poem lies in its tropes, in how convention is turned on this
occasion from the literal to the gurative. Aristotle shows a
distinctive use of language almost immediately, in the epithet
o. uo0o. We have examined the word as a cue to the songs
ethos, but it also bears notice as an element of logos, of Aristotles
diction in response to the poetic tradition. The epithet is not an
odd word in itself, and indeed compound epithets with much-
as the rst element are very common in hymnic proems. But
Aristotle seems to indulge in a slight catachresis, since the usual
meaning of the word, suffering many toils, does not make sense
applied to Virtue. Wilamowitz characterized Aristotles epithet
as venturesome (khn), the rst of several expressions in the
Readi ng 139
poem he called dithyrambic. In this he relied on Aristotles
own insight that dithyrambic poetry had a special liking for long,
compound epithets, even to the point of courting obscurity.
2
Using suchanepithet as a termof praise ina hymnic address thus
challenged listeners, especially since, as far as we can tell, no one
had ever applied that word to Virtue. Hence Aristotles opening
phrase sounded less like a hymnic formula ritually intoned
than like a kenning, propounding a new concept in language
demanding to be claried. A closer look at the words semantics
brings out some of the puzzles it posed. The adjectives second
element implies a verb, to toil (o0 ), andverbal elements in
compounds can have an active or passive sense. The Euripidean
scholar Donald Mastronarde has noted that o. uo0o most
often bears what he calls a quasi-passive sense, implicated in or
experiencing many toils. Muchlabouring is the rst denition
given in the dictionary (LSJ s.v.). In e ,x e o. uo0 y vt
p,ox t, this meaning wouldonly be appropriate as a transferred
epithet belonging to the following humankind; saying that
humanity is immersed in toil would be an unexceptionable
instance of the usual view Greek poetry takes on life, for no
one is without toil (e
o0o y` e, ou c t, Soph. Fr. 410.1 Radt).

But there is no removing o. uo0, the vocative, frome ,x e:
an invoked deity requires a title, and generation has its own
epithet anyway. Mastronarde thus attributes to the word here a
quasi-active sense, occasioning many toils; for this active sense,
two precedents can be cited from Euripides, a lyricist whom
Aristotle admired.
3
The rst comes at the beginning of a choral ode that
has much in common with Aristotles opening: O Ares [who
provides] much toil, why do you deal with blood and death,
out of tune with the feasts of Dionysus? (Phoenician Women
7845:
o. uo0o

A,y, x t o0 et
\
ext / xe`t 0ev ex
xex y B,o tou e, eouoo
\
o,xe t). This invocation to
an abstraction, essentially war personied, resembles Aristotles
opening in rhetorical posture, in rhythm(Euripides dactyls have
the same time signature as Aristotles anapestic opening), and
in giving an active meaning to o. uo0o. The fact that
Euripides epithet is decidedly negative shows what a swerve was
needed to apply o. uo0o to Areta: even turning the voice
of the word into the quasi-active does not convert it into a term
of praise.
A second Euripidean use of active o. uo0o as causing
toil is closer in tone to Aristotle. A speech lamenting the
unpredictability of human life begins (645a TrGF = Fr. 916
Nauck):
o. uo0o ptox` y 0vyxo t,
`t evx`t oe.,` e x toet,

xe`t x` e ` v eu
t, x` e c e o0tv u0t
xou x
oxtv o
\
,o x tvo ou c`t
t o
\
vxtve ,` y x .oet 0vyxo t, 5
.` yv o
\
xev
.0y x,u,` e Lt0v
0ev exou 0 toe x.ux y.
O life of much toil for mortals
how you are unstable at every point,
and some things you make grow, others you make wither;
and there is no clear boundary stone set up
to which mortals know they must steer, 5
except for the miserable, Zeus-sent
culmination that is death.
As in the address to Ares, the vocative epithet has active voice,
but o. uo0o here differs in having adversative force, not
simply invoking the god who causes toil but setting up what
follows: life engenders many troubles for mortals, but especially
Readi ng 141
in being unreliable. Bundy would call this use of o.o0o a
summary foil to highlight the truly dreadful thing about life,
its insecurity. So in Aristotle, Virtue is associated with toil, but
immediately declared the nest things for mortals to seek. This
second Euripidean use of the word, then, is closer to Aristotles
than the prayer to Ares in the way it puts weight on the epithet
and exploits one of its possible senses as a springboard for what
follows.
But there is one aspect of Aristotles o. uo0o that neither
precedent captures, that is, its surprise, its force as a rhetorical
provocation. Our lexicographical analysis can go further if we
consider not just the shades of meaning this word can be
shown to bear but also its roles in discourse genres. From
this perspective, Aristotles use of o. uo0o is most closely
paralleled in the literature of protreptic where, as noted, words
like 0o named the painful price one pays for true excellence.
With this admonitory avor, Aristotles o. uo0o easily
becomes only to be acquired throughmany toils, a conventional
laudation in line with conventional notions of aret, as voiced by
Euripides: who is glorious without toil? (Fr. 240.2 Nauck: x t c
e
o0o u x. y). The closest precedent I nd for this nuance

in the epithet is in an author Aristotle may not have read, and in
a slightly different toil compound,
to0o. But unlike the

other parallels, this text applies
to0o to e ,x e, and this is

but one of a number of points in which it resembles Aristotles
song. Here is a little perorationonexcellence fromthe conclusion
to Bacchylides rst epinician Ode (1.17884 Maehler):
o
\
vxtve xouxexet
0uv cov ouot ,tvet,
o
\
ooov e` v yt ,vov, xvc
.eev xt- 180
ev. e ,x` e c
to0o
v, x].uxe0 toe c o ,0
e vc,`t x]e`t u
x 0 evyt . t-
[t o.u] y.xov u x. te e
[ye.]e.
Whoevers heart is stirred
by ighty ambitions,
may yet, for as long as he lives, 180
win honor [tima]. But excellence [areta] is attended by toil,
yet when brought to its perfect end
provides for a man even after death
the much-envied monument of glorifying speech.
The key point of contact between Bacchylides and Aristotle is
areta (vv. 181/1), around which congregate positively construed
toil (mokhthos, 181/1), the choiceworthy (with poluzlton v.
184, cf. zalotos in 4), fame (vv. 184/17), and some kind of
survival after death (vv. 183/16). Like Aristotles of much toil,
Bacchylides attended by toil (
to0o) is adversative: tima,

respect or honor, canbe acquired by people of small ambition,
but areta inescapably involves effort. Beyond this dictional
afnity, both songs move toward a similar end, concluding with
a complex conceit describing posthumous glory as the ultimate
reward of areta. In Bacchylides, areta is said to give rise to
celebratory speech (u -x. te in v. 184), as in Heracles promise
of a glorious life to Neoptolemos in Philoctetes (u x. e p tov, Phil.
1422). Bacchylides adds the promise that this fame will last after
death in the same way as an e
ye.e, a monument or temple

dedication.
4
Something of the same promise, as we will see, is
implicit in Aristotles assertion that Hermias will be celebrated
in song for his steadfast friendship (p peto, v. 21), an apt
adjective to describe a statue or a monument. On the whole,
it is as if Aristotle has taken the e ,x` e
to0o with which

Bacchylides winds down his epinician and modied it slightly to
Readi ng 143
make an opening for his own song, creating a space he lls in
with Hermias before coming to the same Bacchylidean close.
Now the name Bacchylides is never found in Aristotles
writings, and it would be rash to assume that he is alluding to
or even reacting to this specic text. It is possible that another
poet, Aristotles beloved Euripides, for example, mediated such
language to him. For these texts drawona traditionof moralizing
lyric that offered composers a set of recognizable rhetorical
postures along with a traditional set of themes and lexical
items in which to execute them. It is because of this integrated
tradition that Aristotle, resembling a poet he may not have
directly known, opens his speech with an emphatic toil and
closes it by guratively describing posthumous fame. This same
traditional nexus, I suggest, can be dimly discerned in the
Euripidean discourse above on the many toils in life (Fr.
916). Though its tenor is blame instead of praise, the theme
of immortal glory is quietly present: when Euripides speaker
says that Life makes everything wax (eu
t) and wither
away (e o-0tv ut, v. 3), he means to stress the breadth of
her power, polar expressions being traditional in this mode of
speech. (So a good king makes the crooked straight and the
rough places plain.) But the pair of vegetal metaphors also has
connections with the very ancient poetic theme of immortal
glory, or to use the Homeric trope, of glory that does not
wither (x. o e
-0txov). In portraying Life as controlling what

waxes and withers, Euripides invests her with power over his
own aspirations to poetic fame. We shall see below that Aristotle
drew from this same deep reservoir of imagery when he selected
the other word from that verbal pair to say the Muses will make
Hermias wax or grow (eu evtv) into immortality. Before we
come to this, the concluding guration of Aristotles poem, let us
take a look at the myths in between.
Mythologizing: Hymn to Hermias, vv. 916
Forms of the second-person pronoun acoustically mark off
Aristotles mythic section proving the excellence of Areta. The
two pairs of heroic exempla followed by Hermias are articulated
by forms of you (9, 12, 13), and the entire excursus is
enfolded by a pair of yours, with the one at v. 15 (o e c
\
vxv t. tou o, e) echoing that in anastrophe at v. 3
(o e ,t, e,0 v, o, e). Through these audible cues, the
mythic section is sealed off as a coherent block of argument.
Heracles leads off the myths as bets a son of Zeus, a paternity
emphasized in v. 9. One may suspect that Aristotle had in
addition political and philosophical reasons for beginning with
this gure: Heracles was something like a patron saint for the
house of Macedon, and his labors were used by philosophers
to symbolize philosophy as a heroic quest.
5
In the company of
Achilles and Ajax, however, Heracles is fully intelligible as a gure
from popular ethics embodying excellence attained by toil.
6
A
comedy about education produced in 423 showed an aspiring
sophist being trained to play the What is best? game: when
set the challenge Of the children of Zeus, which do you think
had the most excellent [ariston] soul, and endured the most
toils?, the student, a traditionalist, reveals his conservatism
by declaring No man was better than Heracles (Aristophanes
Clouds 1047 ff.).
For all these reasons, Heracles is perfectly expectable in this
context, but things become more resonant with the addition of
Castor and Pollux: Heracles, of course, ultimately ascended to
Olympus after his labors, and these other sons of Zeus also
achieved immortality of a kind in taking turns, one dwelling
in Hades while the other was on earth. Both Heracles and the
Dioskouroi were used as paradigms for achieving immortality
in the secret initiatory religions that go under the name of
Readi ng 145
Orphism, and their apotheosis entered Alexanders propaganda
as precedents for the kings divinity.
7
In Aristotle, then, the rst
two examples are at least intimations of immortality, though
it should be said that he remains mute about anyones being
divinized: Heracles and the Dioskouroi exemplify toiling for
Virtue (11), not dying for and being resurrected by her. There
is no doubt that the next pair, Achilles and Ajax, die, even in
periphrasis (14). Yet Achilles fate is also potentially ambiguous:
his death in battle is presupposed by Homer and other early
epic, but elsewhere the tradition hints that his goddess mother
Thetis secured for him a special status in the afterlife. The story
that she transported him to the Isles of the Blessed is a mythic
paradigm in an Attic skolion in praise of Harmodius (894 PMG),
the young tyrant-slayer of the end of the sixth century who was
honored by many in Athens as a martyr who ushered in the
democracy. The skolion asserts that Harmodius never died but
dwells in the Isles of the Blest, along with Achilles and Diomedes.
The sequence from Heracles to the Dioskouroi to Achilles thus
might give a listener a hint that heroes somehowovercome death,
or at least prepare us for the award of some form of immortality
to Hermias; but the last hero namedsolid, stubborn, suicidal
Ajaxremains to insist on the fact of death. What joins him
to Achilles is not immortalization but the fact that both were
venerated posthumously with heroic cult.
8
At this point Hermias enters, markedas the culminationof the
argument by the ring-composition that seals the mythic section.
As with Sapphos Anactoria, Hermiass rst appearance in the
poem can be thought of as his epiphany, as the reappearance
of the longed-for one, insofar as language can bring him back.
Aristotle marks the epiphanic moment with a not-immediately-
legible expression, the kenning Atarneus nursling, thereby
making himarrive suddenly andinvisibly, as the hearers solution
to a riddle. Evasive language also marks the mention of his death,
which comes in periphrastically but heroically: depriving the
rays of the sun of his presence (e . tou y,ov eu y e, 16) is
a strong twist (which some have found awkward) on a series of
epic expressions for life as looking upon the light of the sun
and for death as leaving the suns rays.
9
Periphrasis is always
welcome in mentioning death, a word which was not exactly
taboo in speeches over the grave but which could readily be given
up when an ennobling alternative was available.
10
There is no doubt that Hermias dies in Aristotles song.
The speaker never takes leave of his senses and closes not
with his heros deication but with the perfectly traditional
idea of his immortalization in memory. Indeed, the rst allu-
sion to immortality in the poemthe aforementioned epithet
isathanaton for the fruit that Virtue casts before the minds
eye (v. 7)is decidedly qualied: despite having a root sense
that can mean equal to immortal, the exaggeration amounts
to equivocation, precisely implying that equal to is not the
same as immortal. I have identied the closest thing to a
causus belli in the poem as v. 18, where Aristotle comes out
with the simple, unambiguous athanaton, immortal, modifying
Hermias.
11
But in the grammar-plot of that sentence, athanaton
is not an attribute bestowed by the speaker but one that
will be bestowed on Hermias in the future, and by goddesses
who, like Sophocles immortalized Heracles, can be presumed
to know what they are talking about. For his part, Aristotle
offers Hermias only the conventional promise that his name
and exploits will be repeated in song. I cite but one earlier
example of the convention, from Homer, who also seems to
know the trick of referring to his poem as it approaches its end.
In the underworld scene from the last book of the Odyssey, a
deceased Agamemnon predicts that Penelopes excellence will
be rewarded by the gods, and so the glory of her aret will
never die, and the gods will contrive for her a song among
Readi ng 147
mankind, a gracious song for prudent Penelope (24.1968: x ot
\
x. o ou
ox o. txet /

y
\
e ,x y, x uouot c
t0ov tototv
e otc` yv / e 0 evexot e, tooev
,ovt Hyv.o ty). The

word translated gracious (e, tooev) is often understood as
pleasing, suggesting that the song of a faithful wife, reunited
with her husband after many trials, will give pleasure; so it
doubtless does in the formof the Odyssey itself. But Agamemnon
sees this song as a reward for Penelope, and so it is gracious
because enduring glory is how the gods reciprocate (extending
e,t for e,t) her devotion to aret. Clearly, Hermias ts into
the Homeric model of fame. The venerable pedigree of Aristotles
promise of fame is signaled in v. 17 by the word e o tcto,
celebrated in song, a word that occurs just once in Homer
in a memorable scene from the Iliad. Referring to itself with a
directness unparalleled in the rest of the poem, the Iliad shows
Helen complaining that her sad fate will be e o tcto, a subject
of song in the future (Iliad 6.358). Helens prediction is, needless
to say, fullled in the very work that reports it, and the trick of
ironic self-reference accompanies the word when Aristotle takes
it over. Like Homer, he manages to make the promise of songs to
come for Hermias come true in the course of singing about them.
The intricate way in which he does so, however, repays a closer
look at the way fame is spread in the nale.
Immortalizing: Hymn to Hermias, vv. 1721
In shaping an ending for his poem, Aristotle took up a very old
patternfor closing a Greek lyric, a reectiononfame that was also
a self-reective allusion to the poems future re-performances.
The two ideas were connected by the fact that the word for fame,
x. o, meant what people hear, and muchof what people heard
about the past they heard in song and verse.
12
A striking archaic
example may be quoted from Ibycus in the sixth century BCE.
13
An itinerant poet who frequented courts and symposia, Ibycus
addressed one longish ode of uncertain genre to Polycrates,
tyrant of Samos and one of the most inuential patrons of the
arts in his day. The last three verses of the poem (quoted below)
suggest that it was an enkmion, a praise song directed at a living
person.
14
But we must be tentative since we do not know what
form enkmia were expected to have in the sixth century, and
Ibycus seems to be toying with whatever form he has chosen as
his point of departure. The beginning of the poem is missing,
but the 48 verses that survive are almost completely taken up
with summarizing the Trojan War, including a mini-catalog of
Greeks in which Achilles and Ajax are paired as the foremost
spear-ghters (vv. 3234 PMGF). Closing the penultimate triad
with this item, Ibycus leads off the nal one by identifying a
certain Cyanippus as the fairest (x e..toxo, 36) of the Greeks;
it seems he had such a lovely form (o, ev, 45) as to outshine
even Troilus, a Trojan youth celebrated for his beauty. Having set
up, as it were, a beauty contest for the fairest young man at Troy,
Ibycus breaks off his myth and turns to Polycrates in the here
and now: he concludes by promising the prince unwithering
glory alongside Cyanippus and Troilus:
xo t ` v ce x e..o et ` v
xe`t o u, Ho. ux,ex, x. o e
0txov
\
t
\
x` ex e otc` ev xe`t
v x. o.
These will have a share in beauty forever,
andyoutoo, Polycrates, will have a glory that does not wither,
so far as song assures, so far as is assured by my glory.
Ibycuss narrative shift from the martial to the erotic recalls,
in a general way, Sapphos poem on Anactoria, and the shift
Readi ng 149
in the quoted lines from muthos to logos is familiar from both
Sappho and Aristotle. As in Aristotle, a nal conceit about the
fame conferred by poetry caps a mythologizing core. What is
most pertinent about this text, however, is that it nds closure
by turning the age-old topos of the glory conferred by song
on itself. Ibycus concludes not with a general promise that
Polycrates will be celebrated in song but with the assertion that
his future fame will take the formof repetitions of this very song;
Polycrates hope for future remembrance and Ibycuss quest for
poetic glory are jestingly identied: the patrons beauty will be
dateless as long as the singers song is heard. In a similar way,
Aristotles prediction that Hermias will be sung of alongside
heroes (aoidimos) will have been fullled any time his ode, with
its mythological core, is sung. What is distinctive about Aristotles
development of this conceit is that his hopes for his songs being
repeated through time are expressed in the gure of a song
endlessly sung by the Muses; as we will see more fully below,
this connection is reinforced by making their song a sort of echo
of his. At present we may parallel this closing move by adding
Ariphrons hymn to Health and the suggestion near its end, on
my interpretation, of re-performances in future conversations
(oarois). It seems that the end of praise songswe have also
considered the ourish about kleos ending Bacchylides rst
epinicianwas felt to be a place where thinking about kleos and
its ironies was appropriate. The trick was not forgotten, and can
be seen closing many a Renaissance sonnet: So long as men can
breathe, or eyes can see, / So long lives this, and this gives life
to thee.
15
The fact that Aristotle is executing a traditional conceit in
its expected sedes allows him some complexity in developing
the guration of fame; indeed it demands that he be ingenious
in expression to make the song new.
16
So let us press closely
the logic of the sentence, curious though it may seem at points.
Here again is the last movement of the song, after Hermias has
died in the poem:
xoty` e, e o tcto
,yot,
e 0 evexv x tv eu yoouot Mo uoet,
Mveoo uve 0 uyex,, Lt-
v tou o pe eu
ou- 20
oet t. te x y ,e ppe tou.
Hence he will be a subject of song because of his exploits,
and the Muses will grow him into immortality,
those daughters of Memory,
making reverence for Zeus grow, 20
god of guest-friends, and the rewards of steadfast friendship.
A logical particle (xoty e,, cf. y e, in Ariphron) articulates a
break: Hermias, having crowned the list of Virtues devotees,
becomes the subject of a new discourse about the rewards she
bestows. The Muses will create songs about Hermias because
his exploits (
,yot in v. 17) in pursuit of Areta are on a par
with the legendary exploits (
,ye, v. 10) of Heracles and the
other heroes. Applying to Hermias the Homeric word e o tcto
(v. 17) was enoughtopromise enduring poetic glory; but Aristotle
adds that Hermias will thereby become undying (e -0 evexo),
a strongly afrmative swerve from fames traditional epithets
like unwithering (e
-0txo), which might be expected in this

context. Still, this bold predicate is only bestowed in prospect:
undying is proleptic and Hermiass immortality is only an
anticipated result of the Muses assiduous singing.
The Muses are introduced as the daughters of Memory
(Mnamosuna); this too is quite traditional, but also susceptible
of an anthropological interpretation to the effect that Hermiass
immortality will consist in this songs being remembered, that is,
Readi ng 151
re-performed. Aristotle is not obtruding a philosophical doctrine,
but uses the Muses as a gure for a human faculty in order to
express himself in words that an enlightened sort of person
could sing. In the background here must hover the role of
the goddess Memory in Greek religious beliefs concerning the
afterlife. Although these ideas tend to be submerged in our
literary texts, recent years have seen the recovery from tombs
throughout Greece of small gold tablets and other inscribed
tokens that carry instructions for securing a happy existence
in the underworld. In many of these texts, and in the mystery
initiations they point to, Mnamosuna was a powerful goddess who
hadtobe entreatedtoensure the continuedpersonal identity and
status of the deceased.
17
In the context of predicting Hermiass
immortality, mentioning the daughters of Memory could not
have failed to touchonwidely held spiritual hopes. But Aristotles
Muses are also, as the run of the sentence quickly reminds
us, daughters of Zeus; the fathers name places Mnamosuna in
the poetic tradition of Homer and Hesiod, thereby afrming
the Olympian order in which great men are immortalized
through song, not through secret initiations or cult. In fact,
with the mention of Zeuss name, the Muses song glorifying
Hermias morphs in the nal verses into an act of reverence for
their fatherit is Zeus that they venerate. This is also quite
appropriate, for in the canonical picture of the Muses in Hesiods
Theogony, they do nothing so much as delight the mind of their
father with song (Theog. 3637; cf. 11, 4041, 4748, 6871).
Taking things quite literally, however, this song-within-a-
song makes the genre of Aristotles hymn shift a nal time: in
describing the song the Muses will sing as an act of reverence
for Zeus xenios (vv. 1920), Aristotle quotes the keynote epithet
they will give him, in effect echoing their song in advance. But
so potent is the venerable epithet xenios that Aristotles song
to Virtue undergoes inuence, as Harold Bloom might put it,
from the Muses future song, in which praise of Hermias will be
transmogried into a hymn to Zeus. In heaven, Hermias will
change from the dedicatee of the song to Virtue and become
an example in the Muses song: the suggestion is that the rm
friendship he showed will be the story (the muthos) they tell to
exemplify the rewards in store for those who honor the god of
guest-friends. Aristotles own song for Hermias does not entirely
disappear as it is sublimated into a hymn to the god of guest
friendship, for a self-reinforcing irony remains: the heavenly
hymn of the Muses will only be heard on earth to the extent that
Aristotles hymn of praise, predicting that hymn, is performed.
In this way, Aristotles close entangles an encomium for
Hermias with a hymn to Zeus. This overlapping of songs is
brought out by repeating the verb eu evtv in 18 and 20
to describe both song-acts. Some have found the repetition
awkward, and Wilamowitz suggested that the rst (v. 18) be
changed. But both should be kept, as the Didymus papyrus
conrms. The most general sense of eu evtv is to make
increase (cognate with English augment), but relevant here
is the verbs special vegetal sense, to foster or make grow.
(Its Indo-European root is the source of to wax.)
18
The verb
had been used with this sense in connection with areta in
earlier songs of praise: Virtue when praised grows like a tree
declared a praise poet sometimes thought to be Bacchylides
(Fr. dub. 56 Maehler: e ,x` e y` e,
etvo ve c vc,ov
\
`
e xet); and Pindar expressed the idea with a typical increase
in density, Virtue springs up like a tree freshly bedewed when
it is raised among just and wise men to the watery heavens
(Nemean 8.4042: e t
oot c e ,x e, .,e t
,oet /
\
o
\
x c vc,ov, -
v>ooo t e vc, v e ,0 to
v ctxe tot x ,
u
\
y,v et 0 ,e). When Octavian became the rst emperor of
Rome, he exploited this same root to rename himself Augustus
as a fostering tutelary spirit rather than tyrant over his people.
Readi ng 153
Activating the vegetal sense of eu evtv well suits the context
of v. 18, for, as noted above, the metaphor of growing or fostering
is traditionally associated with enduring glory. Again, Pindar
says The Pierian daughters of Zeus [i.e., the Muses] cause glory
to increase far and wide (Ol. 10.95: x, ovxt c u ,` u x. o /
x,et Ht, tc Lt) and Bacchylides says much the same: The
gleam of areta does not die out for men along with their bodies
but a Muse makes it increase (3.9092: A,x e[ y ]` v ou
tv u0t / p,ox v e
\
e [ ]ext yyo, e ..` e / Mo uo e
vtv x,[ t.]). I have noted that vegetal tropes run very deep in
the traditional vocabulary of glorication, as in the epic phrase
x. o e
0txov, glory that does not wither away (e o-0 tvtv).

