Pavlk 2003 English Phoneticsand Phonology

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/354507445
A Theoretical Introduction to English Phonetics and Phonology
Book · January 2003
CITATIONS READS
16 2,709
1 author:
Radoslav Pavlík
27 PUBLICATIONS 76 CITATIONS
SEE PROFILE
All content following this page was uploaded by Radoslav Pavlík on 10 September 2021.
The user has requested enhancement of the downloaded file.

Radoslav Pavlík
A Theoretical Introduction to
English Phonetics and Phonology
Publikáciu finanþne sponzorovalo

Obþianske zdru!enie Lingos
Bratislava 2003
¤ PhDr. Radoslav Pavlík
Recenzenti: prof. PhDr. Ábel KráĐ, DrSc.

doc. PhDr. Pavol Kvetko
Vydal: RETAAS s r.o.

Prvé vydanie
Bratislava 2003
ISBN: 80-89113-08-7
EAN: 9788089113088
Contents
Contents
Introduction 7
1 Main Aspects of Communication 9

1.1 Language and Communication 9
1.2 Concrete and Abstract Sides of Language 13
1.3 Phonetics and Phonology 18
1.3.1 Phonetics 18
1.3.2 Phonology 22
1.3.2.1 The Definition of Phonology 22
1.3.2.2 Methods of Determining Phonemic Inventories 23
1.3.2.3 Phonological Oppositions 26
1.3.2.4 Distinctive Features 28
1.4 Transcription 32
1.4.1 The Definition of Transcription 32
1.4.2 Phonemic Transcription 34
1.4.3 Allophonic Transcription 36
1.5 Physiological Aspects of Speech and the Organs of Articulation 41
1.5.1 Respiratory Organs 42
1.5.2 Phonatory Apparatus 43
1.5.3 Resonators 44
1.5.4 Modulating Organs 45
1.6 Acoustic Aspects of Speech 46
1.6.1 Sound Waves 46
1.6.2 Acoustic Properties of Sounds 50
1.6.2.1 Fundamental Frequency 50
1.6.2.2 Intensity 52
1.6.2.3 Quantity (Duration) 53
1.6.2.4 Quality (Wave Structure) 54
1.7 Auditory Aspects of Speech 59
1.7.1 The Auditory System 59
1.7.1.1 The Peripheral Auditory System 59
1.7.1.2 The Internal Auditory System 61
1.7.2 The Perception of Sound 62
1.7.2.1 Pitch 62
1.7.2.2 Loudness 65
1.7.2.3 Qualitative (Auditory) Impression 66
1.7.2.4 Length 69
3
Contents
2 The Classification and Description of English Speech Sounds 71

2.1 Vowels 71
2.1.1 Definition and Classification of Vowels 71
2.1.2 Cardinal Vowels 74
2.1.3 The Length of English Vowels 75
2.1.4 The Description of English Distinctive Vowels 78
2.2 Diphthongs 83
2.2.1 Definition and Classification of Diphthongs 83
2.2.2 The Length and Prominence of English Diphthongs 85
2.2.3 The Description of English Distinctive Diphthongs 85
2.3 Triphthongs 89
2.3.1 Definition and Classification of Triphthongs 89
2.3.2 The Description of the English Triphthongs 90
2.4 Consonants 91
2.4.1 Definition and Classification of Consonants 92
2.4.2 The Length of English Consonants 95
2.4.3 The Description of English Distinctive Consonants 96
2.4.3.1 Obstruents 96
2.4.3.1.1 Plosives 96
2.4.3.1.2 Affricates 104
2.4.3.1.3 Fricatives 105
2.4.3.2 Sonorants 109
2.4.3.2.1 Nasal Sonorants 109
2.4.3.2.2 Oral Sonorants 112
3 The Word 116

3.1 Suprasegmental Features of Language 116
3.2 The Syllable 118
3.2.1 The Definition of the Syllable 118
3.2.2 Syllable Boundaries 122
3.2.3 Syllabic Consonants 125
3.3 Phonotactics 131
3.4 Elision within the Word 135
3.5 Pronunciation of Suffixes 142
3.5.1 Inflectional Suffixes 142
3.5.2 Derivational Suffixes 146
3.6 Word Stress 147
3.6.1 The Nature and Definition of Stress 147
3.6.2 Types of Word Stress 149
3.6.3 The Influence of Stress on the Clarity of Vowels 152
3.6.4 Stress Pattern of the Word 153
3.6.4.1 Stress in Simple Words 153
4
Contents
3.6.4.2 Stress in Derived Words 156

3.6.4.3 Stress in Compound Words 160
3.6.5 Variation of Word Stress 161
3.6.5.1 Word Class-Maintaining Variation of Stress 162
3.6.5.2 Word Class-Changing Variation of Stress 163
3.7 Variation of Phonemic Structure of Words 165
3.7.1 Word Class-Maintaining Variation of Phonemic Structure 165
3.7.2 Word Class-Changing Variation of Phonemic Structure 167
3.8 Assimilation within the Word 168
3.8.1 Allophonic Assimilation within the Word 169
3.8.2 Phonemic Assimilation within the Word 170
3.8.2.1 Historical and Contextual Phonemic Assimilation 170
3.8.2.2 Progressive Phonemic Assimilation 171
3.8.2.3 Regressive Phonemic Assimilation 172
3.8.2.4 Coalescent Phonemic Assimilation 175
3.9 Linking Within the Word 176
4 Connected Speech 179

4.1 Strong and Weak Forms 179
4.2 Sentence Stress 188
4.3 Shift of Stress 192
4.4 Rhythm 193
4.5 Linking in Connected Speech 195
4.6 Elision in Connected Speech 197
4.7 Assimilation in Connected Speech 200
4.7.1 Allophonic Assimilation in Connected Speech 200
4.7.2 Phonemic Assimilation in Connected Speech 201
4.7.2.1 Progressive Phonemic Assimilation 201
4.7.2.2 Regressive Phonemic Assimilation 202
4.7.2.3 Coalescent Assimilation 205
4.8 Juncture 207
4.9 Intonation 208
4.9.1 The Definition of Intonation 208
4.9.2 The Forms of Intonation 210
4.9.2.1 The Tone and the Tone-unit 210
4.9.2.2 Pitch Behaviour within the Tone-unit 213
4.9.3 The Functions of Intonation 215
4.9.3.1 The Attitudinal Function of Intonation 216
4.9.3.2 The Stress Function of Intonation 218
4.9.3.3 The Grammatical Function of Intonation 221
4.9.3.4 The Discourse Function of Intonation 224
5
Contents
5 Geographical Varieties and Accents of English 226

5.1 British English 227
5.1.1 English English 227
5.1.1.1 RP Accent 228
5.1.1.2 London Accent 233
5.1.1.3 Estuary English 234
5.1.2 Scottish English 235
5.1.3 Welsh English 238
5.2 Irish English 240
5.3 North American English 242
5.3.1 US English 243
5.3.1.1 General American Accent 244
5.3.1.2 Southern American Accent 252
5.3.1.3 Eastern American Accent 255
5.3.2 Canadian English 255
5.4 Australian English 257
6 A Confrontational Analysis of English and Slovak Pronunciation

Systems 260
6.1 Vowels 260
6.2 Diphthongs 263
6.3 Consonants 267
6.4 Stress 272
6.5 Assimilation 274
6.6 Elision 280
6.7 Rhythm 284
6.8 Intonation 286
A Concise English-Slovak Dictionary of Phonetic and Phonological Terms 289
Appendix A. A feature matrix of RP General vowels 301
Appendix B. A feature matrix of RP General consonants 302
References 303
6
Introduction
Introduction
The study of English pronunciation as a subject of scientific attention has been
around for quite some time. Not so long ago, phonetic research was confined to
specialized laboratories, which were both very costly to maintain and accessible to
only a handful of devoted scientists. At the same time, the global spread of
information and the will to share scientific findings were technically limited by the
possibilities then available. Naturally, the number and quality of works published
under these circumstances reflected the reality of those days.
New developments in the area of computer hardware and software, which have
furthered and facilitated rigorous phonetic research, along with the latest trends in the
related linguistic disciplines such as morphonology, psycholinguistics,
sociolinguistics and the like, have lead to an accumulation of an enormous number of
works worldwide in just a couple of years. Not only is it no longer possible to read
even half of what is being published in the field of phonetics and phonology, but the
research has become so highly specialized, that a lucid medium-sized theoretical
introduction to the issues of pronunciation, reflecting a variety of opinions and trends,
is increasingly hard to come by. One can occasionally come across brief introductory
handbooks, but these are either too practically oriented, in which case they cover
theory only very concisely, or too narrowly focused, and do not take into account the
current diversity of linguistic points of view. Moreover, books on English
pronunciation written for a Slovak learner reflecting a contrastive aspect are almost
non-existent.
Under these circumstances, it seemed necessary to bridge the evident gap and
write this book.* It reflects many different (sometimes contradictory) opinions, some
of which, naturally, I felt the need to comment on. Apart from about a hundred and
fifty quotations taken from universally acclaimed works written by the most famous
linguists, I have included some of my own ideas and viewpoints, whether in terms of
the definitions of certain phenomena, or as far as their classification is concerned.
Since Slovak anglicists do not currently devote much attention to the issues
concerning the confrontational analysis of English and Slovak phonic levels, the
chapter dealing with English and Slovak pronunciation similarities and differences
could be taken as a minute step forward in this area. If, therefore, different opinions
concerning this matter should surface, I would gladly welcome any relevant
comments or suggestions for improvement.
One of the aims of this book is to describe the pronunciation of English spoken
in the south of England, namely the accent known as the Received Pronunciation. The
book is intended to serve as a general theoretical introduction to the phonetics and
phonology of English. It was written, above all, for students at Slovak universities
*
This book is a revised and extended version of the earlier Phonetics and Phonology of English: A
Theoretical Introduction (2000).
7
Introduction
who wish to study English as either a single subject study, or in combination with
another subject. Its purpose is to systematically describe, classify and sort available
information related to the subject, and to build a solid basis for the future study of
other linguistic disciplines in a duly scientific, yet reasonably succinct manner.
The book is concerned with articulatory, acoustic, and auditory aspects of
pronunciation. It describes and classifies the speech sounds of the RP accent, it draws
attention to the systems and patterns functioning in language, introduces the basics of
transcription, gives a short account of English accents, and briefly covers the main
differences between English and Slovak pronunciations. Despite the predominantly
theoretical character of the book, one can find frequent cases of exemplification,
either in the form of transcription concretizing the particulars of the covered
phenomena, or in the form of tables, figures, and pictures.
Even though I have tried to keep the scope of the book as broad as I could,
while also trying to pay due attention to detail, it will soon become evident that, no
matter how big one intends to think in terms of scope, only a fragment of information
can be offered on the vast area of study that phonetics and phonology have
indubitably turned into. It must therefore be emphasized again that this book is just a
general introduction, with some specific contrastive features, and other specialized
works should be consulted if one wants to obtain a more complete picture of the
issues discussed.
Finally, I should like to thank all those who have helped me and supported me
in any way in the preparation of this book. My very sincere thanks go to my friend
and colleague doc. PhDr. Pavol Kvetko, who has very meticulously read the
manuscript, made corrections, and suggested improvements. I am also grateful to
prof. PhDr. Ábel KráĐ, DrSc. for many constructive criticisms on the content of the
individual chapters of this book. Next, I wish to express my thanks to professors Peter
Ladefoged and John Wells, whom I have contacted through e-mail on several
occasions, and whose hints and advice proved very helpful. I would also like to
acknowledge my debt to Peter Blight for his detailed reading of the manuscript. All
these people and their invaluable comments have helped me correct many errors and
infelicities. I alone take the responsibility for those that remain. It is my hope that this
book, for all its limitations and imperfections, will serve its purpose, and prove a
useful reference tool for English students and teachers alike.
R. Pavlík
Comenius University, Bratislava
Faculty of Education
Department of English Language and Literature
[email protected]
8
Main Aspects of Communication
1 Main Aspects of Communication
1.1 Language and Communication

One of the basic characteristics of human beings is their ability to
communicate. This communication is usually carried out vocally or/and visually. In
this book, we will be concerned with vocal communication only, that is,
communication carried out by means of sounds. Communication is one of the abilities
animals and humans have in common. For instance, birds can produce a whole range
of different sounds, depending on the situation, apes utter different cries to express
anger, fear or pleasure, bees can indicate where and how far the honey is from the
bee-hive by means of the famous bee-dance, etc. What, then, is the difference
between the way animals and humans communicate? The sounds that animals
produce lack structure. They cannot be divided, like human utterance, into words. If,
in an utterance, we replace one word by another, we can change the meaning of the
utterance. Animals, however, produce single indivisible sounds which always mean
the same. That is why animal communication is very limited. In human
communication, the number of possible utterances is infinite. Thus, human
communication is based on combining a certain limited number of basic elements !
phonemes ! into higher units, e.g. syllables, morphemes, words, clauses, etc., and it
constitutes a system. This system is called language.
Traditional linguistics differentiates between language as an abstract system,
and language as an act of speaking. This distinction was proposed by the Swiss
linguist Ferdinand de Saussure,1 who distinguishes langue (language/abstract system
of language) from parole (speech/speaking/language in use). According to Saussure,
langue is "... a storehouse filled by the members of a given speech community through
their active use of speaking, a grammatical system that has a potential existence in
each brain, or, more specifically, in the brains of a group of individuals. For
language is not complete in any speaker; it exists perfectly only within a collectivity"
(Saussure 1974: 13!14). The term "parole#, on the other hand, refers to the concrete
act of speaking, writing or signing (sign language) in a particular situation. It is the
concrete use of language, not the abstract system.
A Similar distinction was made by the American linguist Noam
Chomsky,2 who distinguishes between competence and performance. Competence is
the knowledge of the system and rules, which means approximately the same as
Saussure#s "langue# (the difference is that Chomsky#s term is on a higher level
of abstraction), whereas performance is the actual use of the rules in speaking, with
1
/ \fd n nd d± s±h\sjh± /
2
/ \n±h±m \jÂmsk /
9
errors, memory limitations, etc., and it can be compared to what Saussure calls
"parole#.
In his famous work Syntactic Structures (1957: 13), Noam Chomsky
formulated this definition of language: "From now on I will consider a language to be
a set (finite or infinite) of sentences, each finite in length and constructed out of a
finite set of elements". In this book, Chomsky laid the foundations of generative
linguistics.
According to the definition proposed above, each language has a finite number
of elements in it, and by combining these sound elements one can create (generate) an
infinite number of sentences.
There are many definitions of language, and they differ from one another in
many respects. But what most of them seem to have in common is the fact that they
stress the communicative function of language, and its importance as a means of
thinking.
Language is a phenomenon only characteristic of humans. It has several
properties which differentiate it from all other forms of communication and which
make it an unparalleled system. Let us now consider some of these properties.
1. Arbitrariness
In human language, there is no direct link between the object, process, or state of the
extra-lingual reality (e.g. dog, walk, pretty, etc.) and the sound sequence denoting that
object. The names of the objects, states and processes were created by humans
arbitrarily, that is, people thought up a particular sequence of sounds for a particular
object, state, or process, and started to use this sequence of sounds (word) as a
community. However, there are some words in language that were not created
arbitrarily. They are the so called onomatopoeic words, such as cock-a-doodle-doo,
splash, ding-dong, etc. The sound form of these words actually resembles the real
sounds, and we say that these words are phonetically motivated.
2. Duality
Language works simultaneously on two levels, or layers. The first level is the set of
distinct sounds of language ! phonemes. Most phonemes are meaningless in isolation,
but when we combine them in specific ways, we build meaningful units ! words,
which form the second level. Thus, at one level we have distinct sounds (phonemes),
on the other level we have distinct meanings (words). In this way, humans are able to
create a very large number of words out of a limited number of phonemes.
In animal communication, the number of distinct sounds is limited, and there is only
one level, carrying distinct sound and distinct meaning simultaneously. Animals are
usually not able to combine these distinct sounds into higher units.
10
3. Displacement
Animals only communicate when they want to send a message concerning the
present, something happening here and now. They are not able to give information
about something that happened in the past, nor can they express their relation to the
future. Humans, on the other hand, are capable of referring to past, present, and future
alike. This property of human language is called displacement.
4. Productivity
Productivity, or creativity, is the ability to create new words in language, and also to
form new sentences, (possibly) ones that have never been formed before. This
property is only typical of human language. Animals only have a limited number of
signals in their repertoire, they are not capable of combining these signals, and they
cannot create new ones.
5. Cultural transmission
The communication signals in animals are genetically inbuilt. If we isolate a kitten
after birth from its mother and let a monkey or a human being raise it, the kitten is
unlikely to chatter like a monkey or speak like a human when it reaches maturity. It
will always meow, no matter where it is raised. Humans, however, are different in this
respect. It is a well-known fact, that children brought up in isolation do not acquire
language. There are cases of children who were raised by animals, and when people
found them, these children could only produce animal sounds. Similarly, if a child
born to an Englishman is raised in Slovakia, and it is sufficiently exposed to the
Slovak language, by the age of six or so it will speak perfect Slovak, not English. The
language is therefore not hereditary;3 it is a matter of cultural transmission ! the
culture and the language are transmitted to an individual.
6. Discreteness
Every language consists of a certain limited number of discrete sounds ! phonemes.
These discrete elements are grouped in a certain way to build higher units ! words,
which have their specific meaning. If we exchange one phoneme for another in a
particular word, we may change the meaning of that word, e.g. pay ! bay. This
difference in meaning is due to the difference between the phonemes / p / and / b /.
Thus, the discrete, but (mostly) meaningless, elements of the language ! phonemes !
can bring about a change in the meaning of the meaningful units of the language !
words. This property of language is called discreteness. Every sound in language
is treated as discrete, that is, each sound must ultimately fall into some category
3
This means a particular language, not language in general. It has been proposed by Noam
Chomsky, that language in general, as a system based on combining a limited number of elements
into a large number of units, is genetically imprinted on humans. He claims that language is such a
complex phenomenon, that it would be impossible to acquire it if there wasn't some innate potential
for language in the first place.
11
(phoneme) if it is intended to participate in meaning. For instance, the first sound in

the word pay can be pronounced differently in different contexts, or even in the same
context. Its pronunciation is never exactly the same. However, it will always fall into
the category of the phoneme / p /, so it is always treated as a distinct sound, because it

influences the meaning.
Language manifests itself in two basic forms ! speech and writing. The primary
form of language is spoken; the written language is secondary and derivative. The
priority of speech over writing is obvious. Every child learns how to speak before it
learns to write. There are many communities, which only communicate by means of
speaking, but there are none which only communicate in written form. In this respect
we speak about four aspects of spoken language taking priority over written (Lyons,
1992).
1. Historical priority ! there are no known communities without the capacity of

speech.
2. Structural priority ! spoken language is structurally more basic than written
language, because fewer sounds (and this is especially true
of English) are, in written form, represented by more
letters ! sounds and letters are not in one-to-one
correspondence, e.g. tough / tYf / ! 5 letters versus 3
sounds.
3. Functional priority ! writing serves as a functional substitute for speech only in
situations where vocal communication is not possible.
4. Biological priority ! all babies start babbling when they reach a certain age, and
it is believed that it is not just imitation, because deaf
babies babble too. Also, there is the fact that usually the
left hemisphere of the human brain, where the processing
of sounds takes place, is dominant after childhood.
When humans communicate, a very complex and complicated process is

triggered. What we have in mind is the transmission of information from one
member of the language community to another. This process involves the use of a
certain code and channel, through which the code signals are transmitted. The
process of selecting certain signals and putting them into channel is called encoding.
The process of identification and interpretation of the coded signals is known as
decoding. The person who encodes the message and sends it down the channel is
called the sender, and the person who receives and decodes the signals is called the
receiver. The channel of the communication can be either the auditory-vocal channel,
visual channel, or sometimes also the haptic channel. The medium used is most often
12
speech or writing. Thus language can be realized in different media ! speech, writing,
etc. This property of language is called medium-transferability (Lyons, 1992).
We will now consider the individual systems of human communication. Since
this book is concerned with the phonic level of language, we will only look at face-to-
face communication, which, of course, employs mainly the auditory-vocal channel.
Table 1. shows the individual systems of communication and their mutual
relationships.
Face-to-face communication
Auditory-vocal channel Visual channel

Haptic channel
Linguistic Non-linguistic
Systems regarded as being part of Features not regarded as being part of
language language
Verbal Non-verbal
Using words and All phenomena conveyed non-verbally
sentences
Verbal Prosodic Paralinguistic Kinesic Standing
features
Words Intonation Voice qualifiers Facial movement Appearance
Clauses Stress Voice Eye gaze Clothes
Sentences Rhythm qualifications Eye contact Bodily adornment
(see 3.1) (see 3.1) Head movement Proxemics
Hand movement Touch
Gesture
Posture
Table 1. Systems of human face-to-face communication
It should be noted that prosodic, paralinguistic, and kinesic features can be

classified in a different manner, and the classification often depends on the preference
of the individual author.
1.2 Concrete and Abstract Sides of Language

The concrete audial manifestation of language is speech. Speech can be broken
down into minimal segments called speech sounds. A speech-sound is a unit of
speech having certain articulatory, acoustic, and auditory characteristics. The
difference between sounds in general and speech sounds is obvious. Sounds, as such,
are any vibrations in the approximate range between 20 and 20 000 hertz, detectable
13
by the organs of hearing. Speech sounds, on the other hand, are only those vibrations,
which constitute a system of elements which can be organized into higher units called
words. Words, in turn, can be grouped into sentences, and sets of sentences make up a
particular language.
Based on the way they are produced, speech sounds can be grouped into two
main groups: vowels and consonants. Although these terms are very familiar ones, it
is not easy to define them exactly. The problem is, that some English consonants,
/ j / or / w / for instance, could, from the acoustic and articulatory point of view, easily
pass for vowels. By definition, a vowel is "... a voiced sound in forming which the air
issues in a continuous stream through the pharynx and mouth, there being no
obstruction and no narrowing such as would cause audible friction" (Jones 1960:
23). Both / j / and / w / can be pronounced in such a way that there is no obstruction
and no audible friction. This makes them, according to the definition, vowels.
However, from the distributional viewpoint, these sounds usually behave like
consonants, that is, they occur at syllable boundaries. Vowels, on the other hand, are
found in the centres of syllables. Because of this difference between the articulatory
and functional classifications, these sounds are often called semivowels.
Consonants can be briefly defined as speech sounds produced by a partial or
complete obstruction of the air stream by any of the organs of articulation. Again, the
precise distinction between vowels and consonants is very difficult to make for the
above-mentioned reasons.
As we have shown, the terms vowel and consonant are used both in acoustic
and functional analysis of speech sounds. The two types of analyses, however, do not
produce identical results ! they do not give us identical groups of sounds. For
instance, if we use the above-mentioned phonetic (articulatory, acoustic and auditory)
definition to determine all vowels in English, we would have to include / r /, / j /, and
/ w / into this group, because they meet the phonetic criteria for vowels. On the other
hand, if we analyze the English sounds from the point of view of their function, / r /,
/ j /, and / w / belong to the group of sounds found at the edges of syllables !
consonants.
To clarify this matter, the terms vocoid and contoid have been invented by the
American linguist Kenneth L. Pike. Both these terms are characterized by phonetic
definition only, unlike vowel and consonant, which cover both phonetic (articulatory,
acoustic and auditory) and phonological (functional) definition. Thus, a vocoid is a
vowel in its phonetic sense without considering any functional criteria. Similarly, a
contoid is a consonant in its phonetic sense without considering any functional
criteria.
When we investigate the concrete speech sounds of a given language from the
point of view of their function, we find out that only a certain limited number of
sound elements can cause a change in meaning of words. Consider the following
example. In British English we have two distinct types of l-sound: clear [ lj ], as in
14
the word "leap#, and dark [ 1 ], as in the word "kill#. These are, acoustically, two
different sounds. Clear [ lj ] lies somewhere between Slovak l and Đ, and dark [ 1 ]
sounds something like Slovak l, but it is darker. If we exchange the clear / l / for dark
/ l / in any English word, the meaning of that word will remain the same. However, if
we do the same thing with Slovak l and Đ, we may change the meaning of the word,
e.g. lavica ! Đavica. Similar case is that of the sounds / n / and / Ì /. No matter which
of these two sounds we use in Slovak word "banka# [ \banka, \baÌka ], we will never
change the meaning of that word. However, if we exchange these two sounds in
English, we may change the meaning of the word, e.g. sin / s n / ! sing / s Ì /.
Sounds like these, which can change the meaning of the word, are called phonemes.
"A phoneme may be described roughly as a family of sounds consisting of an
important sound of the language (generally the most frequently used member of that
family) together with other related sounds which #take its place$ in particular sound-
sequences or under particular conditions of length or stress or intonation" (Jones
1960: 49). This means that a phoneme is "... the smallest contrastive linguistic unit
which may bring about a change of meaning" (Gimson & Cruttenden 1994: 42).
Phonemes are abstract units. Each phoneme represents a set of speech sounds,
each of which is a member of a particular phoneme. The phoneme / p /, for instance,
can be realized differently in different words. It can be aspirated and very clear in the
word "pay#, or almost inaudible in the word "stop#. These members of a phoneme are
called allophones.
The occurrence of allophones in words is predictable. For example, let us take a
look at the above-mentioned case of clear and dark / l /. Clear [ lj ] in RP English can
occur only before vowels and the consonant / j /, e.g. leap [ ljip ], lure [ ljjh± ], etc.
Dark [ 1 ], on the other hand, occurs only when followed by a consonant (except / j /),
or when standing in the word-final position, e.g. felt [ fe1t ], kill [ k 1 ], etc. These two
allophones can never swap places, and they are mutually exclusive. They are said to
be in complementary distribution and they are called combinatory variants of the
phoneme. At the same time, there are cases in which one phoneme of a given word
can have several different realizations (allophones), and the choice of the allophone
depends solely on the speaker. For instance, the phoneme / r / in the English word
"very# can be realized as [ \veÚ ], where [ Ú ] is an approximant (typical RP General
pronunciation), or as [ \ve ], where [ ] is a tap (typical RP Conservative
pronunciation). These allophones can be mutually substituted (exchanged) and they
are said to be in free variation. They are called facultative variants of the phoneme.
Thus, we have established two sources of variation of sound quality within the
phoneme:
(a) The position of the sound within a certain environment

(b) The choice of the speaker
15
(a) The number of positionally determined allophones (combinatory/positional/

conditioned variants) is limited by the number of possible sound environments in
which a sound may occur in a particular accent. For example, the word-initial / k /
preceded by silence can be realized, depending on context, as:
[ hh ] in the word "key# (the tongue is advanced)

[ kh ] in the word "cut# (the tongue is in the central position)
[ ih ] in the word "coo# (the tongue is retracted)
etc.
If we continue with this analysis and examine / k / in all possible positions and
all possible environments (contexts), we will get a large but finite set of the
allophones of the phoneme / k / found in a particular accent (not all combinations are
possible in an accent; see 3.3).
(b) The number of allophones in free variation within a certain accent (free/facultative
variants) is also limited. For instance, a speaker with RP accent (pronunciation found
in the south of England) can in the word "very# use either [ ], which is similar to
Slovak "r#, or he can use a frictionless consonant [ Ú ], but not normally [ ], which is
found in the General American accent. Needless to say, the phoneme / r / in different
positions in words will have several positional variants.
Allophones must be distinguished from phones.