The same complex of ideas sustains Aristotles nearly oxymoronic
phrase immortal-like fruit (xe,v t oe0 evexov) in v. 7.
19
In
this sphere, eu evtv functions as the opposite of 0 tvtv (as
in Euripides Fr. 645a.3, quoted above) and so it is very neatly
used in v. 18 to depict the Muses fostering of Hermias by
using song to prevent his falling into oblivion. Up to Aristotles
time, eu evtv is not commonly found in connection with the
bestowing of x. o, and so its use here may be a slightly fresh
variant for such expected (and tired) verbs as glorify (x. t)
or celebrate (u
\
v ).
20
Wilamowitzs proposal to change the
rst to eu c yoouoet, voicing, could claim as a merit that it
would be a fresh turn on traditional language while remaining
traditional in expressing the often acknowledged truth that the
glory (x. o) bestowed by song is not abstract but depends on
the name of the honorand being heard, being voiced inrepeated
oral performances. (In this respect it surpases Bergks e oxo uoet,
making, adorning, or Crusiuss e
ouoet, venerating.) Yet

I fear Wilamowitz may have been overimproving Aristotle.
Repetition or near-repetition is not something the song tries
to avoid; rather, it seems to resort to repetition consciously for
emphasis and connection.
21
In augmenting Hermiass excellence, then, the Muses will
keep him growing until he becomes immortal in song. The
same verb can be repeated in v. 20 (o pe eu
ouoet) because
by the same act they will make increase the respect owed to
Zeus xenios. Keeping the second instance of eu vtv, we may
note that it had a further range of meanings quite different from
the ones above but relevant here. Inrhetorical contexts eu evtv
(and its nouneu
yot) had the technical sense of amplifying or

magnifying a subject througha discourse of praise.
22
Aristotles
recommendations for rhetorical auxsis appear in his discussion
of prose encomium in Rhetoric 1.9 (esp. 1368a): praise ampli-
es its subjects through examples (paradeigmata), comparisons
(parabola), and rhetorical arguments (enthummata).
23
Aristotle
takes his own advice in framing his praise of Hermias: the
paradeigmata of his song are Heracles and the heroes, the
priamel supplies attering parabola, and the whole is a rhetorical
argument (therefore, 17). This recipe for praise is followed not
only by Aristotles song but within it, in the song the Muses
sing for Zeus. As suggested, their praise would be rhetorically
strong if they found examples of devotees of Zeus xenios, an
obvious possibility being Hermias of Atarneus, who had found a
reward (geras, 21) for being rm in friendship. If in Aristotles
song Hermias is an example of the desirability of Areta, his story
will gure in the Muses song to show the greatness of their
father. The repetition of eu evtv underscores the doubleness,
or echo-effect, with which the song is invested at the end.
A doubling or echo-effect need not lead to confusion, and
the lurking charge of impiety compels us to be clear that
Aristotle does not collapse the distinction between god and
mortal. The twin objects of the second eu evtv are sharply
contrasted in a way that strikes again the note of piety in the
songs opening. The Muses singing will keep alive the worship
(sebas) of Zeus xenios and the reward (geras) that Hermias is
Readi ng 155
owed for his friendship toward Aristotle and all things Greek
(vv. 2021). The direct objects, formally and phonetically parallel
(both are neuter s-stems with preposed genitives), are strongly
antithetical: whereas a geras, reward, is what a hero gets (it is
often used of booty as a reward or prize of honor, sometimes of
last honors), sebas belongs to gods. (So the name Augustus was
translated back into Greek as Sebastos.
24
) The chime makes more
prominent the different prospects open to gods and menas
Pindar put it, the vast difference in power that separates them
(Nemean 6.17). Provided that the crucial gap between mortal
and immortal is not forgotten, the two songs tend to converge,
for one can praise Hermias and honor the divine mandate to
hospitality at the same time.
There is, then, nothing impious in a song that acknowledges
the limits of mortality and the sovereignty of Zeus. Aristotle
today would not be accused of impiety, though some would not
like it that in the end he takes a secular, humanistic attitude
toward the nature of human excellence and the prospects of life
after death. The sequence of noun phrases beginning with the
appearance of the Muses in v. 18 creates a steady diminuendo
away from myth back to reality: the Muses rst appear as the
daughters of Zeus and Memory, as they do in Hesiod; then
they seem to personify an awe-inspired respect for xenia; lastly,
what is in view is the stableness of rm friendship. What will
ultimately keep Hermiass example alive is a combination of the
power of memory, veneration for basic social principles, and
loyalty among friends, especially those who may embody their
friendship in sharing and handing on old songs. In the more
poetic terms of Harold Bloom, this pious poem is not Romantic;
it is not a lie against time. I nd it a courageous poem, one that
discovers within hard limits a reason to praise, a condence to
afrm that praise is worthwhile, and a hope that praise of the
good will last through time and may even reach the one praised.
The ultimate law the text obeys is not one of literary genre, but
the ancient religious and ethical ideal of do ut des, let good be
repaid with good.
Aristotles reference to future glory is not merely a pretty
nal conceit but the real end of his poem (or his prayer).
As long as people sing this song, which commends in words
what Hermias commended in action, the devotee of virtue will
have left behind something lasting, something with a power to
inuence the character of people in the future, a paradoxical
monument or reminder composed of speech, as Bacchylides
put it. Aristotles sense of commemoration is not the old magical
one of intoning the deceaseds name to bring the shade near, of
the name as a metonym for the person. This may explain why
he conspicuously declines to fulll one of the oldest functions of
commemorative verse and at-out name his friend.
25
His vision
of the function of song is a more humanistic and philosophical
one of providing an example to remind the living of the rewards
that friendship brings and to encourage the preservation of this
value with promises that stop just short of being incredible.
The excellence that Hermiass life displayed is reciprocated in
the praise his steadfast friends bestow on him, and, as the
last chapter will show, repeating this praise will sustain the
community of his survivors.
Chapter 9
Endurance
Abook devotedto reading andre-reading a single lyric frommany
angles might end by trying to tie everything together neatly. But
the previous chapter has argued that Aristotle placed Hermiass
immortalization in an endless series of musical re-performances,
and so it would be wrong to impose any reading of this text as its
complete or nal one. I therefore propose by way of conclusion
to connect some episodes in the later career of Aristotles song
with its principal theme, showing how the survival of its words
in fact depended on the friendship it celebrates.
Much is packed into Aristotles nal words of praise of
Hermiass steadfast (p peto) friendship. The same virtue is
cited by Callisthenes when he praises Hermiass courage and
steadfast character (x` yv e vc, tev xe`t x` yv ppet oxyxe x v
x, ov, col. 6.2). In part, this emphasis can be understood
in light of Aristotles ethical writings, especially his discussion
in Nicomachean Ethics of friendship as a form of virtue, as
a mutually shared enjoyment at the exercise of high human
capacities. Scholars have thought of Hermias in connection with
Aristotles discussion of how friendships are possible with those
157
who are outstanding in wisdom or political power.
1
With each
contributing his proper part, such men can have the best kind of
friendship, that based on the excellence of each partner, rather
than their usefulness or the pleasure they afford (NE 8.6). As
in Aristotles relationship with Hermias, we may understand,
virtuous friendships are proof against slander, and they last
(NE 8.6, 1158b89: e ct ep.yxov xe`t ovtov); friendships based
on utility or pleasure are less enduring (1158b4:

y
\
xx ov . . .
vouotv). The steadfastness of Hermiass friendship, therefore,
is reciprocated by Aristotles steadfastness in coming to his aid
even after death. In this way, the word bebaios takes on a second
referent and suggests the lastingness of Aristotles tribute to
Hermias, his song as stable monument. In what follows I will
argue that the last word of the song is not only ethical praise but
also alludes to the afterlife of the song, in line with the assertion
of Plato that the products or offspring of spiritual friendship
are more steadfast, ner, and more immortal than physical
progeny (Symposium209C: t. tev ppetox ,ev, e
\
x xe..t ovv
xe`t e 0evexx ,v e tcv xxotvvyx ox).
Literally, bebaios denotes standing rmly, immovably in space;
it is a good word to describe a monument or a border-marker,
any hard and hardy object like a statue or a stele planted to stay
xed in the ground. In praise discourse, bebaios is frequently
extended metaphorically to time, referring to enduring virtue
and memory. With such a resonance, the metaphor at the end
of Aristotles poem takes on something like its function at
the end of Bacchylides ode quoted above: the virtuous mans
reward after deathto be celebrated in song and have glory
is the best possible monument (e
ye.e) he can leave behind.

Bacchylides agalma, literally an object of delight, commonly
denotes objects dedicated in sanctuaries, often commemorative
objects inscribed with a message to posterity. (The cenotaph or
statue for Hermias in Delphi would be an example.) I suggest
Endurance 159
that Aristotles closing phrase glances at this traditional idea of
spreading glory throughbotha solidmonument anda lasting oral
report: like Bacchylides and many other praise poets, Aristotle
wants his winged words to have the heft and durability of stone.
2
He hints at this by combining the ethical use of bebaiosinwhich
the idea of being xed in place is applied to a moral qualitywith
the use of the term in literary contexts to gure the lastingness
to whichverbal commemorationaspires.
3
Aristotles closing note
has something of the avor of Horaces boast in a poem closing
a lyric collection: I have brought to completion a monument
more lasting than bronze (Exegi monumentum aere perennius,
Odes 3.30.1).
The nexus of ideas attaching to ethical bebaios can be
connected by a short chain of associations to the idea of writing
as speech made lasting by being made xed and tangible. The
best known link in this chain is the use of bebaios in Platos
Phaedrus to describe the presumed stability that writing confers
on speech.
4
That supreme writer undermines the pretension of
writing to give stable form to thought in order to insist that
true understanding is not transcribed in texts but is created
in the souls of philosophers by their dialectical conversations
(their dialogues). This ironic attack on writing in writing is
characteristic of Plato, but neither conned to him nor without
precedent: earlier commemorative poetry in Greek had hit on
the idea that the true monument of a fallen warrior is not the
physical marker over his body but the inspiration his example
leaves behind for the survivors. Hence I need not assert a direct
connection with Platos Phaedrus when I suggest that bebaios
represents Aristotles response to the perennial challenge for
eulogistshow to make plausible the claim that something of
the departed will live on. I suggest that this words background
and emphatic placement in the song indicates that, in promising
that Hermiass heroism will be preserved, Aristotle considered
whether this preservation will be on peoples lips, on stones, or
as scripts. Having seen the handwriting on the wallperhaps
having seen an inscribed version of Ariphrons famous paean
he alluded with his last trope to the question of what shape the
survival he promised might take.
Memorial: Aristotles Elegiacs to Eudemus
That Aristotle gave thought to how his commemorative song
might become a lasting memorial appears if we consider his other
most substantial poetic remnant, three anda half elegiac couplets
quoted by Olympiodorus, a sixth-century Neoplatonist philoso-
pher. The poemis only partially quoted, and in speculating about
its missing part, it is key to bear in mind that it is adduced
as evidence that Aristotle did not quarrel with Plato, contrary
to what some had asserted. I will accordingly interpret the
fragment as if it constitutes, inOlympiodoruss words, Aristotles
encomiuminpraise of Plato (
etv vH. exve
yxt et):
5
.0 ` v c
x.tv` ov Kx,o ty c ecov

u op ov y t. ty t
\
c, uoexo p` ov
e vc,` o o
\
`v ou c et v tv xo tot xexo tot 0 t,
o
\
` ovo y` , xo 0vyx v xex ctv
ve,y
ot x tt x p tt xe`t 0 ocotot . oyv 5
\
e ye0 o x xe`t u ce tv e
\
e y tvxet e v y,
ou v uv c
oxt .ep tv ou cv`t xe ux e ox.
. . . . Coming to Cecrops glorious plain
he piously established an altar to holy Friendship
for a man whom the lowly ought not even to praise,
who alone or rst of men showed for all to see
by his conduct in life and his way of speaking 5
Endurance 161
that a man becomes good and happy at once;
now is it not possible for anyone ever to attain these
things.
Olympiodorus calls the poem the elegiacs to [, o] Eudemus
and once again interpretation hangs on a proper name. The ques-
tionof which Eudemus the poemaddresses affects howwe decide
who it is who is arriving in Athens at the fragments opening and
who is the excellent man commemorated with the altar. In an
excellent exegesis, Jaeger inferred from Olympiodoruss use of
the preposition , o in his title that the addressee was alive at
the time; he accordingly identied him as Eudemus of Rhodes,
a well-known pupil of Aristotles. If this tiny detail in a very
late source is to be pressed, we must discard another far more
dramatic candidate, Eudemus of Cyprus: in his youth a friend of
Aristotle and pupil of Plato, he died in 354 ghting the tyrant of
Syracuse.
6
As this context has been lost to us, we fall back on the
text to try to identify its actors.
To begin with the person who dedicates the altar: an early
view, now not generally believed, proposed that it was Socrates
who showed the path to becoming truly good (agathos, the
adjective of aret), in which case Plato would naturally have
been the dedicator.
7
But coming to Cecrops plain (i.e., Attica)
clearly indicates that a non-Athenian is intended, in which case
it is natural to think that the dedicator was Aristotle, either
subsequent to his arrival to study with Plato in 367 or, more
likely, after 334 when he returned to Athens after his sojourn in
Asia Minor and Macedon. This is the interpretation proposed by
Dring, which I support.
8
It makes sense to think that Aristotle
on his return to Athens would have put up a sign of his close
connections with the thriving Platonic Academy (and possibly
too would have announced that he was back in town). His last
will and testament makes several bequests for monuments to be
erected for friends and family, showing that he was fully involved
with the culture of dedications.
On this view, Aristotles elegiacs recount his own action in
erecting an altar to commemorate his friendship with Plato,
whose philosophy, as Jaeger points out, can be seen neatly
epitomized in the penultimate line identifying being good
and being happy.
9
Now this altar is, for us, an object made
purely of discourse, much like Aristotles Delphic monument to
Hermias; but even imaginary altars offer space for inscription,
and ancient as well as modern readers have taken vv. 23
to be Aristotles paraphrase of what he had inscribed on the
monument. Aristotles biographical tradition records a number
of attempts to reconstruct the original behind the paraphrase,
including a version that would have Aristotle write, on the
grounds of august friendship I dedicate this altar to Plato.
10
But if there were an actual altar behind these verses, Aristotle
would hardly have outed convention and dedicated it to Plato;
great as is the respect he evinces here, one did not offer altars
to mortals.
11
(The Samians dedicated an altar to Lysander along
with the paean they sang for him, but Lysander was being given
extraordinary honors and 404 was an exceptional time.) The
adjective ov y (august, holy) in the paraphrase suggests,
as Jaeger observes, that the abstraction friendship was being
personied and divinized, and this is a far likelier dedicatee of
an altar than a person. Jaeger is convincing when he argues
that the altar, whether we think of it as real or imaginary,
carried only one word, To Friendship.
12
The grammar of
the paraphrase, however, is dense enough (both a man and
friendship being genitives depending on altar) to bring the
two postulated dedicatees together, suggesting that the goddess
Friendship was revealed to mankind in the person and actions of
Plato. (Describing such vivid apparitions is one common use of
the adverb
ve,y in 4.) Such an implication would t with the

Endurance 163
impressionthe complete text gave toOlympiodorus, whoclaimed
it as proof that Aristotle revered (o pt) Plato as a teacher.
13
Jaeger is right that the imagined inscription need not have
named Plato any more than the elegy to Eudemus does (or than
Hermiass song names him). In this case, the fact that Plato was
the proper name underlying man in v. 3 would have to have
been preserved in a para-tradition, either an oral tradition as
friend passed the couplets on to friend, or, if the elegiacs were
inscribed, by those people to be found around monuments and
antiquities who were willing to explain them. In fact, as Jaeger
also observes, Aristotle frames his speech as the discourse of such
an exegete.
14
This is also, one sees on reection, the stance he
took in the couplets on Hermiass monument at Delphi.
The greatest objection to taking Aristotle as the dedicator of
the monument in the poem is that speaking of oneself in the
third person is really quite unparalleled in elegiac poetry until
well into the Hellenistic age. For this reason, Jaeger concluded
that Aristotle describes the dedication of some unknown person,
presumably a pupil of Platos.
15
But it is hard to imagine such a
close friend of Platos passing unmentioned among these giants,
and here we may, as in the case of the possibly faux Delphic
inscription, allow Aristotle to be near the avant garde in verse.
Taking Aristotle as the dedicator in this text gives us a
rather multilayeredelegyhe commemorates inanelegy his own
earlier commemorationof Plato witha monument, itself possibly
bearing a message inelegiacs. Towrite anelegy about aninscribed
altar may seem excessively self-referential and Hellenistic, but
canalsobe seenas a way of broadcasting Aristotles message more
widely: as in the case of the multiform songs praising health, an
ability to translate a message fromone verbal mediumor register
to another extended its reach by multiplying the occasions on
which it could be suitably performed. It is possible that Aristotle
memorializedhis friendshipwithPlato, andwithHermias as well,
both on stone and in verse: in this case, the elegy to Eudemus
serves as the monuments loudspeaker, replaying its message and
extending its reach.
16
As forms of commemoration, then, we may compare the
altar to friendship erected in the words of this poem and
Aristotles ending his song to Hermias on a note of stability
(ppe tou). As the old topos of the funeral oration has it,
Aristotles hymn to areta will be Hermiass true shrine, and his
steadfastness will be matched by the endurance of the song
in remembrance of him. Without imputing postmodern views
to Aristotle, we note that his song ends with poetry making
everything happen, contradicting Audens despairing declaration
in his elegy for Yeats that poetry makes nothing happen.
The power of perpetuating fame that tradition had symbolized
in the Muses really inheres in the song tradition, which is a
prime way that values are grown from one generation to the
next. Aristotles song ends on a bid to join this tradition: on
endless future occasions the prize for steadfast friendship
will be perennially re-awarded to Hermias in a ceremony that
reverence enjoins and memory makes possible. If we take the
song at its word and suppose it continued to be performed in
the Lyceumafter Aristotle died, we can imagine something of the
setting where this monument in words would have resounded.
Theophrastuss will, as recordedinDiogenes, makes provisionfor
refurbishing a shrine in or near the Lyceum called the Mouseion,
andit directs that animage of Aristotle be erectedthere alongside
one of Nicomachus; Diogenes also mentions an altar nearby.
17
As James Redeld remarks in an illuminating study of the rise
of academic institutions in Athens, This is not exactly hero cult,
but it is something like it: the spirit of the founding philosophers,
lingering in the place where they taught, motivates the presence
there of their successors and gives those successors a claim on
the space.
18
In this sense, the Hymn to Hermias would not only
Endurance 165
preserve the patrons memory among his friends but also that of
its author, keeping the cult of Aristotle alive.
19
The survival of the song was assisted both by creating physical
spaces in which it might be re-performed and by investing its
language with hymnic, poetic, and rhetorical powers: Aristotle
claims the right toname the god, topersonify, torecall, toinspire.
These powers can be called divine if for no other reason than that
through them values last for more than a human generation.
The true daughters of Zeusthe true prop and stay of their
fathers lawsare the songs that manage to get remembered.
The immortalization Hermias will obtain depends on a series
of human, social actions; the song will survive by being handed
down among friends, as if it were a valuable guest gift. And as
with guest gifts, each exchange honors both of the principals
while afrming the greatness of Zeus xenios.
20
Thus Aristotles
song ends on a sophisticated, humane thought, but still piously.
Composed, subtle, observing due decorum without forgoing
irony, it reinvents the hymn: adopting elements from hymns
to praise Hermias, it does not deny the distinctions between
mortal and immortal, encomium and hymn. The themes that
Aristotle chose make Hermiass song suitable to serve as a skolion
celebrating fellowship, even as the friendship extends across
generations; at the same time, his scrupulousness in expression
makes the song appropriate to be re-used, for example as a
libation paean, especially when the toasts were to Zeus, the
Olympians, and the heroes. Aristotle composed at a time when
genres were in ux, as they always are; his work as poet was
simultaneously to create a statement greeting something new, to
project an atmosphere of ritually correct form behind the words,
and to inect this statement so as to have an effect, a pointed
meaning, both at its rst presentation and in its hoped-for
afterlife. Exploiting the language of tradition and of his time, he
made an utterance that was at once intelligible and memorable,
well-behaved and distinctive, tightly compact with thought and
free to address the future.
Survival: A Letter from Plato
Inthe collectionof letters ascribedto Plato, the Sixthis addressed
to Hermias of Atarneus. It says it was written in Platos old age
(322D), and so may date from the last decade of the tyrants
life. This brief letter recommends two brothers, Erastos and
Koriskos, who had studied with Plato and now are in Hermiass
area, having returned to their hometown Skepsis in the Troad.
Plato vouches for his former students as talented but not very
worldly and, addressing all three, urges them to form a mutually
benecial friendship. In other words, he asks Hermias to serve
as patron to these promising young philosophers. As a letter of
recommendation it seems to have worked, if Jaeger was right to
restore Koriskos beside Erastos in Didymuss list of Platonists at
Hermiass court (col. 5.5354).
21
Lovers of Plato learn to regard the epistles with fascination
underlined by worry. Some seem too good to be true, most
famously the Seventh, in which Plato discloses the true reasons
for his deliciously complex writing. Treasures like the Seventh
Letter make it worth bearing in mind that a famous writers
letters were as desirable to an Alexandrian librarian as they were
easy to concoct from the authors oeuvre. But there are no strong
grounds for suspecting the authenticity of the Sixth.
22
It has
no obvious, excessive details, and what details it does give do
not conict with what we know from elsewhere: Erastos is easily
identied with the Erastos mentioned in Epistle 13 as having
assistedPlatoina nancial transaction(363B); Philodemus notes
that he wrote a memoir of Plato, and Didymus places him with
Aristotle at Assos.
23
Koriskos is much better known, especially
Endurance 167
fromnumerous incidental references to himin Aristotles works:
they were fellow pupils at the Academy and lifelong friends.
The brevity of the letter and its lack of juiciness count in
its favor. I do not pretend to retry the case, but note that
the language in which the tyrant is addressed is, typologically
speaking, quite in harmony with the courteous didacticism we
have seen in several of our texts. Amid the opening greetings,
the authorwhom I call Platocongratulates Hermias: Now
Hermias power will be increased not by the number of his
horse nor by any military alliance, nor even by the addition of
gold, so much as by steadfast friends of sound character (322d:
\
E, te ` v y` e, ou
x t
\
v . y0o ou
x e
..y o.tx y
oue te ou c eu
,uoo u,ooyvo vouy votx e` v tv

t x` e evxe c uvet. y` t.v ppe tv x xe`t

y 0o
ovxv
u
\
yt ). Addressing a prince, Plato is ceremonious as he teaches,
and so it is not fortuitous that he sounds like Sappho in
deprecating the value of cavalry in a priamel; or that he sounds
like Aristotle in putting the value of wealth below friends.
24
A
little later, Plato comes up with a gnome that might well have
been pronounced by the proto-advisor to tyrants, Simonides,
Nothing is steadfast [bebaion] in human affairs (322e: x` o y` e,
e v0, tvov ou evx eeotv p petov), a truth which Hermias
eventually learned to his cost and of which Greek tyrants at all
times seemed to need reminding.
Steadfast and sound friendship is what Hermias lacked in
Mentor, but this faint topical point does not prove the letter
a ction. What seems more signicant is Platos using the same
word, bebaios, that was used by Hermiass other Platonist friends.
And the word friend itself (philos) could have a special sense
for this circle, for friend was the common manner in which
members of the Academy greeted each other: Theophrastuss will
refers to members of the Peripatetic community as friends.
25
More important than the authenticity of the letter as a work
of Plato is its allusion to the network of friendship that bound
them all. Certainly there is no mistaking the links that joined
Plato to Aristotle, Aristotle to Hermias, Hermias to Koriskos,
and so on. One noteworthy link here is that between the two
Academicians, Aristotle and Koriskos, for it bore fruit of a rare
sort in the next generation: the son of Koriskos was Neleus of
Skepsis, a close associate of Aristotles successor Theophrastus.
It seems that at Aristotles death in 322 Theophrastus took over
Aristotles personal library, and when he died in the 280s, his
will left all the books to Neleus (DL 5.52). The story gets very
Romanticand implausiblefrom there.
Plutarch and Strabo say that after Koriskoss son inherited
Aristotles library from Theophrastus, he moved it from Athens
to Skepsis. They give not wholly harmonious accounts but agree
that the works were hidden away underground with the result
that they were acquired neither by the Alexandrian library nor
by its rival in Pergamon, which rose to prominence in the second
century BCE. Only around 90 BCE did Neleuss successors sell
the collection, to Apellicon of Teos, a wealthy book collector and
devotee of Aristotle. He brought the library back to his home in
Athens, where it fell into Romanhands whenSulla conquered the
city in 86 BCE. Sulla had the books transported to Rome, where
they eventually came to be cared for by Andronicus of Rhodes.
Andronicus published works of Aristotle (and Theophrastus) on
a better and more authoritative basis than theretofore. In this
way, Neleus, the son of Koriskos, contributed to a renaissance in
the study of Aristotle in the rst century BCE.
This story is not as fully eshed out as one could wish for
in dealing with a corpus of such momentous signicance, and
Athenaeus gives a contradictory account, according to which
Neleus sold the books right off to Ptolemy II Philadelphus
(reigned 274246) for the Alexandrian library.
26
What makes
the stories suspicious is that they seem to be taking sides in
Endurance 169
an argument that either Athens or Alexandria was backward in
its knowledge of Aristotle during the centuries after his death.
And they seemto contradict indications that some scholars knew
Aristotles writings inthe Hellenistic age.
27
One likely alternative
channel for knowledge of at least some of Aristotles works in
this period is another close associate, Eudemus of Rhodes; this
Eudemus, possibly the addressee of Aristotles elegiacs on the
altar for Plato, was a member of the Lyceum at the time when
Theophrastus took over, and may have taken copies of some of
the founders works back to Rhodes. He was later known for his
valuable commentaries on Aristotles works and even, in the case
of the eponymously titled Eudemian Ethics, for editorial work.
Yet the story about Neleus may contain a relevant truth.
Even if the collection he safeguarded was not the only one in
circulation, it could have followed more or less the course that
Plutarch and Strabo suggest, passing on to Apellicon in Athens,
a part of the story attested to by the rst-century BCE historian
Poseidonios (cited by Athenaeus 214d-e =FGrH 87 F 36 =F 253
Edelstein-Kidd = F 247 Theiler). One way or another, it seems
that Aristotles books had to spend time in the care of friends
until the Romans rediscovered them. If the sixth Platonic Epistle
has any basis in fact, it was a protg of Hermias whose steadfast
son played a key role in ensuring that Aristotles excellence would
be recognized in the new world to come. The honor of Zeus, who
honors the obligations of friends, was indeed vindicated long
after all the players had died.
The book collector Apellicon also contributed the nal chapter
to the ancient Hermias story: Aristocles knew a book he wrote
about Aristotles and Hermiass friendship and says of it, anyone
who reads this book will soon cease to speak evil of the two
men.
28
As for the song, it survived better than Apellicons book,
though we cannot say who exactly preserved the text so it could
be copied by Didymus, Athenaeus, and Diogenes. It is possible
that Hermippus recorded the words to Aristotles song along
with the trial story; he could have found the text among the
works about Aristotle in Alexandria, if not in the philosophers
corpus itself; but we cannot be sure that Hermippus quoted
the text.
29
Alternatively, it may have found its way into the
record among those lost works on paeans with which Athenaeus
supplemented the material from Hermippus; this was a branch
of study that Didymus knewas well, and to which he contributed
On Lyric Poetry. Finally, Aristotles opponents may have cited the
incriminating lines and so unwittingly kept them alive. I have
stressed that songs by themselves did not automatically get
transferred to texts, and it is not clear that Aristotles lyric
was included with the corpus of his other writings. Diogenes
appends to his Life of Aristotle a list of 146 of his works
(5.27), as indeed Hermippus had included such a list in his
biography. Diogenes list ends with two books of poems: one
is labeled hexameter poems (
y, no. 145), whence he cites
the hymn to the pure goddess (Fr. 671 Rose); the other is
elegiac poems (
.y te, no. 146), identied by O Daughter of