A phone is a concrete acoustic articulation with zero degree of abstraction. It
is "... the smallest perceptible discrete segment of sound in a stream of speech ...%
(Crystal 1991: 258). It is a concrete measurable unit realized in speech at a certain
point in time in certain sound environment by a certain speaker. Phones are
physically unique ! you cannot produce the same phone twice. Such is the nature of
human articulators and the brain. We are not computers, therefore, even if we tried to
pronounce the same sound in the same environment twice, there will always be some
variation of the quality. The number of phones in a given accent is infinite, hence, it
is not possible to express them in transcription, because we would have to use a
different symbol for each phone, and even then we would be making an abstraction of
a certain degree. Phones are not representations of anything, phones are real concrete
units of speech.
An allophone is a group (set/family) of phones with (usually) very similar
articulatory, acoustic, and auditory characteristics. It is an articulatory-acoustic-
auditory abstraction of a group of phones made from a perceptual point of view. An
allophone is not unique ! in identical or comparable sound environments, you can
produce the same allophone repeatedly. For instance, if you repeatedly pronounce the
word "pay# in a certain accent and under certain stable conditions, you will always
produce and perceive the same allophone ! an aspirated plosive [ ph ]. Each time,
16
however, you will produce a different phone, that is, an allophone with a slightly
different quality. These slight differences, nonetheless, will not influence the accent.
In other words, no-one will perceive them as having an effect on the accent; what is
more, they may not even be detectable by the human ear. Since allophones are not
unique units, but rather generalizations of the real physical articulations, they
represent a certain degree of abstraction ! the first-degree abstraction. The number
of allophones in a given accent is finite.
Note
In this book, we equate the term "allophone# with the term "speech sound# (in line with the
classification of Slovak phoneticians KráĐ and Sabol). However, other phoneticians, e.g.
Daniel Jones, define the term "speech sound# as a sound "... which is incapable of variation"
(Jones 1960: 49), which equates it with the term "phone#.
A phoneme, in this threefold system, can be characterized as a purely abstract

unit on the second-degree level of abstraction. A phoneme is a group (set/family) of
phones with the same distinctive function in a given accent. It is an articulatory-
acoustic-auditory (perceptual) abstraction of a group of phones made from the point
of view of meaning (the articulatory, acoustic, and auditory characteristics of the
phones abstracted in this way need not be similar, for instance, they may be members
of different allophones, such as [ n ] and [ ) ] in Slovak, but they belong to the same
phoneme / n /, that is, they fulfil the same distinctive function). The number of
phonemes is limited in every accent. The English language (RP accent), for instance,
has 44 phonemes. This limited number of phonemes can be combined into an
unlimited number of utterances. As we have seen, two separate phonemes in one
language can well be two members of the same phoneme in another language.
To sum up the above, we can say that:
Phones are concrete physical articulations. They can only be detected instrumentally.
Allophones are articulatory-acoustic-auditory abstractions of phones from a
perceptual point of view.
Phonemes are articulatory-acoustic-auditory abstractions of phones from the point of
view of meaning.
phoneme /p/ second-degree abstraction

(phonological abstraction)
allophones [p] [ ph ] [ p ] first-degree abstraction
(phonetic abstraction)
phones h h zero-degree abstraction
ppp p ph p h p p p p p p
etc. etc. etc.
concrete units
Table 2. The mutual relationship of phonemes, allophones and phones (Adapted from KráĐ
& Sabol, 1989).
17
Table 2. shows the relationship between phonemes, allophones and phones.

The phoneme, as a purely abstract unit, can be realized in speech as any one of the
possible allophones of a given accent. Allophones, in turn, are representations of a
group of concrete physical articulations ! phones. In the concrete act of speech, each
phone is traceable upwards to its source, first the allophone, then the phoneme. By the
same token, each phoneme narrows downwards through one of its allophones to a
particular phone.
1.3 Phonetics and Phonology

The terms phonetics and phonology can be found in many books dealing with
the pronunciation of English, even though they are not always sufficiently
distinguished from each other. The confusion about these two terms arises from the
fact, that both phonetics and phonology are concerned with the same subject matter !
speech sounds. However, they approach the matter from different points of view.
1.3.1 Phonetics
Phonetics is a branch of linguistics concerned with the study of speech sounds
and their production, transmission, reception, description, and representation by
written symbols. "Phonetics is interested exclusively in the material aspect of human
speech, i.e. in the production of sounds, in the organs of articulation which are active
when producing them, and in the sounds which are the result of this production, of
their physical, mainly acoustic qualities" (Vachek 1973: 1). Phonetics is "The science
which studies the characteristics of human sound-making, especially those sounds
used in speech, and provides methods for their description, classification and
transcription" (Crystal 1991: 259). The basic elements of the phonetic analysis are
both phones and allophones .
Phonetics studies human speech from physiological, acoustic, and partly from
psychological aspects. It can be divided into three principal sub-branches: articulatory
phonetics, acoustic phonetics, and auditory phonetics.
1. Articulatory (organic) phonetics is concerned with:
(a) the organs of articulation ! it names and describes organs participating in

sound production.
(b) the production of speech sounds ! it describes how sounds are produced
from the point of view of the rate of the vocal-fold vibration, physical effort,
configuration of the articulators, and the timing of articulatory movements.
18
(c) the description and classification of speech sounds ! it describes and

classifies the sounds from the point of view of the rate of the vibration of the
vocal cords, physical effort, configuration of the articulators, and the timing of
articulatory movements (quantity).
2. Acoustic phonetics is concerned with:
(a) the acoustic properties of speech sounds ! it analyzes sound from the
point of view of fundamental frequency, intensity, wave structure (quality) and
duration (quantity).
(b) the transmission of speech sounds ! it describes how sound waves are
transmitted between the mouth and the ear.
classifies the sounds from the point of view of their acoustic properties, i.e.
fundamental frequency, intensity, wave structure (quality) and duration
(quantity).
3. Auditory phonetics is concerned with:
(a) the reception and analysis of speech sounds ! it describes how the sound is
received by the ear, how it is further modulated and processed by the organs of
hearing, and how it is analyzed in the listener#s brain
(b) the anatomical and physiological study of the ear ! it describes the structure
and the function of the ear
classifies the sounds from the point of view of their auditory perception, i.e.
pitch, loudness, qualitative (auditory) impression, and length (quantity).
The relationship among the three sub-branches of phonetics can be illustrated in the
following table:
Categories of study
Articulatory Rate of vibration Physical effort ! Configuration of the Timing of
phonetics of the vocal folds energy exerted articulators articulatory
movements
Acoustic Fundamental Intensity ! Wave structure Duration
phonetics frequency amplitude (periodic/aperiodic)
Auditory Pitch Loudness Auditory qualitative Length
phonetics impression
Table 3. Sub-branches of phonetics and their categories of study.
19
It is evident from the table that all three sub-branches of phonetics study
different aspects of the same categories, e.g. the vocal folds vibrate at a certain rate
(articulatory aspect) which produces a wave of a certain fundamental frequency
(acoustic aspect), and this fundamental frequency is perceived as a certain pitch
(auditory aspect), etc.
Phonetics, however, can be classified from many other points of view. Here are
the main categories:
Based on the degree to which it complies with a given norm, phonetics can be divided
into:
(a) Descriptive phonetics

It describes the phonic level of a particular language as it is, not how it ought to
be.
(b) Prescriptive phonetics
It attempts to lay down the rules of correctness and gives the norms of usage. It
prescribes a model of correct pronunciation.
Depending of how wide or narrow the focus of phonetics is we can distinguish among
these three types:
(a) Descriptive phonetics

It describes the phonic level of one particular language at its present state.
(b) Contrastive phonetics
It describes and contrasts the phonic levels of at least two languages as existing
at a particular point in time, trying to establish the similarities and differences
between the analyzed languages.
(c) General phonetics
It describes the phonic level of language in general and tries to establish
phonetic universals.
From the point of view of the time continuum, phonetics can be divided into two
categories:
(a) Diachronic phonetics

It traces the historical development of the phonic level of a particular language,
and records the changes that have taken place in the pronunciation during the
individual periods of time.
(b) Synchronic phonetics
It presents an account of the phonic level of a particular language at some
particular point in time.
20
In terms of the applicability of the results of phonetic studies we distinguish between

two areas of phonetics:
(a) Theoretical phonetics

It studies the phonic level of language(s) with the intention of determining its
articulatory, acoustic, and auditory features without any regard to possible
applications.
(b) Applied phonetics
It is concerned with the application of the findings of phonetics to a variety of
practical tasks, such as foreign language teaching, correction of speech
impediments, rhetoric, speech recognition, speech synthesis, etc.
Phonetics can also be seen from the point of view of its linguistic relevance. In this
respect, we can speak about two broad types:
(a) General studies of articulation, acoustics, and reception of speech sounds.

This type of phonetics is concerned predominantly with physiological, acoustic,
and psychological phenomena and processes, therefore it could also be termed
non-linguistic phonetics.
(b) Studies of phonetic properties of languages form the functional point of view.
This type of phonetics is concerned with how the individual sounds are used in
the pronunciation system of a particular language. This functional approach to
phonetics is called phonology, and we could also term this type of phonetics
linguistic phonetics.
This twofold character of phonetics creates a problem for linguists: is phonetics

a separate subject or can it be seen as a branch of linguistics? Depending on tradition
and personal opinion, some linguists subsume phonetics under linguistics, others see
it as a separate discipline.
In phonetic research, these types of methods can be found:
1. (a) Subjective methods

They are based on direct observation carried out by our senses, mainly
eyesight and hearing.
(b) Objective methods
These are methods used in articulatory and acoustic research. We do not rely
on our senses but usually use some type of instrument.
2. (a) Mathematical methods

In speech analysis, statistical methods (mean, median, mode, variance, range,
deviation, etc.) are often used.
21
(b) Non-mathematical methods

We may analyse a sound (its wave structure, length), and compare it to other
sounds without using any mathematical methods.
3. (a) Experimental methods

We have a regulated environment and control the individual variables ! we
perform an experiment. For instance, we may experiment with synthetic
speech and its perception, we may deliberately remove some frequencies from
a sound and examine the influence this might have on the sound#s perception,
etc.
(b) Non-experimental methods
We do not alter any variables while analysing sounds.
4. (a) Instrumental methods

An instrumental method is a method involving the use of some instrument,
e.g. a tape recorder, a computer, an artificial palate, etc.
(b) Non-instrumental methods
No instruments are used in speech analysis.
The above method types can be mutually combined, and some of them may overlap to
a certain extent.
1.3.2 Phonology
1.3.2.1 The Definition of Phonology
Phonology (phonemics, functional/linguistic/systemic phonetics) is a branch of
linguistics concerned with the classification of speech sounds from the functional
point of view. It is "...the description of the systems and patterns of sounds that occur
in a language. It involves studying a language to determine its distinctive sounds and
to establish a set of rules that describe the set of changes that take place in these
sounds when they occur in different relationships with other sounds" (Ladefoged
1975: 23). It studies "...the abstract side of sounds of language" (Roach 1991: 43).
Phonology is concerned with language as a system of abstract elements ! phonemes,
while phonetics deals with the formal representations of concrete sounds !
allophones, and the concrete sounds themselves ! phones. "Phonetics gathers raw
material. Phonemics cooks it" (Pike 1947: 57). The main task of phonology is to find
those features of sounds, which enable us to distinguish one word from another (sin !
sing, my ! by, etc.).
Within phonology, two branches of study are usually recognised. Segmental
phonology, which analyzes speech into the minimal functional elements of
22
phonology ! phonemes, and suprasegmental (non-segmental) phonology, which

analyzes functional segments higher than phonemes ! syllables, tone-units,
utterances, etc.
The smallest unit of phonology is the phoneme. The phoneme theory was first
formulated by Jan Baudouin de Courtenay,4 a philologist of Polish origin who
established himself in Russia. He was the first one to point out the distinction between
the "phone# and the "phoneme#. He had been working out this theory since 1868.
Independently of Baudouin de Courtenay, the concept of the phoneme was also
recognized in the 1870s by the English linguist Henry Sweet, although he did not use
the term "phoneme#. His concept of the phoneme theory can be seen in his "broad#
and "narrow# types of phonetic transcription. The definition of the phoneme and the
basic terminology concerning this matter was straightened out by 1918, and the
theory of the phoneme was firmly established in England in the early 1920s. Later, in
America, somewhere about 1940, the terminology regarding the phoneme was further
improved by inventing a new term "allophone#, which replaced the old term "member
of the phoneme# (Jones, 1967). So from now on we can speak of phonemic and
allophonic transcription (the terms are discussed in the following chapter).
1.3.2.2 Methods of Determining Phonemic Inventories

Every accent has a specific phonemic inventory. The phonemic inventory
(phonemic structure/phonemic system) is a set of phonemes found and used in a
particular accent. The RP General accent, as we mentioned earlier, has a phonemic
inventory of 44 phonemes.
The phonemic inventory of a particular accent can be determined through either
of the following procedures, or tests:
(a) minimal pair test

(b) commutation test
(a) Minimal pair test

Let us suppose that we have a speech sound which regularly appears in a given
accent. To determine whether or not this sound is a phoneme we must find such a
word, in which the replacement of the original speech sound by our investigated
sound would change the meaning of that word, and we would produce a minimal
word-pair:
cut / kYt / ! gut / gYt / fit / f t / ! feet / fit /

sin / s n / ! sing / s ) / four / f / ! bore / b /
4
/ \j n \b±hdw n d± \ktn /
23
(b) Commutation test

In a commutation test we look for a set of minimal word-pairs in such a way that one
phoneme is replaced, if possible, by all phonemes of a given accent:
/ p / contrasted with other phonemes
/ p / ! / b / : pour / p / ! bore / b /

/ p / ! / k / : par / p / ! car / k /
/ p / ! / m / : seep / sip / ! seem / sim /
/ p / ! / s / : ape / e p / ! ace / e s /
A very precise set of rules for determining a phonemic inventory was proposed
by N. S. Trubetzkoy5 (a member of the Prague School of linguistics). Here are the
rules (reproduced from Fudge 1973: 60 ! 64):
1. Two sounds of a given language are merely optional phonetic variants of a

single phoneme if they occur in exactly the same environment and are
interchangeable without a change in the lexical meaning of the word.
English: The word "very# in the RP accent can be realized with either a flapped [ ],
or a frictionless [ Ú ]. The mutual substitution of these two speech sounds does not
change the meaning of the word, ergo, they are only allophones of the phoneme / r /.
They are in free variation (they are not positionally restricted):
very [ ¥ve ] ! [ ¥veÚ ]
Slovak: The word "slávny# in standard Slovak can be realized with either a
labiodental approximant [ ], or a bilabial approximant [ w ]. The mutual substitution
of these two speech sounds does not change the meaning of the word, they are only
allophones of the phoneme / v /. They are in free variation (they are not positionally
restricted in this particular case):
slávny [ ¥slani ] ! [ ¥slawni ]
2. If two sounds occur in exactly the same position and cannot be interchanged
without a change in the meaning of the words or without rendering the word
unrecognizable, the two sounds are phonetic realizations of two different
phonemes.
5
/ ¥trhbetsk /
24
English: The substitution of one speech sound for another brings about a change in
the meaning of the word:
pin / p n / ! bin / b n / full / fhl / ! foot / fht /
Slovak: The substitution of one speech sound for another brings about a change in
the meaning of the word:
tok / tk / ! bok / bk / drak / drak / ! dral / dral /
3. If two sounds of a given language, related acoustically or articulatorily, never

occur in the same environment, they are to be considered combinatory variants of
the same phoneme.
English: Clear [ lj ] in RP accent occurs only before vowels and / j /, whereas dark
[ 1 ] can only occur before consonants (except / j /) and in the word final position.
leap [ ljip ] ! fool [ fu1 ]
Slovak: Velar [ ) ] in standard Slovak can occur only before / k / or / g /, while

alveolar [ n ] never occurs before / k / or / g /.
banka [ ¥ba)ka ] ! nos [ ns ]
4. Two sounds that otherwise meet the conditions of Rule 3 can still not be regarded
as variants of the same phoneme if, in a given language, they can occur next to
each other, that is, if they are part of a sound sequence in those positions where
one of the sounds also occurs in isolation.
English: RP accent / w / and / h / are acoustically and articulatorily similar, but

since / w / is a consonant and / h / is a vowel, they never occur in the same
environment (consonants occur at the edge of syllables, vowels are found in the
centre of syllables). Thus they meet the conditions of Rule 3. However, they can
occur next to each other, therefore, they are two separate phonemes.
wool / whl / wood / whd /
Slovak: Slovak / j / and / i / are acoustically and articulatorily similar, but since / j / is
a consonant and / i / is a vowel, they never occur in the same environment (Rule 3).
However, they can occur next to each other, therefore, they are two separate
phonemes.
zmija / ¥zmija / moji / ¥mji /
25
1.3.2.3 Phonological Oppositions

Phonological opposition is a relation between two (sometimes more)
phonemes based on their mutual similarities and differences. Depending on different
points of view, there exist different types of phonological oppositions.
(a) Binary versus ternary

A binary opposition is an opposition of two values only, where the presence of one
value can be regarded as the absence of the other. e.g. voiced vs. voiceless:
/ p / vs. / b / / s / vs. / z / / f / vs. / v /
Ternary opposition is an opposition of three different values, e.g. the three different
positions of the tongue during the articulation of vowels :
high ! / i / mid ! / e / low ! / /
Note
Modern phonology works mainly with binary oppositions.
(b) bilateral versus multilateral

A bilateral opposition is an opposition restricted to two phonemes only. The two
phonemes share all features but one; there is only one parameter of contrast, e.g.
voicing:
/ t / vs. / d / / k / vs. / g / / p / vs. / b /
/ t / and / d / are both alveolar and plosive, and they are differentiated by voicing.
This pair, however, is unique (the opposition is bilateral), because there are no other
phoneme pairs which have the alveolar and plosive features. / k / and / g /, although
both plosive, are not alveolar but velar. Similarly, the / p / and / b / pair is
distinguished from the previous two pairs by being plosive and bilabial.
A multilateral opposition is an opposition where there is more than one parameter of

contrast:
/ d / vs. / f / are differentiated by both voicing (the first is voiced, the second is
voiceless) and the place of articulation (the first is alveolar, the second is labiodental).
26
(c) proportional versus isolated

A proportional opposition is an opposition which is found in other pairs of an accent.
It is an opposition of the type "A vs. B# = "C vs. D#. An example of such an
opposition is voicing:
/ t / vs. / d / / s / vs. / z / / f / vs. / v /
An isolated opposition is an opposition not found in other pairs of an accent. Only

one pair in the whole phonemic inventory is distinguished in a particular way.
/ v / vs. / l / ! no other pair is contrasted in this way, i.e. voiced labiodental fricative
versus voiced lateral.
(d) privative versus gradual versus equipollent

A privative opposition is a binary opposition. Two phonemes differ only by one
feature. The phoneme which possesses the feature is seen as a marked member of the
pair, and the phoneme which lacks the feature is seen as an unmarked member of the
pair.
/ p / vs. / b / / s / vs. / z / / t / vs. / d /
A gradual opposition is an opposition where the phonemes possess different degrees

of the same property (e.g. vowel height or vowel length)
In Slovak, for example, / a / and / a / have the same property ! certain length. They,
however, possess different degrees of this property; / a / is long, / a / is short.
An equipollent opposition is an opposition where the members are equal, or
equivalent, contrasted neither gradually nor by a binary feature. Along with identical
features, the phonemes also have differing features.
/ p / vs. / s / identical features: voiceless, oral, consonantal, etc.

differing features: / p / is bilabial, plosive, etc.
/ s / is alveolar, fricative, etc.
(e) constant versus neutralizable

A constant opposition is an opposition which always exists, whenever and wherever
two phonemes are found in a given accent.
/ p / vs. / b /: The members of this pair will always keep their properties, wherever
they may be found , e.g. pore ! bore, hop ! hob, etc.
A neutralizable opposition is an opposition which exists only in certain positions.

There are positions in which the opposition between two phonemes disappears. This
27
happens, for instance, in phonemic assimilation (see 3.8 and 4.7), where the
opposition becomes neutralized ! both members acquire identical features:
English: The opposition between / t / and / p / in the words "great# / greΙt / and
"grape# / greΙp / may disappear if the word "great# is used in a word group "great
man#. What happens is that the word-final phoneme / t / in the word "great#
assimilates to / p / under the influence of the following word-initial / m /. The word
group "great man# is then pronounced as / ¤greΙp¥m n /.
Slovak: The opposition between / t / and / d / may disappear in the words "plod# and
"plot# followed by silence. The word-final phoneme / d / in the word "plod# will be
neutralized to / t / ! / plt /, so both words will sound alike.
1.3.2.4 Distinctive Features

In the area of phonology, many interesting approaches have been developed
regarding the very basic notion of the phoneme. One of the approaches, which
originated in the Prague School of linguistics, is distinctive feature analysis. It is
based on the principle that phonemes, though still minimal segments of languages, are
not their minimal elements. Here, a phoneme is defined as "...a bundle of distinctive
features which are simultaneously realized" (Vachek 1973: 4). This theory was first
proposed by N. S. Trubetzkoy and later further developed by Roman Jakobson.
According to this theory, every phoneme is a combination of a small number of
different distinctive features. In distinctive feature theories "... the phoneme is not
considered to be a relevant unit of explanation: symbols such as p, b, etc., are seen
simply as convenient abbreviations for particular sets of features. It is the features
which are the minimal units of phonological analysis, not the phonemes" (Crystal
1991: 109). Examples of such features could be place of articulation, manner of
articulation, voicing, nasality, etc. Each phoneme of a given language must be
differentiated from the rest of the phonemes by at least one distinctive feature,
otherwise it could not be classified as a phoneme. For instance, the English
phonemes
/ s / and / z / are only differentiated by one feature ! voicing. Their place and manner
of articulation are exactly the same. Each feature can be labelled + or !, where +
means that the feature is present, ! means that the feature is absent.
In this type of analysis of speech we can make generalizations about the
relationships between sounds in a language, and it also enables us to make inter-
language comparisons.
The choice and number of proposed features differs with different linguists, so
it is difficult to say whether this or that set of features should be considered the
"ultimate# set of contrasting units of speech.
28
We now present a distinctive feature system proposed by Morris Halle6 and

Noam Chomsky in their book The Sound Pattern of English (1968), which was
inspired by the research of N. S. Trubetzkoy and Roman Jakobson. The features of
this system are treated as binary, that is, having either a plus value (the feature is
present), or a minus value (the feature is absent). The features of this system are
universal, which means that they are applicable to other languages, not just English.
Let us now briefly describe the individual features.
SYLLABIC
The feature [+ syllabic] is assigned to phonemes that can form the syllable nucleus
(see 3.2), that is, all vowels and syllabic consonants.
CONSONANTAL
The feature [+ consonantal] is assigned to phonemes that are formed with an
obstruction or narrowing (sufficient to produce friction) in the vocal tract. All
consonants except approximants are [+ consonantal].
SONORANT
The feature [+ sonorant] is assigned to phonemes which are produced with a
relatively free airflow and in which spontaneous voicing is possible. Vowels,
approximants, and nasals are [+ sonorant].
VOICED
The feature [+ voiced] is assigned to phonemes which are produced with a periodic
vibration of the vocal folds (see 1.6.2). All vowels and voiced consonants are
[+ voiced].
CONTINUANT
The feature [+ continuant] is assigned to phonemes which are formed with an
incomplete closure of the vocal tract ! the air flow through the mouth is not blocked.
Note
The term continuant is also used in another meaning (a sound that can be produced
continuously for as long as one has enough air in the lungs). See 2.4.1.
NASAL
The feature [+ nasal] is assigned to phonemes produced with the soft palate lowered,
permitting the airflow pass through the nasal cavity. All nasal consonants are [+
nasal].
6
/ \mÂr s \hl± /
29
STRIDENT
The feature [+ strident] is assigned to phonemes characterized by the high-frequency
turbulent noise.
LATERAL
The feature [+ lateral] is assigned to phonemes in the production of which the tip of
the tongue is partially blocking the airflow which is thus forced to pass along one or
both sides of the tongue. The phoneme / l / is [+ lateral].
DISTRIBUTED
The feature [+ distributed] is assigned to phonemes formed with a stricture which
extends for a considerable distance along the direction of the airstream ! the tongue
has a relatively long region of contact along the roof of the mouth.
AFFRICATE (DELAYED RELEASE)

The feature [+ affricate] is assigned to phonemes produced by a combination of a stop
and a fricative sound functioning as a close-knit unit. All affricate consonants
are [+ affricate].
LABIAL
The feature [+ labial] is assigned to phonemes formed by bilabial or labiodental
contact.
ROUND
The feature [+ round] is assigned to phonemes produced with an extension and
pursing of the lips. All sounds that are [+ round] are also [+ labial], but not all [+
labial] sounds are necessarily also [+ round], e.g. / b / is [+ labial] but [! round].
CORONAL
The feature [+ coronal] is assigned to phonemes formed with the blade of the tongue
raised towards or touching the teeth or the alveolar ridge.
ANTERIOR
The feature [+ anterior] is assigned to phonemes made with the main constriction in
front of the palatoalveolar region. All alveolar, dental, and labial phonemes are
[+ anterior].
HIGH
The feature [+ high] is assigned to phonemes formed with the body of the tongue
raised toward or touching the roof of the mouth.
30
BACK
The feature [+ back] is assigned to phonemes made with the body of the tongue raised
and slightly retracted from the neutral central position in the mouth.
LOW
The feature [+ low] is assigned to phonemes made with the body of the tongue
lowered below the level it holds in the neutral central position in the mouth.
English consonant phonemes

p b k g t d f v * s s z µ h j dµ m n Ì l r j w
Syllabic ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !
+ + + + +
Consonantal + + + + + + + + + + + + + + ! + + + + + + ! ! !
Sonorant ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! + + + + + + +
Voiced ! + ! + ! + ! + ! + ! + ! + ! ! + + + + + + + +
Continuant ! ! ! ! ! ! + + + + + + + + + ! ! ! ! ! + + + +
Nasal ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! + + + ! ! ! !
Strident ! ! ! ! ! ! + + ! ! + + + + ! + + ! ! ! ! ! ! !
Lateral ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! + ! ! !
Distributed ! ! ! ! ! ! ! ! ! ! ! ! + + ! + + ! ! ! ! ! ! !
Affricate ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! + + ! ! ! ! ! ! !
Labial + + ! ! ! ! + + ! ! ! ! ! ! ! ! ! + ! ! ! ! ! +
Round ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! +
Coronal ! ! ! ! + + ! ! + + + + + + ! + + ! + ! + ! ! !
Anterior + + ! ! + + + + + + + + ! ! ! ! ! + + ! + + ! +
High ! ! + + ! ! ! ! ! ! ! ! + + ! + + ! ! + ! ! + +
Back ! ! + + ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! + ! ! ! +
Low ! ! ! ! ! ! ! ! ! ! ! ! ! ! + ! ! ! ! ! + ! ! !
Table 4. Distinctive feature analysis of RP English consonants.
By looking at Table 4. one can see that each phoneme has a different
combination of features. The number of such features, for any language, is usually not
higher than fifteen, and the "...phonemes existing in any language must be ultimately
reducible to some of these distinctive features / though of course not all such features
must necessarily be present in the phonological system of any language / " (Vachek
1973: 4).
31
1.4 Transcription
1.4.1 The Definition of Transcription
One of the main problems of the English pronunciation is the fact that the
elements of the written language ! letters, do not correspond to the elements of the
spoken language ! sounds. When we, as learners of English, attend our first English
lesson, we are usually introduced to written and spoken language simultaneously.
What strikes us as unusual is that, unlike in Slovak, where letters and sounds are often
in one to one correspondence, the number of letters in English rarely indicates the
number of sounds. What is more, the same letters or letter sequences can be
pronounced differently in different words. This problem can be found in many other
languages. Because of this fact, the written form of language is not a reliable guide to
the pronunciation of a particular language. That is why there have been many
attempts in the history of mankind to create a set of written symbols that would
consistently and unambiguously represent the sounds people make when they
communicate. Although the choice of symbols differed from scholar to scholar, the
symbols used most often were the ones taken directly from the Roman alphabet, such
as t, d, k, g, m, n, l, etc. The rest of the symbols were either invented, taken over from
other alphabets, or made up by a combination of the symbols. This whole range of
phonetic symbols came to be called the phonetic alphabet. The basic rule to be
followed was a consistent one-to-one correspondence of sound and symbol. In this
way it was possible to create a one-to-one correspondence between the spoken form
of language and the written one. This special written form of language, where one
written element (symbol) represents one spoken element (sound), or the abstract
representation of the spoken element (phoneme), is called transcription
(notation/script). Transcription is "... a method of writing down speech sounds in a
systematic and consistent way ..." (Crystal 1991: 356). Some symbols can, in some
alphabets, consist of more than one character ! aw, ur, ee, etc., but these characters
are considered to be one unit and they are indivisible.
Transcription can be done from different points of view. If our aim is to
indicate only functionally relevant categories of sound which are significant from the
point of view of the meaning of words, we use a simple set of symbols where one
symbol stands for one phoneme. This type of transcription is called phonemic
transcription. If, however, we want to capture the real, physically articulated sounds,
we use a complicated set of symbols and diacritics which can indicate detailed sound
values, that is, the actual allophones. This type of transcription is called allophonic
(phonetic) transcription. When we use allophonic transcription we enclose the
transcription symbols in square brackets [ ]. Phoneme transcription symbols, which
are used in phonemic transcription, are enclosed within slant brackets / /. In
allophonic transcription, the English word "pen#, for instance, might look something
32
like this: [ phj;n ]. This type of transcription represents the real speech sounds and it
is able to indicate, as we can see, such pronunciation features as aspiration following
the initial consonant, or the nasalization of the vowel. A detailed allophonic
transcription of this sort is called narrow transcription. An allophonic transcription
which uses only the functionally relevant symbols, that is, the symbols representing
phonemes is called broad transcription, and the word "pen# would here be
transcribed as [ pen ]. In phonemic transcription, the word "pen# is transcribed as
/ pen /. Although the last two transcriptions look very much alike, it is important to
realize one thing. "The first is a broad phonetic transcription, representing a
sequence of concrete, physical articulations; the second is a phonemic transcription,
representing a sequence of abstract, functional units, and reflecting a particular
theoretical point of view" (Crystal 1991: 357)7.
Transcription
Allophonic transcription Phonemic transcription
Narrow allophonic transcr. Broad allophonic transcr. Simple phonemic transcr. Comparative phonemic transcr.
h
[ p \;? ] [ p\b ] / pab / / pCb /
Fig. 1. The word "pub# transcribed in different types of transcription.
In phonemic transcription, the quality of such vowels as / Ι / ! / i /, / h / !