a mother with fair children (Fr. 672 Rose = 672 IEG). In his
valuable study of the ancient lists of Aristotles works, Paul
Moraux slips in supposing that Aristotles celebrated hymn
to Hermias would have been inserted somewhere among these
collections.
30
This supposition is excluded by Diogenes genre
terms: certainly the epigramonHermias or the poemto Eudemus
could have been placed in the elegiacs, but the song to Hermias
is formally excluded from that book and from the hexameters
as well. Another list of Aristotles works contains a likelier
rubric under which to place the hymn: the anonymous catalogue
appended to the Hesychian life of Aristotle includes an item that
reads, enkmia or humnoi. Difference (A 180:
yx te y` u
vot
cteo, e). Moraux understands difference (cteo, e) as a gloss,
differentia, taking it as a readers reminder that, as a book on
Endurance 171
distinguishing the meanings of words put it around 100 CE, the
hymn differs [cte ,t] from the enkmion: for the hymn is for
gods and the enkmion for men.
31
The song to Hermias would t
well either as a hymnor enkmion, andits receptionhistory makes
the note about the difference between the two all the more
relevant. But it may be that Aristotle simply did not compose
enough lyric poems to tempt later editors to collect them in a
papyrus roll, or that the ones he did compose did not survive in
sufcient quantity. There is in any case no trace of a book of mel
among his output. It is a long shot but perhaps worth remarking
in conclusion that the song to Hermias might possibly lie under
an otherwise unknown work On Virtue that is reported, out of
alphabetical order and in the context of sympotic literature, as
no. A 163 in the Hesychian list.
By obscure routes such as these, Aristotles song became
available to the scholars and antiquarians of the Roman Empire.
In the end, it was a translation from performance piece into
text that kept the song alive, but in reading it I have found it
helpful to keep in mind that Didymus and company were readers
and writers, while Aristotle and his company were singers and
composers as well as writers. We know the song now not from
hearing it intoned by learned gentlemen at the end of a sociable
gathering but because its words were copied into books, much
as Didymus copied out a passage fromCallisthenes or Athenaeus
preserved the old Attic skolia. Once singing is replaced by writing,
scholarship must take up the tasks of friendship if the song is to
survive.
This may be why I nd a passing comment that Didymus
makes so poignant. In the course of recounting laffaire Hermias,
Didymus mentions the song and says it seems worth recording
here since there are not many copies to hand.
32
Despite
Aristotles brave hope for his songs endurance, it was becoming
rare even in the greatest center of learning in the ancient world.
Aristotle had shaped his song so that any time it was read,
performed, or brought up in discussion, Hermias was repaid once
againfor his friendship, repaid and more thanrepaid inthe lavish
way of the large-minded Greek. But eventually, the community
for which the song was composed died out, and the duty of
reciprocating friendship fell to scholars and antiquarian readers.
The music that had accompanied the song is long gone, its chief
residuum being a pattern of long and short vowels in the text;
the turbulent social circumstances in which it was performed
and received have left some marks on the language, but mostly
in the form of ghostly presences. The song has jettisoned a great
deal in eking out its path to the twenty-rst century, but even at
our great historical distance we have been able to notice certain
powers that Aristotle built into its words from the rst: to draw
upon and compress the vast resources of an integrated poetic
tradition and to focus themon a newevent, giving it meaning for
the community called together by that song. At the same time,
Aristotle composed a lyric able to respond to new circumstances
with fresh appeal and so to renew its promise that, through the
steadfastness of friends, excellence may nd a reward that lasts
through time.
N OT E S
Notes to Preface
1. Quoting the version given by Aelian in his (3rd-century CE) Various
History, 3.36. The reports on Aristotles parting words are collected in
the invaluable work of Dring 1957, T 44a-e.
2. Martindale (2005) urges specicity to the object as the way to make
criticism literary, quoting (170) a memorable sentence from Paters
preface to The Renaissance: To dene beauty, not in the most abstract
but in the most concrete terms possible, to nd not its universal
formula, but the formula which expresses most adequately this or that
special manifestation of it, is the aim of the true student of aesthetics.
Chapter 1
1. Aristotle Fr. 675 Rose/5A Gigon (= 842 PMG, which Davies is in the
process of re-doing as PMGF). Page is the basis of the most scholarly
English translation, by David Campbell in the Loeb Library Greek
Lyric, vol. 5, 214217. Notable editions include Wilamowitz (1892,
2.406), Macher (1914, 21), Wormell (1935, 6263), Plezia (1977, 45),
Marcovich (2009, 308309); others cited at Harding (2006, 154). The
textual questions are relatively minor and will be addressed as they
173
174 Notes
come up. In general, see Renehan (1982), Dorandi (2007), and among
earlier studies, Gerke (1902) and Dring (1957, 5960).
2. P. Berol. 9780 (ed. Diels-Shubart), col. 6.2243. I cite the papyrus
from the edition of Pearson and Stephens (1983).
3. Slightly different colometries are conceivable; see, for example,
Rutherford (2001, 9293), Bermer and Furley (2001, 222), and Dorandi
(2007).
4. West (1982, 139).
5. Notable studies of the poem are: Wilamowitz (1893, 2.403412);
Smyth (1906, 468469 and xxxviii); Wormell (1935, 6165); Bowra
(1938/1953); Jaeger (1948, 117121); Renehan (1982); Guthrie (1981,
2636, 44); andmore recently Santoni (1999); Furley andBremer (2001,
vol. 1.224227, 263266; vol. 2.221228 [commentary]).
6. See Depew (1997) and the fundamental study of Norden (1913,
143176).
7. Russell and Wilson (1981, xvi, drawing on Aristotles Rhetoric
(1.9, 1366a2368a37); similarly, Eudemian Ethics 2.1, 1220a2934. The
article on aret in LSJ sufces to showthat it can be used for goodness
or excellence of any kind (e.g., one can speak of the aret of productive
land), but inpeople it especially refers to manly qualities inits rst attes-
tations, as bets its root ar-, as in e
,oyv, male or masculine; only

later inthe history of the worddo we ndit associatedwithmoral virtue,
andappliedtothe gods andtheir wondrous works. The traditional trans-
lation virtue (descending from Latin virtus by way of vir, man) has
the merit of expressing the Romans awareness that aret encompassed
all the good qualities to which a real man (a vir) ought to aspire.
8. On the importance of reciprocity in Greek religious ideas, Parker
(1998).
9. See Race (1982), developing Bundy (1969).
10. Wilamowitz (1893, 2.405); Page (1981, 1994) calls the charge
absurd, Renehan (1982, 255) a sham, seconded by Furley and Bremer
(2001, 1.265).
Notes 175
Chapter 2
1. In framing my attitude to the historical value of poetic texts I take
inspiration from Mas (2000, 20) discussion of how to use epigraphical
evidence: a royal letter might speak of the subjects loyalty, the kings
benevolence and his benefactions. We do not know if the subjects were
truly loyal, nor whether the king was truly benevolent . . . we do not
even know if promised benefactions actually took place. What we can
say, in all such cases, is that the letter was written and received. That
the king did say these things to the recipients of his missive, that the
recipients, the citizens of that supremely articulate body, a Hellenistic
polis, inscribed the missive and perhaps produced a document of their
own: all these things in themselves are very real historical facts which
deserve attention.
2. Derrida (1974, 158 [originally 1967]); Derrida (1989, 873) returns
to the formula, observing that it can equally be used to demand
contextual readings: An internal reading will always be insufcient.
And moreover impossible. Question of context, as everyone knows,
there is nothing but context, and therefore: there is no outside-the-text
[il ny a pas de hors texte].
3. Beyond Didymus, Athenaeus, and Diogenes Laertius, Hermias is
mentioned by: Diodorus Siculus 16.52, Polyainos Strat. 6.48, Photius
Bibl. 279, p. 350; Himerius Or. (40Colonna); Anon. [Aristotle] Economica
29; Anon. Vita Aristotelis Westermann; Dionysius of Halicarnassus
Letter to Ammaeus 5.262 and the so-called Index of Academic Philosophers
now attributed to Philodemus (PHerc. 1021 col. 5.122; 1018 col. 2.7
Dorandi). The Suda lexicon names a Hermias as the author of a treatise
on the immortality of the soul (= T24 Dring, see his skeptical remarks,
p. 283).
4. For difculties in literary history, see Perkins (1992). It should
be clear that the historical contextualization I pursue has nothing to
do with the genetic fallacy, the dubious attempt to delineate the
set of historical circumstances that generated Aristotles song. This
ill-considered approach to interpretation rightly fell before the New
176 Notes
Criticism: see Wimsatt and Beardsley (1964) and cf. Wellek and Warren
(1956, 241260).
5. Apart from Theocritus of Chios, studied below, cf. Demetrius
of Magnesia apud DL 5.3 (= F 15 Mejer), who carried on the anti-
Hermias campaign in the rst century BCE (RE 4, col. 2814; Dring
[1957, 58]), and probably Theopompus as quoted by Didymus (at col.
4.69, where Pearson and Stephens [1983, 16.24] read Bithynian,
improving Jacoby 115 FGrH F 291). Most scholars infer fromHermiass
career that he was Greek, though possibly born in Bithynia: Harding
(2006, 128, 138), Flower (1994, 206208), Jaeger (1948, 112 n. 2),
Mulvaney (1926, 151).
6. Noticed by Penella (2007, 57) in an excellent exegesis. Using the
adjective entrophos (living in, reared in) as a noun (nursling) seems
to be poetic, and a suggestive precedent is in a lyric from Euripides
Iphigenia at Aulis: Ajax, born when his father had been exiled from
Aegina to Salamis, is styled nursling of Salamis (v. 288: At
e c o
\
Ye.e tvo
vx,oo). Given that Ajax is soon to be mentioned in
Aristotles poem, the gurative expression here for a complex paternity
may be signicant. (Aristotle mentions this play several times in his
Poetics, though Euripides latest editor, Diggle, suspects interpolation
in the passage under discussion.)
7. In his Nicomachean Ethics, Aristotle asserts that a man who spends
his time in the most virtuous activities will bear the chances of
life most nobly and altogether decorously, if he is truly good and
foursquare beyond reproach (NE 1.10, 1100b1921: xe`t x` e x ue
ot
ot x e..toxe xe`t evxy evx
. o
\
y
\
e .y0 e ye0
xe`t xx, eyvo e
vu yyou), quoting Simonides 542.1, 3 PMG.

A little later he adds that, when suffering chances to occur to such
men, their neness shines through, as when they bear with good
temper many great misfortunes not through insensitivity to pain but
through nobility and greatness of soul (NE 1.10, 1100b303: o
\
c` xe`t
v xo uxot cte. et x xe.v,
tc` ev ,y xt u x.
Notes 177
o..` e xe`t y e.e e xu te, ` y ct e ve.yyo tev e ..` e yvv ece ` v
xe`t ye.yuo, tr. Ross, modied).
8. Redeld (1974, x).
9. Lives of Hermias: Wormell (1935, 5592); P. von der Muhl,
RE Suppl. III, col. 11261130. Further bibliography in Trampedach
(1994, 1).
10. That Hermias murdered Eubulus is mentioned only by Demetrius
of Magnesia (apud DL 5.3 = F 15 Mejer). Cf. Ari. Pol. 1267b3137
and Strabo 13.1.57 C610. Rhodes and Osborne 2003 date Hermiass
taking over from Eubulus after the latters death to ca. 350, referring to
Diodorus Siculus 16.52.56.
11. See Weiskopf (1989, 4142).
12. Herodotus (1.160) says Cyrus gave Atarneus to the Chians as
a reward for their having surrendered Pactyes the Lydian to him.
Cf. Theopompuss Letter to Philip (115 FGrH F 250) and his account
of Hermiass execution from the Philippica (F 291) from which Flower
(1994, 8689) infers an attempt to appropriate Chian territories,
connecting it with Hermiass treaty with Smyrna, discussed below.
13. Notably Callisthenes FGrH124F2(discussedbelow) andTheophras-
tus, if he is the source of the interlude favorable to Hermias at Didymus
col. 5.5363 as argued by Milns (1994, 7881), seconded by Bollanse
(2001, 9495) and Harding (2006, 139).
14. See Owen (1983).
15. Aristocles apud Euseb. Praep. Ev. 15.2.9 Dindorf (= Fragment 2.9 in
Chiesara 2004, superseding the edition of Heiland [1925]).
16. Cf. Trampedach (1994, 6667).
17. GHI 165 Tod; I cite the updated text of Engelmann and Merkelbach
1972, no. 9, Vol. 1. 5660 (= no. 68 in Rhodes and Osborne 2003,
3425).
18. Didymus col. 5.5763; Philodemus Index Acad. 5.211 Dorandi.
19. Hammond and Grifth (1979, 158160) compare the companions
of Philip and Alexander (as referred to in, e.g., Aeschines Against
178 Notes
Ctesiphon 89), the Macedonian nobles who served as the kings guard.
Cf. J. R. Ellis in The Cambridge Ancient History, 2nd ed., ed. D. M. Lewis
et al. (Cambridge 1994) 770771. A Pindaric example of hetairos used
this way is Pythian 5.26, for the Dorian autocrat Arkesilaus of Cyrene;
cf. Currie 2005, 251.
20. As an example, Grifth in Hammond and Grifth (1979, 518)
introduces Hermias thus: this interesting man, combining practical and
political skills with (later in life) a willingness to listen to intellectuals
and support them, and even probably put some political theory into
practice.
21. See, e.g., Jaeger (1948, 110111); cf. Lynch (1972, 73). On
Aristotles feuding with the Academy, see Aristocles F 2.12 Chiesara
(=T 58j Dring) and, on Speusippuss election, Lynch (1972, 60).
22. Redeld (n.d.) points out that the Academy is one of the rst
ongoing secular corporations we knowof that was not based on kinship
or residence.
23. Dring (1957, 459) and Chroust (1967) put Aristotles departure
before Platos death. Proponents of the political interpretation may
point to the testimony of Euboulides, a contemporary rival of Aristotle
of the Megarian school (DL 2.168): his attacking Aristotle in a polemical
book because he was not in Athens when Plato died may imply an
earlier departure: cf. Dring (1957, 276 and earlier studies cited at 392
on T 58j). On Demochares reporting that Aristotle acted as a sycophant
after the fall of Olynthus, cf. Dring (1957, 388 on T 58g).
24. Dring (1957, T 58g); more on this speech in ch. 4 below.
25. Strabo 13.1.57 C610, pronounced surely wrong by Flower (1994,
206 n. 73); Owen (1983, 10) suggests the report stems froman attempt
by opponents of the Academy to associate it with tyrants; alternatively,
it may be that a text like Theopompus 115 FGrH F 250 on Hermiass
associating with Platonists was misconstrued. Platos Sixth letter (more
on which below) makes it clear from its tone and from what is said
(at 322E323A) that he has never met Hermias. The explanation of
Wormell (1938, 59), essentially that Hermias did visit the Academy
when Plato was out of town, is a stop-gap.
Notes 179
26. Nicanor therefore would have been named after his maternal grand-
father Nicomachus: Mulvaney (1926, 159), supported by Gottschalk
(1972, 322323) and Dring (1957, 271). The marriage to Pythias may
have occurred after Hermiass death: Gottschalk (1972, 322 n. 1); cf.
(Dring 1957, 267268), Mulvaney (1926, 155). On Pythias junior
and Nicanor in Aristotles will see DL 5.16 (= 4a Plezia [1977]);
for scholarship on this document, whose authenticity is defended in
Gottschalk (1972, 317), see Bollanse (1999a, 298 n. 1).
27. See Lynch (1972, 7072) for the little that is known about the
arrangements of this school at Assos. Apart from a general account in
Philodemus Index Acad. col. 5.211 Dorandi, we depend on a lacunose
list of names beginning at Didymus col. 5.52, and the only fellow
philosopher that can be condently be read is Erastos (5.534, to whom
I will returninthe nal chapter). Jaeger (1948, 115116 and n. 1) boldly
lled in names: neither Koriskos, Erastoss brother, nor Callisthenes (at
Assos according to DL 5.5, 10, 39) is improbable, but objections have
been raised to Jaegers using an unreliable passage fromStrabo (13.1.57
C610, cited above) to include Xenocrates of Chalcedon (destined to
succeed Speusippus as head of the Platonic Academy); Jaeger would
thereby make Aristotles departure from Athens a virtual secession
(111) from the Academy. For skepticism, see Milns (1994, 7273) and
Owen (1983, 410, esp. 7), challenging the restoration of Xenocrates
name.
28. Didymus col. 5.2127 Pearson and Stephens (= Theopompus 115
FGrH F 250); on the date, see Flower (1994, 86).
29. The distinction between Platonist Academics and Aristotelian
Peripatetics arose in later, not altogether clear circumstances: see
Lynch (1972, 7375) with testimonia in Dring (1957, 404411).
30. Flower (1994, 8889) suggests (if Aristotle went to Macedon in
342 and if Theopompuss Letter to Philip followed soon after) that
Theopompus was more worried about the growing inuence of the
Academy at the court of Philip than about the freedom of the Greeks
in Asia.
31. Chroust (1967, 1971).
180 Notes
32. Demosthenes Fourth Philippic 10.3132 with scholia. See Flower
(1994, 86 n. 60). Bosworth (1988, 18 n. 44) holds that this is pure
speculation on Demosthenes part. Various historical accounts positing
some cooperation between Hermias and Philip are surveyed in Harding
(2006, 124125).
33. Kahrstedt, Mentor (6), RE 15, col. 964965. On the account of
Weiskopf (1989, citing Demosthenes 23.154 ff. and Diodorus 16.52),
Artabazus had been installed as satrap of Dascyleum around 363, and
around this time married Mentors sister, thereby making her brother
an important ally.
34. The Fourth Philippic is usually placed between Hermiass capture and
his death: Dring (1957, 276). Most scholars now date Hermiass death
to 342/1: Trampgedach (1994, 6869). Rutherford (2001, 93) dates it
without argument to 345/4, reverting to an older tradition based on
Dionysus of Halicarnassus, Letter to Ammaeus 5.
Chapter 3
1. Diogenes Laertius 5.6 (= Aristotle Fr. 674 Rose). I give the text of
Page (1975, 622626). For discussions, cf. Page (1981, 3132), Wormell
(1935, 61), Dielh (1925, 1.46.3), Wilamowitz (1893, 2.403404, mainly
censuring Aristotles versication).
2. Herman (1987, 26, 129, where the reference to Nemean 7.9092
should be to vv. 6063).
3. Of course, in our broad use we follow a use of the term developed by
the Greek critics. See Ford (2002, ch. 6).
4. Cf. Young (1983, esp. 4042) on inscriptional pote as it applies to
the possibility of Pindaric re-performance.
5. A.P. 7.258 (= Page [1975, 879881]); cf. Page (1981, 268272). The
cowardice of bowmen is a theme that can be found from Homers
Iliad (see 11.385 ff.) through Euripides (Hercules Furens 157164);
cf. Lissarague (1990, 1334). The adjective xoo,o is not negative
when used in archaic and classical Greek poetry for Apollo or Artemis; it
apparently began to be used of Easterners in Herodotus (9.43, quoting
an oracle) and Simonides.
Notes 181
6. Gow and Page (1965, 2.546) note Theocrituss reply as an early
example. Cf. Fantuzzi in Fantuzzi and Hunter (2004, 283291). On the
passage of epigrams from stone to book see Gutzwiller (1998, ch. 3,
esp. pp. 4753), placing the rise of a distinctly epigrammatic aesthetic
toward the end of the fourth century and the beginning of the third
(52), Bing (2009, 116141). To remark the rise of book epigrams is
not to rule out continuing traditions of oral performance and even
impromptu composition: see Cameron (1995, 71103, esp. 7684),
arguing that many were composed for sympotic contexts; discussion
by Bing (2009, 106116, esp. 113115). On deixis in Greek funerary
inscriptions, see Tsagalis (2008, 216224).
7. Theocrituss riposte apart, Diogenes is our sole source for the verse,
though it is referred to in the pseudo-Aristotelian Apology quoted by
Athenaeus, discussed below, and by Himerius Or. 6.6 when he says for
Hermias alone of his friends Aristotle celebratedhis deathwithanelegy
(xe`t
.y t xv 0 evexov [so Penella, for 0 e.eov of the paradosis]

v x v yv, tv
xoyov).
8. My text is taken from SH 738, except that at v. 4, I read po,p,ou
with Diehl (1925, 1.110) rather than Bo,p,ou. For discussion, see
Wormell (1935, 7475), Dring (1957, 35, 381), Page (1981, 9395),
and Harding (2006, 157158) with further bibliography. Translation by
Flower (1994, 88, reading po,p,ou).
9. See Wormell (1935, 74 n. 32); Harding (2006, 22).
10. Page (1981, 31) copes with the evidence by assuming there was
both a statue at Delphi inscribed with the epigram and a cenotaph
(presumably at Assos or Atarneus) to which Theocritus refers.
11. Aristocles F 2.12 Chiesara (= T 58k Dring); Didymus col. 6.4649
(T 15h Dring) cites Bryons On Theocritus probably courtesy of
Hermippus. Plutarch Mor. 603C quotes only from the end of the third
line, Diogenes Laertius 5.11 only the rst two verses.
12. Flower (1994, 88); cf. Dring (1957, 277), Ruina (1986, 533534).
13. Page (1981, 95) cites Plato Phaedo 69C, where the lot of the
uninitiated is to lie in slime in Hades. Evidence of its popularity is
the exclamation of Dionysus in Aristophanes Frogs when he arrives in
182 Notes
the underworld and exclaims, no doubt peering out at the audience,
What a lot of muck and slime! (Et
xe p,po,ov o.` uv / xe`t ox ,

e tvv, 1401). Liapis, per litteras, points me to Asius of Samos, who
abuses a disgusting old parasite appearing at a feast, rising like a hero
from slime (1.4 IEG). He also notes that it is possible to take Borborou
as a proper name of a ctional river (Slime River).
14. The phrase is perhaps mock heroic, resembling a ctional epitaph
for the hero Memnon from a collection of literary epitaphs Aristotle
may have known (Fr. 641.62 Rose): I Memnon, son of Tithonus
and Dawn, here lie / in Syria by the outpourings of the river Belos
(M vv Tt0vo u x xe`t Ho u
v0 ec x tet /
v Yu, ty B y.ou ` e,
oxeo u ,ooe t).
15. Most scholars place the prosecution of Aristotle in the anti-
Macedonianatmosphere following Alexanders death; Rutherford(2001,
9293) would place it in 335, when Aristotle returned to Athens
and Alexander acceded to power, though he leaves open the pos-
sibility that it inuenced Aristotles nal departure from Athens
in 323.
16. See LSJ s. v. and note s. v. II: a synonym for prokhoos, a vessel for
pouring libations.
17. See, e.g., Wormell (1938, 75 n. 34), Ruina (1986, 532, 533).
18. An archetypal example is Odysseus in rags in the Odyssey: see Pucci
(1987, 165172).
19. Ruina (1986, citing Timaeus 73A and Phaedrus 238A); like Page, he
would include Aristotles alleged sexual impropriety in the charge; but
Xenophon Mem. 1.2.1 (cited by Ruina) makes a distinction between sex-
and gastr-greed: Socrates was the most temperate of men as far as
sex and the stomach went (Yx, exy . . . e ,octo tv xe`t yeox,
evxv e v0, v
yx,ex oxexo