/ u /, / Â / ! / /, etc. is not functionally relevant. The only functionally relevant
feature here is the length. That is why the word pairs like sit ! seat, full ! fool, sot !
sought, etc. can be transcribed as / sit / ! / sit /, / ful / ! / ful /, / sot / ! / sot /, etc.
without any confusion about the meaning of these words. Such phonemic
transcription, where the minimal number of symbols is used and the counterparts of
short vowels (long vowels) are created just by adding the length mark to the symbols
for short vowels, without any regard to the possible differences in the quality of such
vowel pairs, is called simple phonemic transcription. When, in phonemic
transcription, we use separate symbols for long and short vowel counterparts, for
instance / Ι / ! / i /, / h / ! / u /, / Â / ! / /, etc., we are indicating, that these vowel
counterparts differ not only in length but also in quality. This type of transcription is
7
Some phoneticians, for the sake of simplicity, do not make any distinction between broad and
phonemic transcriptions. In their view, the narrow transcription means the same as allophonic
transcription, and the broad transcription equals phonemic transcription.
33
called comparative phonemic transcription because "... it allows comparison with

vowels in other languages to be made, even though a phonemic transcription is being
used" (Gimson & Cruttenden 1994: 48).
In this book, we will be using both comparative phonemic transcription and

allophonic transcription. Phonemic transcription will be used in cases requiring
abstract treatment of sounds, that is, it will be used to express certain functional or
structural relations among sounds without particular attention to their acoustic
properties. Allophonic transcription (mostly of broader type) will be used for
showing the actual acoustic properties of sounds or selected acoustic phenomena.
We will enclose such transcription in square brackets.
The selection of the symbols used in comparative phonemic transcription is

based on the International Phonetic Alphabet (IPA). The International Phonetic
Alphabet is a system of symbols, that was established by the International Phonetic
Association in 1886. The system is based on an alphabet developed earlier by the
British phonetician Henry Sweet (1845 ! 1912), and most of the symbols come from
the Roman alphabet. Other symbols of the IPA come from Greek, and some were
created as new ones, particularly for this purpose. The IPA also contains a set of
diacritical marks, which modify the symbols to indicate different realizations of
speech sounds.
1.4.2 Phonemic Transcription

Phonemic transcription is a method of using a system of symbols which
represent the set of phonemes of a particular language, following the rule "one
phoneme ! one symbol#. As we have stated in the previous section, there are two
main types of phonemic transcription: simple phonemic transcription and comparative
phonemic transcription.
Let us now take a closer look at each of the two transcriptions.
SIMPLE PHONEMIC TRANSCRIPTION
Simple phonemic transcription can be found in most of the English-Slovak and

Slovak-English dictionaries of the 60s, 70s, and 80s. It is a simplified version of the
transcription of Daniel Jones, and it follows these principles:
34
1. The transcription reflects a functional point of view.

2. The minimum possible number of symbols is used.
3. Using symbols without diacritics.
4. Long vowels differ from short vowels only by a length mark (the differences in
quality are not shown)
Thus, the difference between simple phonemic transcription and comparative

phonemic transcription lies only in the fact that comparative phonemic transcription
uses, in addition to the length mark, a different symbol for each vowel, which
indicates that English vowel-pairs differ not only in length, but also in quality.
The choice of symbols, for any transcription, depends solely on the author of
the transcription, but he must then use the selected symbols in a systematic and
consistent way. Here is an example of the most frequently chosen vowel symbols
used for the simple phonemic transcription:
Vowels Diphthongs
/ i / sit / i / see / ai / eye / i± / here

/ u / look / u / two / ei / day / e± / there
/ o / hot / o / door / oi / boy / u± / sure
/ a / cut / au / now
/ ± / above / ±u / go
/ a / far
/ ± / burn
/ e / bed / / hat
Notice that all vowel-pairs (the short and long counterparts) use the same symbol, and
they differ only by a length mark. The same symbols are also used in diphthongs.
COMPARATIVE PHONEMIC TRANSCRIPTION
This type of transcription was proposed by A. C. Gimson, and it indicates that

English vowel-pairs do not differ only in length, but also in quality. His comparative
phonemic transcription is based on D. Jones# selection of symbols, with several new
symbols and length marks added to the old system. We could therefore see it as a
more sophisticated version of the transcription of D. Jones. It follows these principles:
1. The transcription reflects a functional point of view.

3. Using symbols without diacritics.
4. Long vowels differ from short vowels both by a length mark and a different
symbol, which reflects different quality.
35
Here is an example of the set of English vowel phonemes transcribed in comparative

phonemic transcription:
Vowels Diphthongs
/ / sit / i / see / a / eye / ± / here

/ h / look / u / two / e / day / e± / there
/ Â / hot / / door / / boy / h± / sure
/ Y / cut / /
/ ± / above
far / ah / now
/ / burn / ±h / go
/ e / bed / / hat
Notice that all vowel-pairs use a different symbol, which symbolizes different
quality. In case of / a /-like sounds, there are three different qualities distinguished in
comparative phonemic transcription: / Y / in cut, / / in far, and / a / of / a / in eye.
1.4.3 Allophonic Transcription

Allophonic transcription is a method of using a system of symbols and
diacritics to represent the concrete speech sounds produced by a speaker of a
particular language. In this case, we are not interested in finding the minimum
number of distinctive sounds. Instead, our aim is to transcribe words in a detailed and
accurate manner. It is important to realize that in allophonic transcription we
transcribe speech sounds ! allophones, whereas in phonemic transcription we only
transcribe phonemes, which are functional units of speech.
Allophonic transcription is based on the International Phonetic Alphabet (IPA),
which is a system of all possibly pronounceable vowel and consonant sounds of the
world, combined with a set of diacritical marks, which can add further detail to the
sounds.
Depending on how much detail we want to indicate in the transcription, we
speak about two basic types of allophonic transcription: broad allophonic
transcription and narrow allophonic transcription.
BROAD ALLOPHONIC TRANSCRIPTION
Broad allophonic transcription represents one extreme of allophonic

transcription, that is, it is the most simple allophonic transcription, with little detail.
Words transcribed in this type of transcription may look very much like the words
transcribed in phonemic transcription, e.g. pen [ pen ]. However, as already
mentioned, the difference is that phonemic transcription reflects only a certain
36
functional point of view, not the real articulation, whereas broad allophonic
transcription represents the actual articulated sounds. Of course, if we transcribe
English words in broad allophonic transcription, we are not able to indicate the actual
English accent. This means that even though we are putting down concrete sounds,
these sounds do not sound like English sounds (at least most of them do not) because
we only use the standard, simple and most frequently used IPA symbols, which,
acoustically, may not represent the speech sounds of English, and they are not
modified by diacritical marks to add further detail.
As we have said, the broadest type of allophonic transcription may look like
phonemic transcription, although the choice of the symbols may be conditioned by
the degree to which we want the transcription to be readable and clear. For instance,
the word "pub# may be transcribed as [ pYb ] or [ p\b ]. Since this is a representation
of real articulation, the vowel in the second transcription ! [ \ ] ! is a better choice
because it is closer to the actual quality of the English / Y /. On the other hand, the
vowel symbol in the first transcription ! [ Y ] might be more readily understood.
Thus, the symbol choice in broad allophonic transcription must necessarily come
down to the issue of clarity versus accuracy.
Broad allophonic transcription is based on these principles:
1. The transcription reflects a formal point of view.

3. We use the standard, simple and most frequent IPA symbols.
4. The symbols are without diacritics
NARROW ALLOPHONIC TRANSCRIPTION
This type of transcription represents the other extreme of allophonic

transcription. It is the most detailed sort of allophonic transcription. In between broad
and narrow transcription, there can be different degrees of complexity. Sometimes,
for the sake of clarity, we can transcribe one particular sound of a word in a very
detailed way (we can specify the degree of centralization, palatalization, nasalization,
etc.), and the rest of the sounds of that word without detail, e.g. pub [ p\;b ],
where the sound / Y / is transcribed in detail, whereas / p / and / b / are not.
Narrow allophonic transcription is based on these principles:
1. The transcription reflects a formal point of view.

2. We use as many symbols as necessary.
3. The symbols are modified by diacritics
37
THE INTERNATIONAL PHONETIC ALPHABET (revised to 1993, updated 1996)
CONSONANTS (PULMONIC)
Bilabial Labiodental Dental Alveolar Postalveolar Retroflex Palatal Velar Uvular Pharyngeal Glottal
Plosive
p b t d c k q
Nasal
m n
Trill
r
Tap or Flap

Fricative
f v s z x ! h "
Lateral
fricative
# $
Approximant
% & ' j )
Lateral
approximant
l * + ,
Where symbols appear in pairs, the one to the right represents a voiced consonant. Shaded areas denote articulations judged impossible.
SUPRASEGMENTALS TONES WORD ACCENTS

CONSONANTS (NON-PULMONIC) \ Primary Stress
]foKnP\tIPn LEVEL CONTOUR
Clicks Voiced implosives Ejectives ] Secondary Stress eg or v Extra high e} or Ê Rising
- Bilabial 1 Bilabial 6 Examples ^ Long ej w High e h Falling
. Dental 2 Dental/alveolar p6 Bilabial _ Half-long en x Mid e º High rising
! (Post)alveolar 3 Palatal t6 Dental/alveolar ` Extra-short eq y Low e b Low rising
/ Palatoalveolar 4 Velar k6 Velar Syllable break &iWkt et z Extra low e ¡ Rising-falling etc.
0 Alveolar lateral 5 Uvular s6 Alveolar fricative d Minor (foot) group { Downstep Global fall
e Major (intonation) group | Upstep Global rise
f Linking (absence of a break)
VOWELS
Front Central Back
Close i •y E•G H•u DIACRITICS Diacritics may be place above a symbol with a descender, e.g.
Voiceless n d º Breathy voiced b¹ a¹ Á Dental tÁ dÀ
IJ K
Close-mid e•L M •N O•o Voiced s t ½ Creaky voiced b¼ a¼ Å Apical tÅ dÄ
P h Aspirated Linguolabial Laminal
th dh ¿ t¿ d¾ É tÉ dÈ
Open-mid Q •R S•T U•V More rounded Labialized Nasalized
V w tw dw Ì eÍ
W X
Less rounded V j Palatalised tj dj n Nasal
dn
Open a •Y ! Z•[ release
Where symbols appear in pairs, the one to the Advanced Velarized l Lateral
right represents a rounded vowel. u¡ t d release
dl
Retracted ! Pharyngealized audible
¦ i¦ t! d! Ð No
release
dÐ
OTHER SYMBOLS
ª Centralised e« Ñ Velarized or pharyngealized lÑ
7 Voiceless labial-velar fricatives < = Alveolo-palatial fricatives
w Voiced labial-velar fricatives > Alveolar lateral flap ¯ Mid-centralised e® Ô Raised eÔ (&Ô = a voiced alveolar fricative)
8 Voiced labial-palatal approximant ? Simultaneous and x ° Syllabic &° Ù Lowered eÙ (Ù = voiced bilabial approximant)
9 Voiceless epiglottal fricative Affricates and double articulations Non-syllabic
´ eµ ß Advanced tongue root eß
: Voiced epiglottal fricative can be represented by two symbols
· Rhoticity P· â Retracted tongue root eâ
; Epiglottal plosive joined by a tie-bar if necessary
kBp tAs
Table 5. The International Phonetic Alphabet

Here is an example of allophonic analysis (narrow transcription) of some

English words in RP General accent (all symbols and diacritic marks are taken from
the International Phonetic Alphabet):
tries / tra z / = [ kwGwQ;fD ]
The phoneme / t / is pronounced with the tongue slightly retracted, because it is

preparing for the pronunciation of the following sound / r /. Thus the / t / is not
alveolar, but post-alveolar. The diacritic symbol expressing retraction is [ p ], and it is
placed under the respective sound symbol ! [ k ]. At the same time, the sound [ k ] is
slightly affricated which is expressed by a small raised [ ], and labialized because it
is followed by / r /.
The phoneme / r / is devoiced as a result of the aspiration of the preceding / t /, and
we can transcribe it as [ Gw ].
The following diphthong / a / starts with a centralized vowel [ Q ] gliding to a
lowered central vowel [ f ] with extra short realization.
The final phoneme / z / does not reduce the length of the first part of the preceding
diphthong, which can be phonetically transcribed as [ Q; ] (voiceless consonants
reduce the length of preceding vowels). The phoneme / z / is partially devoiced in the
final position in words (unless followed by a voiced sound), which is transcribed
phonetically as [ D ].
All diacritic symbols can be found in the diacritics table of the IPA.
cat / k4t / = [ hh t ], [ hh t_ ], [ hhn=t ]
The phoneme / k / is pronounced with the tongue slightly advanced, because it is

preparing for the pronunciation of the following sound / /. The diacritic symbol
expressing advancement is [ q ], and it is placed under the respective sound symbol
! [ h ]. The pronunciation of the consonant [ h ] is accompanied by aspiration [ h ].
The RP phoneme / / is transcribed as [ ] which is reduced (shortened) by the
following consonant / t /. When [ ] is followed by [ = ], as in the third example, the
realization of [ ] may be extra short ! [ n ].
The final phoneme / t / can be pronounced in different ways, depending both on the
speaker"s style, the tempo of speech, and other factors. The first possibility is a
voiceless unaspirated alveolar plosive with audible release ! [ t ]. The second variant
is an unaspirated alveolar plosive with no audible release [ t ]. The third variant is a
glottalized unaspirated alveolar plosive with audible release ! [ =t ], which, however,
is not used in all RP accents, although it is becoming more and more widespread. This
phenomenon is called #glottal reinforcement".
40
comfort / ªkYmf±t / = [ ªkhmPfct ], [ ªkhmPfct_ ], [ ªkhmPf;t ]
The initial phoneme / k / is stressed and therefore it is accompanied by aspiration.

Phonetically we can transcribe it as [ kh ].
The next sound is the vowel / Y /, which is nasalized because of the following nasal
consonant / m /. Phonetic representation of this sound is [ m ].
The following nasal consonant / m / is realized as labiodental variant [ P ] because of
the following labiodental consonant / f /.
The consonant / f / is pronounced as a most frequent variant of this phoneme ! [ f ].
The next vowel / ± / can be realized either as an extra-short central vowel [ c ]
(because it is in an unstressed syllable), or, in fast speech, it can be dropped
completely. In such case the preceding / f / can be slightly lengthened to [ f; ].
The final consonant / t / can be realized in two different ways. First as an unaspirated
alveolar plosive with audible release [ t ], or as an unaspirated alveolar plosive with
no audible release [ t_ ].
All diacritic symbols are found in the diacritics table of the IPA.
Note
The list of the most frequent allophones in the RP General accent can be found in 5.1.1.1.
1.5 Physiological Aspects of Speech and the

Organs of Articulation
When we communicate, we "... make use of organs whose primary
physiological function is unconnected with vocal communication ..." (Gimson &
Cruttenden 1994: 9). It means that all of these organs, except perhaps for the vocal
folds, are primarily used for some other purpose than for speaking. The organs which
we use in communication are called organs of articulation or articulators, and they
can be divided into four main groups:
(a) Respiratory organs (lungs, trachea)

(b) Phonatory apparatus (vocal folds)
(c) Resonators (pharyngeal, oral and nasal cavities)
(d) Modulating organs (mandible, lips, teeth, tongue, alveolar ridge, palate, uvula)
41
Oral Cavity
Nasal
Cavity
Hard
Palate
Soft Palate
(Velum)
Alveolar
Ridge
Uvula
Lips
Pharynx
Teeth
Epiglottis
Blade
Tip
Tongue Front
Back
Root
Larynx
Thyroid
Cartilage
Vocal Folds Trachea
Fig. 1. Organs of articulation.
1.5.1 Respiratory Organs

The main source of energy for the production of speech sounds comes from the
lungs. The lungs are spongy, sac-like organs whose primary physiological function is
to remove carbon dioxide from the blood and provide it with oxygen. In other words,
the lungs participate in breathing. When breathing occurs, the air flows in and out of
the lungs. This is done for biological reasons. However, the air can also be expelled
from the lungs for other than biological reasons, and that is what happens during
communication. The energy of the airstream expelled from the lungs is vital for the
production of most human speech sounds. Our speech is therefore dependent of and
42
limited by the capacity and state of our lungs and the muscles controlling the action of
the lungs.
When the air is forced out of the mouth, we call it an egressive pulmonic
airstream. The sounds produced in this way are called egressive pulmonic sounds.
Virtually all English sounds are made in this way. Sometimes, sounds can be
produced by breathing the air in. Such a type of airflow is called an ingressive
pulmonic airstream, and the sounds produced in this manner are called ingressive
pulmonic sounds. This type of sounds is rare.
When the airstream leaves the lungs, it passes through trachea, or windpipe,
which is a thin-walled cartilaginous and membranous tube carrying air to the lungs.
Then the airstream undergoes important modifications when it passes through larynx.
1.5.2 Phonatory Apparatus

The airstream is modified substantially in the upper part of the respiratory tract,
especially when it passes through the larynx. !The larynx is a casing, formed of
cartilage and muscle, situated in the upper part of trachea. Its forward portion is
prominent in the neck below the chin and is commonly called the #Adam$s apple$%
(Gimson & Cruttenden 1994: 10). Within this casing, there are two thick flaps of
muscle rather like a pair of lips, which can be brought together or parted through
muscular tension. This organ is called the vocal folds (vocal cords/vocal chords). The
opening between the vocal folds is known as the glottis. The glottis can assume
different positions of openness, depending on the type of the sounds produced. There
are four recognisable states of the vocal folds:
(a) Wide apart ! The glottis is open. This occurs during breathing and when
producing voiceless consonants.
(b) Loosely together ! The glottis is slightly opened. This occurs during the
production of the voiceless glottal fricative / h /.
(c) Loosely together and vibrating ! The glottis is slightly opened and the air
passing through it causes vibration. This occurs during the production of all
voiced sounds.
(d) Tightly closed ! The glottis is firmly closed. When we close the glottis during
the speech, we produce a glottal stop.
43
Vocal Folds
Glottis
wide apart loosely together

Thyroid
Cartilage
Arytenoid
Cartilages
loosely together and vibrating tightly closed
Fig. 2. The vocal folds and their four recognizable states (Adapted from Laver, 1994).
When the vocal folds are made to vibrate, we can hear sound. The process of
passing the airstream through the vocal folds resulting in audible vibration (sound) is
called voicing or phonation. The sounds during the production of which the vocal
folds vibrate are called voiced (all vowels and voiced consonants). The sounds
produced without any vocal fold vibration are called voiceless (voiceless consonants).
1.5.3 Resonators
When the airstream leaves the larynx, it is !... subject to further modification
according to the shape assumed by the upper cavities of the pharynx and mouth, and
according to whether the nasal cavity is brought into use or not% (Gimson &
Cruttenden 1994: 12).
The pharynx or the pharyngeal cavity extends from the top of the trachea,
past the root of the tongue, to the back part of the soft palate. The shape and size of
44
this cavity can be modified during production of sounds, therefore it affects the
quality of the sounds.
The airstream can leave the pharynx in one of these three ways:
(a) The soft palate may be lowered slightly, and the air escapes both through the
nose and the mouth. Sounds produced in this way are called nasalized sounds.
French nasalized vowels are produced in this way.
(b) The soft palate is lowered, and the air escapes through the nose only, because
there is a complete obstruction made somewhere in the mouth. The sounds
produced in this manner are called nasal sounds. English nasal consonants / m,
n, Ì / are produced in this way.
(c) The soft palate is held in raised position, and the air escapes only through the
mouth. The sounds produced in this manner are called oral sounds. All English
sounds, with the exception of nasal consonants, are made in this way.
When the air escapes through the nose, as in (a) and (b), it passes through the
nasal cavity. The nasal cavity is of fixed dimensions and shape, and it functions as a
resonator. Sounds which pass through the nasal cavity are acoustically modified, they
are nasalized.
When the air escapes through the mouth, as in (c), it passes through the oral
cavity. It is the cavity whose size and shape is formed by the mouth with all
modulating organs. In the oral cavity, the quality of the majority of our speech sounds
is determined.
1.5.4 Modulating Organs

The modulating organs can be divided into two groups:
(a) The passive (fixed) modulating organs (teeth, alveolar ridge, hard palate)
(b) The active (mobile) modulating organs (tongue, mandible, lips, soft palate,
uvula)
The passive modulating organs
The teeth are a set of hard bone-like structures rooted in sockets in the jaws of
most vertebrates. They are divided into upper teeth and lower teeth. Many speech
sounds are formed by the contact of the tongue and the upper teeth. Sounds made
with the tongue touching the front teeth are called dental.
The alveolar ridge is between the top front teeth and the hard palate. Sounds
produced with the tongue the alveolar ridge are called alveolar.
45
The hard palate, which is often called #roof of the mouth", stretches between
the alveolar ridge and the soft palate. The sounds made with the tongue touching here
are called palatal.
The active modulating organs
The tongue is probably the most important articulator, and it can assume many
different positions and shapes. It is usually divided into five parts ! tip (apex), blade
lamina), front (pre-dorsum), back (post-dorsum), and root (radix). The muscular
tension of the tongue is an important factor in the articulation of speech sounds.
According to this factor, vowels are divided into tense ! produced with strong
muscular tension ! (/ i /, / u /, etc.) and lax ! produced with weak muscular tension
! (/ /, / h / etc.). The consonants are, in this respect, divided into fortis (/ p /, / t /
etc.) and lenis (/ b /, / d / etc.).
The mandible is the lower jaw. Depending on its position, it can influence the
openness of speech sounds.
The lips are two fleshy muscular folds that together surround the opening of
the mouth. They can be pressed together, in which case they produce sounds we call
bilabial. The sounds made by lips touching the teeth are called labiodental.
The soft palate or velum is the soft fleshy part of the back portion of the roof
of the mouth. It plays an important role in determining the quality of sounds (oral,
nasal). The sounds produced by the tongue touching the velum are called velar.
There is a small conical, fleshy mass of tissue suspended from the centre of the
soft palate above the back part of the tongue. This organ is called the uvula. The
sounds formed by a vibration of the uvula against the back of the tongue are called
uvular. French and German r-sounds are pronounced in this way.
1.6 Acoustic Aspects of Speech

In the following sections we will attempt to describe the sounds of language
from the point of view of acoustics. We will examine the way a sound is transmitted
through the air, and we will also try to show how a sound can be analyzed in terms of
its fundamental frequency, intensity, quantity, and quality.
1.6.1 Sound Waves

"The formation of any sound requires that a vibrating medium should be set in
motion by some kind of energy. We have seen that in the case of the human speech
mechanism the function of vibrator is often fulfilled by the vocal folds, and that these
46
are activated by air pressure from the lungs" (Gimson & Cruttenden 1994: 19). The
medium through which speech sounds travel is air. When our vocal folds start
vibrating, they cause variation in the density of the air particles. This variation
manifests itself in compression and rarefaction of these air particles. In other words,
the air (its particles) is alternately squashed and expanded. By this squashing and
expanding movement, the signal (in our case a speech sound) spreads in all
directions. The wave created by compressing and expanding of the air particles
carrying sound is called a sound wave. Sound waves move through air just like
waves on a pond. A sound wave is a type of waveform, and it can be visually shown,
for example, as an oscillogram (see Fig. 5).
When the air comes out of the oral cavity, the vibration we hear can be of two
types.
(a) Regular (periodic) vibration.
(b) Irregular (aperiodic) vibration.
REGULAR VIBRATION
Regular vibration is produced by the vocal folds, providing that they are
vibrating at a constant frequency and with constant intensity. Such periodic vibration
manifests itself as a periodic wave. Periodic waves are called periodic because of the
fact that they consist of certain wave patterns that periodically (at equal intervals)
repeat themselves. These waves can be of two types:
(a) Simple periodic waves

(b) Complex periodic waves
Simple periodic waves are almost non-existent in the domain of naturally

produced sounds. They can be shown as simple sine waves.
Fig. 3. A sine wave. T = one cycle, a = amplitude.
47
Sine waveforms are often called pure tones. "The sound of a tuning fork is one
example of a pure tone; when people whistle they often produce pure tones, and a few
musical sounds, for instance some low notes on the flute, provide further examples. It
is very rarely in nature, however, that pure tones or sinusoidal sound vibrations are
generated. Practically every sound we hear is not a pure tone but a complex tone; its
waveform is not a sine wave but a complex wave" (Fry 1979: 20).
A complex periodic wave is a wave in which several simple sine waves are
combined into one single wave. Sounds produced by the vocal folds are complex
periodic waves. However, one cycle of such a wave is never exactly the same as the
next cycle. Such identical cycles can only be generated by a computer (see Fig. 4).
The reason why the cycles in the periodic waveforms produced by a human being are
not completely identical is because the organs of articulation are in constant
movement during the production of a sound. Nevertheless, when the neighbouring
cycles are approximately the same during a certain period of time, the waveform is
said to be periodic (or quasi-periodic).
Fig. 4. Three simple sine waves of different frequencies (the upper row) combined into a
single complex periodic wave (the lower row). (Speech Analyser 1.5, SIL)
48
In Fig. 4 we see a computer-generated complex periodic waveform in which

three simple sine waves of different frequencies are combined into one. The resulting
waveform shows a repeating pattern which occurs one hundred times in one second.
This means that the complex wave has a frequency of 100 Hz. This frequency is
called the fundamental frequency (Ladefoged, 1996). For more on the fundamental
frequency, see 1.6.2.1.
IRREGULAR VIBRATION
Irregular vibration manifests itself as an aperiodic (random) wave. Aperiodic

waves do not show a regular, periodic pattern. They are composed of a large number
of sine waves of different frequency and intensity, and there is no repetition of any
pattern, no repetitive structure, no regularity. An aperiodic wave can be characterized
by a continuous distribution of energy along a certain frequency scale. This means
that the individual sine waves out of which the aperiodic waveform is composed are
found in different positions on a frequency scale.
An aperiodic wave can be produced when, for example, several people
simultaneously start speaking. There will be no repetition of a pattern of any kind. In
this particular case, the waveforms will be created by the vibration of the vocal folds.
However, aperiodic waves can also be created without the vocal-fold vibration. In
fact, this is the most frequent way of producing an aperiodic wave in the speech of a
single individual. All voiceless fricatives, affricates and plosives, for instance, are
produced in this manner. Here, the aperiodic wave is created by a turbulent airflow
generated in the oral cavity by the mutual contact of articulators. It is not formed in
the larynx by the vocal folds like periodic waves are.
Fig. 5. A section of the oscillogram of the sound / s /. The waveform is irregular. (Speech
Analyser 1.5, SIL)
Fig. 5 shows a section of the sound / s /. As one can see, there is no recurring
pattern, no regular structure.
49
All sounds produced during speech consist of complex periodic and/or

aperiodic waveforms. The sounds which consist of periodic waves are called tones,
the sounds which consist of aperiodic waves are called noises (Malmberg, 1963). All
sounds produced by the periodic vibration of the vocal folds are tones. This means
that all vowels and sonorants (nasal sonorants and oral sonorants) are represented by
tones. Voiceless obstruents (voiceless fricatives, affricates, and plosives) all consist of
aperiodic waveforms and are therefore represented by noises. A little more
complicated is the case of voiced obstruents (voiced fricatives, affricates, and
plosives). During the production of these sounds there are two sources of vibration.
One is the vocal folds, which generate periodic waves, the other is the closure or the
narrowing in the oral cavity which generates aperiodic waves. The two waves
(periodic and aperiodic) are therefore combined into one waveform.
To sum up the above we can say that:
Vowels and sonorants are produced by regular vibration (tones).

Voiced obstruents are produced by both regular and irregular vibration (tones and
noises).
Voiceless obstruents are produced by irregular vibration (noises)
1.6.2 Acoustic Properties of Sounds

Every sound can be examined from the point of view of these four acoustic
properties: fundamental frequency, intensity, quality (wave structure) and quantity
(duration).
1.6.2.1 Fundamental Frequency

Frequency is a technical term and it refers to "... the number of complete
repetitions (cycles) of variations in air pressure occurring in a second" (Ladefoged
1975: 162). As Fig. 3 shows, frequency is the number of cycles that occur in one
second. The shorter the cycles are, the higher the frequency is. The unit of frequency
is the Hertz, and it is usually abbreviated to Hz. The human ear is capable of hearing
the sounds ranging from 20 to 20 000 Hz. However, most human sounds are
recognizable within the frequency scope as narrow as 80 ! 350 Hz. "A male voice
may have an average pitch level of about 120 Hz and a female voice a level in the
region of 220 Hz" (Gimson & Cruttenden 1994: 22). This means that in a male voice,
there are 120 complete cycles of compressing and expanding of the air particles in
one second, because the vocal folds made 120 complete opening and closing
movements in one second.
50
Pitch, unlike frequency, is not an exact measurable property. It is a listener"s

subjective perception of the height of the tone (see Table 3). Pitch is considered in
terms of ranging from low to high. The important factor in deciding the pitch is the
pitch range of the individual speaker, the sex and age of the speaker, and the pitch
environment. For instance, a woman speaking in a low voice may produce a sound
having the frequency of 120 Hz. Since she is a woman, this frequency is perceived as
low pitch. If, however, a man makes a sound with the same frequency, his pitch is
considered as medium.
The pitch range of an individual does not stay constant throughout his life. Fig.
6 shows the change of the fundamental frequency (for more, see the 1.6.2.4) of an
individual with relation to his age.
300
Male
Female
Mean speech frequency in Hz
250
200
150
10 20 30 40 50 60 70 80 90
Age in years
Fig. 6. The change of the fundamental frequency of an individual with relation to his age
(Adapted from Carrell & Tiffany, 1977).
Speech sounds do not consist of just one frequency (sine wave), but of many
different frequencies, that is, several combined sine waves. These complex periodic
waveforms, however, have one basic frequency. It is called the fundamental
frequency ! F0. The fundamental frequency equals the highest common denominator
of the individual constituent frequencies. For example, let us suppose that we have a
complex waveform consisting of three simple sine waves of different frequencies;
51
200, 500, and 800 Hz. The highest common denominator of these three numbers is
100, because it is the highest number by which all three waves can be divided. This
means that the fundamental frequency is 100 Hz. In other words, the repeating pattern
of the complex periodic waveform will be found one hundred times in the time-span
of one second.
The vocal folds vibrate in such a way that, in addition to the fundamental
frequency, they also produce a number of overtones or harmonics. Harmonics are
additional frequencies (sine waves) to the fundamental frequency, and they are simple
multiples of the fundamental frequency (O"Connor, 1973). For instance, if the
fundamental frequency is 100 Hz, the harmonics will be found at whole-number
multiples of the fundamental frequency ! 200 Hz, 300 Hz, 400 Hz, etc. We will
return to this issue in the section 1.6.2.4.
Fundamental frequency is of particular importance in stress and intonation,
because the pitch of the voice depends mainly on this frequency.
1.6.2.2 Intensity
Intensity is the amount of energy which is carried by a sound wave. It depends
on the variation of air pressure coming from the lungs. Intensity is usually measured
in decibels, which are abbreviated to dB. Decibels are not absolute quantities like
meters or grams. They express relative values, that is, ratios between the individual
intensities. "When one sound has an intensity 5 dB greater than another, then it is
approximately twice as loud" (Ladefoged 1975: 163). Some typical values can be
found in Table 6.
Intensity relates to amplitude of vibration. The higher the intensity, the bigger
the amplitude of a wave (see Fig. 3). Intensity is the term used in acoustic phonetics
and it corresponds to loudness, which is an auditory term (see Table 3). Loudness
and intensity are in a similar relation as pitch and frequency. Loudness, although it
depends on intensity, is only a subjective perception of intensity. A sound with the
intensity of 100 dB sounds loud to a normal person, but it may sound quiet to
someone who is hard of hearing. Similarly, the loudness of the sound may depend on
the loudness of the sounds produced before and after it. If someone whispers in your
ear and then suddenly produces a voiced sound of average intensity, you may
perceive this sound as loud. Thus, the factors influencing loudness are numerous: the
intensity of the sound, the state of the listener"s hearing mechanism, the distance from
the sound source, the sex and age of the speaker, the sound environment, etc.
52
Intensity
(dB) Sound
__________________________________________________
130 4-engined jet aircraft, 120 ft
120 Threshold of pain; pneumatic hammer, 3 ft
110 Boilermakers" shop; #rock" band
100 Car horn, 15 ft; symphony orchestra, fortissimo
90 Pneumatic drill, 4 ft; lorry, 15 ft
80 Noisy tube train; loud radio music
75 Telephone bell, 10 ft
70 Very busy London traffic, 70 ft
60 Conversation, 3 ft; car, 30 ft
50 Quiet office
40 Residential area, no traffic; subdued conversation
30 Quiet garden; whispered conversation
20 Ticking of watch (at ear); broadcast studio
10 Rustle of leaves
0 Threshold audibility
Table 6. Decibel scale of common sound intensities (Fry, 1979).
It is a fact that two sounds which have identical amplitude of vibration may not
be equally sonorous. This means that although they have an identical intensity, one of
them sounds to the human ear louder than the other. This is usually due to their
inherent quality ! sonority. Sonority is the inherent distinctiveness of sounds. It is "...
the overall loudness of a sound relative to the others of the same pitch, stress, and
duration" (Crystal 1991: 320).
Some sounds appear more sonorous than others. Usually, vowels are more
sonorous than consonants. These facts are closely connected with the sonority
hierarchy theory, which we will discuss in later chapters.
1.6.2.3 Quantity (Duration)

Quantity is the duration of the vibration of a particular sound. Duration is an
acoustic term. In auditory phonetics it corresponds to length (see Table 3). Although
duration is a measurable property, it is very difficult to measure. One problem lies in
the segmentation of the utterance, which is a string of speech sounds influencing one
another considerably. The other difficulty lies the amount of stress imparted to
individual sounds, and also in the fact, that an utterance can be spoken in a fast or
53
slow manner. Although there are terms such as long and short used for describing the
length of sounds in the language, it will be demonstrated in the later chapters that
these terms are, in absolute acoustic sense, misleading. This, for instance, applies to
English vowels, the real length of which depends on the type of the following sound.
On the other hand, if we consider length as a part of certain system, we can, and we
do use terms short and long in describing the sounds of language, since length
functions as an important factor in distinguishing one word from another. Thus, we
can speak about two categories of length:
(a) Absolute phonetic duration, which is a measurable property.

(b) Functional phonological (linguistic) length, which can bring about a change in
meaning of the word.
It has been measured that in connected speech there are about 6 to 20 sounds
pronounced in one second, but this number can be much lower and much higher. "The
time required for the recognition of a sound will depend upon the nature of the sound
and the pitch, vowels and consonants differing considerably in this respect, but it
seems that a vowel lasting only about 4 msec may have a good chance of being
recognized" (Gimson & Cruttenden 1994: 25).
1.6.2.4 Quality (Wave Structure)

The quality (timbre/tambre) is defined by the American Standards Association
as follows: Quality (timbre) is "... that attribute of auditory sensation in terms of
which a listener can judge that two sounds similarly presented and having the same
loudness and pitch are dissimilar" (American Standards Association, 1960). Quality
depends on many factors. When vocal folds vibrate, they produce a complex, but
regular, vibratory motion. The result of this vibration is a glottal tone (it is called
glottal because it is formed in glottis). The glottal tone is a complex waveform ! a
waveform composed of several sine waves (Broshnan & Malmberg, 1970). Let us
now take a look at what exactly happens. The vocal folds vibrate at a certain basic
frequency. This frequency is called the fundamental frequency. As we mentioned
earlier, the fundamental frequency is the rate at which a regular pattern of a complex
periodic wave repeats itself. Since the fundamental frequency is the number of cycles
per second, it is perceived as the pitch of a sound. However, when the vocal folds
vibrate, they vibrate not only as a whole (at a frequency dependent on their length,
mass, and tension), but also, the individual sections of the vocal folds vibrate. These
sections of the total length of the vocal folds vibrate at simple multiples of the
fundamental frequency. To give an example, let us suppose that the whole length of
the vocal folds vibrates at 150 Hz. The sections will vibrate in such a way as to
produce whole-number multiples of the fundamental frequency, that is, new sine
waves are generated with the frequencies 300 Hz, 450 Hz, 600 Hz, etc. These new
54
frequencies (sine waves) are called harmonics or overtones. The total sum of these
frequencies is called the harmonic structure.
A fully developed harmonic structure can be seen in Fig. 8. It is a spectrogram
of the Slovak vowel / a /. #Fully developed" in this case means that the harmonics are
found at all whole-number multiples of the fundamental frequency, that is, 2, 3, 4, 5,
etc. A look at the spectrogram will, indeed, show that this is the case. The
fundamental frequency is 155 Hz, and the harmonics are found at all whole-number
multiples of the fundamental frequency: 310 Hz, 465 Hz, 620 Hz, etc.
Fig. 7. The spectrogram of a complex periodic waveform (200, 500, and 600 Hz)
decomposed into individual component frequencies. The sine waves with higher amplitude
are shown as darker lines, whereas the sine waves with lower amplitude are presented as
lines of lighter colour. (Speech Analyser 1.5, SIL)
All sounds produced by the vibration of the vocal folds have a fully developed
harmonic structure. This complex periodic waveform, however, does not distinguish
vowels from one another. It is the resonators above the larynx ! pharyngeal, oral and
nasal cavities, that modify the sound and give it typical vowel qualities. Here"s what
happens:
The sound with a fully developed harmonic structure generated in glottis
travels through several cavities of the vocal tract. The shape and size of these cavities
differ from one individual to another. Also, every speaker can modify this shape and
size depending on the position and height of the tongue, the shape of the lips, and the
position of the soft palate. When sound passes through these cavities, the cavities act
as a filter which selectively amplifies (intensifies) certain frequencies of the source
signal and dampens (reduces) others. In this way, the cavities add a typical colouring
to the original signal generated in the vocal folds. In other words, some harmonics
55
(frequencies) are intensified, others are reduced. This is how different types of vowels
come into existence.
A group of adjacent high-intensity harmonics (harmonics amplified by the
resonators), which is clearly marked off from the harmonics with lower intensity, is
called a formant.
Both formants and harmonics can be shown graphically in a spectrogram as
horizontal lines of varying intensity. A spectrogram is a visualization of a sound. It
is a picture of a complex waveform decomposed into a set of simple sine waves with
different frequencies and amplitudes. The spectrogram offers an analysis in three
dimensions. The first (horizontal) dimension represents time, the second (vertical)
dimension represents frequency, and the third dimension (distinguished according to
the colour intensity of the vertical lines in the spectrogram) represents amplitude.
There are two basic types of spectrogram ! #narrow-band" and #broad-band".

Narrow-band spectrograms have a higher resolution in the frequency
dimension (the frequency is analyzed more finely) and a lower resolution in the time
dimension (the time window is larger and therefore analyzed with less detail). We use
the narrow-band spectrogram if we want to inspect separate harmonics (the whole-
number multiples of the fundamental frequency) of a sound.
Fig. 8. A narrow-band spectrogram of the Slovak vowel / a /. The separate horizontal black
lines with equal distance in between each other are the harmonics. (Speech Analyser 1.5,
SIL)
56
Broad-band spectrograms, on the other hand, have a higher resolution in the

time dimension (the time window is smaller and therefore analyzed with more detail)
and a lower resolution in the frequency dimension (the frequency is analyzed only
coarsely). This coarse analysis of the frequency causes the adjacent harmonics
to merge, because the resolution is so low that it is difficult to separate the individual
frequencies. The result of this is that groups of adjacent higher-intensity harmonics
(shown in the darker colour) can be clearly distinguished from the other, less
intensive harmonics (shown in the lighter colour), and they are no longer visible as
separate harmonics, but as thick black horizontal bars. These bars are the formants.
The presence of formants in a spectrogram is a characteristic feature of most voiced
sounds. Formants, like harmonics, are numbered from bottom to top.
Fig. 9. A broad-band spectrogram of the Slovak vowel / i /. The thick black horizontal bars
are the formants. Formants determine the quality of vowels. (Speech Analyser 1.5, SIL)
Voiceless consonants do not have formant structure. The reason for this is that
they don"t have a fundamental frequency, thus no harmonics in the traditional sense
are formed. In spectrograms, voiceless consonants show up as a comparatively
random pattern of energy placed over the whole range of the frequency scale, but
usually with a focus of energy in a certain part of this scale.
The individual formants may play different role in the overall make-up of a
sound. Vowels, for instance, are mainly distinguishable by the first and the second
formant ! F1 and F2 . The position and shape of these two formants are very similar
for all speakers of the same accent. The position and shape of the other formants vary,
and the formants above the fourth formant determine the speaker"s voice quality.
Every formant depends on all parts of the vocal tract. The degree of the
influence of the individual parts of the vocal tract on a particular formant varies, and
it depends on the overall configuration of the vocal tract.
57
Fig. 10. A spectrogram of the voiceless consonant / /. Notice the random noise pattern
with the focus of energy starting in the region of about 2,000 Hz and extending upwards.
(Speech Analyser 1.5, SIL)
While the spectrogram analyses the sound in three dimensions (time,

frequency, amplitude), the #spectrum" is an analysis of the sound in two
dimensions only: frequency and amplitude. In other words, a spectrum is a two-
dimensional representation of a sound, with frequency displayed along the horizontal
axis and intensity along the vertical axis. It shows a static picture, a snapshot, as it
were, taken at a certain point in time. A spectrogram, on the contrary, gives a
dynamic picture of a stretch of sound.
Fig. 11. The spectrum of the Slovak / a /. The sharp peaks correspond to the individual
formants. (Speech Analyser 1.5, SIL)
58
1.7 Auditory Aspects of Speech
1.7.1 The Auditory System

The human auditory system is a complex of hearing organs whose function is
to receive acoustic stimuli (sound waves) and convert them into neural impulses
analysable by the brain. It can be divided into the peripheral auditory system and the
internal auditory system.
1.7.1.1 The Peripheral Auditory System

It is convenient to treat the peripheral auditory system as a mechanism
composed of three parts ! the outer ear, the middle ear, and the inner ear.
Auditory ossicles
Oval window
Malleus Incus Stapes
Pinna
Auditory nerve
Eardrum
Cochlea
Meatus
Eustachian tube
Fig. 12. The peripheral auditory system (Adapted from Clark & Yallop, 1995).
The outer ear consists of the externally visible part called the pinna (auricle),
and the meatus (the ear canal). The pinna does not make much contribution to our
basic hearing ability, but it may be of help in terms of sound localization ! it enables
us, to a certain extent, to determine what direction the sound comes from. The ear
59
canal is a short air-filled tube of about 3 cm in length, and it has two main functions.
First, it serves as a passageway channelling sound waves to the eardrum. The second
function is to act as an acoustic resonator which amplifies the frequencies between
about 2,000 and 4,000 Hz. This means that the sounds with the frequencies between
2,000 and 4,000 Hz are much louder (have greater intensity) at the eardrum than they
are outside of the ear. "This effect enables us to detect sounds that would be
imperceptible if the eardrum were located at the surface of the head" (Denes &
Pinson 1963: 66). For example, it is especially helpful for the perception of fricative
sounds.
The middle ear is a mechanism whose purpose is to "... transform the sound
pressure variations in air that arrive at the outer ear into equivalent mechanical
movements" (Clark & Yallop 1995: 304). It consists of the eardrum and the
auditory ossicles. The eardrum is a small membrane at the end of the ear canal
connected to the auditory ossicles. The auditory ossicles are three interconnected
bones called the malleus (hammer/mallet), the incus (anvil), and the stapes (stirrup).
The auditory ossicles perform two main functions ! they increase the amount of
acoustic energy transferred to the inner ear (this amplification enables us to hear very
weak sounds), and protect the inner ear from very loud sounds. The middle ear is
connected to the pharynx by the Eustachian tube. It is a narrow tube ensuring that
"... the air pressure in the normally enclosed middle ear can be equalized with
outside air pressure quickly and easily if they become very different, as when an
aircraft gains height" (O"Connor 1973: 97).
The inner ear is a small intricate system of cavities. The most important of
these cavities is called the cochlea. The cochlea is a fluid-filled organ "... responsible
for converting mechanical movement into neural signals: the mechanical movement
conveyed to the oval window by the auditory ossicles is transformed into neural
signals that are transmitted to the central nervous system" (Clark & Yallop 1995:
304). Enclosed within the cochlea there is a membrane called the basilar membrane.
This membrane transmits the vibrations received from the auditory ossicles to the
Organ of Corti, which is a complex structure containing many minute sensory
receptors ! hair cells. These hair cells transform the vibrations of the basilar
membrane into neural signals.
60
Reissner"s
membrane
Organ
of Corti
Basilar
membrane
Fig. 13. Cross-section of the cochlea (Adapted from Denes & Pinson, 1963).
1.7.1.2 The Internal Auditory System

Man"s nervous system is composed of a large number of specialized cells
called neurons. Neurons are interconnected and form a highly complex network. The
places where neurons are connected to other neurons are called synapses. In addition
to neuron-to-neuron junctions, synaptic junctions also exist between neurons and
receptor cells and neurons and effector cells. "Receptor cells, such as the hair cells in
the Organ of Corti, receive sensory information from their environment and help to
code this information into the electrochemical pulses that are transmitted and
processed in the nervous system. Effector cells, such as those in muscle fibers,
respond to the electrochemical pulses sent to them along nerve fibers" (Denes &
Pinson 1963: 95).
The auditory nerve connected to the cochlea contains about 30,000 neurons.
"Under a stimulus of an intensity above its threshold value, a neuron responds by
firing, that is, by transmitting ... a series of impulses or waves of activity of electro-
chemical nature. Normally a stimulus activates a number of neurons and the
responses in these set up similar responses in other neurons across regions of contact
known as synapses" (Broshnan & Malmberg 1970: 165).
It is still not fully understood what happens when the acoustic signals from the
ear, converted into neural signals, reach the brain. What is clear, however, is that the
activity of the brain connected with speech is localized. It is mainly the left
61
hemisphere of the brain that is responsible for controlling our language functions,
more specifically, there are three important areas: Wernicke"s area, Broca"s area, and
the supplementary motor area. This process of the hemispheres of the brain
specializing in certain functions is known as lateralization.
Wernicke"s area, it is claimed, is responsible for our understanding of the
signals received. This means that the signal is broken down into the smallest
functional units of speech ! phonemes, and then reconstructed into higher units !
syllables, words, sentences, etc. The result of this analysis may not always be
identical to the original signal. For instance, the original word my may be analyzed,
for some reason, as buy, or pie, etc.
When the speaker wants to react to the signal received, he needs to form his
own signal. This process is done in Broca"s area. Here, the reply to the original
signal is formulated and constructed into words. After that, a signal is sent to the
motor area, which is responsible for the actual physical articulation of the reply.
This, of course, is a very simplified version of what may actually take place. It
is important to view the brain as a network where signals may spread in all directions,
rather then just linearly. So far, neither the mechanism of transmitting, receiving, and
processing of the sound stimuli, nor the process of reacting to the stimuli have yet
been fully explained. However, the fact remains that speech recognition and
production is normally located in the left hemisphere of the brain.
1.7.2 The Perception of Sound

On the level of perception, we are dealing with experiences that are purely
subjective. To obtain any experimental data on sound perception means exposing a
human subject to some auditory stimulus, and asking him to tell us what (s)he hears.
Due to the fact that the auditory sensitivity of different people is not necessarily the
same, the data collected in this way is often variable.
The perception of sound can be analysed in terms of these four auditory
categories: pitch, loudness, qualitative impression, and length.
1.7.2.1 Pitch
Pitch is a subjective sensation primarily connected with frequency. There are,
however, other factors than frequency influencing our perception of pitch. "For
example, the pitch of a tone depends to some extent on the intensity at which it is
presented to a listener. This is particularly noticeable at either very high or at very
low frequencies. If we strike a low frequency tuning fork (say, about 150 cps1), its
pitch decreases noticeably as the fork is brought closer to the ear" (Denes & Pinson
1
cps (cycles per second) = Hz
62
1963: 84). This is true mainly for simple sine waves. The pitch of sounds with
complex waveforms varies only slightly as the intensity changes.
The unit of pitch is the mel. Mels are subjective units determined through
experiments. It has been found out that our perception of pitch is not proportional to
frequency. For instance, a tone with a pitch of 1,000 mels has a frequency of 1,000
Hz, but a tone with a pitch of 2,000 mels (that is, a pitch sounding twice as high as a
1,000 mel tone) has a frequency of 4,000 Hz. This is another reminder of the
distinction we must make between the subjective sensations (pitch) and the physical
property of sound (frequency).
3000
2500
SUBJECTIVE PITCH IN MELS
2000
1500
1000
500
0
10 20 50 100 200 500 1000 2000 5000 10,000
FREQUENCY IN HZ
Fig. 14. The mel scale of pitch, showing how pitch is related to frequency for pure tones
(Adapted from Denes & Pinson, 1963).
When we combine several simple sine waves into a complex periodic

waveform, the pitch is perceived as the greatest common denominator of the sine
wave frequencies. Let us suppose that the complex periodic wave is composed of
these frequencies: 300 Hz, 400 Hz, 700 Hz and 1000 Hz. The greatest common
denominator (the greatest number by which all of them can be divided) is 100.
Therefore the perceived pitch (and the fundamental frequency) of this complex
periodic waveform will be that equalling 100 Hz (Ladefoged, 1996).
63
All voiced sounds are in fact periodic waves (with or without a noise
component) and they can be evaluated in terms of their pitch. Voiceless sounds are
noises without the fundamental frequency, and their definite pitch cannot be detected.
Despite this fact, it would be unfair to say that we have no idea of the pitch of these
sounds. The fricative consonants / s / and / /, for instance, can easily be
distinguished from each other by our general impression of their pitches ! the first
one is higher-pitched than the second one (O"Connor, 1973).
The range of frequencies perceptible by the human ear lies between 20 and
20,000 Hz. The actual frequency range that one can hear depends, to a great extent,
on the age of the listener. Generally speaking, the older we are, the narrower our
range of frequency perception gets. Fig. 15 shows the relation between age and the
audible frequency range. It should also be noted that our general hearing acuity lies in
the frequency range between 500 and 4,000 Hz, which is the range of frequencies
carrying most cues for the phonological identification of sounds produced in speech
(Clark & Yallop, 1995).
20-20,000
Audible frequency range in Hz
30-18,000
40-16,000
40-14,000
40-12,000
40-10,000
50-8,000
0 10 20 30 40 50 60 70 80
Age in years
Fig. 15. The relation between age and the audible frequency range.
In connection with the fundamental frequency of a sound a question arises of

what the minimal difference of the pitch is that a human listener is capable of
noticing. This minimal difference, also called the just-noticeable difference (JND), or
difference limen (DL), appears to be about 1 Hz for the frequencies from 80 to 160
Hz (Laver, 1994), 2 to 3 Hz for the frequencies of about 1,000 Hz (KráĐ & Sabol,
1989), 6 Hz for the frequency of 3,000 Hz, and 14 Hz for the frequency of the order
of 7,000 Hz (Ladefoged, 1996).
64
1.7.2.2 Loudness
Loudness is a perceptual dimension which is primarily related to sound
intensity. As with pitch and frequency, however, the relation is not simple. The unit
of loudness level is the phon. Experiments have been carried out to determine the so
called loudness level contours, as presented in Fig. 16. The loudness level of a given
tone is defined as the intensity (measured in decibels) of a 1,000 Hz tone that sounds
equal in loudness to the given tone (Denes & Pinson, 1963). For instance, any tone on
the 20 phon contour sounds equally loud to the human ear, even though the actual
intensity of these tones is different ! a 200 Hz tone must have an intensity of 40 dB to
be judged equally as loud as a 1,000 Hz tone, which only has the intensity of 20 dB.
It is obvious that very high and very low tones need to be of much greater intensity to
be perceived as equally loud in comparison with the tones in the middle of the ear"s
frequency range.
INTENSITY LEVEL IN DECIBELS
FREQUENCY IN HERTZ
Fig. 16. Loudness level contours plotted against intensity and frequency (Adapted from
Pierce & David, 1958).
The voiced sounds we produce in speech are composed of a number of

frequency components of different intensity. These differences in intensity,
nevertheless, may not necessarily be perceived in terms of loudness. "Rather we
perceive them as differences in quality; if the intensity of the second formant of a
vowel is weakened we do not react by saying that the second formant is less loud or
65
even that the whole vowel is less loud. Our reaction is to say that the vowel has
changed its quality, its colour" (O"Connor 1973: 102). The intensity is therefore
perceived in two ways. On the one hand we relate it to loudness (changes in intensity
bring about changes in loudness), on the other hand we relate it to quality (changes in
intensity may cause changes in quality).
The difference limen for overall vowel intensity is about 1.5 dB. For voiceless
consonants the limen appears to be of the order of 0.4 dB for sounds louder than 30
dB (Flanagan, 1972).
1.7.2.3 Qualitative (Auditory) Impression