y v).
20. Aristocles reading is also the basis for the version of Theocritus
in Michael Apostolius, a fteenth-century Paroemiagrapher (6.38a
Leutsch-Schneidewin). Didymus apparently read, o
\
` yeox, xt v
e
voov uotv t
\
.xo ve ttv, a milder expression.
Notes 183
21. Archelaus is also said to have lured Euripides to his court at the end
of his life, but for doubts on this tradition see Lefkowitz (1981, 103),
Scullion (2003).
22. Simonides actual social and economic situationwhich there is no
reasontodoubt may have beenremarkable inthe mobile sixthcentury
is too obscurely known; he truly is, for us, no more than a gure of
discourse, but one that is beautifully opened up in a study by J. M. Bell
(1986).
23. Cephisodorus, noted as a detractor of Aristotle in Athenaeus 354b,
is cited by Aristocles F 2.7 Chiesara (= T 58h Dring). Similar abuse in
Pseudo-Aristippus, fromthe rst book of the treatise On Ancient Luxury,
is mentioned below.
24. 115 FGrH F 210, 242; further tales are given in [Demetrius] On Style
293. Owen (1983, 1516) refers it to stereotype invective.
25. Mulvaney (1926, 155).
26. Fragments 61517 Rose. See Christesen (2007, 179190).
27. SIG
3
275 (= 187 Tod = Callisthenes FGrH 124 T 23 = 80 Rhodes
and Osborne). See Tod (1985, 2.246248, esp. 248), and Rhodes
and Osborne (2003, 395), who date the inscription to 337327 BCE.
A similar instance of iconoclasm occurred in the wake of Philips
aggression in 340, which provoked the Athenians to smash the stone on
whichtheir recent treaty withMacedonhad beeninscribed (Philochorus
FGrH 328 F 55A and 55B; cf. Diodorus Siculus 16.77.2).
Chapter 4
1. Wilamowitz (1893, 2.405 with n. 3).
2. Wormell (1935, 76); followed by Dring (1957, 277). The tradition
of Callisthenes is questioned by Bosworth (1970), but hypercritically:
Lynch (1972, 72), Fox (1986, 112 n. 55, citing Chares 125 FGrH T 15).
3. W. Crnerts
yx tov in Rheinisches Museum 62 (1907), 383 is

printed by, inter alios, Jacoby in FGrH 124 F 2, Wormel (1935, 75) and
Dring (1957, 274). It is defended by Harding (2006, 144145, noting
184 Notes
that Crnert was anticipated by Blass in Archiv fr Papyrusforschung 3
[1906] 290) on the grounds that it ts a tendency elsewhere in Didymus
to dene works by their genres.
4. On prose encomia, see Velardi (1991); on fourth-century experi-
ments in prose, Ford (2008).
5. So Evagoras begins (1:
\
O, v,
Ntx ox.t, xt vx e o x` ov x eov

xo u ex, o). On encomiazing the excellence of a man in prose, see
the commentary by Nicolai (2004, 8890).
6. Momigliano (1993/1971, 82) and Russell and Wilson (1981, xv) on
Agesilaus. One may suggest that, as if in compensation for giving up
stirring poetic devices, prose eulogists ventured some rather extreme
topoi of praise, such as Isocrates declaration in Evag. 72 that, If anyone
was ever a god among men, Evagoras deserves that title. The one bit
of content from Speusippuss work that we have is also religiously bold,
claiming that Platos real father was Apollo: cf. DL 3.2, also mentioning
an Encomium of Plato by Clearchus (2a Wehrli).
7. Didymus col. 5.646.18 (FGrH 124 F 2). Supplements are generally
as in Pearson and Stephens, but for legibilitys sake I have thought
it safe to follow Hardings readings (2006) at lines: 5.69 (reverting to
Diels-Schubarts eu x t x t 0ev ext for eu x[ t x t ctvox ex]t of
P-S), 5.71 (e, for Krtes , t), 6.8 (ctx ev Harding), and 6.9 (e,
eu x t). On Pearson-Stephens see the discussion by Rusten (1987).
8. Jacoby, Komm. to 124 F 2, p. 416.
9. Didymus col. 6.5059 reports the controversy, citing (at col.
6.5153) Hermippuss On Aristotle apropos the tradition that Hermias
died in prison (44 Wehrli = FGrH 1026 F 31); cf. Bollans ee (2001,
8389). Flower (1994, 209) nds it likely that Callisthenes, the eulogist,
euphemizes. See further Rusten (1987, 268) and Harding (2006, 151).
The Athenian Constitution that was produced in Aristotles school has in
ch. 18a grisly account of the torture anddeathof the heroic Aristogeiton
by Hippias, the tyrant he had been conspiring against.
10. Rusten (1987, 268269); Harding (2006, 152). Jaeger (1948, 117)
suggests Hermias summoned Artaxerxes himself to be his witness.
Notes 185
11. In an anthropological account integrating Achaemenid practices of
torture into their religious, legal, and political ideals, Lincoln (2007,
9091) notes that a show of benecence on the part of the torturers
could be a signicant part of the spectacle.
12. On the terms, see Ford (1992, 13).
13. Cf. Furley and Bremer (2001, 2.228): The very fact that Aristotle
thought this allusive naming [sc. nursling of Atarneus] to be sufcient,
is ample proof that he composed this poemfor private use only (though
one may query their only).
14. On mimetic hymns, Hopkinson (1984, 3243), Depew (1997).
15. Guthrie (1981, 44) thinks the detail that the song to Hermeias
was sung daily at dinner was added for the sake of artistic verisimil-
itude. But while daily may be an exaggeration, performance in the
Lyceum was apt to have been seized on by the prosecution as an
inammatory detail, suggesting that Aristotles school was a secret,
pro-Macedonian cabal. It plausible in itself that Aristotle would have
had the song performed at common meals (sussitia) he took care to
provide for the Lyceum. (Athenaeus lists Aristotle among philosophers
who held symposia: 186b, cf. Lynch [1972, 112]). In his Politics he
recommended sussitia as something a well-run city ought to support:
7.10 (1330a1315); cf. 7.11.1 (1331a19 ff.) and Ford (2006, 313).
By the time of Theophrastuss successor, Strato, the school property
included furniture for sussitia, couch-covers, and cups (as reported
by DL 5.62: xexe. t c eu x xe`t x` e ptp. te evxe, .` yv
\
v eu xo`t
yy, eev, xe`t x` eox uy evxexex` ex` oouoo txtov xe`t x` eox, exe
xe`t x` e ox y,te).
16. Cited in the scholium on Plato Gorgias 451E, quoted below. On the
etymological question: Teodorsson (1989) and Liapis (1996).
17. Athen. 694b-c: xyvtxe uxe y` e, x v oo v
\
xeoxov
c yv xtve
xe.` yv t oov to ,tv y
touv. xe.` yv c` xe uxyv

.yov
\

e,e tvotv xe`t yv yv
ouoev x p t ,yo tyv.
18. On paeanic idiom, Rutherford (2001, 94 n. 10).
186 Notes
19. Making the refrain obligatory: [Plut.] De Musica 1134d-e, with
Harvey (1955, 172173).
20. On the use/absence of refrains in paeans, see Kppel (1992, 1013,
6570, 8485, with his chart on 6667); Rutherford (2001, 6883, 96).
21. Evidence may appear in the transmission of Aristophanes Wasps
where the strophe of a hymn petitioning Apollos help ends in an extra
metrum add-on, t yt Het ev (874), but the antistrophe (890) does not.
Cf. Willi (2003, 4547) and Ford (2006, 286268).
22. Duris FGrH 76 F 71 and F 26; see Habicht (1970, 36 and 2434)
for epigraphical evidence that a Lysandreia continued to be celebrated
in the fourth century.
23. Alexinus F91Dring (=40SH). Pace Jacoby, Krateros, RE XI 2, col.
1617 who takes the Craterus here as the son of the general (321250).
Bollans ees (2001, 73n. 17) compendious discussionincludes suggestive
speculation about the possibility of festival performances of the
paean for Crateros at Apollos site, Delphi. This musical show of
magnicence was matched with one in stone by Craterus the younger,
who commissioned the famous statuary Lysippus to represent his
father and Alexander in a lion hunt (despite Plutarch Alexander 40.45,
who attributes the dedication to the elder Craterus). The work was,
like Hermiass cenotaph, dedicated at Delphi, and its inuence did
not stop there: the image seems to have been reproduced back in
Macedonia in a mosaic found in Pella. See discussion in Bollans ee.
Another notable contemporary case of memorialization in speech and
stone is by Mausolus of Caria, who in 353/2 took occupancy of the so-
called Mausoleum, one of the Seven Wonders of the Ancient World. His
funeral was graced by a contest in prose eulogy among three orators of
Isocrates schoolTheopompus, Theodectes, and Naucrates.
24. H,`t .u,tx v otyx v. See Schmidt (1854, 386396). On
Athenaeuss sources here, cf. Wilamowitz (1893, 2.403), Bowra (1938,
182), Wehrli (1974, 75). Hermippuss fragments are now available in
the superb continuation of Jacobys Die Fragmente der Griechischen
Historiker as 1026 in Bollans ee (2001), FGrHist IV A 3.
Notes 187
25. On Aristotle is FGrH 1026 F 2833; Athenaeus 696a-697b is given
as Hermippus FGrH1026F30(=Fr 48Wehrli =Test. 7Kppel), though
Bolans ee (2001, 7475) concludes that Hermippus was not Athenaeuss
sole source.
26. Hermippus has often been identied as Didymuss chief source,
perhaps through intermediaries: Diels-Schubart (1904, xxxvixliii,
esp. xxxviii), Wormell (1935, 7882); Dring (1957, 275). Reservations
in Wehrli (1974, 7576), Yunis (1997, 1052), Bollans ee (1999b, 6465
with n. 135 for bibliography), and Harding (2006, 35, 3738).
27. Dring (1957, 79) calls Hermippus a main source for Diogenes,
but Mejer (1978, 3234) cautions about oversimplifying the tradition.
Against the assumption of Wilamowitz (1893, 2.403 with n. 1) that,
what is common to Diogenes Laertius and Athenaeus goes back to
Hermippus see Bollans ee (1999b, 6569 and 56 n 1).
28. On Hermippuss reputation: Diels in Diels-Schubart (1904, xxxviii
xli), Heibges in RE 8, Wehrli (1974, 75), Bollans ee (1999b, xiixv, 54).
29. Bollans ee (1999a, 312); critiques of Hermippus in Wehrli (1974, 75)
and Momigliano (1993, 79, 114) are possibly overdone: Mejer (1978,
32 n. 1).
30. Derenne (1930, 198). We knowof Socratic Ao.oy tet composed by
Plato, Xenophon, Crito, Lysias, Theodectes, and Demetrius of Phalerum
(an associate of Aristotles); a version of the Accusation of Socrates was
produced by the rhetorician Polycrates: Ford (2008, 3233).
31. o e is Kaibels supplement to the text of Athenaeus, which gives
xoo yoexo (A)/
x ooouv (C, E); without an object for the verb the

grammar is a little rough, and o e (better than Drings [1957] 281
x ooyoe eu x ov) reinforces clausular balance (
x ooyoe x` o o e
recalls, chiastically, the end of the rst clause v ye xexox ueov)
and structures its own clause with a uotv /o e opposition.
32. DL 5.5, just before citing Favorinus on Demophilus, on which see
below. Cf. Dring (1957, 374), Bollans ee (2001, 6970 n. 6).
33. Aristocles F 2.8 Chiesara (=T 58i Dring, possibly fromHermippus:
Dring [1957, 464]). On the tale, cf. Wormell (1935, 87) and Mulvaney
188 Notes
(1926, 156157), who suggests it may contain some truth. It may
be relevant that in his will Aristotle provided for a Demeter to be
dedicated on his mothers behalf at Nemea (DL 5.16.12 = 4b Plezia
[1977]). On Lycon of Iasos, a rather obscure gure, see 1110 FGrH
F 1 (= 57 T 4 DK) and W. Capelle, Lykon (15) in RE s. v. col. 2308
2309. Onthe complicatedtestimonia about Aristotles marital relations,
Bollans ee (1999a, 298304).
34. Athenaeus may be attempting to reconcile varying traditions. In
his version (696b), the hierophant Eurymedon stirred the politician
Demophilus to act. Diogenes Laertius says (5.5) that Aristotle was
indicted for impiety by Eurymedon, on account of both the paean
and the inscription, but adds that Favorinus in his history named
Demophilus as the indicter (= F 68 Barrigazzi / 36 Mensching). We
do not know whether (a) Diogenes and Athenaeus both used Favorinus
or (b) whether Favorinus and Athenaeus used the same source: Dring
(1957, 279). Clinton (1974, 21) notes that a Eurymedon is recorded as
chairman of the hieropoioi of the Eleusinian Boul in 329/8. Demophilus
may be the one named by Plutarch (Phocion 38.1) as one of the accusers
who in 318 brought about the execution of Phocion, an Athenian
associate of Demetrius of Phaleron, for advising accommodation
with Macedon. In general, the attention to the teamwork between
Eurymedon and Demophilus recalls the use of Anytus and Meletus
in Socratic literature to represent different constituencies within the
prosecution.
35. DL 5.34 (= T 10e Dring): A, toxto c
v x , x H,`t
e.et e x,u y yotv
,eo0 yvet x` ov A,toxox .yv e..ex tco

xo u
\
E, tou. xo u c` ouy, yoevxo
yy x eu x` yv xe`t
0uv
u
\
,e t,v x yuve t,
\
A0yve tot x y E.uotv te L yyx,t x
x
\
E, te et eve
y,eyv, o
\
`
vcov y y,exet. I take it that
Diogenes paraphrase extends through
y,eyv at least, because of
the close way that x x
\
E, te is linked to x yuve t. That is, I take
it that pseudo-Aristippus mentioned Hermias as well as the sacrice.
At the end of the quoted block, it is safer to attribute which is written
out below to Diogenes (since he actually goes on to transcribe the text)
than to Aristippus, but the possibility that he quoted it as well cannot
Notes 189
be excluded. On the dating of pseudo-Aristippus see Wormell (1935, 85
n. 59); cf. Dring (1957, 268).
36. Something like the reconstruction attempted in this paragraph
must be, as far as I can see, the basis of Wilamowitzs summary
declaration (1893, 2.403): Before Hermippus, the poem had been
noticed in the defense speech of Aristotle that he himself suspected
(Ath. 979a), in pseudo-Aristippus (Diog. 5.4), and perhaps in Lycon
the Pythagorean from Iasos . . . from which one can suppose that
the scoundrel who charged Aristotle with impiety had not so much
abused the poems as the fact of their existence. (Vor Hermippos
hatten die gedichte bercksichtigung gefunden in der von diesemselbst
bezweifelten verteidigunsrede des Aristoteles (Ath. 697a), bei dem
falschen Aristippos (Diog. 5.4) und vielleicht dem Pythagoreer Lykon
von Iasos . . . denen man wohl so viel glauben kann, dass der schurke,
der den Aristoteles wegen religionsfrevels belangte, nicht sowohl die
gedichte als die tatsache ihrer existenz misgebraucht hatte.) I differ
fromWilamowitz only in his nal point, for the passage fromAeschines
Against Timarchus 135137, discussed below, shows that the actual
words of a defendants poetry could have been thrown in his face during
a political wrangle.
37. Rose (1886, 599). See Wormell (1935, 86 n. 62).
38. Aristocles F 2.5 Chiesara (= T 58f Dring).
39. Mulvaney 1926, 156, adopted by Dring (F 60) and Dring (1957,
388) with n. 124; Vayos Liapis achieves the same result more simply by
deleting
\
as dittography (per litteras). Wormell (1935, 86) considers
it probable that Euboulides refers to forged incriminating evidence
that was produced in reaction to laffaire Hermias.
40. As noted by Chiesara (2001, 72).
41. Wehrli (1974, 76). Wehrli also thinks one may doubt whether
Hermippus took over the entire text of the song in his work, see
below.
42. So Dring (1957, 343344), citing Derenne (1930, 200). Cf.
OSullivan (1997, 136139).
190 Notes
43. As Lord (1986) notes, citing: Diodorus Siculus 20.45.25, DL 5.78
79, Strabo 9.1.20; skeptical are Pfeiffer (1968, 87104) and Bagnall
(2004, 349351).
44. DL 5.38. More on Sophocles law: Athenaeus 610b-f (including
a telling fragment from the comic poet Alexis); Pollux 9.42 (3689
Bekker). Cf. Regenbogen RE Suppl. 7, col. 1360.
45. Aristocles F 2.6 Chiesara (= T 58g Dring). On Demochares, cf.
Athenaeus 215c, 508b, 610f. On the speech: Dring (1957, 343344,
388); Lynch (1972, 103104, 117118); Chroust (1973, 1.149151);
Habicht (1988); Brunt (1993, 332334); Bollans ee (1999a, 505 n. 221).
Chapter 5
1. The charge was, in effect, as Kevin Clinton (1974, 21) puts it
worshipping in public a god whose cult was not ofcially authorized by
the state, citing Rudhardt (1960, 9293); cf. Parker (1996, 214217,
276277).
2. This is a main theme of Ford (2002).
3. Harvey (1955). My sketch is also indebted to the ne account of Lowe
(2007, 167176).
4. See Heaths 1985 very rich study of receiving the kmos in epinician.
Pindar Nem. 4.78 (
tvtx totov e otce t) and Bacchylides Ode 2.13 are

likely to have provided the scholars with the idea for the technical term.
Note that at Ol. 8.75 veoo uvev . . .
tv txtov refers in the rst

instance to the memory of victory that Pindars song will arouse.
5. Harvey (1955, esp. 162164) on skolion and enkmion; Lowe (2007,
169). An early example of an enkmion in prose is Platos Symposium,
where the littrateur Phaedrus implicitly contrasts enkmia with songs
for divinities: we have numerous hymns and paeans composed by
poets to the other gods, but for Eros we have not even an encomium
(177A-B). The underlying distinction between songs for gods and songs
for mortals is elaborated in Laws 700A-E, discussed below.
6. Cf. Eudemian Ethics 2.1, 1219b8ff., esp 1219b89:
xt c ot
\
etvot
y e ,x y ct ` e x e
,ye, xe`t x` e
yx te x v
,yv.
Notes 191
7. On the canonical Alexandrian editions of Pindar (by Callimachus,
Aristophanes of Byzantium and his predecessors including Apollonius
the eidographer), Rutherford (2001, 152158).
8. Though the paean was traditionally associated with Apollo and his
sister Artemis, from the classical period one nds paeans generalized
to other gods, as Sophocles paean to Sleep in Philoctetes 827832. See
Kppel (1992, 341349), Schrder (1999, 2231), Ford (2006, 285).
9. Laws 700C: (pex uovx xe`t e..ov xo u c ovxo xexvot
u
\
y
\
cov y, ` o
\
o `
bo . . . xe`t evxe t evxe ouv eyovx. For a discussion,
see Kppel (1992, 3638) and Ford (2002, 258260). Nagy (1990,
109110) insightfully places such responses in the context of the
infusion of Panhellenic lyric genres into the classical Athenian theater.
For recent perspectives on the musical revolution see, e.g., Csapo
(2004) and LeVen (2008), with bibliography.
10. For the interplay between cult occasions and song types, cf. the
anecdote (noted by Chroust [1973, 146 n. 7]) about Xenophanes of
Colophon cited by Aristotle in the Rhetoric (2.23, 1400b58): when
asked by the Eleans whether or not they should sacrice and sing
dirges to their Leucothea, he advised them that if they supposed her
a god not to sing a dirge, and if they supposed her mortal, not to offer
sacrice. (o
t
\
ov vo evy E. exet
,x otv t 0 uot x y Auxo0 e

xe`t 0,yv otv y` y, ouvpo u.uv, t ` v 0v u
\
o.ep evouotv, ` y
0,yv tv, t c e
v0,ov, ` y 0 utv).
11. Poetics 1448b2527: ot
\
` v y` e, ovx,ot x` e xe.` e ` to uvxo
, et xe`t x` e x v xoto uxv, ot
\
c` u x. ox,ot x` e x v e u.v,
, xov yyou oto uvx,
\
o,
\
x,ot u
\
vou xe`t
yx te.
12. Aristotle uses paian as a metrical term in the Rhetoric (3.8), where it
seems to be a neologism, for he says that orators began using the paean
from the time of Thrasymachus, but without being able to say what its
name was (so I take1409a1-3: . txet c` et ev,
\

, vxo ` v e
O,eou eou e , evot, ou x
tov c` . ytv x t

y v).
13. Dichaearchus On Musical Competitions Frr. 8889 Wehrli; cf.
Reitzenstein (1883, 5), Rutherford (2001, 51), Aristoxenus Fr. 125
192 Notes
Wehrli; cf. Reitzenstein (1893, 313, 16 n. 28), Frber (1936, Pt. I,
5763), Harvey (1955, 162163, 174).
14. On Aristotles library, see Dziatzko (1899, 408409); testimonies in
Dring (1957, T 42a-d and T 66), esp. Strabo 13.1.54 C608 and Plutarch
Sulla 26 with Athenaeus 3a-b.
15. On Sappho in Alexandria, Yatromanolakis 1999 and Ferrari (2010,
116-123) on the epithalamia.
16. See Lowe 2007 for details. Simonides epinicians may have been
classed primarily by event, as Lobel inferred from P. Oxy. 2431.
17. Barthes (1988, 85).
18. Proclus apud Photius Bib. cod. 239.319b ff. Proclus to some extent
relied on Didymus: cf. Schmidt (1854, 390), Severyns (1938, 114),
Rutherford (2001, 105107 n. 39). Russell and Wilson (1981, 227228)
compare the very similar (though not identical) list of hymnic genres
that is found (possibly interpolated) in Menander Rhetor, I: Division,
331.20332.7.
19. On the persistent hymns-gods/encomia-mortals distinction in the
rhetorical tradition, cf. R. Wnsch, Hymnos RE9.1col. 181.28182.52;
on Didymus, see 181.64ff. Lowe (2007, 172) notes that Procluss
distinction between prosodia sung in procession and humnoi per-
formed while standing (as, e.g., around an altar) may derive from
Didymuss use of the mobile/static opposition to distinguish prosodia
and humnoi.
20. Pfeiffer (1968, 127134), Bagnall (2004, 356, n. 36). Moraux (1951,
221222) argues that Hermippus relied on the Pinakes. Bollanse
(1999b, 114) stresses Hermippuss association with Callimachus over
the tradition that would make him a (distant) student of Aristotles.
21. Dithyramb Fr. 23 Maehler, who associates the song with a
Bacchylideandithyramb mentionedby the scholiumonPindar Pythian
1.100; cf. Porphyry on Horace Odes 1.15.
22. P. Oxy. 2368 (= Bacchylides p. 128 Snell-Maehler = Test. 3 Kppel
1992): Aristarchus says this song is dithyrambic because it includes the
story of Cassandra, and titles it Cassandra; and he says that Callimachus
Notes 193
erred in ranging it among the Paeans because he did not understand
that the refrain is shared between dithyrambs [sc. and paeans]. I give
the text as in Callimachus Fr. 293 SH: xe uxyv x] ` yv
tc` yv A, toxe,
(o) [. . . . . ct0]u,eptx` yv
t [ve t yot]v ct ` e x e,t. y[0et
v
e]u x y x` e ,`t Keo[o evc,e,]
ty, et c eu x` yv [ . . . K eoo]evc,ev,
.evy[0 vxe c eu x` yv xexex eet [
v xo t H]et eot Ke.. teov

[. . . . . . .] ou ouv vxe o
\
xt [. . t0]
.
{y
.
}e xotvv
[oxt xe`t c]t0u, epou

xx.. Rutherford (2001, 9798) supports Lobels restoration, as does the
logic of the argument: see Kppel and Kannicht (1988) on the text and,
on the debate generally, Kppel (1992, 3842), contra Schrder (1999,
110119). Kppel (2000) discusses Platos obiter dictum in Rep. 394B-C
that associates narrative and dithyramb.
23. Jaeger (1948, 117).
24. For Simonides, see Ford(2002, 110111and128130) onSimonides
531 PMG; see Fantuzzi (2001) and Boedekker (2001) on the new
Plataea elegy of Simonides (11 IEG), which pushes the tradition by
making the recently dead Greek soldiers equivalent to Homeric heroes.
The most important passage of that song for present purposes is
its transitional invocation to the Muses in which Simonides, having
recounted the death of Achilles and mentioned its glorication by
Homer, turns to those who showed valor (aret) in the battle of
Plataea and prays for help in making their fame immortal (11.28
IEG). For Pericles, see Thucydides 2.43 and Stesimbrotos of Thasos, a
contemporary, who recalled him saying in an encomium for those who
had died in Samos that they had become immortal like the gods: for we
do not see these but reckonthat they are immortal fromthe honors they
receive and the benets they bestow (107 FGrH F 9 = Plut. Pericles 8.5:
ou c` y` e,
x tvou eu xo` u o
\
, v, e ..` e xe t xte t e
\
`
ouot xe`t
xo t e ye0o t e
\
` e, ouotv e 0ev exou
t vet xxet,0e).
25. Momigliano (1987, 97). Ruler cult appears not to be directly
dependent from hero cult, which is attested for living gures in the
fth century: e.g., the cult for Brasidas in 422 at Amphipolis; cf.
Cleomedes, the mad boxer who was ca. 500 given cult at Astypulaea
after disappearing in a homicidal bout.
194 Notes
26. Momigliano (1987, 94), quoting Simon Price. Cf. Habicht (1970,
1725, 243, 245). On Philip, see Fredricksmeyer (1979). Ma (2000,
219226) carries the analysis further with a case study fromHellenistic
Teos.
27. See Bosworth (1977).
28. Arrian, Anabasis 4.1012, 1314. See Rhodes and Osborne (2003,
395) who put it in the context of the conspiracy of pages in 327.
29. Aelian, Various History 2.19, cf. Athenaeus 251b. On the divinity
of Alexander, a trenchant discussion is Bosworth (1988, 278290);
cf. Hammond and Wallbank (1988, 82). It is controversial whether
Alexander actually, as was said, wrote to the Greeks demanding divine
cult or merely let it be known that he would regard such gestures on the
part of Greek cities with favor: see, e.g., Badian (1981), Fredricksmeyer
(1981), Cawkwell (1994), and Bosworths discussion (1988, 288).
30. Athenaeus 251b, Aelian Various History 5.12. The charge may have
been a graph paranomos: see OSullivan (1997, 138139), Derenne
(1930, 185188; to be used with caution). On such trials generally,
which go back to the fth century, see Parker (1996, 206207).
31. The quotation is fromHermocles ithyphallic processional: Powell,
CA pp. 173175, quoted by Athenaeus 253d-f; cf. 697a citing Philocho-
rus (328 FGrH 165) on a contest in composing paeans for Demetrius
and his father. On poems of this kind (e.g. SH 491, 492), cf. Mikalson
(1998, ch. 3), Gelzer (1993). Michael Flower points me to a powerful
statement of Athenian resentment of such practices from around 322.
Hyperides Funeral Oration deplores Macedonian arrogance for having
imposed on Greece such spectacles as sacrices being performed for
men, the statues and altars and temples of the gods being neglected
while those for men are carefully established, and we are forced to
honor the servants of these men as heroes (6.21, the nal reference
being to Hephaestion, Alexanders comrade who died in 324 and for
whom Alexander instituted heroic cult).
32. Currie (2005, esp. 159172) is a rich study that succeeds at least in
showing that Pindars epinicians often raise the idea of heroic honors
Notes 195
as a reward to victors, even if actual cult for living athletes remained
exceptional.
33. Duris of Samos FGrH 76 F 71 (=Plut. Lysander 18.24): , x
` v y e,,
\
t
\
oxo, t Lo u,t,
\
E.. yvv
x tv po` u et
\
.t
e v oxyoev
\
0 xe`t 0uo te
0uoev, t , xov c` et ev
y
o0yoev,
\
v
\
v e ,` yv e ovyov uouot xot evc. This evidence
has been subject to dispute, notably by Badian (1981), but see the
discussion of Flower (1988, 131133), and Habicht (1970, 244245,
271) for epigraphical evidence for a fourth-century Lysandreia. Plutarch
(Lysander 18.7) names a number of poets who composed such paeans
to mortalsfor lavish rewards at times: Antilochus, Antimachus, and
Niceratus the Heracleote, all in SH 51, 325, 565.
34. I take the verb as a short-vowel subjunctive, as commonly in hymnic
song; the meaning is not much different if we take it, also in conformity
with hymnic use, as a performative future whose resolve is being
fullled at the moment it is expressed: cf. Bundy (1969, 21).
35. Derenne (1930, 190) compares the bringing to trial of the courtesan
Phryne for violating the Eleusinian mysteries and of Demades in 324/3
for proposing that Alexander be made the thirteenth Olympian. See
Fredricksmeyer (1979, 5960), OSullivan (1997) and Worthington
(2004, 266267). Clinton (1974, 21) nds it more probable that
Eurymedon was acting from political or personal reasons than that
Aristotle in any way actually denigrated the Eleusinian cult. On the type
of religious prosecutor, see Gagn (2009).
36. Aeschines Against Timarchus 1357. For discussion, see Ford
(1999a, 251).
37. Demosthenes On the False Embassy 162163, cf. 128 and Rutherford
(2001, 67, 93). On the Crown 287 attacks Aeschines paean-singing
after Chaeronea. Rutherford (2001, 9293) supposes the prosecutions
case was that the hymn was a sympotic paean, and as part of the
preliminary rites belonged among the praises of the gods. The evidence
we have suggests that the procedures for holding a proper symposium
were adaptable to different households and occasions and so I doubt
196 Notes
that the prosecution would have based their charge on such a ne point
of party-planning.
38. Though Hermippus may have proposed that Aristotles song was
a skolion, its additional qualication as a skolion sui generis may be
Athenaeus: in the context of the Deipnosophists, the added point makes
sense as an explanation of why the song differs in meter and dialect
from the Attic skolia and the poem to Hybrias recited just before.
39. Cf. Wormell (1935, 85); Dring (1957, 58 on DL 5.4); Bollanse
(1999a, 312 n. 50).
40. Wilamowitz (1893, 405). So too Chroust (1973, 146).
41. Reitzenstein (1893, 42).
42. Smyth (1899, 469).
43. Jaeger (1934, 108). Cf. Derenne (1930, 192), Wormell (1935, 76),
Guthrie (1981, 33, 44), Bollanse (2001, 72 n. 16).
44. Extensively argued in LeVen (2008, ch. 4) and by Vayos Liapis
per litteras. Cf. Furley and Bremer (2001, I.265). While recognizing in
Aristotle many common epinician elements (ranging from dialect and
meter to the use of mythic exempla, themes of ponos, commemoration
in song, etc.), I hold the difference in postulated occasions is decisive:
Hermias has won no victory. (It would obviously be anachronistically
Christian to take Hermiass death as a triumph.) The similarities are due
to the fact that both songs are lyrics honoring mortals and so exploit
traditional praise forms. It is also worth noting that epinician in lyric
meter seems to have been a moribund formafter Pindar (apart fromthe
dactylo-epitritic epinicianto Alcibiades tentatively ascribed to Euripides
by Plurarch, Alibiades. 11.3 = 755 PMG); Callimachuss poetic notices of
victory are in elegiacs (the Victoria Berenicis, 254268C SH, and the
victory of Sosibios, Fr. 384 Pf.) or iambics (Iambi 8), on which see
Fuhrer (1993).
45. Renehan (1982, 254) quoting Harvey (1955, 173, cf. 153). Critiqued
by Kppel (1992, 17).
46. See Rutherford (2001, 95).
47. So Santoni (1991, 194195).
Notes 197
48. Santoni (1991).
49. Renehan (1982); cf. Bowra (1938), seconded by Dring (1957,
5758).
50. For a discussion of the critical price paid by uncompromising anti-
intentionalism, see Hinds (1998, esp. 4751).
Chapter 6
1. Schrder (1999) unpersuasively denies that Ariphrons song is a
paean, onthe grounds that it has more hymnische Sprechhaltung than
gebetshafte Sprechhaltung. Contra, see Kppel (1992, 68), Rutherford
(2001, 1923), Furley and Bremer (2001, 1.225).
2. Lucian, A Slip of the Tongue in Salutation 6.26: x yv,t xexov
x tvo xe`t eot ct ` e oxexo. For the sources of the text, see Wagman
(1995, 149159) and Furley and Bremer (2001, 1.224227, 2.175180).
3. The similarities were rst extensively noted by Wilamowitz (1893,
2.406) and examined by Bowra (1938). For commentary, Wagman
(1995, 160178).
4. Cf. West (1985, 76).
5. Hesiod, Works and Days 760763, much-discussed lines (e.g.,
Aeschines 1.127130, 2.144), see A. Rzachs Hesiodi Carmina, editio
maior (Leipzig, 1902). For personied Health on vases, cf. Shapiro
(1993, 125131), Furley and Bremer (2001, 1.225) and, on new gods,
Parker (1996, 152187, 227237). Simonides, who had helped prepare
the way for Aristotle in praising a personied Areta, also sang in praise
of health, though not necessarily personied: 604 PMG.
6. Foe , optoxo with the sense oldest, cf. its use as an epithet of
Eros as primordial deity in Platos Symposium178C-E and of Earth in the
Homeric Hymn to Gaia (both discussed below). For this reason I would
not atten out the epithet in Ariphron to august, as Wagman (1995,
173) suggests. A later example of the epithet with the sense oldest
is Mesomedes hailing The beginning and source of all, oldest Mother
of the universe (Fr. 35.2 CA): A,` e xe`t evxv y vve, / ,op toxe
Koou ex,).
198 Notes
7. As Aristotle says in Metaphysics (983b32: xtt xexov v y` e, x
,op uxexov). Euripides praises moderation (sophrosun) by declaring
that nothing is more venerable (presbuteron) than she (Fr. 959 Nauck
= Kannicht).
8. Furley and Bremer (2001, 1.47), tracing the tradition back to the
hymns to Anagk (Necessity) in Euripides Alcestis 9621005 and to
Hosia (Holiness) in Bacchae 370433. Looking forward, Russell and
Wilson (1981, 230231) point to the rhetorical category of ctitious
hymns to personied abstractions, for which Mendander Rhetor cited
as a prototype a hymn Simonides addressed to Tomorrow (Aurion,
615 PMG).
9. See Ford (2009, 139141).
10. Santoni (1991, 187 n. 42) cf. Aeschylus Choe. 372. Pindar Isthmian
5.13 refers to the popularity of leading off priamels with gold.
11. Pace Pages comment (vix credibile), the importance of noble
ancestry as a topos of encomiastic rhetoric suggests that we take
gonen (v. 11) in this way. Cf. Aristotle Pol. 1283a7 where eugeneia
is dened as excellence of breeding, aret genous; so, e.g., Jaeger
(1948), Santoni (1999, 189191). Bowra (1938) takes it as referring to
children; Renehan(1982, 260262) toparents. Biographismapproaches
absurdity here if we consider howtactless praise of progeny would have
been if Hermias were in fact a eunuch.
12. InAristotle, the epic terminationof sleeps epithet, e.exeuy yxoto
(v. 8, Atticized to ou in Didymus), may bring a subtle association with
sexual pleasure by recalling the priamel on satiety in Il. 13.636637:
there is a time when one has had enough of everythingof sleep
and love, and sweet singing and fair dancing ( evxv ` v x,o
ox`t xe`t u
\
vou xe`t t.xyxo / o. y x y.ux, y xe`t e uovo
o ,y0o o). Pindar has a compressed version at Nem. 7.5253: x,ov
c
t / xe`t .t xe`t x` e x ,v e
v0 A,oc tote (there is enough

even of honey and of the delightful owers of Aphrodite) and closes a
strophe by mentioning sleep (Pythian 9.2425). Cf. the association of
sweetest sleep (e` v e.ex xexe xe0 ucot) with being with young
boys (etctxo t o
\
t. v) climaxing the priamel inProdicus (Xen. Mem.
Notes 199
2.1.24.1). The dithyrambic qualities of e.exeuy yxoto are analyzed
below.
13. The syntax of Ariphrons v. 9 may be easier if we follow Wagman
(1995, 177) and read o
e,ot in a schema Pindaricum.
14. With the rhetoric of Ariphrons v. 10, compare again Prodicus (in
Xen. Mem. 2.1.32.1), who has Aret say: there is no ne work of man or
god without me (
,yov c` xe.v ou
x 0 tov ou
x e v0, tov ,`t
o u y tyvxet). Cf. Pindars Olympian 14.910, studied in the following

section.
15. See Pulleyn (1997, 4255) on the distinction. An example of a paean
withattestedcult functionis the so-calledErythraeanpaeantoAsclepius
(934 PMG=1367 CA), onwhichsee Kppel (1992, 189200, 370374)
and Graf (1985, 250257).
16. Such a scenario is suggested by Demochares (75 FGrH F 2) when
he sneers that the Athenians were so obsequious that they sang a
hymn to Demetrius the Besieger (CA 173174, noted above) not just in
public but in their homes (ou cyoo te vov, e ..` e xe`t xex ot x tev:
Athenaeus 253b).
17. E.g., Tukh (Fortune) beginning Ol. 12, Hsukhia (Peace) in Pyth. 8,
Theia (Divinity) in Isth. 5; such abstractions in Pindar, called by Bundy
(1969, 36) hypostasizations of aspects of success, are listed in Farnell
(1932, 467); cf. Furley and Bremer (2001, 1.265); Burnett (2005, 93 ff).
18. The text is from B. Snell and H. Maehler, commentary in McLachan
(1993, 4255). For the cult of the Graces, see Farnell (1909, 5.427
431); on their festival at Orchomenos, the Charitesia, Schachter (1981,
1.140144).
19. With Pindars o` uv y` e, u
tv (v. 5), cf. Ariphrons x` e o to (8);

with his ou c` Xe, txv e
x, (8), cf. Ariphrons o 0v c` ,`t ou
xt
(10).
20. The wider fame of the Graces is already acknowledged by the epithet
e o tctot in v. 3: possibly they are called sung of because Hesiod sang
their praises in Theogony 9079, where the three Graces have the same
names and appear in the same order as in Pindar. Various traditions
about their number and names are recorded in Pausanias 9.35.
200 Notes
21. On possible Pythagorean overtones in e veov, see Ford (2002,
108111). It may also be, as scholars have suggested, that the image
of owing prepares Pindars turn to Asopichos, who was named after
another river of Orchomenos, the Asopos.
22. For a possible parallel, consider the third stasimon of Euripides
Medea in which the rst strophic pair (824845) contains general praise
of Athenian culture usable at any time, whereas the second pair (846
865) ruminates, in a slightly different meter, on the specic problem of
how Athens can take in the infanticidal heroine.
23. The mention of Pythian Apollo has been thought to point to a
fact on the ground: ancient commentators on v. 10 say that in Delphi
statues of the Graces (enthroned?) were situated next to Apollo. But
such information may be speculation derived from Pindars text or a
text like the Homeric Hymn to Apollo 194196, which represents the
Graces dancing in Apollos train at Delphi.
24. E.g. Lucian, A Slip of the Tongue 6.26, who reasons that if Health
is presbist, so her work of healing must be ranked before all other
goods (
\
ox t ,op toxy
ox`tv u
\
y tte, xe`t x
,yov eu x y x
u
\
yte tvtv ,oxexx ov x v e
..v e ye0 v). Maximus of Tyre (7.1.e-f)

also justies the epithet. A similarly ambiguous use of the superlative is
Iliad 4.5961, where Hera (who has the epithet , ope inepic) seems to
declare herself the eldest as well as most venerable (,opux exyv,
4.59) of Cronuss daughters, both in birth and because I am the
wife of Zeus (e x,ov yv y x xe`t ou
\
vxe o` y e, exotxt /
x x.yet, 4.6061). The idea of Hera as eldest has been thought
to be a mirage created by interpolation from her more intelligible
claim (in 17.3656) to be the best (e , toxy) among goddesses in
yv y (descent) and marriage. See the Lexicon des Frgriechischen Epos
s. v. ,op ux,o, 2ap. Solmsen (1960) argues that this passage
generated a response in the fth Homeric Hymn to Aphrodite (possibly
the earliest composition in the collection): there Hestia is the one called
, opt,e (H. Hom. 6.32; cf. Hymn 29.3) and Heras aretology (6.40
42) seems expanded in compensation. Cf. Faulkner (2008, 101102,
125126), who is skeptical.
Notes 201
25. See Calame (2005, 1935) for an overview of the genre.
26. Bowra (1953, 30). On the date of Homeric Hymn 30, Janko (1982,
275 n. 12). My attention was drawn to this poem by LeVen (2008).
27. V. 19 is exactly repeated in To Demeter 495, To Apollo 546, etc.;
similar formulae: To Hermes 580, To Aphrodite 21, etc. The third Platonic
epistle (315B) quotes a hexameter line that began a hymn composed
by Platos tyrant-patron, Dionysius of Syracuse. Plato notes that the
hymn was intended for performance at Delphi, and claims to know it
from having been informed by those who were attending the festival
(
\
y
yyt.ev ot
\
xx 0,o uvx). Plato quotes the line to quibble
with it for deploying, as hexameter hymns could, the e t, formula as
anintroduction: Rejoice, and keep safe the life of the tyrant inpleasure
(e t, xe`t y
\
cvov p toxov ct eo xu, evvou). The epistle discourses
on the inaptness of this traditional way of greeting a god; Lucian (Slip of
the Tongue 4) summarizes: Plato reproves Dionysius for commencing
a hymn to Apollo with Rejoice, which he maintains is unworthy of the
Pythian, and not t even for men of any discretion, not to mention
Gods (tr. Fowler, adapted). Plato proposes to say do well instead
(u
, exxtv, 315A). (Doubting the authenticity of the letter, Bury in

the Loeb aptly compares a similar discussion of what is wrong in the
khaire formula in Charmides 164D-E, where note the context is also
Delphic piety.)
28. The epithet was pliant enough that Sophocles could use it to
mean all mother, mother through and through (Sophocles Ant.
1282).
29. On the variant in Iliad 14.259, see Janko (1992, ad loc). In
accord with cosmogonies in which Night gured as the rst thing
created, Zenotodus, Aristophanes, and a papyrus call Night the mother
( yxt,e) of gods and men. Callimachuss yxt,e(v) in SH 301
seems to conrm the form in the Homeric hymn as a variant of
mother, even though etymologically e yxt,e might be construed
as all knowing. (Just to keep things interesting, the form ce ext,e,
subduer, cannot be ruled out from Iliad 14.259 since it is also found
in Callimachus, Fr. 267 Pf.).
202 Notes
30. Theogony 116117. Aristotle and his contemporaries pored over
these Hesiodic lines to get clues about the rst principles in earlier
cosmogonies: Physics 4.1 (208b2733); Metaphysics A4 (984b23
32); so too Plato, Symposium 178B. Details in Ford (2009,
141144).
31. Artemis: Od. 5.123, 18.202, and 20.71, where e
\
yv y is taken as
virginal by the Aristotelian Problems 894b34. Demeter: Works and
Days 465. Aristotle knew e
yvo as the name of the chaste tree (Vitex

Agnus-castus, History of Animals 627a9), the branches of which had a
ritual function for matrons during the Thesmophoria.
32. The same passage of Diogenes (2.42) also quotes from Socrates
Aesopic poem, revealing that it was in elegiac couplets ([1] IEG). See the
testimonia of West on Socrates in IEG for variation in the tradition in
references to Socrates paean or prooimion to Apollo.
Chapter 7
1. Politics 8.3 (1337b23 ff.), cf. 7.1315 and Ford (2004, 313). Two
independent but convergent analyses of this discourse genre are Ford
(1999b) and Pelliccia (2002).
2. Gorgias 451E: ot
oet y e, o e xyxo vet
v xo t ouoo tot
e cvxv e v0, v xo uxo x oxo.tv,
v
\
xexe,t0o uvxet e
covx
o
\
xt u
\
yte tvtv ` v e
,toxv
oxtv, x c` c ux,ov xe.v yv o0et,

x, txov c ,
\
yotv o
\
otyx` y xo u oxo.to u, x .oux tv e c.
Doddss great commentary on Gorgias opened up the signicance of this
verse for me.
3. 890 PMG. Cf. Athenaeus 693f-694c. For other sympotic examples,
see Schmid (1964, 105109).
4. Rhetoric 2.21, 1394b13, with oral variants in the text.
5. Aristotle Eudemian Ethics 1.1, 1214a5; cf. Nicomachean Ethics 1.8.14,
1099a27.
6. Fr. 16 Voigt (= 16 L-P, with Pages supplement at v. 8). I take the
song to be complete after verse 20. For a recent analysis see Calame
(2005, 5569).
Notes 203
7. As Most (1991) notes in a valuable analysis, Aristotle cites Sapphos
use of the Helen myth here in his Rhetoric as an example of the proof
fromauthority; the woman who excels all others in beauty (kallos, v. 7)
can be assumed to know what is kalliston.
8. A prose example that Plato devised to be spoken in the persona
of Socrates is worth quoting for its great renown, and for its broad
relevance to the milieu for which Aristotles poetry was suited: I say,
and it happens to be true, that the greatest good thing for a humanbeing
is this: to spend time each day discussing aret and the other things you
have heardme talking about, toclarify my ownideas andthose of others;
the unexamined life is not worth living for a human being (. y o
\
xt
xe`t xuy evt ytoxov e ye0v o` v e v0, xo uxo,
\
x eoxy y
\
,e
,`t e ,t y xo` u .you ot to0et xe`t x v e
..v ,`t
\
v u
\
t
o u e xo ux cte.yo vou xe`t
euxv xe`t e
..ou
x eovxo, o
\
c` e v xeoxo p to ou ptx e v0, , Apology 38A).
9. Page (1955, 135 n. 1). Anagora, but not Anactoria, is listed in the
Suda lexicon (S 107, iv 322s Adler) among Sapphos female students
(e0 yx,tet). Ferrari (2010, 37 n. 16) would emend. Apart fromSappho
16 Voigt, we hear of Anactoria in Ovid, Heroides 15.17, and Maximus of
Tyre 18.9.
10. Bundy (1969, 5); cf. Race (1982, 6364).
11. Euripides personied areta at the beginning of a choral ode in
Orestes 807808, though in the nominative, not vocative case.
12. As whenCastor, inanepiphany, declares that eventhe gods feel pity
for toiling mortals (o
t xxo 0vyx v o.u0v, Eurpides Electra