Our perception of quality is related to the spectral pattern of sound ! the
number of waveform components, their frequencies, their intensities, and their mutual
composition within the frequency range.
A. VOWELS
Naturally produced vowels contain many different formant frequencies. It has

been confirmed by experiments with artificial speech that for the accurate perception
of vowels only the first two formants ! F1 and F2 ! are needed. The formants above
the first two do not seem to contribute much to the phonological structure of a sound,
but they may be important for the identification of the speaker and determine his
voice quality.
200
300
400
F1 (Hz)
500
600
700
800
900
2,600 2,200 1,800 1,400 1,000 800 600
2,400 2,000 1,600 1,200
F2 (Hz)
Fig. 17. F1/F2 plot of the Australian English monophthongs (Reproduced from Clark &
Yallop, 1995).
66
Experiments have also shown that there is a great degree of variability among
supposedly identical vowels, and, more importantly, that the concrete realizations of
different vowel phonemes may overlap (see Fig. 17). In addition to this there is
further diversity to be found in the vowel qualities depending on the age of the
informants (children, adults, old people), the sex of informants (men, women), and
the present state of informants" organs of articulation, to name but a few. In spite of
this great acoustic variability, we are still able to distinguish the individual vowel
phonemes from each other. This perceptual identification of vowel segments (or of
any sounds, for that matter) depends on two main factors:
(a) The acoustic parameters of the segment itself ! its wave structure
(b) The context.
The context for the perceptual identification of a sound can be of three types.
1. Phonetic context ! speech sounds often depend for their identification on the
dynamic co-articulatory transitional information contained in the neighbouring
sounds. In other words, we identify a sound not only on the basis of its own qualities,
but also on the basis of the qualities of the preceding and succeeding sounds.
2. Phonological context ! the reason we are able to disregard the immense variation
in the quality of vowels is because we perceive them within a certain system, where
the quality of each vowel is judged against all the other vowels functioning within a
closed phonological system.
3. Morpho-lexical context ! we may adjust our perception of sounds in order to be
able to arrive at meaningful units on morphological and lexical levels. Put differently,
we try to make sense of what we hear, and we disregard any possible variation not
influencing meaning.
B. CONSONANTS
Since the production of consonants involves a great variation in terms of the

manner and place of articulation, there are great differences in the way we perceive
them. Generally, however, we can place them into two main perceptual categories:
obstruents2 ! noise (+ tone) consonants, and sonorants ! noise consonants.
1. Obstruents ! Noise (+ tone) consonants
This group of consonants gives a general impression of a noise component

being present in the sound. This noise component may, or may not, be combined with
2
The term itself is indicative of an obstruction, which places it in the realm of articulatory
phonetics, but we will use it here as an auditory term.
67
a tone component. Obstruents in English may be divided into plosives, fricatives, and
affricates.
(a) Plosive consonants
Auditorily, plosive consonants consist of a very brief period of silence followed

by a short momentary burst of noise. These two perceptual phases correspond to three
main articulatory phases ! the closure and compression phase (silence), and the
release and post-release phase (noise). The term #plosive" is, in auditory terms,
indicative of an explosion. In many cases, however, no burst of noise may be present,
e.g. the plosive / t / in the phrase break it may not be released at all, and no noise
component need necessarily be present. Nevertheless, this sound is identifiable as / t /
because the offset stage of the preceding vowel carries the information about the
following segment.
(b) Fricative consonants
The term #fricative" is an auditory term. It indicates that some friction is

present. According to the auditory impression (the colouring of fricative noise), we
can distinguish between two main types of fricatives: sibilant fricatives and non-
sibilant fricatives.
Sibilant fricatives are perceived as sharp fricative noise of high frequency.
They can be further subdivided into hissing sibilants / s, z / and hushing sibilants
/ , / (O"Connor, 1973).
Non-sibilant fricatives are all other fricative consonants, such as [ f, v, , , h,
], etc.
(c) Affricate consonants
An affricate is a compact realization of a plosive and a fricative consonant. In

auditory terms it means that there are three main stages: a brief period of silence, a
momentary burst of noise, and a short period of friction, all realized in a tight
succession. As in the case of fricatives, affricates in English and Slovak can be
divided into hissing affricates, e.g. Slovak / ts, dz /, and hushing affricates, e.g. / t,
d /.
Note
Approximants and trills with a friction component present can also be placed into this
category.
68
2. Sonorants ! Tone consonants
Sonorants are all consonants without any perceptual indication of the presence
of noise. This group of consonants can be divided into nasal sonorants (sonorants
with nasal resonance), and oral sonorants (sonorants with oral resonance).
(a) Nasal Sonorants
This perceptual group of consonants contains all consonants produced with the
participation of the nasal cavity. Nasals are produced without any plosion, and they
could be characterized auditorily as a kind of humming sound ! hence, they could be
called #humming consonants", but this is not generally done (O"Connor, 1973).
(b) Oral Sonorants
Sonorants with oral resonance contain no noise component, and the air is
released through the oral cavity, which gives them a different colour compared to the
nasal sonorants. This group comprises all approximants, trills, flaps, and taps without
a noise component.3
1.7.2.4 Length
Length is a perceptual category usually characterizing speech sounds by such
auditory terms as long, half-long, and short. In general, under comparable
conditions, long sounds are about twice as long as short sounds. The perception of
length depends primarily on the acoustic duration of a sound. The minimal difference
of duration that a human listener is capable of noticing ! the difference limen !
appears to be between 10 and 40 ms for sounds of a duration between 30 and 300 ms
(Lehiste, 1970). "Certain investigations made recently in the field of phonetic
quantity have, however, demonstrated that what we perceive subjectively as a
difference of quantity or of length is in reality often something else. A long subjective
duration is often accompanied by falling intonation which, at least in certain cases, is
the only difference that can be noted objectively between it and the "short", which is
characterized in its turn by a rising or level intonation (Malmberg 1963: 76). A
similar type of influence has been noticed not only for pitch, but also for intensity and
quality. This indicates that length as a subjective attribute depends not only on the
real physical duration of a sound wave, but also on such factors as pitch, intensity,
and quality (wave structure). In addition to this, such phenomena as tempo of speech
and the number of syllables in a word also tend to influence the length of sounds.
3
In some languages, approximants trills, flaps, and taps under certain conditions (or in general) may
contain some degree of friction, which would place them, perceptually, into the category of
obstruents.
69
Generally speaking, the faster the tempo, the shorter the sound, and, by the same
token, the greater the number of syllables in a word, the shorter the sounds of that
word.

To conclude, any auditory stimulus analysis carried out by the ear and the
human brain must take into account these factors:
1. The state of the listener"s hearing mechanism.

The organs of hearing must be able to receive the stimulus. If the organs are
not functioning properly, they may not be able to receive the stimulus, e.g.
someone may be hard of hearing or deaf.
2. The acoustic properties and clarity of the stimulus.

An important part in sound reception is played by these four factors:
fundamental frequency, intensity, wave structure (quality), and duration
(quantity) of the stimulus. These condition the auditory perception of the
sound, i.e. its pitch, loudness, qualitative impression, and length. The fact
whether the sound is clear or garbled in some way is also of significance, as
are the distance of the sound source from the listener and the acoustics of the
environment (small room, large room, outside) through which the sound
stimulus travels.
3. The listener"s awareness of the phonological system of the language

analyzed.
In order to be able to distinguish the (right) meaning of words, the listener
must be aware of the number and type of the phonemes of the language
analyzed.
4. The extent of the listener"s knowledge of the language analyzed and its
context.
The listener does not analyze the stimulus only on the basis of what he hears,
but also on the basis of what he knows about the lexis, grammar, culture, etc.
of the language analyzed, and also on the basis of the context of the situation.
5. The listener"s mental qualities.

The listener"s ability to concentrate, his intelligence, his memory, etc., also
play a role in the stimulus analysis.
70
The Classification and Description of English Speech Sounds
2 The Classification and Description of

English Speech Sounds
In this chapter we will attempt to describe and classify the speech sounds used
in the pronunciation of people living in England, more specifically, the accent found
in the south of England which is often referred to as Received Pronunciation (RP).
For the purposes of comparison, selected speech sounds used in other English accents
will also be occasionally mentioned. For a more complete picture of English accents
see chapter 5.
2.1 Vowels
In English language (RP General accent), there are twelve distinctive
(relatively) pure vowels:
/ /, / i /, / e /, / /, / Y /, / /, / Â /, / /, / h /, / u /, / ± /, / /
2.1.1 Definition and Classification of Vowels

Before we move on to the definition of the vowel, we must explain another
term related to this phenomenon, the term monophthong. Monophthong is a "... term
used in phonetic classification of vowel sounds on the basis of their manner of
articulation: it refers to a vowel (a pure vowel) where there is no detectable change
in quality during a syllable ..." (Crystal 1991: 222). The term vowel therefore covers
also diphthongs, because diphthongs comply with the acoustic definition of the
vowel. What is different then between monophthongs and diphthongs is the fact that
in the production of diphthongs there is a noticeable change in the quality, that is,
there are two recognizable vowel sounds in a diphthong. Since the following sections
are concerned with the sounds where there is no dramatic change in quality during
their pronunciation, the more proper name for the chapter should be !Monophthongs",
instead of !Vowels". However, in many traditional works on phonetics and phonology
the term vowels is often used for such types of sounds, and the two terms are not
differentiated in any particular way. In other books, on the contrary, these terms are
clearly differentiated. To avoid possible confusion about this, we will stick to the
traditional terminology, but we should keep in mind that monophthongs are pure
vowels, whereas diphthongs are gliding vowels.
71
Here are some definitions of vowels:
(A) Phonetic definitions
A vowel is "A speech sound produced by the relatively free passage of breath
through the larynx and oral cavity, usually forming the most prominent and central
sound of a syllable" (Webster"s II New Riverside University Dictionary 1984: 1295#
1296).
"... vowels are sounds in which there is no obstruction to the flow of air as it passes
from the larynx to the lips" (Roach 1991:10).
"A vowel is defined as a voiced sound in forming which the air issues in a continuous
stream through the pharynx and mouth, there being no obstruction and no narrowing
such as would cause audible friction" (Jones 1960: 23).
"This category of sounds (vowels) is normally made with a voiced egressive

airstream, without any closure or narrowing such as would result in the noise
component characteristic of many consonantal sounds; moreover, the escape of the
air is characteristically accomplished in an unimpeded way over the middle line of
the tongue" (Gimson & Cruttenden 1994: 34).
"They (vowels) are voiced and are oral resonant sounds. The different sound quality
of vowels depends upon variations in the shape of the oral cavity" (Calvert 1986:
105).
(B) Phonological definitions
"... vowels are those units which function at the centre of syllables" (Crystal 1991:
377).
"A vowel can be regarded as the nucleus of a phonological syllable" (Laver 1994:
34).
"A vowel is the nucleus or the central part of the syllable ..." (Abercrombie 1967: 39).

During the articulation of vowels, several organs of articulation are involved. It is

mainly the vocal folds, resonating cavities, soft palate, lips and tongue. A description
of vowels can therefore be done according to:
72
(a) the length of the vocal fold vibration.

Depending on how long the vocal folds vibrate during the production of
vowels, we distinguish between (relatively) short and (relatively) long
vowels. The absolute length of English vowels depends on the nature of the
sounds that follow them (for more, see 2.1.3).
(b) the position of the soft palate.

When the soft palate is raised (closing the entrance to the nasal cavity), the
vowels produced are purely oral. When the soft palate is lowered, the vowels
are nasalized.
(c) the shape of the lips.

The shape of the lips or lip-rounding is another important variable of vowel
quality. We distinguish three stages of lip-rounding. Lips can be rounded, as
for / u /, spread, as for / i /, or neutral, as for / ± /.
(d) the part of the tongue which is raised.

(the horizontal position)
The tongue, or, more specifically, its raised part, can assume many different
positions in the oral cavity. This mobility of the tongue is usually done in
horizontal terms, that is, it can move from back to front and vice versa.
According to this horizontal movement, we distinguish three distinct tongue
positions # front, central and back. Vowels made in these positions can be
respectively described as front, central and back.
(e) the position of the mandible and the height of the tongue.
(the vertical position)
The mandible, or lower jaw, can assume different degrees of openness. These
degrees play an important role in the production of vowels. Based on the
degree of openness, we distinguish four different vowel types: close e.g. / i /,
close-mid (or half-close) e.g. / /, open-mid (or half-open) e.g. / e /, and
open e.g. / /.
(f) the muscular tension of the tongue.

According to the muscular tension of the tongue, English vowels can be
divided into tense and lax. "Tense vowels are those which are supposed to
require considerable muscular tension on the part of the tongue; lax vowels
are those in which the tongue is supposed to be held loosely" (Jones 1960:
39). The term tense and lax usually applies only to the case of close vowels.
73
2.1.2 Cardinal Vowels
It is clear that the description of vowels is a very complicated matter. There are
many different vowel sounds in the languages of the world. Some of these sounds can
be grouped, according to what they have in common, into sets of, let us say, o-like, or
i-like sounds. It is quite obvious that these vowels, although phonologically related,
may be quite different from the phonetic viewpoint (they can have different degrees
of openness, rounding, nasalization, etc.). Therefore, a more objective and
independent system of description of vowels is needed. The most satisfactory, in this
respect, seems to be the cardinal vowel system devised by Daniel Jones.1 It allows
us to define accurately any vowel of any language. The vowels which form this
system are called cardinal vowels. "These cardinal vowels are a standard reference
system, and people being trained in phonetics have to learn to make them accurately
and recognize them correctly. If you learn the cardinal vowels, you are not learning
to make English sounds, but you are learning about the range of vowels that the
human vocal apparatus can make, and also learning a useful way of describing,
classifying and comparing vowels" (Roach 1991: 13). Cardinal vowels are enclosed
in square brackets [ ], because they indicate the real physical articulations # sounds.
Cardinal vowels are located on a specially shaped vowel diagram
(quadrilateral), and they can be divided into primary cardinal vowels and
secondary cardinal vowels. Primary cardinal vowels are very similar to the vowels
in many European languages. They are shown in Fig. 18. Cardinal vowel no.1 has the
symbol [ i ], and it is produced by the front of the tongue raised as close as possible to
the palate without any friction noise. It is the most front and close vowel that it is
possible to make. Cardinal vowel no. 5 has the symbol [ ¥ ], and it is produced by the
whole of the tongue as low as possible in the mouth, with a very slight raising at the
extreme back. It is the most open and back vowel that it is possible to make. Cardinal
vowel no. 8 has the symbol [ u ], and it is articulated by the back of the tongue raised
as close as possible to the velum. It is a fully close and back vowel. Cardinal vowel
no. 4 has the symbol [ a ], and it is a fully open and front vowel. These four cardinal
vowels represent extreme points in the pronunciation of vowels. After establishing
these extreme points, it is possible to put in intermediate points # vowels no. 2, 3, 6,
and 7. "Many students when they hear these vowels find that they sound strange and
exaggerated; you must remember that they are extremes of vowel quality" (Roach
1991: 13).
1
/ d njl d nz /
74
zX
Front Central Back
Close L z \
Close-mid H z 2 )zR
Open-mid (z ¡ ¥ zo
Open D z $ zc
Fig. 18. Cardinal vowels.
The secondary cardinal vowels can be made by alternating the primary cardinal
vowels in terms of lip rounding. Such secondary series is denoted by the following
numbers and symbols: no. 9 [ y ], no. 10 [ u ], no. 11 [ s ], no. 12 [ t ], no. 13
[ Â ], no. 14 [ Y ], no. 15 [ v ], no. 16 [ Ñ ] (see Fig. 18).
The set of cardinal vowels combined with a set of diacritics available in the
IPA alphabet (the diacritics can express different qualities of vowels such as voicing,
aspiration, centralization, nasalization, palatalization etc.) can describe objectively,
and with a high degree of precision, any vowel of any language.
2.1.3 The Length of English Vowels

In general terms, we can talk about two basic vowel lengths and, consequently,
about two types of vowels # long vowels and short vowels. However, these terms are
very relative because the real length of English vowels depends on the type of the
sounds following them. Nevertheless, from the practical point of view, it is
convenient to divide English vowels into (relatively) short and (relatively) long
vowels, because we can demonstrate certain phonetic and phonological relationships
between them, as in the following examples:
sit # seat / ! i / don # dawn / Â # /

cut # cart / Y ! / for (weak form) # fur / ± # /
full # fool / h ! u / bet # bat / e # /
75
Here we have tried to group the English vowels according to the opposition of
length. In all cases the opposition is based on both quality and quantity. That is why,
in addition to length marks, we also use a different symbol for each member of the
pair. Special attention should be paid to the sound / / which, compared to the
length of the rest of English vowels, is sometimes classified as neutral, that is,
neither short nor long.
As we have already mentioned, the real length of English vowels depends on
the type of sounds following them. It has been measured that the quantity (length) of
vowels increases when they stand in a word-final position or when they are followed
by a voiced consonant. When the vowels are followed by a voiceless consonant, their
quantity decreases. This rule applies to both short and long English vowels. Thus we
can speak of four lengths of English vowels, such as we find in words sot, sod, sort
and sword.
As regards the above definition of some sounds influencing the real length of
English vowels, a few facts must be clarified. It is not absolutely fair to say that the
length of English vowels increases when these are followed by a voiced consonant or
when they stand in a word-final position. It is more correct to say that these vowels,
when followed by a voiced consonant, are not influenced, that is, they retain their
original length # they are unclipped (Wells, 1990). When they stand in the word-
final position followed by a silence, it is quite obvious that their length is not
influenced, and this (uninfluenced) length should be taken as a basis for what we may
call !original length". On the other hand, this original length is clipped (shortened) if
the vowels are followed by a voiceless consonant. To sum it up we can say that:
(a) voiceless consonants decrease the length of short vowels # sot [ sÂt ]
(b) voiced consonants do not influence the length of short vowels # sod [ sÂ;d ]
(c) voiceless consonants decrease the length of long vowels # sort [ s;t ]
(d) voiced consonants do not influence the length of long vowels # sword [ sd ]
As far as the long vowels are concerned, it is easy to determine their original,
uninfluenced length, because they can be found in the word-final positions. Short
vowels, however, are never found in the word-final position in stressed syllables.
There is no such word in English, that would end with / / or / Â /, or any other
short vowel ( / ± / and / / occur finally only in unstressed syllables ). Therefore when
we say that voiced consonants do not influence the length of short vowels we take for
granted, that if there was a short vowel in stressed final position in words, it would
have to have the same length as a short vowel followed by a voiced consonant.
The total duration of the short vowel plus a voiced consonant equals the total
duration of the short vowel plus a voiceless consonant. The same applies to long
vowels (see Fig. 19).
76
; d t
Fig. 19. Spectrograms of the English words !plod" and !plot". The overall length of each of
the words (plod # plot) is approximately the same.
The influence of voiced and voiceless consonants on the duration of preceding

vowels can be found in all languages, and it has come to be considered a universal
feature. That is to say, vowels in all languages are phonetically shorter before
voiceless consonants and longer before the voiced ones. This phenomenon is also
observable in Slovak. However, the difference between Slovak vowels followed by
voiced and voiceless consonants is much smaller than in English, and it is practically
inaudible to the human ear. The difference between the duration of English vowels in
such environment, on the other hand, can be clearly heard, and therefore it should be
practised by Slovak learners studying English pronunciation.
The duration of vowels can also be affected by other factors. Their length ''may
be determined by the nature of the segment itself, that is, by its point and manner of
articulation'' (Lehiste 1970: 18). The duration of the vowel determined by its
phonetic quality, without any influence of the context (preceding and following
sounds), is termed intrinsic duration (intrinsic length). The intrinsic duration of
vowels appears to be correlated with tongue height: ''other factors being equal, a high
vowel is shorter than a low vowel'' (Lehiste 1970: 18). This means, for example, that
in comparable sound environment, vowels / i / or / u / will always be shorter than
/ /.
77
2.1.4 The Description of English Distinctive Vowels
i u

e

Fig. 20. English distinctive vowels.
Vowel / /
Example words: in, pin, happy

Description: "The short RP vowel / / is pronounced with a part of the tongue
nearer to centre than to front, raised just above the close-mid position; the lips are
loosely spread; the tongue is lax, with the side rims making a light contact with the
upper molars" (Gimson & Cruttenden 1994: 99). It can be found in all positions in
words.
Main features: short, spread, front-central, close-mid, lax.
Variants: In the word-final unaccented position, e.g. city, ready, lazy, etc., / / is
often replaced by [ i ] (short version of / i /) by many speakers, e.g. [ ¥s ti ], [ ¥redi ],
[ \le zi ]. In penultimate and ultimate unaccented syllables, but not word finally, e.g.
quality, peril, positive, etc., / / tends to be replaced by / ± / # / \kwÂl±t /, / \per±l /,
/ \pÂz±t v /. Suffixes -less and -ness are pronounced / s / or / ±s / (/ ±s / being
preferred by the younger generation).
Vowel / i /
Example words: eve, beat, fee

Description: !The front of the tongue is raised to a height slightly below and
behind the front close position; the lips are spread; the tongue is tense, with the side
78
rims making a firm contact with the upper molars# (Gimson & Cruttenden 1994: 97).
It can be found in all positions in words.
Main features: long, spread, front, close, tense.
Variants: This type of vowel tends to be frequently diphthongized, especially in
the word-final position. A slight glide from a position near to [ ] is common
amongst RP speakers, and it is used more frequently than a pure vowel. Words like
tea, me, or lead could thus be transcribed as [ t i ], [ m i ], [ l id ]. However, any glide
starting in the central area, e.g. tea / t±i /, is dialectal. On the other hand, the use of a
pure vowel in the word-final position may be typical of an over-cultivated
pronunciation.
Vowel / e /
Example words: any, let, breath

Description: "For the short RP / e /, the front of the tongue is raised between the
close-mid and open-mid positions; the lips are loosely spread and are slightly wider
apart than for / /; the tongue may have more tension than in the case of / /, the
side rims making a light contact with the upper molars" (Gimson & Cruttenden 1994:
101). It can never stand in the word-final position.
Main features: short, spread, front, mid.
Variants: In General RP, / e / is pronounced closer to the cardinal vowel [ w ],
whereas the cardinal vowel [ e ] is typical of Conservative RP. A variety of / e / can
be found in Australian English and in Cockney, where it may glide towards [ ].
Vowel / /
Example words: act, back, rash

Description: "The mouth is slightly more open than for / e /; the front of the tongue
is raised to a position midway between open and open-mid, with the side rims making
a very slight contact with the back upper molars; the lips are neutrally open. This
traditionally short vowel is now generally longer in RP than the other short vowels
/ , e, Y, Â, h /" (Gimson & Cruttenden 1994: 103). It never occurs in the word-final
position.
Main features: short, neutral, front, open.
Variants: In Conservative RP (older speakers) we can find a closer variant of / /,
which is close to the cardinal vowel [ w ], and it is usually slightly diphthongized
to [ w± ]. In Advanced RP (younger speakers) we can find a more open variant similar
to the cardinal vowel [ a ].
79
Vowel / Y /
Example words: other, duck, enough

Description: "The short RP / Y / is articulated with a considerable separation of
the jaws and with the lips neutrally open; the centre of the tongue is raised just above
the fully open position, no contact being made between the tongue and the upper
molars" (Gimson & Cruttenden 1994: 104). It never occurs in the word-final position.
Main features: short, neutral, central, open-mid.
Variants: Conservative RP has a variant of / Y / which is more of a back vowel.
Cockney has a variant of / Y / which is more front and more open, approaching the
cardinal vowel [ a ]. In the north of England there is no contrast between / Y / and
/ h /, so look and luck sound alike # / lhk /.
Vowel / /
Example words: aunt, bath, car

Description: "This normally long vowel is articulated with a considerable
separation of the jaws and the lips neutrally open; a part of the tongue between the
centre and back is in the fully open position, no contact being made between the rims
of the tongue and the upper molars" (Gimson & Cruttenden 1994: 106). It occurs in
all positions in words.
Main features: long, neutral, central-back, open.
Variants: The type of / / used in Conservative RP is close to the cardinal vowel
[ ¥ ]. In Australian English and in some cities in the north of England, a variety close
to the cardinal vowel [ a ] is used. In General American and in the north of
England / / is replaced by / /, e.g. laugh, pass, etc.
Vowel / Â /
Example words: odd, fog, because

Description: "This short vowel is articulated with wide open jaws and slight, open
lip-rounding; the back of the tongue is in the fully open position, no contact being
made between the tongue and the upper molars" (Gimson & Cruttenden 1994: 108).
It never occurs in the word-final position.
Main features: short, rounded, back, open.
Variants: In Conservative RP and in Cockney, a small number of words, e.g. off,
cloth, across, etc., prefer / / instead of / Â /. In General American, some words,
e.g. bomb, hot, not, etc., have / /, and some words, e.g. cross, dog, soft, etc., have
/ /. In Scottish English almost all words with RP / Â / use long / /.
80
Vowel / /
Example words: all, call, law

Description: "This relatively long RP vowel is articulated with medium lip-
rounding; the back of the tongue is raised between the open-mid and close-mid
positions, no contact being made between the tongue and the upper molars" (Gimson
& Cruttenden 1994: 110). It occurs in all positions in words.
Main features: long, rounded, back, close-mid.
Variants: The contrast between / / and / ± /, which was used in the past in the
words like saw # sore, is not made nowadays. A number of words which had / h± / in
the past, now have an alternative pronunciation with / /, e.g. sure, tour, poor.
Vowel / h /
Example words: oops, look, butcher

Description: "The short RP vowel / h / is pronounced with a part of the tongue
nearer to centre than to back raised just above the close-mid position; it has,
therefore, a symmetrical back relationship with the front vowel / /; the tongue is
laxly held ..., no firm contact being made between the tongue and the upper molars.
The lips are closely but loosely rounded" (Gimson & Cruttenden 1994: 112). It rarely
occurs in the word-initial position, and never in the word-final position.
Main features: short, rounded, central-back, close-mid, lax.
Variants: In the north of England, / h / is used instead of / Y /. In Scottish English,
both / h / and / u / are centralized and sound like centralized cardinal vowel [ ü ].
Vowel / u /
Example words: ooze, fool, through

Description: "RP long / u / is a close back vowel, but the tongue-raising is relaxed
from the closest position and is somewhat centralized from true back; its relationship
with / h / is similar to that between / i / and / /, the articulation of / u / being tense
compared with that of / h /, though no firm contact is made between the tongue and
the upper molars. The lips tend to be closely rounded" (Gimson & Cruttenden 1994:
114). It occurs in all positions in words.
Main features: long, rounded, back, close, tense.
Variants: Within General RP, two variants are common; a centralized long
vowel [ J ], and a gliding vowel [ hu ]. In Conservative RP, a fully back and round
long cardinal vowel [ u ] can be found. Cockney uses a diphthong [ ±ü ].
81
Vowel / ± /
Description: "/ ± / has a very high frequency of occurrence in unaccented

Example words: above, woman, colour
syllables. Its quality is that of a central vowel with neutral lip position, having in non-
the velar consonants / k, g / and / Ì /, however, the tongue may be slightly more
final positions a tongue-raising between open-mid and close-mid; in the vicinity of
raised and retracted. But in final positions, the vowel may be articulated in the open-
mid central position" (Gimson & Cruttenden 1994: 117). This type of vowel is
usually called schwa or mixed (murmured/indefinite) vowel. It occurs in all positions
in words. However, it never occurs in stressed syllables.
Main features: short, neutral, central, mid.
Variants: In Conservative RP, word-final / ± / may be realized as [ \ ]. In General
American, this type of vowel is realized with a strong r-colouring, which is
symbolized by [ H ].
Vowel / /
Example words: earn, serve, fur

Description: "RP / / is articulated with the centre of the tongue raised between
close-mid and open-mid, no firm contact being made between the tongue and upper
molars; the lips are neutrally spread" (Gimson & Cruttenden 1994: 116). The quality
of / / is similar to that of / ± /, the main difference is in the length. / / occurs in
Main features: long, neutral, central, open-mid.
Variants: In Conservative RP, a variant close to [ ] can be found. In General
American, / / carries strong r-colouring, often symbolized by [ ].
82
2.2 Diphthongs
In English there are eight distinctive diphthongs:
/ e /, / a /, / /, / ±h /, / ah /, / ± /, / e± /, / h± /
2.2.1 Definition and Classification of Diphthongs

Here are some definitions of diphthongs:
(A) Phonetic definitions:
"A diphthong is defined as an independent vowel-glide not containing within itself

either a $peak% or a $trough% of prominence. By a vowel-glide we mean that the
speech-organs start in the position of one vowel and move in the direction of another
vowel. By $independent% we mean that the glide is expressly made, and is not merely
an unavoidable concomitant of sounds preceding and following" (Jones 1960: 58).
"In some vocalic sounds, the glide component is so prominent that the vowel no
longer has a single identifying vowel target value, even though it is still heard as a
single sound. Such sounds are DIPHTHONGS" (Clark & Yallop 1995: 34).
Diphthongs are "... sounds which consist of a movement or glide from one vowel to
another" (Roach 1991: 20).
"The sequences of vocalic elements included under the term $diphthong% are those
which form a glide within one syllable" (Gimson & Cruttenden 1994: 119).
"Phonetically,..., the diphthong is defined as a combination of two vowels in one and

the same syllable ..." (Vachek 1973: 19).
"Diphthongs are considered single phonemes, each having a sequence of two

different vowel positions. One of the positions is the dominant nucleus (sometimes
called the !radical#) with greater duration; the other position, the glide (or
!vanish#), is of reduced duration and stress. Both positions are taken in a single
syllable" (Calvert 1986: 128).
A diphthong is "A term used in the phonetic classification of vowel sounds on the
basis of their manner of articulation: it refers to a vowel where there is a single
83
(perceptual) noticeable change in quality during a syllable, as in English beer, time,

loud" (Crystal 1991: 105).
(B) Phonological definition:
A diphthong is a distinctive sound unit which forms the centre of syllables.

The diphthongs / e /, / a /, / /, / ±h /, / ah /, / ± /, / e± / and / h± / fully

comply with the above definitions. There is, however, one more vowel glide which,
at first sight, meets the criteria for a diphthong. This vowel glide is the
combination / u /. It can be found in such words as ruin, fluid, doing, or suet. In
these cases, however, the glide is usually divided into two syllables. Therefore the
sequence / u / is not classified as a diphthong.
According to the type of sounds they consist of, diphthongs can be divided into
closing diphthongs and centring diphthongs. Closing diphthongs can be further
divided into closing diphthongs ending in / / ( / e /, / a /, / / ), and closing
diphthongs ending in / h / ( / ±h /, / ah / ). Centring diphthongs are diphthongs
ending in / ± / ( / ± /, / e± /, / h± / ).
DIPHTHONGS
CLOSING CENTRING
Ending in / / Ending in / h / Ending in / ± /
/e / /a / / / / ±h / / ah / / ± / / e± / / h± /
Fig. 21. The division of English diphthongs.
The closing diphthongs are called closing because of the fact, that during their
pronunciation the lower jaw makes a closing movement. Therefore also the vowel
these diphthongs end with is a close vowel. In other words, "... a glide from a
relatively more open towards a relatively more close vowel is produced" (Roach
1991: 21).
The centring diphthongs are called centring because they glide towards the
vowel / ± /, which is a central vowel.
84
2.2.2 The Length and Prominence of English

Diphthongs
English distinctive diphthongs are equivalent in length to the English long
(relatively pure) vowels. They consist of two vowel sounds, one of which is usually
longer and more prominent. "The greater prominence may be due either to greater
inherent sonority or to stronger stress or to a combination of the two" (Jones 1960:
58). When the first part of the diphthong is longer and more prominent, the
diphthong is said to be falling (descending). When the second part of the diphthong is
longer and more prominent, the diphthong is said to be rising (ascending). Most of
the English distinctive diphthongs are falling, but there are some exceptional cases
(/ ± /, / h± /) in which diphthongs can be, depending on the context, either rising or
falling.
In addition to the already unequal length of the individual elements of the
English diphthongs, the length may be further influenced by the type of the following
sound. Similarly as vowels, English diphthongs are clipped (shortened) when
followed by a voiceless consonant and unclipped when followed by a voiced
consonant or standing in the word-final position. It is usually the first element of the
diphthong that is influenced by the following sound. When a diphthong is reduced,
both elements have approximately the same length. When a diphthong is non-
reduced, usually the first element sounds much longer than the second element, e.g.
ice [ a s ] # eyes [ a; z ], lout [ laht ] ! loud [ la;hd ], etc.
2.2.3 The Description of English Distinctive

Diphthongs
English Closing Diphthongs

e

a a
Fig. 22. English closing diphthongs.
85
Diphthong / e /
Example words: eight, make, day

Description: "The glide begins from slightly below the close-mid front position and
moves in the direction of RP / /, there being a slight closing movement of the lower
jaw; the lips are spread" (Gimson & Cruttenden 1994: 120). It occurs in all positions
in words.
Variants: Older speakers may have a starting-point nearer to the cardinal vowel
[ e ]. In Cockney, [ ] variant can be found. In Scottish English, General American,
and Northern English, / e / can be replaced by monophthongal [ e ].
Diphthong / a /
Example words: ice, mine, fly

Description: The glide of RP / a / begins at a point slightly behind the front open
position, and moves in the direction of the position associated with RP / /; the glide
is much more extensive than that of / e /, the closing movement of the lower jaw
being obvious. The starting point may be similar to the articulation used in RP / Y /.
The lips change from neutral to a loosely spread position" (Gimson & Cruttenden
1994: 122). It occurs in all positions in words.
Variants: In Conservative RP, a very back starting-point is common # [ ¥ ], which
may sometimes change into a monophthong close to / /. In Cockney and Australian
English, / a / has a back starting-point and it sounds close to [ Â ]. In Southern
American accent, / a / may be pronounced as an open monophthong [ a ].
Diphthong / /
Description: "For RP / / the tongue glide begins at a point between the open-
Example words: oil, noise, ploy
mid and open back positions and moves in the direction of / /. The tongue movement
extends from back to centralized front, but the range of closing in the glide is not as
great as for / a /; the jaw movement, though considerable, may not, therefore, be as
marked as in the case of / a /. The lips are open rounded for the first element,
changing to neutral for the second" (Gimson & Cruttenden 1994: 123). It occurs in
Variants: Conservative RP may centralize the starting-point of / / into [ ¥ ].
Diphthong / ±h /
Description: "The glide of RP / ±h / begins at a central position, between close-

Example words: old, toast, flow
mid and open-mid, and moves in the direction of RP / h /, there being a slight closing
86
movement of the lower jaw; the lips are neutral for the first element, but have a
tendency to round on the second element. The starting-point may have a tongue
position similar to that described for / / " (Gimson & Cruttenden 1994: 125). It
occurs in all positions in words.
Variants: In Conservative RP, / ±h / is often realized as [ 4h ]. In Cockney and
Australian English we may find the variant [ h ]. Scottish English and General
American have a variant close to [oh ].
Diphthong / ah /
Example words: out, sound, now

Description: "The glide of RP / ah / begins at a point between the back and front
open positions, slightly more fronted than the position for RP / /, and moves in the
direction of RP / h /, though the tongue may not be raised higher than the close-mid
level. The glide is much more extensive than that used for / ±h / and is symmetrically
opposed to the front glide of / a /. The lips change from a neutrally open to a weakly
rounded position" (Gimson & Cruttenden 1994: 127). It occurs in all positions in
words.
Variants: In Cockney and Australian English, a [ ] or [ ;Þ ] variety can be
heard. In Conservative RP, a diphthong close to [ h ] is frequent.
English Centring Diphthongs

e
Fig. 23. English centring diphthongs.
87
Diphthong / ± /
Description: "The glide of RP / ± / begins with a tongue position approximately

Example words: eerie, fierce, dear
the same as that used for / /, i.e. close-mid and centralized from front, and moves in
the direction of the more open variety of / ± / when / ± / is final in the word; in non-
the / ± / element being of a mid type. The lips are neutral throughout, with a slight
final positions, e.g. in beard, fierce, the glide may not be so extensive, the quality of
movement from spread to open" (Gimson & Cruttenden 1994: 131). As we have
already mentioned in a previous chapter, this diphthong is not always falling, that is,
it does not always have prominence and length on the first element. To be more
specific, in unstressed syllables, the / / element is weaker, and it equals / j /.
Therefore in words like period / \p ±r ±d /, easier / \¬z ± /, hideous / \h d ±s /, idiom
/ \ d ±m / etc., the / ± / diphthong is close to [ j± ] or [ ,± ], and it constitutes a rising
diphthong. The diphthong / ± / occurs in all positions in words.
Variants: In Conservative RP, / ± / can sound like / j / or / j /, in the word-
final position e.g. here / hj / = [ I ], dear / dj /. In both these cases it forms a
rising diphthong. The form with / / is usually characterized as an affectation. In
General American there is no / ± / diphthong. The sounds / / or / i / + / r / are used
instead (in the words which have an !r" in the spelling). In Australian English this
diphthong can be monophthongized to [ Z ].
Diphthong / e± /
Description: "The glide of RP / e± / begins in the open-mid front position, and

Example words: airway, carefree, dare
moves in the direction of a more open variety of / ± /, especially when the diphthong
is final; where / e± / occurs in a syllable closed by a consonant, the / ± / element
tends to be of a mid / ± / type. The lips are neutrally open throughout" (Gimson &
Cruttenden 1994: 133). It occurs in all positions in words.
Variants: In General RP, a long monophthong [ w ] is completely acceptable.
Conservative RP has a diphthong close to [ 5 ]. In General American there is
no / e± / diphthong. The sounds / e / + / r / are used instead (in the words which have
an !r" in the spelling).
Diphthong / h± /
Example words: Urdu, cruel, sure

Description: "RP / h± / glides from a tongue position similar to that used for / h /
towards the more open type of / ± / which forms the end-point of all three centring
diphthongs with, again, a somewhat closer variety of / ± / when the diphthong occurs
in a closed syllable. The lips are weakly rounded at the beginning of the glide,
88
becoming neutrally spread as the glide progresses" (Gimson & Cruttenden 1994:
134). Similarly as / ± /, this diphthong can also be a rising diphthong, when
occurring in unstressed positions, as we can see in words like influence / \ nflh±ns /,
valuable / \v ljh±b /, jaguar / \dµ gjh± / etc., where the prominence and length are
on the second element, and phonetically we could transcribe this diphthong as [ h± ].
Diphthong / h± / only rarely occurs in the word-initial position, e.g. Ur / h± /, Urdu
/ \h±du /, etc.
Variants: Younger speakers tend to replace / h± / with / ± / or the monophthong
/ /. Thus words sure and shore may sound alike # / /. This monophthongization
usually occurs in the word-final position, although it can also be found in the word-
medial position # Europe / \jr±p /, curious / \kjr ±s / etc. In Australian English,
/ h± / tends to be monophthongized to [ h ] or / /. In General American, / h± / is
replaced by / u / or / h / + / r /.
2.3 Triphthongs
There are five triphthongs in English:
/ e ± /, / a ± /, / ± /, / ±h± /, / ah± /
2.3.1 Definition and Classification of Triphthongs

There are words in English, in which we can find a sequence of three vowels.
These three vowels in succession can create five combinations, which regularly
appear in English words. The regularity of the combinations rests in the fact that all
of these five vowel sequences are formed by adding the / ± / vowel to a closing
diphthong. A closing diphthong is therefore a part of every such three-vowel
sequence. That is why there have been different approaches to classification of this
phenomenon. Some call these vowel sequences triphthongs, others see them as a
mere combination of a closing diphthong plus schwa. Triphthongs are not considered
phonemes in the traditional sense.
Here are some definitions:
"A triphthong is a glide from one vowel to another and then to a third, all produced
rapidly and without interruption" (Roach 1991: 23).
89
A triphthong is "A compound vowel sound resulting from the succession of three
simple sounds and functioning as a unit" (Webster"s II New Riverside University
Dictionary 1984: 1235).
A triphthong is "A term used in the phonetic classification of vowel sounds on the
basis of their manner of articulation: it refers to a type of vowel where there are two
English fire and tower / fa,± / and / t¥h± /" (Crystal 1991: 362).
noticeable changes in quality during a syllable, as in common pronunciation of
(B) Phonological definition:
A triphthong is a unit of sound functioning as the centre of syllable.

In the present-day English, there are tendencies towards weakening of the

central elements (/ / and / h /) of triphthongs. Because of this, the central elements of
triphthongs are often not realized at all. We will cover this matter, with particular
cases, in the following section.
2.3.2 The Description of the English Triphthongs

Triphthong / e ± /
Example words: player, layering, greyer

Description: This triphthong is formed by a combination of the RP closing
diphthong / e / plus / ± /. It occurs in the word-medial and word-final positions.
Variants: In General RP, the central element of this triphthong becomes very
weak, therefore it sounds very close to / e± /. Thus word pairs like layer # lair or
payer # pair become neutralized: / le± /, / pe± /. In General American, there are no
triphthongs. These three-vowel sequences are replaced by a combination of a closing
diphthong plus / r /.
Triphthong / a ± /
Example words: iron, society, fire

diphthong / a / plus / ± /. It occurs in all positions in words.
Variants: In General RP, this diphthong may approach [ a± ], or even / /. Thus,
word pairs like fire # far or byre # bar may be neutralized: / f /, / b /.
90
Triphthong / ± /
Example words: oy, joyous, employer

diphthong / / plus / ± /. It rarely occurs in the word-initial position.
Variants: In General RP, the central element of the triphthong may be weakened so
that the sound / ± / is produced.
Triphthong / ±h± /
Example words: mower, lowering, slower

diphthong / ±h / plus / ± /. It never occurs in the word-initial position.
Variants: The triphthong produced in General RP may be monophthongized
into / /. Thus word-pairs like mower # myrrh or slower & slur may be neutralized:
/ m /, / sl /.
Triphthong / ah± /
Example words: hourglass, coward, flower

diphthong / ah / plus / ± /. It occurs in all positions in words.
Variants: In General RP, / ah± / may be reduced to [ ± ]. In Conservative RP it
may further be reduced to / /. Thus word-pairs like tower & tar or our & are may be
neutralized: / t /, / /.
2.4 Consonants
There are twenty four distinctive consonants in English:
/ p /, / b /, / t /, / d /, / k /, / g /, / t /, / dµ /, / f /, / v /, / * /, / s /,
/ s /, / z /, / /, / µ /, / h /, / m /, / n /, / Ì /, / l /, / r /, / j /, / w /
91
2.4.1 Definition and Classification of Consonants

A consonant is "a speech sound produced by a partial or complete obstruction of the

air stream by any of various constrictions of the speech organs" (Webster"s II New
Riverside University Dictionary 1984: 302).
Consonants are "... (i) all sounds which are not voiced (e.g. p, s, ³ ), (ii) all sounds in
the production of which the air has an impeded passage through the mouth (e.g. b, l,
rolled r ), (iii) all sounds in the production of which the air does not pass through the
mouth (e.g. m ), (iv) all sounds in which there is audible friction (e.g. f, v, s, z, h )"
(Jones 1960: 23).
"Phonetically, they (consonants) are sounds made by a closure or narrowing in the

vocal tract so that the airflow is either completely blocked, or so restricted that
audible friction is produced" (Crystal 1991: 74).
"Sounds characterized by a constricting or a complete (momentary) closing of the air

passages are called consonants" (Malmberg 1963: 32).
(B) Phonological definitions:
"... consonants are those segments which, in a particular language, occur at the
edges of syllables ..." (Gimson & Cruttenden 1994: 27).
"From a phonological point of view, consonants are those units which function at the
margins of syllables, either singly or in clusters" (Crystal 1991: 74).

Consonants can be described from different points of view. This usually

involves characterization according to:
(a) the source of the energy.

The source of energy for the production of the majority of English sounds comes
from the lungs. Sounds produced by the energy of the airstream from lungs are called
pulmonic. Sounds in the production of which there is no involvement of the lungs are
called non-pulmonic.
92
(b) the direction of the airstream.

When the airstream is forced out of the lungs, the sounds produced are called
egressive. Sounds made by the airstream sucked inwards are known as ingressive.
(c) the vocal fold activity.
When the vocal folds are active (vibrating), voiced sounds are produced. When there
is no vocal fold activity (not vibrating), voiceless sounds are made.
(d) the position of the soft palate.
When the soft palate is raised, the airstream is released through the mouth only.
Sounds produced in this manner are called oral. When the soft palate is lowered
slightly, the airstream is released both through the oral and nasal cavities. Sounds
made in this way are called nasalized. When the soft palate is lowered and the oral
passage is blocked completely, the airstream escapes solely through the nasal cavity.
Sounds produced in this manner are called nasal.
(e) the place of articulation.

Organs of articulation can make different contacts in different places. The points of
these contacts are the following:
1. Bilabial # the upper and lower lips make contact, e.g. / p, b, m /
2. Labiodental # the lower lip and the upper teeth make contact, e.g. / f, v /
3. Dental # the tongue makes contact with the upper teeth / *, s /
4. Alveolar # the tongue makes contact with the alveolar ridge, e.g. / t, d, l, n, s, z /
5. Post-alveolar # the tongue makes contact with the rear part of the alveolar
ridge, e.g. RP / r /
6. Retroflex # the tip of the tongue is curled back and makes contact with the
hard palate, e.g. General American / r /
7. Palato-alveolar # the tongue makes contact with the alveolar ridge and hard
palate, e.g. / , µ, t, dµ /
8. Palatal # the tongue makes contact with the hard palate, e.g. / j /
9. Velar # the back of the tongue makes contact with the soft palate (velum), e.g.
/ k, g, Ì /
10. Uvular # The back of the tongue makes contact with the uvula, e.g. French / r /
11. Glottal # the vocal folds are narrowed and make friction with or without
vibration, e.g. / h /, / /
(f) the manner of articulation.

The closure or narrowing during the articulation of consonants can be of different
types. Each type produces characteristic groups of consonants:
1. Occlusives: complete occlusion
(a) single occlusion # with oral air-flow # [ p, b, t, d, c, , =, ] etc.
# with nasal air-flow # [ m, n, , , ] etc.
(b) repeated (intermittent) occlusion # [ r ] etc.
93
2. Semi-occlusives: complete occlusion + constriction # [ ts, dz, t, d ] etc.

3. Constrictives: narrowing with friction # [ s, z, f, v, , , , ] etc.
4. Approximants: narrowing without friction
(a) with central air-flow [ j, w, , Ú, ] etc.
(b) with lateral air-flow [ l, ] etc.
Note
The speech sounds [ ] and [ r ] in Slovak are often realized without a complete occlusion
(Dvonþová, Jenþa & KráĐ, 1969; Li$ka, 1971b; Dvonþová, 1980; KráĐ & Sabol, 1989), and
they are therefore classified as produced by narrowing without friction, which places them
in the category of approximants. However, in line with other sources (Pike, 1943;
Malmberg, 1963; Abercrombie, 1967; O"Connor, 1973; Laver, 1994), we will classify [ ]
and [ r ] as occlusives.
(g) the articulating organ

Depending on the active articulating organ used, we distinguish these classes of
speech sounds:
1. Labial
(a) Bilabial # [ p, b, m ] etc.
(b) Labiodental # [ f, v, , ]
2. Lingual
(a) Apical # [ t, d, n, l ] etc.
(b) Laminal # [ s, z ] etc.
(c) Pre-dorsal # [ j, c, , , ]
(d) Post-dorsal # [ k, g, , x, ]
(e) Radical # [ , ] ([ ] can be found, for example, in Arabic, or Hebrew)
3. Glottal # [ h, , = ]
(h) the auditory impression

Consonants can be classified according to the presence or absence of a noise
component. Those consonants during the production of which there is an audible
noise component are called obstruents (plosives, fricatives and affricates). Those
consonants during the production of which there is no audible noise component are
classified as sonorants.
1. Obstruents # Noise (+ tone) consonants
(a) Plosives
(b) Fricatives
(c) Affricates
2. Sonorants # Tone consonants
(a) Oral sonorants # central air-release [ j, w, , r, Ú, ] etc.
# lateral air-release [ l, ] etc.
(b) Nasal sonorants # [ m, n, , ] etc.
94
(i) the muscular effort.

It has been suggested that some consonants require more energy and muscular effort
to produce than others. English consonants have been therefore divided into two
groups. One group consists of voiced consonants, which tend to be articulated with
little muscular effort, and the consonants made in this manner are called lenis
consonants. The other group comprises voiceless consonants, which are articulated
with relatively strong muscular effort. This group of consonants is called fortis
consonants.
(j) the possible length.
Some consonants can be held on continuously without change of quality. That means
that !... you can continue making them without interruption as long as you have
enough air in your lungs# (Roach 1991: 47). Such consonants are called
continuants. They include nasal, lateral, fricative and trilled consonants. Vowels are
also continuants, but the term is used predominantly for consonants.
(k) the amount of high-frequency energy.
Some consonants have more acoustic energy (greater loudness) at high frequencies.
These sounds are called sibilants. We have six sibilants in English # / s, z, , µ, t,
dµ /.
2.4.2 The Length of English Consonants

The length of consonants in English is not a distinctive feature. That means that
the length of consonants does not influence the meaning of words. Primarily, all
consonants in English are short (unlike in Slovak, where we have long Ě and Ď,
although they do not have a phonemic status). However, the length of English
consonants varies according to the type of the preceding and following sound. Only
continuant consonants can change their length. These are the main cases in which
consonants can change their length:
1. All continuant consonants standing in the word-final position of one-syllable

words are lengthened when preceded by a short vowel, e.g.
come [ kYm; ] # calm [ km ] Chris [ kr s; ] # crease [ kris ]
sin [ s n; ] # scene [ sin ] buzz [ bYz; ] # bars [ bz ]
2. Lateral and nasal consonants can often be lengthened for the sake of emphasis
(usually emotional reasons), when they are preceded by a short vowel and followed
by a voiced consonant, e.g.
splendid [ \splen;d d ] hungry [ \hYn;gr ]

numbers [ \nYm;b±z ] seldom [ \sel;d±m ]
95
3. When two identical continuant consonants meet at a word boundary, they may
merge into a single longer one, e.g.
some men [ ©sYm;ªen ] this site [ ©s s;ªa t ]

tough fight [ ©tYf;ªa t ] ten nights [ ©ten;ªa ts ]
It is important to realize that the real length expressed by a one-point length

mark is relative, and it depends on many factors such as tempo of speech, stress,
sound environment, etc.
2.4.3 The Description of English Distinctive

Consonants
Auditory (Qualitative) Impression
Obstruents Sonorants
Nasal Oral
Plosive Affricate Fricative Lateral Central
Bilabial p b m w
Labiodental f v
* s
Place of Articulation
Dental
Alveolar t d s z n l
Post-alveolar r
Palato-alveolar t dµ µ
Palatal j
Velar k g Ì
Glottal h
Table 7. Chart of English distinctive consonants.
2.4.3.1 Obstruents
Obstruents are consonants containing a noise component. This group comprises
plosives, affricates, and fricatives.
2.4.3.1.1 Plosives
There are six distinctive plosive consonants in English # / p, t, k, b, d, g / and

the glottal plosive [ = ] which occurs frequently (as an alternative for / p, t, k / in
certain contexts), but it does not change the meaning of words, that is, it is not a
distinctive consonant. The term !plosive" comes from the fact that this type of
96
consonant produces plosion, which is the escape of the compressed air from the
mouth. Plosive consonants are sometimes called stops, because during their
articulation they form a complete obstruction in the mouth and stop the airstream
issuing from the lungs.
A plosive is a consonant, the production of which consists of three main phases

(voiceless plosives may also have an additional, fourth phase):
1. Closure phase (Closing stage). During this phase the articulators move to form
an obstruction (stricture).
2. Hold phase (Compression stage). During this phase the air compresses behind
the closure.
3. Release phase (Release stage). During the release phase !... the organs forming
the obstruction part rapidly allowing the compressed air to escape abruptly ...#
(Gimson & Cruttenden 1994: 139). During the release the escape of air produces an
audible noise called plosion.
4. Post-release phase (Post-release stage). During this phase the air escaping
through the vocal folds makes a sound similar to / h /. This additional puff of air is
called aspiration. The post-release phase is noticeable only in some contexts (stressed
voiceless plosives / p, t, k /).
English plosives can stand in three positions in words # initial position, medial
position and final position.
1. Word-initial position
All plosive consonants in English standing in the initial position, e.g. part, boat,
keen, etc., produce plosion # audible escape of air. Voiceless plosives produce
stronger plosion than voiced plosives. Voiceless plosives / p, t, k / in stressed initial
position followed by a vowel or diphthong are accompanied by aspiration.
Aspiration is an additional puff of air (audible release of breath) accompanying a
sound"s articulation. It is "... a period of voicelessness that follows the voiceless
closure phase of a stop2" (Abercrombie 1967: 148). It is symbolized by a small raised
[ h ] which follows the main symbol, in the case of English, / p, t, k /. When these
plosives are followed by / l, r, j, w /, the aspiration is manifested as devoicing of / l, r,
j, w /. Voiceless plosives preceded by / s / in a stressed position, e.g. stone, spy, sky,
etc., lose their aspiration. Voiced plosives are never aspirated.
It has been found out that in the word-initial position, it is mainly aspiration
which distinguishes voiceless plosives from voiced ones. The initial voiced
plosives / b, d, g / are often pronounced with very little voicing (Lisker & Abramson,
1964), so the main factor distinguishing them from voiceless plosives / p, t, k / is
aspiration.
2
stop = plosive
97
2. Word-medial position
The pronunciation of English plosives in medial positions depends to a large
extent on the presence or absence of the primary stress, and also on the preceding and
following sounds. All plosives in the stressed medial position produce audible
plosion, and all voiceless plosives in this case are aspirated or devoice the following
/ l, r, j, w /. Voiceless plosives preceded by / s / in stressed medial position, e.g.
astound / ±ªstahnd /, respect / r ªspekt /, discourage / d ªskCr dµ /, etc., lose their
aspiration. English voiceless plosives occurring word-medially in unstressed syllables
are not aspirated, e.g. later / ªle t± /, barking / ªbCk Ì /, apron / ªe pr±n /, etc.
3. Word-final position
The plosion of all English plosives in the final position, e.g. stop / stÂp /,
lick / l k /, bag / b g /, etc., is very weak and often not audible. Voiced plosives
standing finally have very little voicing. Therefore the main factor distinguishing
voiced plosives from voiceless ones in final position is the length of preceding
vowels (see 2.1.3).
The process of plosion of English plosives can be realized in many different

ways, whether the plosion occurs initially, medially or finally. It depends largely on
the type of the following sound. The main division is according to the direction of the
release, which can be done either through the nasal passage # nasal plosion, or
through the oral cavity # oral plosion (Abercrombie, 1967).
1. Oral plosion (Oral release)

(a) Central plosion (Central release)
This type of plosion is the most frequent one in English. It occurs when a
plosive is followed by a vowel, diphthong, triphthong, fricative (in certain contexts),
or approximant, or when it stands in the word-final position, e.g. pit, bake, power,
hops, quick, next, etc. After the opening of the closure, the air is released centrally
through the mouth.
(b) Lateral plosion (Lateral release)

When English alveolar plosives / t / and / d / are followed by lateral
consonant / l /, the air is released laterally, over one or both sides of the tongue. Such
lateral release occurs both when / l / is syllabic, e.g. cattle, muddle, little, etc., and
when / l / is followed by a vowel or a diphthong, e.g. atlas, regardless, reckless, etc.
Lateral plosion also occurs at word boundaries.
(c) Incomplete plosion (Incomplete release)

"Sometimes plosive consonants are not fully pronounced. This happens in
normal English when a plosive consonant is immediately followed by another plosive
consonant or by an affricate" (Jones 1960: 155). In clusters of two consonants
combined in this way, either within word or at word boundary, the first plosive
98
consonant has no audible plosion (or incomplete plosion). Here are some of the
possible combinations:
/p/+/t/ kept, sleep tight

/b/+/d/ robbed, cab driver
/t/+/p/ right place
/d/+/b/ good bye
/k/+/t/ blocked, like tennis
/g/+/d/ hugged, big day
When two identical plosives or affricates occur at word boundaries, no separate

release of the first consonant is made (its plosion is incomplete). Such case is called
gemination. Here are a few examples:
/ t / + / t / great time
/ d / + / d / good day
/ t / + / t / much cheese
The first phoneme of the geminating pair tends to be elided (omitted) in rapid
casual speech. For more detail, see the respective chapters on elision.
To understand the phenomenon of incomplete plosion it is important to explain

how sounds are produced. The production of every sound can be divided into three
stages: onset stage (on-glide/approach), medial phase (articulation proper/hold) , and
offset stage (off-glide/release).
During the onset stage, the articulators approach their target position for the
articulation of a sound. The medial stage is considered the main part of the sound,
where the sound manifests its most characteristic features. The final, offset, stage is
the movement of the articulators away from the target position towards the position
required for the production of the next sound (or towards a position of rest).
Graphically, it can be expressed by the following picture:
medial stage
onset stage offset stage
What happens during incomplete plosion is that the off-glide of the first sound
overlaps with the on-glide of the second sound. The extent of overlap may differ
according to the type of the two sounds. If the two sounds are homorganic (their
place of articulation is identical), the first sound has no off-glide and the second
sound has no on-glide. Graphically it can be expressed by the following picture:
99
first sound second sound
Examples of this phenomenon (the overlapping of off-glide and on-glide) are not just
combinations of plosive consonants, but all combinations of homorganic sounds:
/ n / + / n / as in unknown
/ n / + / t / as in ant
/ Ì / + / k / as in sink
etc.
If two plosives or a combination a plosive and an affricate are not homorganic, the
amount of overlap depends on the place of articulation of the sounds combined; the
closer the points of articulation of the two sounds, the greater the amount of overlap:
first sound second sound
Examples: / k / + / t / as in fact
/ t / + / t / as in hatched
/ b / + / d / as in robbed
etc.
2. Nasal plosion (Nasal release)

When a plosive is followed by the homorganic nasal consonant (with the same
place of articulation), the oral closure is retained and the air is released through the
nasal cavity. Such nasal release occurs both when a nasal consonant is syllabic, e.g.
cotton, maiden, etc., and when it is followed by a vowel or diphthong, e.g. submerge,
fortnight, etc. Nasal plosion also occurs at word boundaries. Here are the possible
combinations:
/ p / + / m / topmost, happen (assimilated)
/ b / + / m / submerge, ribbon (assimilated)
/t/+/n/ fitness, cotton
/ d / + / n / madness, sudden
/ k / + / Ì / thicken (assimilated)
/ g / + / Ì / pagan (assimilated)
100
Bilabial plosives / p, b /
Example words: / p / peace, apple, stop