1330).
13. Athenaeuss version of v. 12, o` ev e y, uovx c uvetv (tracking
down your power), reprises the metaphor, if right: Diogenes Laertius
has e veyo, uovx; Didymus is unclear. Renehan(1982, 257) compares
a Euripidean phrase, to track down excellence is a great thing ( ye
xt 0y, utv e ,x ev, Iphigeneia at Aulis 568), noting it has a sophistic,
we can say protreptic, background. The metaphor is mock-heroic in
Ariphrons periphrasis for sex, longings we hunt with Aphrodites
sweet and secret snares (0v ou
\
x,u tot A,oc txe
\
,xotv
204 Notes
0y, uov, 5). The attempt by Crosset (1967, 148 n. 7) to make the
hunting metaphor the burden of the song seems to me a warning
example of excessive New Criticism.
14. On philosophical protreptic at this time, see Slings (1995 and 1999,
6795).
15. It was popularly thought that repeated exposure to Homers
encomia of heroic warriors, for example, inculcated in the young
an appetite for valorous action (aret): Isocrates Panegyricus 4.159;
cf. Lycurgus Against Leocrates 102103 and cf. Aristophanes Frogs
10341036 for Homer as a teacher of aret.
16. This is the message, often not fully appreciated, of Politics 78; see
Ford (2004).
17. On epinician ponos, see Bundy (1969) and Kurke (1991, indexes
s.v.). A biographical reading might suggest that Aristotle has chosen
mokhthos because, encompassing more passive travails than ponos, it is
a broad enough term to include Hermiass tortures among his pursuit
of Virtue. Of the two uses of 0o in Homer, Il. 2.723 concerns the
suffering Philoctetes, on whom see below.
18. Jaeger (1948, 118, 140, 153), followed by Crosset (1967, 148149)
(who yet usefully compares the hymn to beauty in Isocrates Helen
5458); so too Wilamowitz (1893, 2.410411). Against seeking out
such issues in the poem, Bowra (1938, 188), followed by Dring (1957,
5960), Renehan (1982, 268274), Guthrie (1981, 34).
19. Norden (1913, 159160).
20. With Aristotles placement of the vocative, cf. the hymn-style
opening of Pindar Isthmian 5.6: ct ` e x ev,
veooe, xt` ev.
21. On reading xe,v here, see Renehan (1982, 259260), Dorandi
(2007, 2425).
22. For the semantics of t oe0 evexov, cf. the common epic epi-
thet t o0o, godlike, used only of mortals; for the formation, cf.
Ariphrons t ooce tovo peot.y tco e , e (v. 4) of royal power a
fortune like that of the gods. Lyric antecedents include t ooce tv
Notes 205
peot. u in a lament for the Great King in Aeschyluss Persians 634
(cf. t o0o Le, to of the decidedly mortal Darius at 857).
23. Aristotle refers to the play at NE 1146a20 and 1150b9.
24. Text by Lloyd-Jones and Wilson (1990), except that I have retained
ou y` e, yu
\
o pte ouv0v yoxt at 1443, see the discussion.
25. A euphemismfor kill comparable to Aristotles left the rays of the
sun bereft (v. 16).
26. In the LSJ, Philoctetes 1420 is cited as the earliest example
(followed by Platos Symposium 208D, quoted below) of aret being
used in an extended sense not for excellence but for excellences
rewarddistinction, fame, glory.
27. E.g. Isocrates To Demonicus (4950): Zeus made [Heracles] immor-
tal because of his aret while subjecting [Tantalus] to the greatest
torments because of his baseness. (xv ` v ct ` e x` yv e ,x` yv e 0 evexov
o tyov, xv c` ct ` e x` yv xex tev xe t y toxet xt, tet
x.eov).
Cf. Evagoras 70: those in the past who have become immortal on
account of their aret (xtv x v yyvy vv ct e ,x` yv e 0 evexot
yyveotv); also On the Peace 94, and Hyperides Funeral Oration 19.10.
28. Jebb (1932) notes on v. 1420 that the seer Diotima makes the same
point in Platos Symposium 208D-E: this prose encomium resembles
Aristotles in offering a trio of heroic exemplars who nobly accepted
death (Alcestis, Achilles, capped by the Athenian Codrus) to argue
that everyone is moved by the hope of immortal aret and a glorious
(eu-kleous) reputation of the sort we here now sustain [i.e. by recalling
these very stories] (u
\
` , e ,x y e 0ev exou xe`t xote uxy cy
u x.o u y
\
` v v uv y
\
t
ov).
29. Webster (1970, 157) takes it as the excellence of an immortal
as distinct from a mortal, suggesting Heracles appeared as divinely
youthful; similarly, Kamerbeek (1980, 195).
30. On Asclepius, Jebb (1932, 205, 221) gives details (on vv. 1333,
1437); cf. Kamerbeek (1980 ad loc.). In the background may be, as
Mitchell-Boyask (2007, 85114) argues, a reference to the cult of
Asclepius, introduced in Athens a decade earlier. The ancient biography
206 Notes
of Sophocles reports that he served as a ministrant of the cult in
recognition of which he was given heroic (NB not divine) honors after
his death.
31. In place of y
\
y` e, u o pte at 1443, Lloyd-Jones and Wilson
(1990) and other recent editors print Dawess ou y` e, yu
\
o pte.
Kamerbeek (1980) has the fullest discussion, supporting the conjecture
and Linforths interpretation.
32. Linforth (1956, 149 n. 30), approved by Lloyd-Jones and Wilson: it
is the fame resulting from piety that the poet has in mind.
33. Wilamowitzs explanation is cited by Kamerbeek (1980, 191);
similarly, Campbell (1881, vol. 2. 477) (Follows men in their Death).
A parallel to the eschatological thought is Sophocles Electra 291292,
where I take Clytemnestra to say, may you [Electra] perish miserably,
and may the gods belownever release you fromyour present lamenting
(pace Jebb and others; cf. Finglass [2007: 186]).
34. Aristotle approaches this idea in his ethical works, declaring
that no function of man has so much permanence as virtuous
activities; these are thought to be more durable even than scientic
knowledge (NE 1.10 1100b1214: ,`t ou c` v y` e, ou
\
x u
\
e,t
x v e v0, tvv
,yv ppetxy
\
,`t x` e
v,y te x` e xex
e ,x yv ovt x,et y` e, xe`t x v
toxy v e
u
\
xet coxo uotv
t vet).
35. Jebb (1932, 221222) argues that Sophocles should say the effect
of piety is imperishable, bringing happiness to the pious in Hades and
anexample to those who survive. Rejected by Kamerbeek (1980 ad loc.).
36. Sophocles composed a renowned paean to Asclepius (Philostratus
The Life of Apollonius of Tyana 3.17 = T 73a TrGF Radt; see Testimonia
6773) and, as noted, experimented in Philoctetes by directing a paean
to Sleep (827832): see Haldane (1963).
37. Pickard-Cambridge (1988, 276), citing Theophrastus Char. 15,10,
27.2; see also Aeschines 1.168; Menander Epitrepontes 767768.
38. On Sophoclean epiphanies, see Pucci (1994) and Parker (1999,
1113); on tragic epiphanies, see Sourvinou-Inwood (2003, 459
512) who, like Pucci, characterizes the Philoctetes as Euripidean in
Notes 207
its stress on the disjunction between the divine and mortal worlds
(482484).
Chapter 8
1. For divining resemblances see Poetics 1459a5-8: "The most
important thing by far [for a poet] is to be good at metaphor. For this
alone is not possible to get from another person but is a sign of inborn
talent. Making good metaphors is the ability to discern similarities
(o.` u c` ytoxov x xeo,txv
tvet. vov y` e, xo uxo ou

x e,
e
..ou
oxt .ep tv u ut
e x oy tv
oxt x y` e,
u xe ,tv x
x o
\
otov 0, tv
oxtv). So too Rhetoric 3.8 (1409a8-10). For

deviations from ordinary language (vtxv c` . y y. xxev xe`t
xeo,` ev xe`t
xxeotv xe`t ev x e,` e x x u,tov, Poet. 1458a22

23) see Poetics 1458a18-1459a17.
2. Aristotle Poetics 1459a19, Rhetoric 3.3, 1406b2. Wilamowitz (1893,
1.407), followed by Smyth and Bowra. On compound epithets in the
new dithyramb of the later classical age, see Ford (forthcoming). The
poems most outstanding example is the epithet soft-beamed sleep
(e.exeuy yxoto, v. 8). The word seems to be a coinage blending a
more straightforwardlyric compoundlike soft-eyed (e.exexo,
929(g)1 PMG) with the kind of expression to be found in the
dithyrambist Licymnius, a poet whose style Aristotle admired (Rhet.
3.11, 1413b1216); Licymnius said that Sleep used to put Endymion to
sleep with his eyes open because he so enjoyed the beams of his eyes
(o exv eu ye t, 771.2 PMG).
3. Mastronarde (1994), on Euripides Phoenissae v. 784. Santoni (1991,
180 n. 4 and 181 n. 9) collects earlier uses, noting that, outside of a few
uses in Sophocles and Euripides, o. uo0o is common in Orphic
hymns and Sibylline oracles.
4. Bundy (1969, 91) notes that a number of epinician odes turn in
conclusion to the inevitability of death in order to dene and illuminate
the essence of areta, citing Bacchylides lines among the examples and
characterizing his as the least stylizedandallusive versionof the topos.
As Bundy notes, the underlying ethical idea is expressed in Aristotles
208 Notes
Rhetoric (1.9, 1367a1-3): the preeminently praiseworthy actions are
those which it is possible for a man to possess after death rather than
during his lifetime, for the latter involve more selshness.
5. On Philips interest in Heracles, see Hammond and Grifth (1979,
514515) and Markle (1976). Among the evidence is Isocrates To Philip
(5.7677, where the hero is cited as an ancestor of Philip), and the
prominence of Heracles in the Letter to Philip written by Speusippus
of the Academy (Epist. Socrat. 30). Heracles had been ethicized in
writings of Prodicus and Antisthenes and, at the end of the fth
century, Herodorus of Heraclea Pontica (e.g., FGrH 31 F 14where it is
noteworthy that Virtue is symbolized by the apples of the Hesperides).
Plato used Heracles as a gure for Socrates life (and death?): Apology
22A; cf. Theaet. 175A, Lysis 205C.
6. Galinsky (1972, 107) speaks of Aristotles song as continuing the
traditional Pindaric praise of Herakles as the embodiment of aret.
7. On these myths in mystery religions, Boyanc (1937, 307); on
the Alexander historians, Harding (2006, 155), citing Bosworth. Of
course, philosophers had their own interpretations of the stories, e.g.,
Xenophons Socrates mythologizing at a symposium that Heracles
and the Dioskouroi are symbolic of beauty of mind and that of body
(Symposium 8.2830). These exempla, and the elusiveness with which
they suggest immortality, were still in use much later: Menander Rhetor
(414.2327) advises eulogists to cite traditions (they say) that Helen,
the Dioskouroi, and Heracles share the community of the gods, and on
that basis to praise [the deceased] as a hero, or rather bless him as a
god. For commentary, see Russell and Wilson (1981, ad loc.).
8. On Achilles cult in the Black Sea area, Burgess (2009, 111131);
Hooker (1988) is skeptical; on Ajax, Henrichs (1993, 175177).
9. For look upon, Il. 16.188; Mimnermus 1.8 IEG; Theognis 426.
Leaving the light of the sun, Il. 18.11, cf. Od. 11.93. Smyth (1906,
471) suggests Aristotles y,ov is equivalent to a middle (bereft
himself), but in favor of taking it transitively is a passage Smyth
cites fromEuripides Cyclops 440, rightly characterizing the expressions
dithyrambic extravagance.
Notes 209
10. E.g., in Pindar Pyth. 5.96 the periphrase .evx At
cev (winning
Hades as their lot) coexists with the explicit and emphatic (triad-
ending) x txet 0ev v in 5.93, describing the heroized founder of
Cyrene, Battos. For the motif in fourth-century Attic tombstones, see
Tsagalis (2008, 7785).
11. Aristotles accusers could have construed e 0 evexv x tv
eu yoouot Mo uoet in 18 as will make him increase until he is
immortal, and could have compared Pindar Pythian 9.35, which
describes immortalization by applying nectar and ambrosia to the lips
as making him deathless (0 yoovxe t x vtv e 0 evexov).
12. See Ford (1992, 5967).
13. Ibycus S151 PMGF; 282(a) PMG, on which see Budelman (2009)
and Hutchinson (2001). For other self-referential signatures (seals,
sphrgides) closing a lyric, cf. Pindar Pyth. 4. 298299, Bacchylides
3.9698.
14. The enkmion is usually assumed to be a genre for solo performance,
while the triadic composition of Ibycuss song is assumed to indicate
choral performance; but I see no reason why these short triads could
not have been performed by a soloist; we simply do not know what
forms lyric encomium might have taken before our rst (fragmentary)
examples from Pindar.
15. Sonnet 18. Shakespeares irony is in drawing attention to the media
in which the song will be preserved: for the object of praise to survive
requires breath to sing the sonnet or eyes to read it (read, not behold,
because in the future the only object of beauty actually before admiring
eyes will be the text). With Ibucyss blending praise of patron and self,
cf. Pindar Paean 6.61.
16. Santoni (1991, 187) aptly compares the conclusion of the speech of
Aret in Prodicuss Heracles at the Crossroads (Xenophon Memorabilia
1.2.3334) which strikes many of the same themes as Aristotle:
And when the appointed time may come, they [Virtues devotees]
do not lie unhonored and in oblivion, but through memory they
ourish in songs (u
\
vo uvot 0 e..ouot) for all time. If you toil
after such things (cteovyoe v), Heracles, you stripling of noble
210 Notes
parents, you may come to obtain the most blessed form of happi-
ness (o
\
xev c
.0y x , vov x .o, ou x` e . y0y e
xtot
x tvxet, e..` e x` e v yy xv e`t ,vov u
\
vo uvot 0 e..ouot.
xote ux e oot,
e t xox v eye0 v
\
H, ex.t,
oxt cteovyoe v
x` yv exe,toxox exyv u cetov tev xxx yo0et).
17. For representative texts and discussion, see Cole (2003, esp.
202207), and Johnston in Graf and Johnston (2007, esp. 117120).
18. See The American Heritage Dictionary of the English Language, 4th ed.
(HoughtonMifin, 2000), Appendix s.v. aug-. At Athens there was a cult
to Aux, as a form of the Graces: Pausanias 9.35, Farnell (1909, 428).
19. A metaphorical fruit withers in a noteworthy phrase by Pindar
Isthmian 8.48:
v c` xe, ou xex 0tv: the fruit of [Zeus]

words did not wither away (i.e. prove ineffectual).
20. Cf. Tsagalis (2008, 149150) for the (infrequent) use of eu evtv in
early Greek funerary inscriptions; a notable late example is CEG 599.4.
21. There is not a great deal of difference between the force of
y vt p,ox t in v. 1 and p t in v. 2, and the song repeats a number
of roots: euy- 8, 16, x.y- 5, 11, e0evex- 7, 18, o,- 3, 15,
,y- 11,
17; and possibly the hunting metaphor in 1 and 12. Note that Ibycus
151S PMGF also exhibits a good deal of repetitiveness in diction. Other
arguments against Wilamowitzs change: Renehan (1982, 267268).
22. Furley and Bremer (2001, 2.43) note that eu evtv in Aristonooss
hymntoHestia (CA1634, v. 10) combines a musical sense (raise a song
about) with an encomiastic one (raise in stature, cf. Lat. magnicare).
23. Cf. the Rhetoric to Alexander 3.35. For a neat and trenchant
discussion of Rhetoric and Lyric Poetry see Race (2007).
24. Cf. Cassius Dio 53.16: Octavian was styled Augustus, as if he were
a being superior to the mortal race. For all things [among the Romans]
which are considered most honorable and sacred are called august,
wherefore the Greeks rendered the word Augustus by sebastos, as if
venerable [quasi venerandum dicas]. The account of Augustuss names
in Suetonius differs in some aspects but conrms the association of
Augustus with the divine, as does Ovid (Fasti I. V. 609).
Notes 211
25. As Renehan (1982, 255) notes, The only explicit reference to
[Hermias] is oblique[Atarneus nursling]scarcely an honoric
description appropriate to a god. Nor is Hermias named in Aristotles
dedicatory monument at Delphi (though Theocritus knew to whom it
referred). On the role of names in real and ctional grave epigrams, see
the fascinating discussion of Fantuzzi in Fantuzzi and Hunter (2004,
291306).
Chapter 9
1. NE 7.8, esp. 1158a27b5: see Broadie and Rowe (2002, 413).
2. On the connection, see Ford (2002, 115119). Cf. Pulleyn (1997,
55): a hymn is a sort of negotiable e
\
ye.e [offering], which generates
e,t [a feeling of reciprocity] whereas a [prayer] is not.
3. The idea of excellence leaving behind a deathless memorial is a
clich of encomiastic rhetoric: e.g., Isocrates Panegyricus 84 (x y c
e ,x y e 0 evexov x` yv v yyv
o tyoev); Lysias Epitaphios 81 (
tc` y
0vyx v o exv
xuov, e 0 evexov v yyv ct ` e x` yv e ,x` yv eu
\
x v
xex .tov) and Xenophon Agesilaus 6.2.
4. The word bebaios recurs whenPlatodenies there canbe any plainand
stable meaning (xt oe` xe`t p petov) in a piece of writing (Phaedrus
275C) and says it is impossible to leave behind a written text with
stability and clarity (277D: ppet oxyxe . . . xe`t oe yvtev).
5. Olympiodorus In Gorg. 41.3 (= Aristotle Fr. 673 Rose = IEG); also
partly quoted in the Vita Marc. 26 (= T 34c Dring). Cf. Jacoby FGrH III
B 2, p. 482. For claims that Aristotle quarreled with Plato, see Guthrie
(1981, 25 n. 1).
6. Jaeger (1927, 14). Further discussion in Renehan (1991, 256258).
7. Bernays (1878), rejected by Wilamowitz (1893, 2.413).
8. Dring (1957, 315). So too (Immisch 1906). Renehan (1991, 258)
withholds judgment.
9. Jaeger also notes that this line recalls the interpretation that Plato
gave an old poemon aret by Tyrtaeus (cf. 12 IEG) cited in his last work,
Laws 660E: Shall we not enjoin poets to say that the good man, if he
212 Notes
be temperate and just, is also happy and blessed, no matter if he be big
and strong or small and weak, rich or not (Laws 601E: xo` u otyx` e
e veyx ex . ytv
\
o
\
` v e ye0` o e v` y, o ,v ` v xe`t c txeto
u ce tv
ox`t xe`t ex e,to,
evx ye xe`t t ou,` o
evx tx,` o
xe`t e o0v` y
y , xe`t
` ev .oux y xe`t y). Thus Aristotles epitome of

Platonic philosophy in v. 7 would have pleased his teacher.
10. The vulgar Lives give different reconstructions, usually incorporat-
ing something fromvv. 23. See testimonia printed at Wests IEG, 2.45,
where note that a late commentary on Porphyry by David identies
Aristotle as the dedicator in his version of the inscription: p` ov
A,toxox .y t c, ucexo x ovc H. exvo. Cf. Immisch (1906, 1112),
Jaeger (1948, 107 n. 2).
11. The reconstruction by Wilamowitz (1893, 413414), followed by
Immisch (1906, 1516), is rejected by Renehan (1991, 260261), who
points to, inter alia, Aristotles inclusion of Plato among mortals
(0vyx v, v. 4). I agree, and note that the reasoning is reminiscent of the
importance to Athenaeus and others of the mention of Hermiass death
in the alleged paean to him.
12. Jaeger (1948, 109) (adopting a suggestion of Blass). Pace
Wilamowitzs objection (1893, 2.414), who would take t. ty as a
genitive of cause.
13. Jaeger (1948, 108). I would construe the end of the poemdifferently
from Jaeger: I doubt it refers to a philosophical conception of the
impossibility of embodying the ideal (see 109, with n. 2); putting weight
on rst in v. 4, the more widely intelligible statement would be that
Plato can never be displaced as the rst discoverer of the moral truth
enunciated in v. 7. So Renehan (1991, 265266).
14. Jaeger (1948, 109) rightly remarks that Aristotles poem is phrased
a ctional speech interpreting the inscription in the manner of a pious
exegete before a sacred object.
15. Jaeger (1927, 14). Boyanc (1937, 250257) posits a stranger from
the East, since he sees the poemand altar as rendering religious homage
to Plato.
Notes 213
16. There may possibly have been a prose version as well: Olympiodorus
remarks that Aristotle not only composed an encomium in praise of
him[sc. Plato] but also praises himin the elegiacs for Eudemus. Dring
(1957, 317), who does not have a high opinion of Olympiodorus as a
historian, suggests he might be confusing Callisthenes encomium for
Hermias.
17. DL 5.512. Cf. Lynch (1972, 100105 and 114115) on the (not
technically religious) term mouseion for such schools.
18. Redeld (n.d.) refers to Speusippuss aforementioned Funeral-feast
of Plato in showing that Hero cult is indeed one of the many
institutional ancestors of the philosophical schools.
19. Notwithstanding my view that the poem should not be treated as
versied philosophy, it is hardly surprising that compatible ideas can be
found in the famous chapter from Nicomachean Ethics on the best life
(NE 10.7, 1177b31ff.). Aristotle argues that since the mindis something
godlike inus, a life livedinaccordance withit makes humanlife godlike
(t c` y 0 tov o
\
vo u ,` o x` ov e
v0,ov, xe`t o
\
xex` e xo uxov p to 0 to
,` o x` ov e v0, tvov p tov); he thus recommends going beyond the
archaic advice that mortals should think mortal thoughts and as far
as possible become immortal (
o
\
oov
vc xet e 0evex ttv). That

pursuing arte is the way to do so is clear from a passage like NE 7.1
(1145a1525), in which Aristotle opposes to the vice of brutishness
(0y,t oxy) a heroic and godlike excellence (e ,x yv y
\
,tx yv xtve xe`t
0 tev) like the ultimate excellence that people suppose allows men
to become gods (t . xe0 e, eo tv.
e v0, v y tvovxet 0o`t ct

e ,x y u
\
,po. yv).
20. Herman (1987, 69): ritualised friendship was thought to outlast
the individual actors and, conspicuously mimicking kinship ties, pass
on to their descendents. Thus, a person could die, but the role of xenos
could not.
21. See discussion in ch. 2 above.
22. Against authenticity: Trampgedach (1994, 7072). Defending it:
Wilamowitz (1920, 2.281), Wormell (1935, 5961), Guthrie (1975, 400
401), Post (1921, 127), andHarward(1932, 186), observing that a forger
214 Notes
might have been expected to drag Aristotle into the letter. As noted in
my earlier reference to the letter, Strabos unsupported claim (13.1.57
C610) that Hermias knewPlato at Athens does not condemn the letters
denial that Platos had personal knowledge of Hermias (323A).
23. DL 3.46; Philodemus Index Acad. col. 6.10 Dorandi; Didymus col.
5.5354.
24. Cf. Eudemian Ethics 7,2 1238a11 ff.
25. Lynch 1972, 76; cf. 87.
26. Athenaeus (3a-b). Discussions include: Gottschalk (1972, 335342
and 1987, 10831088); Moraux (1973, 394), reviewed by Leonardo
Tarn in Gnomon 53 (1981, 721750, esp. 725731); Barnes (1997, esp.
116), and Bollanse (1999b, 234243); brief remarks in Fraser (1972,
1.320 and 2a.473 n. 100), with critique in Tarn (above, 727), and
Guthrie (1981, 5965). For possible connections between Neleus and
the Attalids of Pergamon, see Kosmetatou (2003). Finally, Lord (1986,
140145) sees behind the story credible measures for safekeeping the
collection in light of the Peripatetics changing political fortunes in the
later fourth century.
27. Lynch (1972, 144149).
28. Trans. Chiesara, Aristocles F 2.13 Chiesara (= T 581 Dring).
29. Diels-Schubart (1904, 2425) suggest Hermippus may have trans-
mitted the song to Didymus; see further Bollanse (1999b, 313314,
and 2001). But Wehrli (1974, 75), citing Jacoby FGrH 3b Suppl. 1,
Text 329, notes that there is no evidence that Hermippus included long
verbatim quotations; so too Harding (2006, 35, 3738). If Athenaeus
copied the song from Hermippus, the differences between his text and
the one in Didymustabulated by Gerke 1902may be due to the
latters having got his version fromanother source, Favorinus according
to Wilamowitz (1892, 2.403 n.1); cf. Harding (2006, 154).
30. Moraux (1951, 144145). For the list inDiogenes, see text inDring
(1957, 4150, 6769). It seems that the catalog preserved in Diogenes
is not closely related to what Hermippus did: Moraux (1951, 211233),
Bollanse (1999b, 163182).
Notes 215
31. [Ammonius] 482 Nikau: -u
\
vo
yx tou> cte ,t. o

\
` v y` e,
u
\
vo
ox`t 0 v, x` o c`
vx tov e v0, v. See Moraux (1951,

261262) and cf. R. Wnsch, Hymnos RE 9.1 col. 181. On Aristotles
On Virtue (A 163), see Moraux (1951, 254, 269). For the appendix
Hesychiana (Morauxs lAnonyme de Mnage), see Dring (1957, 8292).
32. Didymus col. 6.2122: xou x e
v [
]
.
[o]t e u. eu x` ov
e
vey, e[yet ct] ` e x` o ` y o..o t ,` o t,` o (
t vet).
B I B L I O G R A P H Y
Badian, Ernst. 1981. The Deicationof Alexander the Great. InAncient
Macedonian Studies in Honor of C. F. Edson, ed. H. J. Dell, 2771.
Thessaloniki: Institute for Balkan Studies.
Bagnall, Roger S. 2004. Library of Dreams. Proceedings of the American
Philosophical Society 146: 348362.
Barigazzi, Adelmo. 1966. Favorinus. Opere. Florence: Le Monnier.
Barnes, Johnathan. 1997. Roman Aristotle. In Philosophia Togata.
Vol. II, Plato and Aristotle at Rome, ed. Johnathan Barnes and
Miriam Grifn, 169. Oxford: Oxford University Press.
Barthes, Roland. 1988. The Old Rhetoric: An Aide-mmoire. In the
Semiotic Challenge, tr. R. Howard, 1194. NewYork: Hill and Wang.
Bell, J. M. 1978. Kimbix kai sophos: Simonides in the Anecdotal
Tradition. Quaderni Ubinati di Cultura Classica 28: 2986.
Bernays, Jacob. 1878. Aristoteles Elegie an Eudemos. Rheinisches
Museum 33: 232237.
Bing, Peter. 2009. The Scroll and the Marble: Studies in Reading and
Reception in Hellenistic Poetry. Ann Arbor: University of Michigan
Press.
Bloom, Harold. 1975. AMap of Misreading. NewYork: Oxford University
Press.
217
218 Bi bl i ography
Boedeker, Deborah. 2001. Heroic Historiography. In The New
Simonides, ed. D. Boedeker and D. Sider, 120134. Oxford: Oxford
University Press.
Bollanse, Jan. 1999a. Hermippos of Smyrna. Die Fragmente der
Griechischen Historiker Continued IV. A. Leiden: Peeters.
Bollanse, Jan. 1999b. Hermippos of Smyrna andhis Biographical Writings,
a Reappraisal (Studia Hellenistica 35). Leiden: Peeters.
Bollanse, Jan. 2001. Aristotle and the Death of Hermias of Atarneus:
Two Extracts from Hermippos Monograph on Aristotle. Simblos
3: 6798.
Bosworth, A. B. 1970. Aristotle and Calisthenes. Historia 19:
407413.
Bosworth, A. B. 1977. Alexander and Ammon. In Greece and the
Eastern Mediterranean in History and Prehistory, ed. K. H. Kinzl,
5175. Berlin: W. de Gruyter.
Bosworth, A. B. 1988. Conquest and Empire: The Reign of Alexander the
Great. Cambridge: Cambridge University Press.
Bowra, C. M. 1938. Aristotles Hymn to Virtue. Classical Quarterly 37:
182189. (Reprinted in Problems in Greek Poetry, Oxford: Oxford
University Press, 1953, 138150).
Boyanc, Pierre. 1937. Le Culte des muses chez les philosophes grecs. Paris:
E. de Boccard.
Broadie, Sarah, and Christopher Rowe. 2002. Aristotle: Nichomachean
Ethics. Oxford: Oxford University Press.
Brunt, P. A. 1993. Studies in Greek History and Thought. Oxford: Oxford
University Press.
Budelman, Felix, ed. 2009. The Cambridge Companion to Greek Lyric.
Cambridge: Cambridge University Press.
Bundy, E. L. 1969. Studia Pindarica. Berkeley: University of California
Press. Reprint, 1986.
Burgess, Jonathan S. 2009. The Death and Afterlife of Achilles. Baltimore:
The Johns Hopkins University Press.
Calame, Claude. 2005. Masks of Authority: Fiction and Pragmatics in
Ancient Greek Poetics, tr. Peter M. Burke. Ithaca andLondon: Cornell
University Press.
Bi bl i ography 219
Cameron, Alan. 1995. Callimachus and His Critics. Princeton: Princeton
University Press.
Campbell, David. 1993. Greek Lyric, Vol. 5. Loeb Classical Library.
Cambridge, Mass.: Harvard University Press.
Campbell, Lewis. 1881. Sophocles. 2 vols. Oxford: Oxford University
Press.
Cawkwell, G. L. 1994. The Deication of Alexander the Great: A Note.
In Ventures into Greek History, ed. Ian Worthington, 292306.
Oxford: Oxford University Press.
Chiesara, Maria Lorenza. 2001. Aristocles of Messene: Testimonia and
Fragments. Oxford: Oxford University Press.
Christesen, Paul. 2007. Olympic Victor Lists and Ancient Greek History.
Chroust, Anton-Hermann. 1966. Aristotles Flight from Athens in the
year 323 B.C. Historia 15: 185192.
Chroust, Anton-Hermann. 1967. Aristotle Leaves the Academy. Greece
and Rome 14: 3944.
Chroust, Anton-Hermann. 1971. Aristotles Sojourn in Assos. Historia
21: 170176.
Chroust, Anton-Hermann. 1973. Aristotle: New Light on His Life and on
Some of His Lost Works. 2 vols. Notre Dame: University of Notre
Dame Press.
Clinton, Kevin. 1974. The Sacred Ofcials of the Eleusinian Mysteries.
Transactions of the American Philosophical Society, v. 64, pt. 3.
Philadelphia: American Philosophical Society.
Cole, Susan G. 2003. Landscapes of Dionysus and Elysian Fields.
In Greek Mysteries: The Archaeology and Ritual of Ancient Greek
Secret Cults, ed. Michael B. Cosmopoulos, 193217. London and
New York: Routledge.
Csapo, Eric. 2004. The Politics of the New Music. In Music and the
Muses: The Culture of Mousike in the Classical Athenian City, ed.
P. Murray and P. Wilson, 207248. Oxford: Oxford University
Press.
Currie, Bruno. 2005. Pindar and the Cult of Heroes. Oxford: Oxford
University Press.
220 Bi bl i ography
Crosset, John. 1967. Aristotle as Poet: The Hymn to Hermeias.
Philological Quarterly 46: 145155.
Davies, M., ed. 1991. PoetarummelicorumGraecorumfragmenta. Oxford:
Oxford University Press.
Depew, Mary. 1997. Reading Greek Prayers. Classical Antiquity 16:
229258.
Derenne, Eudore. 1930. Les procs dimpit intents aux philosophes
Athnes au Vme et au IVme sicles avant J.-C. Lige: H. Vaillant-
Carmanne. Reprint, New York: Arno Press, 1976.
Derrida, Jacques. 1974. Of Grammatology, tr. G. Spivak. Baltimore: The
Johns Hopkins University Press.
Derrida, Jacques. 1989. Biodegradables: Seven Diary Fragments,
tr. Peggy Kamuf. Critical Inquiry 15: 812873.
Diehl, E. 1925. Anthologia lyrica Graeca. 2 vols. Leipzig: Teubner.
Diels, H., and W. Schubart. 1904. Didymos Kommentar zu Demosthenes.
Berliner Klassikertexte I. Berlin: Weidmann.
Dindorf, G., ed. 18671871. Eusebius. 4 vols. Leipzig: Teubner.
Dodds, E. R. 1959. Plato: Gorgias. Oxford: Oxford University Press.
Dring, Klaus, ed. 1972. Die Megariker. Kommentierte Sammlung der
Testimonien. Studien zur antiken Philosophie, Bd. 2. Amsterdam:
Grner.
Dorandi, T. 1991. Storia dei loso. Platone e lAcademia (PHerc. 1021
e 164). Naples: Bibliopolis.
Dorandi, T. 1994. Storia dei loso. La Sto da Zenone a Panezio (PHerc.
1018). Philosophia antiqua, v. 60. Leiden and New York: Brill.
Dorandi, T. 2007. Note sulla tradizione e sul testo del poema di
Aristotele in onore di Ermia di Atarneo. Zeitschrift fr Papyrologie
und Epigraphik 161: 2126.
Drachmann, A. B., ed. 1969. Scholia vetera in Pindari carmina. 3 vols.
Amsterdam: Adolf M. Hakkert.
Dring, Ingomar. 1957. Aristotle in the Ancient Biographical Tradition.
Gteborg: Elander.
Dziatzko, K. F. O. 1899. Bibliotheken. RE 3 col. 40923.
Engelmann, Helmut, and Reinhold Merkelbach. 19721973. Die
Inschriften von Erythrai und Klazomenai. Bonn: R. Habelt.
Bi bl i ography 221
Fairbanks, Arthur. 1900. A Study of the Greek Paean. Cornell Studies in
Classical Philology XII. New York: Macmillan.
Fantuzzi, Marco. 2001. Heroes, Descendants of Hemitheoi: The
Proemium of Theocritus 17 and Simonides 11
2
W. In The New
Simonides, ed. D. Boedeker and D. Sider, 232241. Oxford: Oxford
University Press.
Fantuzzi, Marco, and Richard Hunter. 2004. Tradition and Innovation in
Hellenistic Poetry. Cambridge: Cambridge University Press.
Frber, Hans. 1936. Die Lyrik in der Kunsttheorie der Antike. Munich:
Neuer Filser-Verlag.
Farnell, L. R. 1909. The Cults of the Greek States. 5 vols. Oxford: Oxford
University Press.
Farnell, L. R. 1932. The Works of Pindar II: Critical Commentary on Pindar.
London: Macmillan.
Faulkner, Andrew. 2008. The Homeric Hymn to Aphrodite. Oxford:
Oxford University Press.
Ferrari, Franco. 2010. Sapphos Gift: The Poet and her Community, trans.
B. Acosta-Hughes and L. Prauscello. Ann Arbor: Michigan Classical
Press.
Finglass, P. J., ed. 2007. Sophocles: Electra. Cambridge: Cambridge
University Press.
Flower, Michael A. 1988. Agesilaus of Sparta and the Origins of the
Ruler Cult. Classical Quarterly 38: 123134.
Flower, Michael A. 1994. Theopompus of Chios. Oxford: Oxford
University Press.
Ford, AndrewL. 1992. Homer: The Poetry of the Past. Ithaca and London:
Cornell University Press.
Ford, Andrew L. 1999a. Reading Homer from the Rostrum: Poetry
and Law in Aeschines, In Timarchus. In Performance Culture and
Athenian Democracy, ed. S. Goldhill and R. Osborne, 281313.
Ford, Andrew L. 1999b. Odysseus after Dinner: Od. 9.211 and the
Traditions of Sympotic Song. In Euphrosune: Studies in Ancient Epic
and Its Legacy in Honor of Dimitrios Marinatos, ed. A. Rengakos and
J. Kazazis, 109123. Stuttgart: Franz Steiner.
222 Bi bl i ography
Ford, Andrew L. 2002. The Origins of Criticism: Literary Culture and
Poetic Theory in Classical Greece. Princeton: Princeton University
Press.
Ford, Andrew L. 2004. Catharsis: The Power of Music in Aristotles
Politics. In Music and the Muses: The Culture of Mousike in the
Classical Athenian City, ed. P. Wilson and P. Murray, 309336.
Ford, Andrew L. 2006. The Genre of Genres: Paeans and Paian in Early
Greek Poetry. Poetica 38: 277296.
Ford, Andrew L. 2008. The Beginnings of Dialogue: Socratic Discourse
and Fourth-Century Prose. In The End of Dialogue in Antiquity, ed.
Simon Goldhill, 2944. Cambridge: Cambridge University Press.
Ford, Andrew L. 2009. Platos Two Hesiods. In Plato and Hesiod, ed.
G. R. Boys-Stones and J. H. Haubold, 133154. Oxford: Oxford
University Press.
Ford, Andrew L. (forthcoming). The Poetics of Dithyramb. In
Dithyramb and Society: Texts and Contexts in a Changing Choral
Culture, ed. Barbara Kowalzig and Peter Wilson. Oxford: Oxford
University Press.
Fox, R. Lane. 1986. Theopompus of Chios and the Greek World. In
Chios: A Conference at the Homereion in Chios, ed. J. Boardman and
C. Vaphopoulou-Richardson, 111120. Oxford: Oxford University
Press.
Fredricksmeyer, E. A. 1979. Divine Honors for Philip II. Transactions
of the American Philological Association 109: 3961.
Fredricksmeyer, E. A. 1981. On the Background of the Ruler Cult.
In Macedonian Studies in Honor of Charles E Edson, ed. H. J. Dell,
145156. Thessaloniki: Institute for Balkan Studies.
Fuhrer, Therese. 1993. Callimachus Epinician Poems. In Callimachus,
ed. M. A. Harder, R. F. Retguit, andG. C. Wakker, 7997. Groningen:
Forsten.
Furley, William D., and Jan Maarten Bremer. 2001. Greek Hymns. A
Selection of Greek Religious Poetry from the Archaic to the Hellenistic
Poetry. Vol. 1: The Texts in Translation. Vol. 2: Greek Texts and
Commentary. Tbingen: Mohr Siebeck.
Bi bl i ography 223
Gagn, Renaud. 2009. Mystery Inquisitors: Sacrilege and Authority at
Eleusis. Classical Antiquity 28: 211-47.
Gaiser, Konrad. 1966. Die Elegie des Aristoteles an Eudemos. Museum
Helveticum 23: 84106.
Galinsky, G. Karl. 1972. The Herakles Theme: The Adaptations of the Hero
in Literature from Homer to the Twentieth Century. Oxford: Oxford
University Press.
Gelzer, Thomas. 1993. Transformations. In Images and Ideologies: Self-
denition in the Hellenistic World, ed. A. W. Bulloch et al., 130151.
Berkeley: University of California Press.
Gercke, A. 1902. Die berlieferung des Diogenes Laertios. Hermes
37: 424425.
Gibson, Craig A. 2002. Interpreting a Classic: Demosthenes and His Ancient
Commentators. Berkeley: University of California Press.
Gigon, O. 1958. Interpretationen zu den antiken Aristotles-Viten
Museum Helveticum 15: 147193.
Gigon, O., ed. 1962. Vita Aristotelis Marciana. Berlin: W. de Gruyter.
Gottschalk, H. B. 1972. Notes on the Wills of the Peripatetic
Scholarchs. Hermes 10: 314342.
Gottschalk, H. B. 1987. Aristotelian Philosophy in the Roman World.
Aufstieg und Niedergang der rmischen Welt 36 pt. 2: 10891112.
Graf, Fritz. 1985. Nordionische Kulte. Rome: Schweizerisches Institut in
Rom.
Graf, Fritz, and Sarah Iles Johnston. 2007. Ritual Texts for the Afterlife:
Orpheus and the Bacchic Gold Tablets. London and New York:
Routledge.
Guthrie, W. C. K. 1975. Platothe Man and His Dialogues: Earlier Period.
(AHistory of Greek Philosophy IV). Cambridge: Cambridge University
Press.
Guthrie, W. C. K. 1981. Aristotle: An Encounter. (A History of Greek
Philosophy VI). Cambridge: Cambridge University Press.
Gow, A. F. S., and D. L. Page. 1965. The Greek Anthology. Hellenistic
Epigrams. Cambridge: Cambridge University Press.
Gutzwiller, Kathryn J. 1998. Poetic Garlands: Hellenistic Epigrams in
Context. Berkeley: University of California Press.
224 Bi bl i ography
Habicht, Christian. 1970. Gottmenschentum und griechische Stdte,
2nd ed. Munich: Book.
Habicht, Christian. 1988. Hellenistic Athens and Her Philosophers.
Princeton: Princeton University Press.
Haldane, J. A. 1963. A Paean in the Philoctetes. Classical Quarterly
13: 5356.
Hammond, N. G. L., and G. T. Grifth. 1979. A History of Macedonia II.
Hammond, N. G. L., and F. W. Wallbank. 1988, A History of Macedonia
III. Oxford: Oxford University Press.
Harding, Phillip. 2006. Didymos: On Demosthenes. Clarendon Ancient
History Series, translation, text, and commentary. Oxford: Oxford
University Press.
Harvey, A. E. 1955. The Classication of Greek Lyric Poetry Classical
Quarterly 49: 157175.
Harward, J. 1932. The Platonic Epistles. Cambridge: Cambridge
University Press.
Heath, Malcolm. 1988. Receiving the Kmos, the Context and
Performance of Epinician. American Journal of Philology 109:
180195.
Heibges, J. S. 1912. Hermippos (6). RE 8.1 col. 848852.
Heiland, H. 1925. Aristoclis Messenii reliquiae. Giessen: Otto Meyer.
Henrichs, Albert. 1993. The Tomb of Aias and the Prospect of Hero
Cult in Sophokles. Classical Antiquity 12: 165180.
Herman, Gabriel. 1987. Ritualised Friendship and the Greek City.
Hinds, Stephen. 1998. Allusion and Intertext: Dynamics of Appropriation
in Roman Poetry. Cambridge: Cambridge University Press.
Hooker, J. T. 1988. The Cults of Achilles. Rheinisches Museum
131: 17.
Hopkinson, N. 1984. Callimachus: Hymn to Demeter. Cambridge:
Cambridge University Press.
Hutchinson, G. O. 2001. Greek Lyric Poetry: A Commentary on Selected
Larger Pieces. Oxford: Oxford University Press.
Immisch, O. 1906. Ein Gedichte des Aristoteles. Philologus 65: 123.
Bi bl i ography 225
Jaeger, W. 1927. Aristotles Verses in Praise of Plato. Classical
Quarterly 21: 1317.
Jaeger, W. 1948. Aristotle: Fundamentals in the History of His Develop-
ment, trans. R. Robinson, 2nd ed. Oxford: Oxford University Press.
Janko, Richard. 1982. Homer Hesiod and the Hymns: AStudy in Diachronic
Development. Cambridge: Cambridge University Press.
Janko, Richard. 1992. The Iliad: ACommentary. Books 1316. Cambridge:
Cambridge University Press.
Jebb, R. C. 1932. Sophocles. The Plays and Fragments VI: The Philoctetes.
Kppel, Lutz. 1992. Paian: Studien zur Geschichte einer Gattung.
Untersuchungen zur antiken Literatur und Geschichte 37. Berlin:
W. de Gruyter.
Kppel, Lutz, and R. Kannicht. 1988. Noch einmal zur Frage
Dithyrambos oder Paian? Zeitschrift fr Papyrologie und
Epigraphik 73: 1924.
Kamerbeek, J. C. 1980. The Plays of Sophocles VI: The Philoctetes. Leiden:
E. J. Brill.
Kosmetatou, Elizabeth. 2003. The Attalids in the Troad An Addendum:
An Episode in the Perils of the Aristotelian Corpus. Ancient Society
33: 5360.
Kurke, Leslie. 1991. The Trafc in Praise. Ithaca and London: Cornell
University Press.
Lefkowitz, M. R. 1981. The Lives of the Greek Poets. Baltimore: The Johns
Hopkins University Press.
LeVen, Pauline. 2008. The Many-Headed Muse: Tradition and Innovation
in Fourth-Century B.C. Greek Lyric Poetry. PhD. dissertation,
Princeton University / Paris IV, Sorbonne.
Liapis, V. 1996. Double Entendres in Attic Skolia: The Etymology of
Skolion. Eranos 94: 111122.
Lincoln, Bruce. 2007. Religion, Empire, and Torture: The Case of
Achaemenian Persia, with a Postscript on Abu Ghraib. Chicago:
University of Chicago Press.
Linforth, I. M. 1956. Philoctetes the Play and the Man. University of
California Publications in Classical Philology 15.3. 95156.
226 Bi bl i ography
Lissarrague, F. 1990. Lautre guerrier. Paris: La Dcouverte.
Lord, Carnes. 1986. On the Early History of the Aristotelian Corpus.
American Journal of Philology 107: 137161.
Lowe, N. J. 2007. EpinicianEidography. InPindars Poetry, Patrons, and
Festivals, ed. Simon Hornblower and Catherine Morgan, 167176.
Lloyd-Jones, Hugh, and P. Parsons. 1983. Supplementum Hellenisticum.
Berlin: W. de Gruyter.
Lloyd-Jones, Hugh, and N. G. Wilson. 1990. Sophoclis Fabulae. Oxford:
Clarendon Press.
Lobel, E., and D. L. Page, eds. 1955. Poetarum Lesbiorum fragmenta.
Oxford: Clarendon Press.
Lynch, J. P. 1972. Aristotles School. A Study of a Greek Educational
Institution. Berkeley: University of California Press.
Ma, John. 2000. Antiochus III and the Cities of Western Asia Minor.
Macher, Engelbert. 1914. Die Hermiasepisode imDemostheneskommentar
des Didymos. Brnn.
Maehler, H. 19821997. Die Lieder des Bakchylides, 2 vols. Leiden:
E. J. Brill.
Markle, Minor M., III. 1976. Support of Athenian Intellectuals for
Philip: A Study of Isocrates Philippus and Speusippus Letter to
Philip. Journal of Hellenic Studies 96: 8099.
Martindale, Charles. 2005. Latin Poetry and the Judgment of Taste: An
Essay in Aesthetics. Oxford: Oxford University Press.
Mastronarde, D. 1994. Euripides. Phoenissae. Cambridge: Cambridge
University Press.
McLachan, Bonnie. 1993. The Age of Grace: Charis in Early Greek Poetry.
Princeton: Princeton University Press.
Mensching, E. 1963. Favorin von Arelate, Vol. 1 Berlin: W. de Gruyter.
Mejer, Jrgen. 1978. Diogenes Laertius and His Hellenistic Background.
(Hermes Einzelschriften 40.) Wiesbaden: Steiner.
Mikalson, Jon D. 1998. Religion in Hellenistic Athens. Berkeley:
University of California Press.
Bi bl i ography 227
Mitchell-Boyask, Robin. 2007. The Athenian Aesklepion and the End of
the Philoctetes. Transactions of the American Philological Association
137: 85114.
Milns, R. D. 1994. Didymea. In Ventures into Greek History, ed. Ian
Worthington, 7088. Oxford: Oxford University Press.
Moraux, Paul. 1951. Les listes anciennes des ouvrages dAristote. Louvain:
ditions universitaires de Louvain.
Moraux, Paul. 1973. Der Aristotelismus bei den Griechen I: von Andronikos
bis Alexander von Aphrodisias. Die Renaissance des Aristotelismus
im I. Jh. v. Chr. Berlin: W. de Gruyter.
Momigliano, Arnaldo. 1987. On Pagans Jews and Christians. Scranton,
Pa.: Wesleyan University Press.
Momigliano, Arnaldo. 1993. The Development of Greek Biography, 2nd
ed. Cambridge, Mass.: Harvard University Press.
Most, Glenn. 1981 Sappho Fr. 16.67 L-P. Classical Quarterly
31: 1117.
Mulvaney, C. M. 1926. Notes on the Legend of Aristotle. Classical
Quarterly 66: 155167.
Nagy, Gregory. 1990. Pindars Homer: The Lyric Possession of an Epic Past.
Baltimore: The Johns Hopkins University Press.
Nickau, K. ed. 1966. [Ammonius] Gramm., De adnium vocabulorum
differentia. Leipzig: Teubner.
Nicolai, Roberto. 2004. Studi su Isocrate. Quaderni dei Seminari Romani
di Cultura Greca 7. Rome: Quasar.
Norden, E. 1913. Agnostos Theos: Untersuchungen zur Formengesschichte
religiser Rede. Leipzig: Teubner.
OSullivan, L.-L. 1997. Athenian Impiety Trials in the Late Fourth
Century B. C. Classical Quarterly 47: 136152.
Owen, G. E. L. 1983. Philosophic Invective. Oxford Studies in Ancient
Philosophy 1: 125.
Page, Denys. 1955. Sappho and Alcaeus. Oxford: Clarendon Press.
Page, Denys, ed. 1962. Poetae melici Graeci. Oxford: Clarendon Press.
Page, Denys, ed. 1975. Epigrammata Graeca. Oxford: Clarendon Press.
Page, Denys. 1981. Further Greek Epigrams. Cambridge: Cambridge
University Press.
228 Bi bl i ography
Parker, Robert. 1996. Athenian Religion. Oxford: Oxford University
Press.
Parker, Robert. 1998. Pleasing Thighs: Reciprocity in Greek Religion.
In Reciprocity in Ancient Greece, ed. C. Gill, N. Postlethwaite, and
R. Seaford, 105125. Oxford: Oxford University Press.
Parker, Robert. 1999. Through a Glass Darkly: Sophocles and the
Divine. In Sophocles Revisited, ed. J. Grifn, 1130. Oxford: Oxford
University Press.
Pavese, C. 1961. Aristotele e i loso ad Asso. La parola del passato
16: 113119.
Pearson, Lionel, and Susan Stephens. 1983. Didymi in Demosthenem
commenta. Stuttgart: Teubner.
Pelliccia, Hayden. 2002. The Interpretation of Iliad 6.1459 and the
Sympotic Contribution to Rhetoric. Colby Quarterly 38: 197230.
Penella, Robert J. 2007. Man and the World: The Orations of Himerius.
Berkeley: University of California Press.
Perkins, David. 1992. Is Literary History Possible? Baltimore: The Johns
Hopkins University Press.
Pfeiffer, Rudolph. 1968. History of Classical Scholarship I: From the
Beginnings to the End of the Hellenistic Age. Oxford: Oxford
University Press.
Pickard-Cambridge, A. W. 1988. The Dramatic Festivals of Athens, 2nd
ed. revised with a new supplement by J. Gould and D. M. Lewis.
Oxford: Clarendon Press.
Plezia, Marianus, 1977. Aristotelis privatorum scriptorum fragmenta.
Leipzig: Teubner.
Post, L. A. 1921. Thirteen Epistles of Plato. Oxford: Clarendon Press.
Powell, J. U., ed. 1925. Collectanea Alexandrina. Oxford: Clarendon
Press.
Pucci, Pietro. 1987. Odysseus Polutropos: Intertextual Readings in the
Odyssey and the Iliad. Ithaca and London: Cornell University Press.
Pucci, Pietro. 1994. Gods Interventions and Epiphany in Sophocles.
American Journal of Philology 115: 1516.
Pulleyn, Simon. 1997. Prayer in Greek Religion. Oxford: Oxford
University Press.
Bi bl i ography 229
Race, William. 1982. The Classical Priamel from Homer to Boethius.
Leiden: Brill.
Race, William. 2007. Rhetoric and Lyric Poetry. In A Companion to
Greek Rhetoric, ed. I. Worthington, 509525. Oxford: Blackwell.
Redeld, James. 1974. Nature and Culture in the Iliad. Chicago:
University of Chicago Press.
Redeld, James. n.d. Theophrastus Will and Platos Academy.
Reitzenstein, R. 1883. Epigram und Skolion. Giessen: Ricker.
Renehan, R. 1982. Aristotles as Lyric Poet: The Hermias Poem. Greek
Roman and Byzantine Studies 23: 25174.
Renehan, R. 1991. Aristotles Elegiacs to Eudemus. Illinois Classical
Studies 16: 255267.
Rhodes, P. J., and R. Osborne. 2003. Greek Historical Inscriptions
404323 BC. Oxford: Oxford University Press.
Rose, V. 1863. Aristoteles pseudepigraphus. Leipzig: Teubner.
Rose, V. 1886. Aristotelis qui ferebantur librorum fragmenta. Leipzig:
Teubner.
Rudhardt, Jean. 1960. La Dnition du dlit dimpit daprs la
lgislation attique. Museum Helveticum 17: 87105.
Ruina, D. T. 1986. Theocritus of Chios Epigram Against Aristotle.
Classical Quarterly 36: 531534.
Russell, D. A., and N. G. Wilson. 1981. Menander Rhetor. Oxford:
Clarendon Press.
Rusten, Jeffrey. 1987. Review of Pearson-Stephens, Classical Philology
82: 265269.
Rutherford, Ian. 2001. Pindars Paeans: A Reading of the Fragments with
a Survey of the Genre. New York: Oxford University Press.
Santoni, Anna. 1991. LInno di Aristotele per Ermia di Atarneo.
In La Componente Autobiographica nella poesia Greca e Latina, ed.
G. Arrighetti and F. Montanari, 179195. Pisa: Giardini.
Severyns, A. 1938. Recherches sur la Chrestomathie de Proclos, vol. 2.
(Bibliothque de la facult de philosophie et lettres de luniversit
de Lige fascc. 78). Paris: E. Droz.
Schachter, Albert. 1981. Cults of Boiotia. 4 vols. London: Institute of
Classical Studies.
230 Bi bl i ography
Schmid, Ulrich. 1964. Die Priamel der Werte im Griechischen von Homer
bis Paulus. Wiesbaden: Harrassowitz.
Schmidt, Maurice. 1854. Didymi Chalcenteri Grammatici Alexandrini
Fragmenta. Leipzig: B. G. Teubner.
Schroeder, Otto. 1925. Aristoteles als Dichter. Neue Jahrbcher fr
Wissenschaft und Jugendbildung, n.s. 1: 3135.
Schrder, Stephan. 1999. Geschichte und Theorie der Gattung Paian.
Stuttgart: Teubner.
Scullion, S. 2003. Euripides and Macedon, or the Silence of the Frogs.
Classical Quarterly 53: 389400.
Shapiro, H. A. 1993. Personications in Greek Art. Zrich: Akanthus.
Slings, S. R. 1995. Protreptic in Ancient Theories of Philosoph-
ical Literature. In Greek Literary Theory after Aristotle, ed.
J. G. J. Abbenes, S. R. Slings, and I. Sluiter, 173192. Amsterdam:
VU University Press.
Slings, S. R. 1999. Plato Clitophon. Cambridge: Cambridge University
Press.
Smyth, H. W. 1906. Greek Melic Poets. LondonandNewYork: Macmillan.
Solmsen, F. 1960. Zur Theologie im grossen Aphrodite-Hymnus.
Hermes 88: 113.
Sourvinou-Inwood, Christiane. 2003. Tragedy and Athenian Religion.
Lanham, Md.: Routledge.
Teodorsson, S. T. 1987. The Etymology of Scolion. Eranos 87: 127132.
Tod, Marcus N. 1985. Greek Historical Inscriptions, 2 vols. (Arno reprint
of the 2nd ed., Oxford, Clarendon Press, 1946).
Trampedach, Kai. 1994. Platon, die Akademie und die zeitgenssische
Politik. Stuttgart: Steiner.
Tsagalis, Christos. 2008. Inscribing Sorrow: Fourth-century Attic Funerary
Epigrams. Berlin: W. de Gruyter.
Velardi, Roberto. 1991. Le origini dellinno in prosa. In L inno tra
rituale e letteratura nel mondo antico (AION 13), Albio Cassio ed.,
205231. Roma: Gruppo Editoriale Internazionale.
Vlker, Harald. 2003. Himerios, Reden und Fragmente. Wiesbaden:
Reichert.
von der Muhl, P. 1918. Hermias. RE Suppl. III col. 11261130.
Bi bl i ography 231
Webster, T. B. L. 1970. Sophocles Philoctetes. Cambridge: Cambridge
University Press.
Wehrli, F. 1974. Hermippos der Kallimacheer. Die Schule d. Arist.,
Supplbnd. 1. Basel: Schwabe.
Weiskopf, Michael, 1989. The So-called Great Satraps Revolt, 366360
B.C. Historia Einzelschriften 63. Stuttgart: F. Steiner.
Wellek, Ren and Warren, Austin. 1956. Theory of Literature, 3rd ed.
New York: Harcourt, Brace & World.
West, M. L. 1982. Greek Metre. Oxford: Clarendon Press.
West, M. L., ed. 1992. Iambi et elegi Graeci. 2nd ed. Oxford: Clarendon
Press.
Wilamowitz-Moellendorff, Ulrich von. 1893. Aristoteles und Athen.
Vol. 2. Berlin: Weidmann.
Wilamowitz-Moellendorff, Ulrich von. 1920. Platon. Berlin: Weidmann.
Willi, Andreas. 2003. The Languages of Aristophanes. Oxford: Oxford
University Press.
Wimsatt, W. K. and Monroe Beardsley. 1964. The Intentional Fallacy.
In The Verbal Icon, 121. Lexington: University of Kentucky Press.
Wagman, Robert S. 1995. Inni di Epidauro. Biblioteca di studi antichi 75.
Pisa: Giardini.
Wormell, D. E. W. 1935. The Literary Tradition Concerning Hermias of
Atarneus. Yale Classical Studies 5: 5792.
Yatromanolakis, Dimitrios. 1999. Alexandrian Sappho Revisited.
Harvard Studies in Classical Philology 99: 179195.
Young, David C. 1983. Pindar Pythians 2 and 3: Inscriptional Pote and
the Poetic Epistle. Harvard Studies inClassical Philology 87: 3148.
Yunis, Harvey. 1997. What Kind of Commentary is the peri
Dmosthenous of Didymus? Archiv fr Papyrusforschung 3:
104955.
Zeller, Eduard. 1897. Aristotle and the Earlier Peripatetics. 2 vols. London
and New York: Longmans, Green.
Zuntz, Gnther. 2005. Griechische philosphische Hymnen, ed. H. Cancik
and L. Kppel. Tbingen: Mohr Siebeck.
G E N E R A L I N D E X
abstractions, hymns to 87, 91,
934, 97, 122, 140, 162,
197 n. 8, 199 n. 7
Academy 214, 35, 40, 47, 161,
1678, 177 n. 21
Achilles 2, 57, 1213, 127,
1445, 193 n. 24,
205 n. 28
agalma 142, 158; see also
memorials
Alexandrian library, see Library
of Alexandria
altars 83, 93, 132, 160, 164,
169, 194 n. 31
Anactoria 1201, 145, 148,
203 n. 9
aoidimos (celebrated
in song) 7, 142, 147,
149, 150
Apellicon of Teos 168, 169
aphthitos (unwithering) 143,
148, 150, 1534
apologia (defense speech) 603,
65, 66, 85
of [Aristotle] 601, 66,
181 n. 7, 188 n. 39
areta (aret), denition of 4,
174 n. 7
Aristocles of Messene 19, 378,
40, 41, 627, 169
[Aristippus] On Ancient Luxury
62, 86
Asclepius 129, 131, 206 n. 36
Artaxerxes II 18, 25
Artaxerxes III Ochus 25, 51
Atarneus 2, 6, 1015, 1820,
225, 39, 42, 47, 567,
82, 145, 154, 166
Athenaeus 2, 12, 5463, 67, 69,
70, 77, 79, 829, 115,
125, 1689, 1701,
186 n. 21, 195 n. 38
Attic skolia 546, 59, 69, 70,
734, 7780, 869, 96,
233
234 General I ndex
Attic skoliaContd
101, 114119, 145,
165, 171
name of 55
Augustus, name of 152, 154,
155, 210 n. 24
auxanein 143, 1524, 209 n. 20,
210 n. 22
Bacchylides 73, 789, 138,
1413, 149, 152, 153,
156, 158, 159, 190 n. 4,
192 n. 22
bebaios (steadfast, stable)
142, 1579, 164, 167,
211 n. 4
book epigrams, see epigrams
Callimachus 53, 59, 789,
109, 192 n. 20,
196 n. 44; see also, hymns,
mimetic
Callisthenes of Olynthus 20,
234, 4353, 82
Castor and Pollux, see
Dioskouroi
Craterus of Macedon 57, 58,
186 n. 23
defense speech see apologia
deictics 32, 345, 38, 43,
53, 103
Delphi 18, 29, 31, 33, 38, 43,
53, 57, 61, 81, 83, 158,
162, 163
Demades 83, 195 n. 35
Demeter 94, 108
rites of 62, 187 n. 33,
201 n. 37
Demetrius of Phaleron 66,
188 n. 34
Demetrius the Besieger
(Poliorcetes) 667, 83,
194 n. 31, 199 n. 16
Demochares 22, 67, 189 n. 45
Demosthenes 3, 22, 2526, 67,
80, 85
Didymus 3, 12, 1724, 32,
368, 45, 46, 48, 59,
634, 86, 126, 152, 166,
169171, 186 n. 21, n. 27
On Lyric Poetry 59, 170
Diogenes Laertius 3, 12, 23, 29,
3440, 59, 61, 62, 86,
10810, 125, 164, 169,
170, 181 n. 7
Dioskouroi (Castor and Pollux)
6, 1445, 208 n. 7
dirge (thrnos) 71, 75, 81, 867,
191 n. 10
dithyramb 749, 87, 193 n. 23
dithyrambic style 1256,
133, 1389, 207 n. 2,
208 n. 9
do ut des 5, 156
elegiacs 29, 31, 33, 36, 513, 87,
109, 1167, 119, 1603,
169, 170, 196 n. 44
enargeia (vividness) 53, 93, 162
encomium (enkmion) 67, 43,
457, 724, 76, 78, 89,
General I ndex 235
94, 148, 152, 154, 160,
165, 1701, 190 n. 5,
193 n. 24, 209 n. 14
name of 70, 73
of Hermias (?) 457
of Plato 47, 160, 184 n. 6
epigrams 2730, 33, -36 81, 170
book 346, 4043, 180 n. 6
epinician 7, 723, 78, 87, 97,
99, 101, 124, 141, 142,
149, 190 n. 5, 204 n. 17
name of 73
epiphany 121, 125, 127, 1334,
145, 206 n. 38
epitaphios logos 61, 210 n. 3
name of 61
epitaphs 302, 359, 52
ctional 38, 181 n. 14,
210 n. 25
epithalamia (wedding songs)
77, 191 n. 15
epithet 5, 33, 61, 84, 91, 934,
100, 105, 1078, 114,
1212, 1246, 1303,
13841, 146, 1501,
207 n. 2
transferred 84, 139
epos/melos distinction 312,
5153, 110, 126; see also
verse
Erastos 23, 166
ethos 14, 55, 1134, 126, 134
Euboulides, of Megara (?)
6365
Eubulus 18, 35, 37, 38,
176 n. 10
Eudemus of Rhodes 1604,
169, 170
Euripides 12, 133, 13841, 143,
153, 182 n. 21, 197 n. 8
fame, see kleos
friend ( philos), friendship
19, 20, 31, 50, 64 142,
152, 1545, 1578,
1624, 1679, 172;
see also xenia
genre 4, 6, 28, 34, 35, 38, 46,
50, 5460, 6979, 868,
97, 100, 104, 110, 111,
114, 124, 134, 141, 148,
151, 156, 165, 170
dependence on context
879
Heracles 2, 5, 6, 12, 82, 116,
123, 12534, 1426,
150, 154, 207 n. 5,
208 n. 7
Hermippus of Smyrna
7, 54, 5760, 63, 65, 79,
86, 170
hero cult 82, 83, 145, 164,
194 n. 32
Hermotimus of Pedasa 412
hexameters, dactylic 7, 31, 53,
100, 10410, 170
humne (to sing) 84, 153,
210 n. 2
husteron proteron 1516
236 General I ndex
hymn (n.) (humnos) 4, 6, 756,
171; see also humne
Homeric 100, 1056,
mimetic 53
name of 80
impiety trials 8, 569, 61,
62, 657, 1545,
194 n. 30
inscription 9, 1720, 29, 31, 33,
36, 43, 53, 162, 163
integration of poetic tradition
105, 143, 172
kleos (fame, glory) 33, 130,
142, 1479, 153
kmos (revel song) 6, 73,
190 n. 4
Koriskos 23, 1668
lament see dirge
Library of Alexandria 3, 59, 66,
714, 77, 79, 89, 108,
166170
Lyceum 23, 54, 81, 123,
164, 169
common meals (sussitia) 54,
56, 86, 87, 185 n. 15
Lycon the Pythagorean 623,
188 n. 36
lyric poetry 2, 10, 14, 27, 28, 33,
512, 59, 70, 85, 86, 110,
113, 138, 159
genres of 7180, 8990,
191 n. 9
transcriptions of 3, 103, 113,
159; see also song,
epos/melos distinction
Lysander of Sparta 57, 58, 80,
834, 162
melos (song) 51, 73, 76; see also
song, epos/melos
distinction
memorials 34, 378, 45, 61, 88,
129, 156, 160, 186 n. 23,
210 n. 3
memory personied, see
Mnamosuna
Mentor of Rhodes 25, 30
mimsis 76
mnma / mnmaion, see
memorials
Mnamosuna (Memory)
1501, 155
monuments see memorials
morph (shape) 124, 144
Muses 2, 7, 81, 143, 149155,
164, 193 n. 24
names, proper 1214, 39, 45,
120, 161 , 163, 181 n. 13
naming 16, 33, 156, 184 n. 13,
210 n. 25
Neleus of Skepsis 168, 169
paeans 8, 5562, 69, 70, 72,
7591, 947, 101, 109,
110, 116, 160, 162, 165,
170, 206 n. 36
at libations 96, 165,
195 n. 37
General I ndex 237
name of 72
sympotic 88, 101
phthinein (withering) 143
Pindar 7, 20, 73, 78, 97105,
152, 190n7, 198 n. 12
Plato 1924, 40, 41, 47,
646, 71, 7480, 858,
94, 97, 110, 1147, 124,
15869
Symposium 94, 158
Poetics of Aristotle 756, 137
polumokhthos (of much toil)
13841
priamel 5, 6, 91, 95, 100,
11920, 154, 167,
198 n. 10, n. 12
Prodicus 94, 207 n. 5
Heracles at the Crossroads
1235, 130
protreptic 121, 1234, 126,
133, 141
Proxenus of Atarneus 223
Pythias, daughter of Aristotle,
23, 42
refrains 5758, 72, 79, 86,
185 n. 19, 192 n. 22
re-performance 4, 316, 71, 97,
1014, 14751, 1567,
165, 171, 180 n. 4
reverence (eusebeia/sebas)
2, 99, 132, 150, 151, 154,
1634
Sappho 778, 1134, 11721,
149, 167
seal (sphrgis) 208 n. 13
sebas, see reverence
Simonides 29, 3334, 41, 81,
116, 122, 123, 167
skolion 55, 59, 74, 80, 86, 165,
190 n. 5, 195 n. 38; see
also Attic skolia
song (melos) xvi, 3, 28,
312, 512, 54, 70,
81, 101
books 117, processional 104
Sophocles of Sounion 67
Speusippus (of Athens) 212,
47, 177 n. 21, 184 n. 6
sphrgis see seal
suggramma ([prose]
composition) 46, 49,
50, 52
sussitia (common meals) see
under Lyceum
symposia 47, 54, 88, 89, 101,
1146, 148, 185 n. 15,
195 n. 37
literary 47, 94
Theocritus of Chios 27, 29,
3542
Theognis 1167
Theophrastus of Eresos 23,
667, 164, 1679
Theopompus of Chios 18, 21,
245, 39, 42, 47, 50
thrnos see dirge
toil 92, 1214, 130, 13845,
204 n. 17
unwithering fame, see aphthitos
238 General I ndex
verse (epos) 117, 1634;
see also epos/melos
distinction
vividness see enargeia 53, 162
virtue, see areta
waxing, see auxanein
withering, see phthinein
xenia (guest friendship), xenos
(guest-friend) 20,
3031, 155; see also Zeus
xenios
Zeus xenios (god of
guest-friends) 30, 151,
154, 165
I N D E X O F P A S S A G E S D I S C U S S E D
AELIAN
2.19: 51
14.1: 43
ALEXINUS (SH)
Fr. 40: 186 n. 23
AMBRYON (Bryon?)
On Theocritus: 37
ANON. skolia (PMG)
890: 1156
894: 145
ARIPHRON (PMG)
813: 5758, 9197, 116
ARISTOCLES of Messene
(Chiesara)
Fr. 2.5: 6365
Fr. 2.7: 41, 182 n. 23
Fr. 2.8: 62
Fr. 2.9: 19
Fr. 2.12: 378
Fr. 2.13: 169
ARISTOPHANES
Frogs
50: 41
1401: 181 n. 13
10346: 203 n. 15
ARISTOTLE
Eudemian Ethics
1219b89: 190 n. 6
Nicomachean Ethics
1100b1214: 206 n. 34
1100b1921: 176 n. 7
1100b303: 176 n. 7
Fragments (Rose)
Fr. 61517: 43
Fr. 641: 181 n. 14
Fr. 645: see [Aristotle]
Fr. 671: 108109, 170
Fr. 672: 109, 170
Fr. 674: 2933
Fr. 673: 1604
Fr. 675 (= 842 PMG): passim
Politics
1283a7: 198 n. 11
Poetics
239
240 I ndex of Passages Di scussed
ARISTOTLEContd
1447a1427: 76
1448b427: 76
1448b257: 76
1449b2150b20: 75
1454b2: 131
458a1859a17: 206
1459a58: 137, 206 n. 1
1459a19: 139
1458a2223: 206 n. 1
1459a58: 137, 206 n. 8
Rhetoric
1366a2368a37: 4, 124
1367a13: 207 n. 4
1367b269: 74
1368a: 154
1388b21: 74
1394b13: 116
1400b58: 191 n. 10
1406b2: 139
1409a13: 191 n. 12
1413b1216: 207 n. 2
[ARISTOTLE]
Fr. 645 Rose: 6063, 66
ARISTOXENOS (Wehrli)
Fr. 125: 77
ATHENAEUS
610f: 67
692f: 54,
693f-694c: 55
694a: 54, 55
694c-695e: 55
695f-696a: 55
696a-697b: 55, 56
696e: 5658
697a: 57
BACCHYLIDES (Maehler)
Epin. 1.17884: 1413, 158
2.13: 190 n. 4
3.9092: 153
Fr. 23: 192 n. 21
Fr. dub. 56: 152
CALLIMACHUS (SH)
293: 192 n. 22
CALLISTHENES of Olynthus
(124 FGrH)
Fr. 2: 4851
CLEARCHUS (Werhli)
2a: 184 n. 6
DEMETRIUS of Magnesia
(Mejer)
Fr. 15: 175 n. 5, 174 n. 10
DEMOCHARES (75 FGrH)
Fr. 2: 199 n. 16
DEMOSTHENES
Fourth Philippic 32: 25
DICHAEARCHUS (Wehrli)
Fr. 8889: 767
DIDYMUS
On Demosthenes
(Pearson-Stephens)
4.6065: 18
5.23: 18
5.534: 166
5.5363: 177 n. 13
5.63: 22, 64
5.64: 46, 183 n.3
6.2: 49, 157
6.1516: 20
6.2122: 171, 214 n. 32
6.4649: 378
I ndex of Passages Di scussed 241
DIOGENES LAERTIUS
2.42: 10910
3.2: 184 n. 6
3.45: 188 n. 35
5.3: 40
5.5.12: 164
5.6: 2930
5.6.10: 86
5.27: 108, 170, 175
DURIS of Samos (76 FGrH):
Fr. 71: 834, 194 n. 33
Erythraean paean to Asclepius
(943 PMG): 198 n. 15
EURIPIDES
Iphigeneia at Aulis
288: 176 n. 6
568: 203 n. 13
Orestes
8078: 203 n. 11
Medea
82445: 199 n. 22
Phoen.
7845: 13940
Fragments (TrGF)
Fr. 645a: 1401, 153
Fr. 734: 133
Fr. 916: 140, 143
EURIPIDES (?) PMG
755: 196 n. 44
EUSEBIUS
Praep Ev. (Mras)
15.2.11:19
GHI (Tod)
165: 1920
187: 43, 183 n. 27
HERMIPPUS of Smyrna (1026
FGrH)
Fr. 30: 186 n. 25
Fr. 31: 184 n. 9
Fr 65: 59
Fr 67: 59
HERMOCLES (CA)
pp. 1735 Powell: 83,
194 n. 31
HERODOTUS
1.8687: 51
8.1046: 423
HESIOD
Theogony:
116117: 108, 201 n. 30
117: 107
120: 94
9079: 199 n. 20
Works and Days:
28792: 1212
76063: 93, 197 n. 5
HIMERIUS (Colonna)
40.23: 12, 20
40.40:13
HOMER
Iliad
4.5961: 200 n. 24
6.358: 147
9.189: 7
13.6367: 198 n. 12
Odyssey
24.1968: 1467
[HOMER] Hymns
30 (to Gaia): 1059
HYPERIDES
Funeral Oration
6.21: 194 n. 31
242 I ndex of Passages Di scussed
IBYCUS (PMG)
S151: 1489
ION of Chios (PMG)
742: 93, 95
ISOCRATES
Evagoras
8: 467
70: 205 n. 27
To Demonicus
4950: 205 n. 27
Panegyricus
159: 203 n. 15
LICYMNIUS (PMG)
771: 207 n. 2
LUCIAN
A Slip of the Tongue in
Salutation
6.26: 92
LYCUYRGUS
Against Leocrates 1023:
203 n. 15
MENANDER Rhetor (Russell
and Wilson)
331.20332.7:
192 n. 18
414.237: 208 n. 7
MESOMEDES (CA)
Fr. 35: 197 n. 6
PINDAR
Isth. 8.48: 209 n. 19
Nem. 4.78: 190 n. 4
8.4042: 152
Ol. 10.95: 153
Ol. 14: 97104
PLATO
Apol.
19C: 66
38A: 202 n. 8
Laws
601E: 211 n. 9
700A-E: 745
700C: 75, 190 n. 9
Phaedo
60C-61B: 110
69C: 181 n. 13
Phaedrus
275C-277D: 211 n. 4
Symp.
177A-B: 94, 190 n. 5
178A-C: 945
208D-E: 204 n. 26, 205 n. 28
209C: 158
PLATO (?) Epistles
3.315B: 200 n. 27
3.322D: 167
3.322E: 167
6: 1669
PLUTARCH
On Exile
603C: 39
[PLUTARCH] De Musica 1134d-e:
185 n. 19
PRODICUS
Choice of Heracles (84 B 2
DK), see Xenophon,
Mem. 2.12134
PROCLUS
Chrestomathy (apud
Proclus Bib.)
319b: 78
I ndex of Passages Di scussed 243
SAPPHO (Voigt)
16: 11721
SIMONIDES (IEG/PMG)
11 (IEG): 193 n. 24
531 (PMG): 81, 193 n. 24
579 (PMG): 122
604 (PMG): 197 n. 5
615 (PMG): 197 n. 8
651 (PMG): 116
SIMONIDES (?) A.P. 7.258 334
SOCRATES (?) 1 IEG: 202 n. 32
2 IEG: 109110
SOPHOCLES
Test. 73a (TrGF): 206 n. 36
Philoctetes
82732: 190 n. 8, 206 n. 36
14091471: 12734
1420: 1301
1437: 131
1442: 130, 142
14434: 132
STRABO
C608: 77, 191 n. 14
C610: 22, 178 n. 25
THEOCRITUS of Chios (SH)
Fr. 738: 541
THEOGNIS (IEG)
2556: 1167, 119
THEOPOMPUS (115 FGrH)
Fr. 250: 18, 245, 39
Fr. 291: 18, 50
TYRTAEUS (IEG)
12: 211 n. 9
XENOPHON
Memorabilia
1.2.1: 182 n. 19
2.1.2134: 123
2.1.24: 198 n. 12
2.1.32: 198 n. 14
2.1.334: 209 n. 16
Symposium
8.2830: 208 n. 7