/ b / bin, able, rub
Description: Both / p / and / b / have the same place and manner of articulation.
The soft palate is raised and the obstacle to the airstream is formed by the lips. When
the air behind the closure is released, an audible noise # plosion is produced.
Generally, plosion produced by / p / is stronger than the plosion of / b /. The force of
plosion depends on the position of the plosive consonant in words. / p / is aspirated
when occurring in stressed initial and medial position followed by vowels,
diphthongs and triphthongs. The only difference between / p / and / b / is voicing #
/ p / is voiceless, / b / is voiced.
Fig. 24. Section of bilabial plosives / p / and / b /.
Variants: No important variants of / p / and / b / occur, / p / may vary slightly in

the amount of aspiration, / b / may vary a little in terms of voicing.
Alveolar plosives / t, d /
Example words: / t / tip, attack, bet

/ d / dim, idle, red
Description: The manner and place of articulation is identical for both / t / and
/ d /. The soft palate is raised and the obstacle to the airstream is formed by the
closure made between the tongue and the alveolar ridge. When the compressed air
behind the closure is released, an audible noise # plosion # is produced. / t / is
aspirated in stressed initial and medial positions followed by vowels, diphthongs or
triphthongs. These two sounds are differentiated by voicing only # / t / is voiceless,
/ d / is voiced.
101
Variants: In General RP, / t / is often replaced by [ = ] at word boundaries e.g. not

there [ ©nÂ= ªse± ], what now [ ©wÂ= ªnah ]. In Cockney, / t / can be replaced by / = /
not only at word boundaries, but also intervocalically (when / t / stands in between
two vowels), e.g. better [ ªbe=± ], a lot of [ ± lÂ= ±v ]. In American English / t / in
an unstressed intervocalic position, e.g. better, cut it, city, etc., or preceded by a
vowel and followed by / l /, e.g. battle, cattle, rattle, etc., is generally realized as a
voiced tap # [ ].
Fig. 25. Section of alveolar plosives / t / and / d /.
Velar plosives / k, g /
Example words: / k / kind, acre, lock

/ g / gain, eager, bog
Description: The place and the manner of articulation of / k / and / g / is identical.
The soft palate is raised, the obstacle to the airstream is formed by a closure made
between the back of the tongue and the soft palate. When the compressed air behind
the closure is released, an audible noise # plosion # is produced. / k / is aspirated in
stressed initial and medial positions followed by vowels, diphthongs or triphthongs.
These two sounds are differentiated by voicing only # / k / is voiceless, / g / is voiced.
Variants: The quality of / k / and / g / depends to a large extent on the following

vowel. When / k / and / g / are followed by / i /, e.g. key, geese, etc., the closure will
be almost palatal. When / k / and / g / are followed by / Â /, e.g. cot, got, etc., the
velar closure will be of the most retracted (back) type.
102
Fig. 26. Section of velar plosives / k / and / g /.
Glottal plosive [ = ]
This glottal plosive (glottal stop/glottal catch) is not a distinctive consonant in

English. Nevertheless, it is so frequently used that it deserves a brief mention.
Description: The obstruction to the airstream is formed by the closure of the vocal
folds. The production of one sound is abruptly stopped by closing the glottis, and
after a short period of silence, another sound continues suddenly. The glottal plosive
is usually considered voiceless. However, the classification of this plosive is of
somewhat problematic nature. Although it is classified as a sound, it manifests no
acoustic features comparable with other sounds. It is, in fact, nothing but an abrupt
cessation of sound occurring between two neighbouring sounds. It is classified as
neither voiced nor voiceless.
Usage: For many RP speakers, [ = ] serves as a syllable boundary marker in those

cases, when the initial sound of the second syllable is a vowel. This situation can
occur within word, e.g. reaction [ r =4k³±Q ], co-operate [ k±h=Âp±re t ], or at
word boundary, e.g. day after day [ de =©Cft±de ], it%s empty [ ts=ªempt ].
For some RP speakers, [ = ] can also serve as a replacement for / p /, / t / and
/ k /. Most frequently it occurs at a word boundary, where final / p /, / t /, and / k / are
followed by a word beginning with a consonant, e.g. not there [ ©nÂ= se± ], soap
powder [ ªs±h=©pahd± ]. The replacement of final / p / and / k / by [ = ] is much less
frequent than the replacement of final / t / by [ = ]. The glottal plosive [ = ] can
sometimes substitute / t / within a word. This can happen when / t / is followed by a
consonant e.g. Scotland [ ªskÂ=l±nd ], football [ ªfh=bl ], nutmeg [ ªnY=meg ], etc.
103
This is still acceptable in RP. However, when / t / is replaced by [ = ] intervocalically

within word, e.g. better [ ªbe=± ], butter [ ªbY=± ], etc., it is not acceptable in RP.
Fig. 27. Section of the glottal plosive [ = ].
2.4.3.1.2 Affricates
There are two distinctive affricate consonants in English: / t / and / dµ /
!Affricates are rather complex consonants. They begin as plosives and end as
fricatives# (Roach 1991: 47). However, not all sequences of plosive plus fricative can
be termed affricates. The first criterion which an affricate must meet is that plosive
and fricative must be homorganic, that is, made with the same articulators. Eight
sound sequences comply with this first criterion # / t /, / dµ /, / tr /, / dr /, / ts /,
/ dz /, / t* /, / ds /. The second criterion is the distribution of affricates, which means
that an affricate should: (a) regularly occur in all positions in words # initially,
medially and finally, and (b) in an intervocalic position they should behave like
simple consonants, that is, "...without separation of the elements between the
syllables (cf. pitches, pities; aged, aided)" (Gimson & Cruttenden 1994: 158). Thus
the number of the possible sequences is reduced to two # / t / and / dµ /, since none
of the other sequences can occur in all three positions, or function as simple
consonants at syllable boundaries.
Even though they consist of two different sounds, affricates are considered to
be single independent phonemes in English unlike the rest of the above mentioned
sound sequences, which are viewed as two separate phonemes rather than one.
104
Palato-alveolar affricates / t, dµ /

Example words: / t / change, feature, much
/ dµ / just, urgent, age
Description: The soft palate is raised and the obstacle to the airstream is formed by
a closure made between the tongue and alveolar ridge and palate at the same time.
/ t / may be slightly aspirated in all positions where / p /, / t / and / k / are. The
difference between these two affricates lies primarily in voicing # / t / is voiceless,
/ dµ / is voiced.
Fig. 28. Section of palato-alveolar affricates / t / and / dµ /.
Variants: No important variants of / t / and / dµ / occur. However, some careful

speakers use / t / + / j / instead of / t /, and / d / + / j / instead of / dµ / in some words,
e.g. Christian / ªkr,stj±n /, soldier / ªs±hldj± /.
2.4.3.1.3 Fricatives
There are nine distinctive fricative consonants in English:
/ f /, / v /, / * /, / s /, / s /, / z /, / /, / µ /, / h /
!In the articulation of a fricative consonant, two organs are brought and held
sufficiently close together for the escaping airstream to produce local air turbulence
(friction); fricatives are, therefore, like plosives and affricates, characterized by a
noise component# (Gimson & Cruttenden 1991: 162). All of the English distinctive
fricatives, with the exception of / h /, can be found in all positions in words #
105
initially, medially and finally. There is no word in English which would end with
/ h /. The degree of voicing of voiced fricatives may vary according to context.
Labiodental fricatives / f, v /
Example words: / f / fight, after, laugh

/ v / very, over, dove
Description: The soft palate is raised and the lower lip makes contact with the
upper teeth. The escaping air produces friction. The difference between / f / and / v /
is in the voicing only. / f / is voiceless, / v / is voiced.
Fig. 29. Section of labiodental fricatives / f / and / v /.
Variants: No important variants of / f / and / v / occur. The degree of voicing of / v /

may vary according to context. In fast speech, / v / can be omitted in unstressed forms
of words like of, have, e.g. waste of time / we st ± ta m /, he could have done it / hi
k±d ± dYn t /.
Dental fricatives / *, s /
Example words: / * / thick, method, earth

/ s / this, other, soothe
Description: The soft palate is raised and the tip of the tongue touches the edge of
the upper teeth. With some speakers, the tip of the tongue may protrude between the
teeth. The escaping air produces friction. / * / is voiceless, / s / is voiced.
106
Fig. 30. Section of dental fricatives / * / and / s /.
Variants: No important variants of / * / and / s / occur. In Cockney and southern

American, / * / and / s / may be replaced by / f / and / v / respectively, e.g. throw
/ fr±h /, mother / ªmYv± /.
Alveolar fricatives / s, z /
Example words: / s / some, concert, boss

/ z / zoo, easy, was
Description: The soft palate is raised and the tip (and blade) of the tongue makes
light contact with the alveolar ridge. The air escapes through a narrow groove
between the tongue and alveolum and creates friction. Again, the main difference
factor of these two fricatives is voicing. / s / is voiceless, / z / is voiced.
Fig. 31. Section of alveolar fricatives / s / and / z /.
107
Variants: The degree of voicing of / z / may vary in different contexts. Word

final / s / and / z / readily assimilate when followed by / / and / j /, e.g. horse
shoe / ªhu /, miss you / ªm (j)u /. For more detail, see the chapter on
assimilation.
Palato-alveolar fricatives / , µ /
Example words: / / sheep, ashore, rush

/ µ / genre, pleasure, mirage
Description: The soft palate is raised and the tongue makes contact with the
alveolar ridge and the hard palate at the same time. The escaping air creates friction.
The pair is produced with the same manner and in the same place of articulation. It is
distinguished by voicing only. / / is voiceless, / µ / is voiced.
Fig. 32. Section of palato-alveolar fricatives / / and / µ /.
Variants: No important variants of / / and / µ / occur. There may be slight

differences in the degree of palatalization (involvement of the hard palate) and lip-
rounding.
Glottal fricative / h /
Example words: huge, hot, behave

Description: !The place of articulation of this consonant is glottal. This means
that the narrowing that produces the friction noise is between the vocal folds ...#
(Roach 1991: 50). The problem with defining this fricative lies in the difference
between its acoustic qualities and its distribution. The thing is that / h / always takes
on the quality of the following sound, which is most often a vowel. Thus, from the
108
phonetic point of view, / h / can be classified as a voiceless vowel with the quality of
the vowel that follows it. Phonologically, however, / h / is a consonant.
Fig. 33. Section of the glottal fricative / h /.
Variants: Although / h / is primarily classified as a voiceless sound, it can occur as

a voiced variant in the word-medial intervocalic position, e.g. behind / b ha nd /,
anyhow / ªen hah /, etc. In many types of popular regional speech in England, Wales
and Australia, / h / is omitted both in content and function words, e.g. house / ahs /,
his / z /, etc. In RP, omission of / h / in content words is unacceptable (for more
detail, see 4.1).
2.4.3.2 Sonorants
The perceptual group of sonorant consonants can be divided into sonorants
with nasal resonance, and sonorants with oral resonance.
2.4.3.2.1 Nasal Sonorants
There are three distinctive nasal sonorant consonants in English: / m /, / n /, / Ì /
"Nasal consonants resemble oral plosives in that a total closure is made within
the mouth; they differ from such plosives in that the soft palate is in its lowered
position, allowing an escape of air into the nasal cavity and giving the sound the
special resonance provided by the naso-pharyngeal cavity" (Gimson & Cruttenden
1994: 176). All nasal consonants are continuants, that is, they can be produced as
109
long as one has enough air in the lungs. They differ from the previously analyzed
consonants in that they do not contain any noise component in their production.
Therefore they belong among sonorants. All English nasals are voiced.
Bilabial nasal / m /
Example words: mother, summer, come

Description: The lips form a closure similar to that of / p / or / b /, the soft palate is
lowered, so the air escapes through the nasal cavity. / m / is a voiced consonant, but it
can be partially devoiced by a preceding voiceless consonant, e.g. smoke, topmost.
When / m / is followed by / f / or / v /, the closure may be labiodental, e.g. comfort,
triumph, etc.
Fig. 34. Section of bilabial nasal / m /.
Variants: There are no important accent variants of / m /.
Alveolar nasal / n /
Example words: no, dinner, fan

Description: The tongue touches the alveolar ridge and forms a closure similar to
that of / t / or / d /. The soft palate is lowered and the air escapes through the nasal
cavity. / n / is a voiced consonant, but it can be partially devoiced by a preceding
voiceless consonant, e.g. snitch, chutney. When / n / is followed by / f / or / v /, the
closure may be labiodental, e.g. infant, in vain.
Variants: There are no important accent variants of / n /.
110
Fig. 35. Section of alveolar nasal / n /.
Velar nasal / Ì /
Example words: singer, anxious, hang

Description: The back of the tongue touches the velum and forms a closure similar
to that of / k / or / g /. The soft palate is lowered and the air escapes through the nasal
cavity. / Ì / is voiced, but occasionally it can be partially devoiced in rare cases of
assimilated syllabic / Ì /, e.g. bacon [ ªbe ky ], thicken [ ª* ky ], etc.
Variants: In many regional types of speech, [ Ìg ] is used instead of / Ì /, e.g.

singing [ ªs Ìg Ìg ]. In many areas, the final -ing / Ì / of the present participle is
pronounced as / n /, e.g. morning [ ªmn n ], doing [ ªdu n ].
Fig. 36. Section of velar nasal / Ì /.
111
2.4.3.2.2 Oral Sonorants
This group of consonants comprises two main types of consonants: central oral
sonorants and lateral central sonorants.
A. Central Oral Sonorants
There are three distinctive central oral sonorants in English: / r /, / j /, / w /
All central oral sonorants in English are approximants. "An approximant, as a

type of consonant, is rather difficult to describe; informally, we can say that it is an
articulation in which the articulators approach each other but do not get sufficiently
close to each other to produce a !complete# consonant such as a plosive, nasal or
fricative" (Roach 1991: 59). Thus, any close vowel could be classified as an
approximant, but the term !approximant" is usually used only for consonants. During
the production of central oral sonorants (approximants) the airstream escapes through
a relatively narrow opening in the mouth without any significant friction. All central
oral sonorants, as the term indicates, are voiced.
Post-alveolar central oral sonorant / r /
Example words: rat, very, arrive

Description: There are many variants of this sound in English, but the most
common and recommended allophone of RP / r / is a voiced post-alveolar
approximant. During the production of this sound, the soft palate is raised and the
tip of the tongue is held, without contact, close to the back part of the alveolar ridge.
The airstream escapes freely, without friction, and we can continue producing this
sound as long as we have enough air in the lungs, therefore we call it frictionless
continuant. Phonetically, this speech sound is very similar to vowels, phonologically
it is a consonant. RP / r / occurs only before a vowel or a voiced consonant, e.g.
arrow, agree, etc., or in the word-initial position, e.g. rat, real, etc. When / d /
precedes / r /, / r / becomes a fricative, e.g. boundary, bedroom, etc. / p /, / t / and /
k / preceding / r /, devoice / r / completely, e.g. cry, price, etc. The devoiced / r / is
phonetically transcribed as [ G ].
Variants: In RP, the approximant may be replaced by an alveolar tap [ ], when

standing in intervocalic positions, e.g. serious, very, etc. In Scottish English and in
RP, a lingual trill (roll) can sometimes be heard. This trill is similar to Slovak ! Ě ". In
American English, the tongue is curled backwards towards the hard palate and the
sound produced is called retroflex / r /, which is phonetically transcribed as [ ]
112
Fig. 37. Section of post-alveolar approximant / r /.
Palatal central oral sonorant / j /
Example words: yard, music, refuse

Description: The soft palate is raised. The tongue assumes the position for a close-
mid or close vowel and moves immediately into the position of the following sound.
The lips are neutral or spread. / j / may be devoiced when preceded by / p /, / t /, / k /
or / h / in stressed syllables, e.g. pew, tune, accuse, huge, etc.
Variants: In many cases of RP and especially in American English, the sequence
/ j / + / u / is reduced to / / when preceded by / t /, / d /, / n /, / l / and / s /, e.g.
tune, due, new, lewd, Sue, etc. When / j / is preceded by an aspirated / p, t, k /, it is
devoiced to [ I ]. The words pew, tune, queue can thus be transcribed as [ pIu ],
[ tIun ], [ kIu ] respectively.
Fig. 38. Section of palatal approximant / j /.
113
Labial-velar central oral sonorant / w /
Example words: wait, twice, swim

Description: The soft palate is raised. The tongue assumes the position for a back
close-mid or close vowel and moves immediately into the position of the following
sound. The lips are rounded. When / w / is followed by / u /, e.g. woo, wound, etc.,
or preceded by / g / or / k /, e.g. guan, quick, etc., the articulation is labial and velar at
the same time, forming two equal strictures. In all other contexts, the tongue is held
in a slightly lower position. When / w / is preceded by / k / or / t / in stressed
syllables, / w / becomes completely devoiced, e.g. queen, twitch, etc.
Fig. 39. Section of labial-velar approximant / w /.
Variants: Amongst careful Conservative RP speakers and in Scottish and

American English, words beginning with !wh", e.g. when, why, what, etc., are
pronounced with the voiceless labial-velar fricative [ Z ]. In such speech we can
differentiate between words like weather & whether, witch & which, wine & whine,
etc., which makes the labial-velar fricative / Z / (= / z /) a separate phoneme.
B. Lateral Oral Sonorant
There is only one distinctive lateral consonant in English: / l /
"A lateral consonant is one in which the passage of air through the mouth does
not go in the usual way along the centre of the tongue; instead, there is complete
closure between the centre of the tongue and the part of the roof of the mouth where
contact is to be made ..." (Roach 1991: 58). The air is, therefore, released along the
114
sides of the tongue. There are several varieties of / l / in English, but for practical
purposes it is sufficient to distinguish two # clear [ lj ]and dark [ 1 ].
Lateral / l /
Example words: leap, glad, kill

Description: The soft palate is raised and the tip of the tongue is in contact with the
alveolar ridge. The air escapes along both sides of the tongue. For clear [ lj ], the front
of the tongue is raised towards the hard palate. For dark [ 1 ], the front of the tongue is
depressed and the back of the tongue is raised slightly towards the soft palate.
clear [ lj ] dark [ 1 ]
Fig. 40. Sections of clear and dark / l /.
Variants: In RP, there are two main variants # clear [ lj ] and dark [ 1 ]. Clear [ lj ]
occurs when vowel or / j / follows, e.g. leap, lure, etc.; dark [ 1 ] occurs when
followed by a consonant (other than / j /), e.g. build, felt, etc., and in the word-final
position, e.g. mill, battle, etc. In Cockney, word final / l / is realized as a centralized
vowel [ ö ], e.g. sell [ seö ], girl [ gö ], etc. In American English, Scottish English,
Australian English and New Zealand English, as well as in many parts of the north of
England and north Wales, dark [ 1 ] may occur in all positions.
115
The Word
3 The Word
In the preceding chapters we focused our attention on the individual sounds of

the English language from both phonetic and phonological points of view. We were
concerned with the segments out of which words are built, that is, we were dealing
with segmental features of the language. In the following chapters, on the other hand,
we will concentrate predominantly on the way the individual segments are grouped
together to form words and sentences. We will be dealing with more than one
segment; we will be looking at suprasegmental features of English language
pronunciation.
3.1 Suprasegmental Features of Language

Although the term !suprasegmental features" has been around for some time,
and it has been used extensively by a number of linguists, no unanimous definition of
the term has yet been agreed upon. We will define suprasegmental features
(plurisegmental/non-segmental/superfix/suprafix features) as features stretching
over more than one sound segment (phoneme) in an utterance. Examples of such
features are stress, assimilation, elision, intonation, etc., that is, all the phenomena
covered in the chapters 3 and 4.
In connection with suprasegmental features, the terms prosodic features and
paralinguistic features are often mentioned.
Prosodic features of pronunciation are the features that "... generally extend
over stretches of utterances longer than just one sound and are hence often referred
to as suprasegmentals ..." (Cruttenden 1986: 1). Although they are often equated with
suprasegmental features, they should be viewed as a subcategory of suprasegmental
features. They are features which, in addition to the fact that they stretch over
sequences larger than one phoneme, also participate (or may participate) in the
meaning of the utterance. Here are the main prosodic features:
A. INTONATION
1. Tone (Pitch direction). The tone-unit can be given any of the five tones
found in English pronunciation # fall, rise, fall-rise, rise-fall, and level. The tones may
signal doubt (fall-rise), surprise (rise-fall), question (rise), etc.
2. Pitch range. The width of the pitch range of individual speakers differs, and
they usually do not use the whole width for communication. The position of the pitch
range within the overall pitch spectrum is different for different speakers. The pitch
range can signal such things as sex (a man or a woman), age (a child or an adult), the
physiological properties of the speaker"s articulators, etc.
116
The Word
3. Key. Key is a particular part of the pitch range. Speakers may use high key
(the upper part of the pitch range), mid key (the central part of the pitch range), or
low key (the lower part of the pitch range). The usage of a particular key may signal
excitement, good mood (higher key), grumpiness (lower key), etc.
4. Pause. An utterance may contain different number of pauses of varying
length, signalling hesitance (many pauses, longer pauses), shortage of time (few or no
pauses, short pauses), etc.
5. Loudness. The individual syllables of the utterance can receive different
loudness, and also the utterance as a whole can be loud, quiet, or the loudness can be
increasing or decreasing. It can signal anger, joy (loud), tenderness (soft, quiet), etc.
B. STRESS
1. Pitch. Pitch or the movement of pitch stretches over one or more syllables
and determines the meaning of the word or word group.
2. Loudness. The individual syllables of the utterance can receive different
loudness, which can contribute to the prominence of the individual units of the
utterance.
3. Length. Length, in combination with pitch and loudness, contributes to the
overall prominence of syllables.
C. RHYTHM
1. Tempo (Speed). The tempo of the utterance is another prosodic feature,
which may signal shortage of time, fear (fast), fatigue (slow), etc., or the type of
personality, e.g. phlegmatic type (slow), choleric type (fast), etc.
2. Rhythmicality. The utterance or its parts can be rhythmical or arhythmical.
Rhythmical speech may signal composure, arhythmical speech may signal hesitance,
unrest, etc.
3. Pause. An utterance may contain different number of pauses of varying
length, signalling hesitance (many pauses, longer pauses), shortage of time (few or no
pauses, short pauses), etc.
Paralinguistic features of pronunciation are features relevant to the act of

speaking but they cannot be regarded as components of speech. There are two such
features: voice qualifiers and voice qualifications. Some linguists also include kinesic
features in this group, but we prefer to put them into a separate category (see Table
1.).
1. Voice qualifiers. The speakers voice can be husky, creaky, resonant, guttural etc.
2. Voice qualifications. The utterance is sometimes accompanied with such vocal
effects as laugh, giggle, sobbing, crying, etc.
117
The Word
3.2 The Syllable
3.2.1 The Definition of the Syllable

When we pronounce or hear words, we are intuitively able to divide them into
smaller acoustic units, which do not necessarily carry lexical or grammatical meaning.
These units can be represented by just one segment of speech # one speech sound, or
by the whole sequences of speech sounds. The exact number of the speech sounds
within the unit, however, is often questionable. Thus, the English word !many"
/ ¥men /, can be divided into two distinct units, / me / and / n /, without any
grammatical or lexical grounds. The units of this sort are called syllables, and we can
find them in all languages. They are "... typically larger than a single sound and
smaller than a word" (Crystal 1991: 338).
A precise definition of the syllable is a difficult task. Several theories attempt to
explain what the syllable is, both from the phonetic and phonological point of view.
Let us take a look now at some of them.
(A) Phonetic viewpoint.
There have been several attempts to define syllables on the basis of the
articulatory effort needed in order to produce them. The pulse theory is one of them.
It was proposed by the psychologist R. H. Stetson. He claimed that every syllable is
produced by the air released from the lungs as a series of chest pulses # contractions
of the muscles of the rib cage that push more air from the lungs. He made a number of
observations of the way the respiratory system works, but his theory was later proven
wrong. According to his theory, the English word !going" would have to be produced
with two chest pulses, because, as we know, it consists of two syllables. In reality,
however, this word is usually said in a single muscular effort # one chest pulse.
Another phonetic approach tries to explain the nature of the syllable in auditory
terms. It attempts to define the syllable from the point of view of the inherent sonority
of each sound. Sonority1 could be defined as a relative loudness or distinctiveness of
the sound, that is, loudness depending on the subjective impression of the listener, not
only the physically measurable intensity. The sonority of sounds depends on the way
they are produced. Various classes of sounds have therefore different degree of
sonority. The vowel / C /, for instance, has more sonority than the consonant / n /,
which in turn has more sonority than the consonant / t /. Thus, we can set up
1
It is important to distinguish between sonority and prominence. Sonority is the inherent quality of
the sound and concerns its relative distinctiveness. Prominence is the overall impression of
distinctiveness brought about by conscious application of the following factors: quality (which
includes sonority), quantity, loudness, and pitch.
118
The Word
a sonority scale or hierarchy which ranks the sounds from the most sonorous ones to
the least sonorous ones. The sonority scale is shown in the following figure.
High sonority open vowels

close vowels
laterals
nasals
approximants
trills
fricatives
affricates cannot constitute the peaks of sonority
Low sonority plosives and taps
Fig. 41. The sonority scale
The sounds placed the highest on the sonority scale constitute the peaks of
sonority, whereas the sounds in the lower part of the scale constitute the troughs
(valleys) of sonority. The theory based on the sonority of the sounds is called the
sonority hierarchy or prominence theory.2 The main argument of this theory is that
"The number of syllables in an utterance equates with the number of peaks of
sonority ..." (Gimson & Cruttenden 1994: 49).
Although the English approximants (/ r /, / j /, / w /) have greater sonority than
fricatives, affricates, and plosives and taps, only / r /, when syllabic, can constitute the
syllable nucleus. The other two, / j / and / w /, are never found as syllable nuclei. In
other words, from the phonetic point of view, all approximants could constitute the
peak of sonority, but from the phonological point of view, / j / and / w / cannot
constitute the peak of sonority in English words, because they are never found in the
centre of syllables.
The above scale reflects only phonetic criteria, that is, it classifies the sounds as
being or not being able to constitute the peak of sonority on purely acoustic grounds.
If we were to reflect both phonetic and phonological criteria, we could create two
groups of sounds: the sounds that can function as a syllable nucleus and the sounds
that cannot function as a syllable nucleus.
2
The term is somewhat misleading because prominence is brought about mainly by stress, and it
functions primarily at a word and sentence level, having certain lexical and grammatical functions.
Of course, sonority too can play a certain part in the prominence of a syllable # the more sonorous a
sound is, the more prominent is the syllable. Sonority, however, functions primarily at a level lower
than that of a word, and it determines its syllable structure.
119
The Word
1. Sounds that can function as a syllable nucleus: vowels (diphthongs, triphthongs)

and syllabic consonants
2. Sounds that cannot function as a syllable nucleus: obstruents and oral sonorants
/ j / and / w /.
Let us take a look now at concrete examples. The English word !letter", for
instance, contains two peaks and two troughs of sonority.
___________________
___________________
l e t ±
As we can see, the number of syllables of the word !letter" equals the number
of the peaks of sonority. However, there are cases where the number of the contours
created by the highest peaks does not correspond to the number of syllables. Let us
take the word !spit" as an example.
________________
________________
s p t
The contour suggests two syllables, but we know intuitively that the word !spit"
consists of one syllable only. This means that sounds below a certain level of sonority
cannot function as peaks of sonority. These are classes of sounds from fricatives
downwards, although the cut-off point may be different in different languages.
(B) Phonological viewpoint.
"Phonological views of the syllable ... focus on the way sounds combine in
individual languages to produce typical sequences" (Crystal 1991: 339). In this
respect we distinguish two classes of sounds:
(1) Sounds that can either occur on their own, or at the centres of a sequence of
sounds. This class of sounds comprises vowels, diphthongs, triphthongs, and syllabic
consonants.
(2) Sounds which cannot occur on their own, or which occur at the margins of
syllables. This group is usually referred to as consonants.
A definition of the syllable in phonological terms is concerned with the
structure of the syllable and the function of its elements. From the structural point of
view, a syllable can be defined as "A unit of spoken language consisting of a single
uninterrupted sound formed by a vowel or diphthong alone, of a syllabic consonant
120
The Word
alone, or of either with one or more consonants" (Webster"s II New Riverside