Aristo Tile As A Poet

Uploaded by

Copyright:

Available Formats

Aristo Tile As A Poet

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Aristo Tile As A Poet

Uploaded by

Copyright:

Available Formats

ARISTOTLE

e x' A tceo cou

vx,oo e . tou y,ov eu y e.

v xe t evyy u,otv e yv txet

xxtvv H,o v xoo,v peot. u,

x yxuov etv o).

yx tov would t the size of the gap and make sense,

yx tov, this is not of course

.].e x(e`t) [xeu-] 65

o tyo c[t ` e x` yv] e ,x( yv) y

] ox[yy c' e]u x t ,` o[] xo` u[ ] t- 15

xxx yvexo A. tvo o

txe tot e` v xte t

..v ,`t xo u y eou xe`t x y ,` o

..v , opv xv et eve, y

couv, u o pouv, e ..' ou x y

v oo tyv). Rutherford is right, I think, to infer

' eu x [t Het]e v); Athenaeus

oe), and argues that it

..e 00v e v0, otot x ,yt y` vv

t vet); being very

vxet ` evxe p,oxo t,

x ce txe e ..` e evxv xe tet

v ou ,ev , ,uoxoov 0 vet e,` e 10

exoo tx v uv, Oe. te x 15

,eo to., t co toe xvc x ov

ox evo xuc tv e 0.v x,o tot e txev.

ouoet, v. 20). Pindar

0ove evxe e,oxt.

v0e e.0ex` e o ty, 15

.o' Ou ,evo u e ox,vxo,

..y v yoo' e otc y.

u0 0.ov, which is found only here in Greek literature.

.y te, no. 146) and quotes

vx tve xo` u xo u At o ou .you xe`t x t

..ot xtv ). The poem cited by

,toxov e vc,`t 0vex ,

v eo0y, e ..` e e, eyey eu

y u0v, e t Ho tevxo 1410

xx,t0`t ox,ex uexo, 1425

o tot voot t p tou,

x.ep ` v ox,ex uexo,

..e evxe c ux, y

o0o y` e, ou c t, Soph. Fr. 410.1 Radt).

o. uo0o ptox` y 0vyxo t,

`t evx`t oe.,` e x toet,

o0o u x. y). The closest precedent I nd for this nuance

to0o. But unlike the

to0o to e ,x e, and this is

to0o) is adversative: tima,

ye.e, a monument or temple

to0o with which

-0txov). In portraying Life as controlling what

,ovt Hyv.o ty). The

-0txo), which might be expected in this

0txov, glory that does not wither away (e o-0 tvtv).

ouoet, venerating.) Yet

yot) had the technical sense of amplifying or

ye.e) he can leave behind.