University Dictionary 1984: 1172).
Syllables can consist of different sequences of vowels and consonants, and
their structure can be defined in the following terms (Roach, 1991):
(a) The syllable can consist of a single vowel, diphthong, triphthong, or a syllabic
consonant. This type of syllable is called a minimum syllable. It is preceded and
followed by silence . Examples: are / /, or / /, err / /, eye / a /, our / ah± /,
hmm [ ], etc.
(b) The syllable can start with one or more consonants, and end with a vowel,
diphthong, or triphthong. This type of syllable is called an open (free) syllable. It has
an opening segment(s) # onset (one or more consonants), and the centre # nucleus
(peak) (a vowel, diphthong, triphthong or a syllabic consonant). Examples: far / f /,
tree / tri /, screw / skru /, day / de /, hire / ha ± /, etc.
(c) The syllable can start with a vowel, diphthong, or triphthong, and end with one
or more consonants. This type of syllable is called a closed (checked) syllable. It has
the peak (a vowel, diphthong, or a triphthong), and the closing segment(s) # coda
(one or more consonants). Examples: is / z /, aim / e m /, east / ist /, asked / skt /,
etc.
(d) The syllable can start with one or more consonants, continue with a vowel,
diphthong, triphthong, or a syllabic consonant, and end with one or more consonants.
It consists of an onset, peak, and a coda. Examples: late / le t /, duck / dCk /, scraped
/ skre pt /, blocked / blÂkt /, etc.
The peak and coda are known as the rhyme, where coda is optional (in rhyme),
as in the word far / f /, so rhyme can consist of a single vowel, diphthong, or a
triphthong.
syllable
rhyme
onset peak coda
Fig. 42. The structure of the syllable. Onset and coda are optional in the syllable.
Examples: back / b k / / b / = onset, / / = peak, / k / = coda, / k / = rhyme
late / le t / / l / = onset, / e / = peak, / t / = coda, / e t / = rhyme
None of the definitions stated above can explain the nature, structure, and
function of the syllable in a completely satisfactory way, which is reflected in the
121
The Word
following definition proposed by Peter Ladefoged.3 He defines the syllable as "A unit
of speech for which there is no satisfactory definition. Syllables seem to be necessary
units in the mental organisation and production of utterances" (Ladefoged 1975:
281).
3.2.2 Syllable Boundaries

In the preceding section we have shown the possible structural make-up of the
syllable. Let us now take a look at the relation among several syllables within one
word. In polysyllabic words (words consisting of more than one syllable) the problem
arises of how to separate syllables from each other. In many languages the syllable
division is a relatively straightforward process. In English, however, it is not. If we
take, for instance, the sonority hierarchy theory, we can relatively easily identify the
number of syllables, because we know that the number of syllables equals the number
of peaks of sonority. Thus we are able to establish the centres of syllables, which are,
in fact, the peaks of sonority. By analogy, we could expect the troughs of sonority to
form the margins of syllables, and so, telling us where the boundaries between
syllables are. Unfortunately, the sonority hierarchy theory itself does not tell us
whether to place the trough consonant with the peak that precedes it, or with the one
that follows. The matter is further complicated by the existence of the non-peak
sounds (fricatives, affricates, plosives, and flaps), which can never form the peaks of
sonority, but they may not necessarily function as the troughs of sonority. Study the
following examples:
_______________________ _________________________
_______________________ _________________________
h p l n s C l t Ì
________________________
________________________
e k s t r ±
In the first two cases, peaks and troughs alternate regularly. In the third case,
however, there are two troughs without a peak in between.
3
/ pit l d fg d /
122
The Word
If we want to separate these words into syllables, we have to decide to which

peak we add the intervocalic consonant or consonants. There are two criteria for the
syllable division. The first one is phonetic, the second one phonological.
1. The phonetic realization of the segments of the possible syllables must
be the same as the phonetic realization of the segments within the word. This
means that if the word contains a particular allophone, the syllable must contain this
allophone too. For instance, the word !basket" [ bskΙt ] could be divided into [ b ]
# [ sk t ], or into [ bs ] # [ kh t ], or into [ bsk ] # [ t ]. If we apply the phonetic
criterion, we must rule out the second possibility [ bs ] # [ kh t ], because the [ k ]
in the word basket is not aspirated, but in the syllable [kh t ] it is. So the correct
syllable division based on the phonetic principle should be either [ b ] # [ sk t ], or
[ bsk ] # [ t ].
2. The sequences of segments of the possible syllables must correspond to
the sequences of segments found at the beginnings and ends of words. This means
that the syllables must be divided in such a way, that the created sequences are the
same as the beginning and ending sequences of English words. For instance, the word
!letter" / let± / could be divided into / le / # / t± /, or into / let / # / ± /. If we apply the
phonological criterion, we must rule out the first possibility / le / # / t± /, because no
English word ends with the sequence / le /. The only correct syllable division based
on the phonological principle is the division into / let / # / ± /.
Let us now apply the phonetic and phonological criteria to the three words
analyzed earlier # happily, insulting, and extra.
happily / ªh p l /:
Theoretical possibilities: / h - p - l /, / h p - - l /, / h p - l - /, / h -pl- /
1. Phonetic solution: [ h p - - l ].
Incorrect syllables: / p / has an aspirated [ ph ], which is not aspirated in the
word 'happily', / l / has a dark [ 1 ], while the word happily
has a clear [ l ], / p l / has both aspirated [ ph ] and dark [ 1 ],
the word happily does not have any of these allophones.
2. Phonological solution: / h p - - l /, / h p - l - /.
Incorrect syllables: / h /. No word in English ends with this sequence.
insulting / nªsClt Ì /
Theoretical possibilities: / n - sCl - t Ì /, / n - sClt - Ì /, / n - sC - lt Ì /,
/ ns - Cl - t Ì /, / ns - Clt - Ì /.
1. Phonetic solution: [ n - sClt - Ì ], [ n - sC - lt Ì ], [ ns - Clt - Ì ].

Incorrect syllables: / t Ì / has an aspirated [ th ], which is not aspirated in the
word !insulting",
123
The Word
2. Phonological solution: / n - sClt - Ì /, / ns - Cl - t Ì /, / ns - Clt - Ì /.

Incorrect syllables: / sCl /. No word in English ends with this sequence.
/ sC /. No word in English ends with this sequence.
/ lt Ì /. No word in English begins with this sequence.
extra / ªekstr± /
Theoretical possibilities: / e - kstr± /, / ek - str± /, / eks - tr± /, / ekst - r± /,
/ ekstr - ± /
1. Phonetic solution: [ e - kst{± ], [ ek - st{± ], [ eks - t{± ], [ ekst{ - ± ]

Incorrect syllables: / r± / has a voiced approximant [ Ú ], while the word !extra"
has a fricative [ { ].
2. Phonological solution: / ek - str± /, / eks - tr± /, / ekst - r± /3

Incorrect syllables: / e /. No word in English ends with this sequence.
/ ekstr /. No word in English ends with this sequence.
In cases where the phonetic and phonological solutions coincide, the choice of
the syllable division is obvious. There may even be more than one identical solution.
However, there are cases where no identical solutions are found. Here the phonetic
and phonological criteria actually rule themselves out. One such case is the word
!petrol".
petrol / ªpetr±l /
Theoretical possibilities: / pe - tr±l /, / pet - r±l /, / petr - ±l /.
1. Phonetic solution: [pe - t{±l ], [ pet{ - ±l ].

Incorrect syllables: / r±l / has a sonorant [ Ú ], while the word !petrol"
has a fricative [ { ].
2. Phonological solution: / pet - r±l /4.

Incorrect syllables: / pe /, no word in English ends with this sequence
/ tr±l /, no word in English begins with this sequence
As we can see, the phonetic and phonological solutions do not have a common
syllable division pattern. In this particular case, most dictionaries prefer the
phonological solution.
4
The sequence / r±l / occurs mainly in American English in words like !rely" / r±ªla /, !relieve"
/ r±ªliv /, etc., although the tendency to obscure clear vowels in unstressed syllables preceding the
main stress is becoming increasingly common also in British English.
124
The Word
3.2.3 Syllabic Consonants

Vowels are generally the central and the most important part of syllables.
However, there are syllables in which no vowel is found. In such a case the peak of
the syllable is formed by one of the voiced continuant consonants / m, n, Ì, l, r /. A
syllabic consonant can thus be defined as a consonant functioning as the syllable
nucleus. The syllabicity of a consonant is usually indicated by a small vertical mark
put under the consonant symbol # [ , , , , ].
Syllabic consonants can be found mainly in the final unstressed syllable of
roots, although sometimes they are found in the middle of roots (in an unstressed
position), e.g. Argentine [ ªdµta n ]. They occur after another consonant and
usually depend on the type of this preceding consonant. The main problem which
arises here, and which applies to all syllabic consonants, is the realization of the
sequence containing the syllabic consonant. There are two possibilities and
graphically they could be expressed like this:
1. / C + ± + C / e.g. mutton / ªmYt±n /

2. [ C + Cs ] e.g. mutton [ ªmYt ]
where C = consonant, ± = optional vowel / ± /, and Cs = syllabic consonant.
It is clear that if the sequence is realized as / C + ± + C /, the final consonant is

not syllabic, because it is preceded by the vowel / ± /, which forms the peak of the
syllable. However, if we omit the vowel / ± /, the final consonant becomes (very
often) syllabic: [ C + Cs ]. The reasons for keeping the vowel / ± / in the sequences of
this type can be either slow tempo of speech, the preceding consonant, or,
sometimes, as it will be shown, the particular spelling of the word.
Syllabic consonant [ ]
Syllabic [ ] is the most frequent syllabic consonant in English, although it is

not found in all accents. The syllabic [ ] in RP is found in words ending with -le, -
al, -el, -il (yl), -ol, -ul, and their inflected and derived forms. This group of words can
be divided into:
(a) Two or more syllable roots ending with -le. These words always have the
pronunciation [ C + ].
(b) Two or more syllable roots ending with -al, -el, -il (yl), -ol, and -ul. These
roots have an optional / ± / in the sequence, and can be realized either as the
sequences in (a), or with the sequence / C + ± + C /. The choice depends
predominantly on the tempo of speech (higher tempo results in dropping the / ± /).
125
The Word
Here is an analysis of the possible combination of the English distinctive

consonants with the syllabic [ ]:
A: Sequences with syllabic [ ] only B: Sequences with a choice between [ ±l ] or [ ]

(words ending with -le ) ( words ending with -al, -el, -il (yl), -ol, and -ul )
[ p ] couple, people [ p ] or [ p±l ] papal, gospel, peepul

[ b ] table, possible [ b ] or [ b±l ] global , label, symbol
[ t ] cattle, title [ t ] or [ t±l ] metal, chattel, pistol
[ d ] coddle, cradle [ d ] or [ d±l ] pedal, model, idol
[ k ] tickle, ankle [ k ] or [ k±l ] logical, nickel
[ g ] giggle, mingle [ g ] or [ g±l ] regal, bagel, argol
[ j ] # does not exist [ j ] or [ j±l ] satchel, Rachel
[ dµ ] # does not exist [ dµ ] or [ dµ±l ] angel, vigil
[ f ] rifle, piffle [ f ] or [ f±l ] offal, duffel, awful
[ v ] # does not exist [ v ] or [ v±l ] oval, novel, cavil
[ * ] # does not exist [ * ] or [ *±l ] lethal, brothel, menthol
[ s ] # does not exist [ s ] or [ s±l ] betrothal
[ s ] nestle, gristle [ s ] or [ s±l ] colossal, vessel, fossil
[ z ] fizzle, drizzle [ z ] or [ z±l ] accusal, easel
[ ] # does not exist [ ] or [ ±l ] marshal, bushel
[ µ ] # does not exist [ µ ] or [ µ±l ] usual, ambrosial (very rare)
[ h ] # does not exist [ h ] or [ h±l ] # does not exist
[ m ] # does not exist [ m ] or [ m±l ] normal, camel
[ n ] # does not exist [ n ] or [ n±l ] final, tunnel, vinyl
[ Ì ] # does not exist [ Ì ] or [ Ì±l ] # does not exist
[ w ] # does not exist [ w ] or [ w±l ] equal, sequel, tranquil
[ r ] # does not exist [ r ] or [ r±l ] oral, mongrel, carol
[ j ] # does not exist [ j ] or [ j±l ] # spaniel, Daniel
[ l ] # does not exist [ l ] or [ l±l ] # does not exist
In connected rapid speech, RP speakers usually drop / ± / in the sequences

in the column B. It is therefore advisable to prefer [ C + ] sequence in the
transcription of connected speech.
When a suffix or another root is added to a root ending with a syllabic
consonant, these rules apply:
1. If the added suffix or root is, or begins with, a consonant (in pronunciation),
the syllabic consonant remains syllabic:
kettle [ ªket ] kettles [ ªketz ] the added suffix is / z /

able [ ªe b ] able-bodied [ ªe b©bÂd d ] the added root begins with / b /
126
The Word
2. If the added suffix or root is, or begins with, a vowel (in pronunciation), the
syllabic consonant may change into non-syllabic:
battle [ ªb t ] battling [ ªb t Ì ] or [ ªb tl Ì ] the added suffix begins with / /

battle [ ªb t ] battle-axe [ ªb t ks ] or [ ªb tl ks ] the added root begins with / /
There are several pairs of words in English which are differentiated by syllabic
and non-syllabic / l /. In these cases the syllabicity is a distinctive feature, and the
two / l / sounds function as separate phonemes. However, since this contrast is not
present in all accents of English and the number of such pairs is very limited, these
two types of / l / are not considered as phonemes in the traditional sense. Here is an
example of such word pair:
coddling / \kÂd Ì / (three syllables) # codling / \kÂdl Ì / (two syllables)
This is the most frequently used nasal syllabic consonant. It is found mainly in
the words ending with -an, -en, -on, -in, -un, and their inflected and derived forms. As
in the previous case, there are different ways of realizing the sequences containing the
syllabic consonant, but unlike in the previous case, the spelling of the word does not
influence the choice of segments in the sequences. The important factor, however, is
the type of the preceding consonant. The possible sequences for the consonant [ ]
are: [ C + ], or / C + ± + n /.
The words ending with the above mentioned sequences fall into these four
categories:
1. If the ending sequence of two or more syllable words begins with any of the
following consonants / t, d, s, z, f, v, /, the sequence pattern [ C + Q ] is preferred.
[ t ] mutton, Britain [ d ] leaden, burden

[ s ] listen, person [ z ] raisin, reason
[ f ] often, stiffen [ v ] heaven, driven
following consonants / p, b, k, g, *, s, , µ, r, j, dµ /, any of the sequence
patterns [ C + ] or / C + ± + n / can be used. Here, the choice depends mostly on
the tempo of speech. In rapid connected speech the first pattern is used more often.
127
The Word
[ p ] or / p±n / happen, weapon [ ] or / ±n / nation, ashen

[ b ] or / b±n / ribbon, urban [ µ ] or / µ±n / Asian, vision
[ k ] or / k±n / bacon, African [ r ] or / r±n / baron, children
[ g ] or / g±n / pagan, wagon [ j ] or / j±n / Christian, kitchen
[ * ] or / *±n / python, earthen [ dµ ] or / dµ±n / region, pigeon
[ s ] or / s±n / heathen, smoothen
following consonants or consonant sequences / j, l, m, nt, nd /, only the sequence
pattern / C + ± + n / is used.
/ l±n / sullen, fallen / nt±n / lantern, wanton

/ j±n / stallion, bunion / nd±n / London, abandon
/ m±n / omen, Walkman
The sequences / h±n / ! [ h ], / Ì±n / ! [ Ì ] and / w±n / ! [ w ] never

occur in English words.
When a suffix or another root is added to a root ending with a syllabic
consonant, these rules apply:
listen [ ªl s ] listened [ªl sd ] the added suffix is / d /
heaven [ ªhev ] heaven-sent [ ©hevªsent ] the added root begins with / s /
syllabic consonant may become non-syllabic:
listen [ ªl s ] listening [ ªl s Ì ] or [ ªl sn Ì ] the added suffix begins with / /
The occurrence of the syllabic [ ] in English is relatively rare and the word-
final sequence / C + (±) + m / appears to follow these rules:
1. If the ending sequence of two or more syllable roots begins with any of the
following consonants / *, s, s, z, /, the sequence pattern [ C + ] is preferred.
[ * ] anthem, Gotham [ s ] blossom, ransom

[ s ] rhythm, algorithm [ z ] spasm, socialism
[ ] petersham, Evesham
128
The Word
2. If the ending sequence of two or more syllable roots begins with a consonant
other than / *, s, s, z, ∫ /, the sequence pattern / C + ± + m / is used.
/ p±m / wampum / j±m / Beecham

/ b±m / album / dµ±m / stratagem
/ t±m / item / r±m / fulcrum
/ d±m / seldom / n±m / venom
/ k±m / caecum / m±m / maximum
/ g±m / lingam / v±m / ovum
/ j±m / William / l±m / hoodlum
Sequences / f±m /, / µ±m /, / h±m /, / Ì±m / and / w±m / never occur in English
words.
3. If a word ends in the sequence / p±n /, [ p ] or / b±n /, [ b ], this sequence

can, as a result of simultaneous elision and assimilation,5 change into [ p ] or
[ b ] respectively.
happen [ ªh p±n ] → [ ªh p ] → [ ªh p ]
ribbon [ ªr b±n ] → [ ªr b ] → [ ªr b ]
When a suffix or another root is added to a root ending with a syllabic consonant,
these rules apply:
blossom [ ªblÂs ] blossomed [ ªblÂsd ] the added suffix is / d /

rhythm [ ªr s ] rhythm-based [ ©r sªbe st ] the added root begins with / b /
syllabic consonant may become non-syllabic:
blossom [ ªblÂs ] blossoming [ ªblÂs Ì ] or [ ªblÂsm Ì ] the added suffix begins

with / /
This syllabic consonant occurs in English only as a result of simultaneous

elision and assimilation. Here is the basic rule:
5
For more on assimilation and elision, see 3.4 and 3.8.
129
The Word
If a word ends in the sequence / k±n /, [ k ] or / g±n /, [ g ], this sequence can,

as a result of simultaneous elision and assimilation, change into [ k ] or [ g ]
respectively.
thicken [ ª* k±n ] → [ ª* k ] → [ ª* k ]
pagan [ ªpe g±n ] → [ ªpe g ] → [ ªpe g ]
Syllabic consonant [ | ]
The syllabic / r / is very common in American English. It is, in fact, the sound
that most American linguists transcribe as [ H ] or [ H ], which is an r-coloured vowel.
The words perceive or teacher may be, in American English, transcribed as [ pHªsiv ]
or [ ªtijH ]. In British English, the syllabic [ | ] can be found in the following cases:
If a word consisting of three or more syllables contains the sequence / C +

±r + V /, and this sequence does not carry primary stress, the / ± / in the sequence
may be omitted. / r / in such cases may become syllabic. (C = Consonant, V =
Vowel)
/ C + ±r + V / → [ C + | + V ]
secretary / ªsekr±t±r / → [ ªsekr±t| ]

bakery / ªbe k±r / → [ ªbe k| ]
history / ªh st±r / → [ ªh st| ]
flavouring / ªfle v±r Ì / → [ ªfle v| Ì ]
adventurous / ±dªvenj±r±s / → [ ±dªvenj|±s ]
In most of the words meeting these criteria the syllabic [ | ] can be replaced either by
/ ±r /, or by a non-syllabic / r / without adding / ± /, which reduces the number of the
syllables in these words by one:
bakery [ ªbe k| ] = three syllables [ ªbe kÚ ] = two syllables

history [ ªh st| ] = three syllables [ ªh stÚ ] = two syllables
In some word pairs the / r / and / / can function as separate phonemes, since they
differentiate the meaning of these words. Here are some examples:
Hungary / ªhCÌg / hungry / ªhCÌgr /

adulterous / ±ªdClt±s / adulteress / ±ªdCltr±s /
130
The Word
However, there is no word in English where it would not be possible to substitute the
syllabic [ | ] by a non-syllabic / r / or / ±r /.
Syllabic consonants in combination
There are cases in English where we can find two syllabic consonants
following each other. They usually form the last two syllables of a word. Examples of
such combination are:
national [ ªn ] visionary [ ªv µ| ]

literal [ ªl t| ] veteran [ ªvet| ]
Of course, the above words can be used with only one, or no syllabic
consonant. In rapid speech, the words with two syllabic consonants tend to keep only
one consonant syllabic, whereas the other is used as non-syllabic, e.g. national
[ ªn n ], visionary [ ªv µÚ ], etc. Nevertheless, the extent of the usage of syllabic
consonants is often a matter of individual pronunciation style.
3.3 Phonotactics
The phonemes of English, or any other language, can be theoretically arranged
into an immense number of combinations. In reality, however, only certain
combinations are possible in a given language. These combinations, or sequences,
are, of course, different in different languages. This sequential arrangement (tactic
behaviour) of phonemes occurring in language is called phonotactics.
Let us take a look now at the possible word-initial and word-final phoneme
sequences in English (RP accent). Most of them are taken from Gimson & Cruttenden
1994, with several alterations and corrections. They are grouped into categories
according to their vowel consonant pattern (V = vowel or diphthong, C = consonant):
1. V These vowels and diphthongs represent monosyllabic words:

/ i / (the letter) E
/ / (the interjection) ah, are
/ / or, ore, oar, awe
/ / err
/ u / (the interjection) ooh
/ ± / (the weak form of the indefinite article) a
/ e / (the letter) A
/ a / (the letter) I, (the pronoun) I, eye, aye
131
The Word
/ / (the interjection) oy
/ ±h / (the letter) O, (the interjection) oh, owe
/ ah / (the interjection) ow
/ ± / ear
/ e± / air, heir, Ayer, Eyre
/ h± / Ur (an ancient Sumerian city)
2. Initial V
All vowels and diphthongs occur initially
3. Initial CV
/ Ì / does not occur initially
/ µ / occurs initially before / / (gigolo), / / (jabot), and / / or / / (genre)
The rest of the consonants occur before vowels and diphthongs.
4. Initial CCV
/ p / + / l, r, j / play, pray, pew
/ t / + / r, j, w / try, tube, twice
/ k / + / l, r, j, w / claw, cry, cube, quite
/ b / + / l, r, j / bleed, breed, beauty
/ d / + / r, j, w / dry, dew, dwell
/ g / + / l, r, j, w / glance, great, gules, Guam
/m/+/j/ mew
/n/ +/j/ new
/l/+/j/ lewd
/ f / + / l, r, j / fly, free, few
/ v / + / l, r, j / Vladimir, vroom, view
/ * / + / r, j, w / three, thew, thwack
/ s / + / l, r, j, w, p, t, k, m, n, f, v / slop, Sri, suit, swell, spit, stop, sky, smear, snow,
sphere, svelte
/ / + / l, r, v, w, t, m, n / schlock, shrine, Schwerin, schwa, schtuck, schmuck,
schnitzel
/h/+/j/ huge
5. Initial CCCV
/ s / + / p / + / l, r, j / splash, spray, spew
/ s / + / t / + / r, j / street, stew
/ s / + / k / + / l, r, j, w / sclera, screech, skew, squash
/s/+/m/+/j/ smew
132
The Word
6. Final V These vowels and diphthongs occur word-finally
/ i / see /a / my / ± / fear
/ / far /e / day / e± / bear
/ / bore / / boy / h± / sure
/ u / two / ±h / go
/ / fur / ah / now
/±/ sofa
/ / happy
The rest of the vowels never occur finally. Word-final / u / and / i / can be reduced
in length to [ u ] and [ i ], but these sounds do not function as phonemes.
7. Final VC
/ r, h, j, w / never occur finally in RP
/ µ / occurs finally only after / , u, e / # camouflage, rouge, beige
/ Ì / occurs finally only after / , , C, Â / # sing, sang, sung, song
The rest of the consonants can occur finally after any vowel.
8. Final VCC
/ p / + / t, *, s, l, n / kept, depth, cops, ripple, happen
/ t / + / *, s, l, n / eighth, pots, cattle, cotton
/ k / + / t, s, l, n / raked, brakes, tickle, thicken
/ b / + / d, z, l, n / robbed, ribs, able, ribbon
/ d / + / z, l, n, * / heads, ladle, hidden, width
/ g / + / d, z, l, n / begged, rugs, ogle, pagan
/ j / + / t, l, n / hatched, Rachel, kitchen
/ dµ / + / d, n / grudged, region
/ m / + / p, d, f, *, z, l / jump, aimed, triumph, warmth, gums, normal
/ n / + / t, d, j, dµ, *, s, z, l / aunt, end, lunch, lunge, tenth, chance, guns, tunnel
/ Ì / + / k, d, z / think, winged, songs
/ l / + / p, t, k, b, d, j, dµ, m, n, f, v, *, s, z / gulp, felt, elk, bulb, bald, belch, bulge,
realm, kiln, elf, delve, wealth, else, feels
/ f / + / t, *, s, l, n / left, fifth, laughs, awful, often
/ v / + / d, z, l, n / lived, gives, novel, oven
/ * / + / t, s, l, m, n / toothed, moths, lethal, Gotham, python
/ s / + / d, z, l, m, n / scathed, clothes, betrothal, rhythm, heathen
/ s / + / p, t, k, l, m, n / lisp, last, husk, nestle, blossom, listen
/ z / + / d, l, m, n / buzzed, fizzle, spasm, reason
/ / + / t, l, m, n / rushed, marshal, petersham, nation
/ µ / + / d, l, n / sabotaged, ambrosial, vision
133
The Word
Other word-final VCC combinations are not possible in RP English.
9. Final VCCC
/ p / + / s / + / t, m / elapsed, Epsom
/ t / + / s / + / t, l / blitzed, pretzel
/ k / + / s / + / t, l / axed, axle
/d/+/s/+/t/ amidst
/m/+/p/+/t/ kempt
/ n / + / s, j / + / t / chanced, clinched
/ Ì / + / s, k / + / t / amongst, flunked
/ l / + / s, p, k, j / + / t / whilst, helped, mulct, belched
/ s / + / p, k / + / t / lisped, masked
/ n / + / dµ, z / + / d / lunged, cleansed
/ l / + / dµ, m, v / + / d / bulged, helmed, delved
/ p / + / t, * / + / s / corrupts, depths
/t/+/*/+/s/ eighths
/k/+/t/+/s/ acts
/ k / + / l / + / z, d / freckles, tickled
/k/+/w/+/l/ equal
/ m / + / p, f / + / s /
/ n / + / t, * / + / s /
glimpse, triumphs
/ Ì / + / k / + / s, l, * /
hints, tenths
/ l / + / p, t, k, f, * / + / s /
stinks, uncle, length
/ f / + / t, * / + / s /
helps, hilts, skulks, gulfs, healths
crafts, fifths
/ s / + / p, t, k / + / s / lisps, lists, basks
/ n / + / d / + / z, l / hands, spindle
/ n / + / t, dµ, s, / + / l / mantle, angel, tinsel, provincial
/ l / + / b, d, m, n, v / + / z /
/k/+/s/+/*/
bulbs, holds, elms, kilns, delves
/n/+/d/+/*/
sixth
/n/+/t/+/*/
thousandth
/n/+/*/+/m/
thousandth
/l/+/f/+/*/
anthem
twelfth
10. Final VCCCC

/ -mpts / prompts
/ -mpst / glimpsed
/ -mpsn / Simpson
/ -lkts / mulcts
/ -lpts / sculpts
/ -lf*s / twelfths
/ -ksts / texts
/ -ks*s / sixths
134
The Word
/ -nd*s / thousandths
/ -nt*s / thousandths
/ -nnt / ancient
In the last two categories we have groups of three and four consonants. In
connected speech, where consonant sequence of one word can combine with a
consonant sequence of another word, the number of consonants following each other
can increase to even a higher number, but usually not exceeding six consonants in a
row. These sequences of three and more consonants are called consonant clusters.
Consonant clusters are liable to elision # omission of sounds # in rapid speech,
especially the clusters consisting of more than three consonants. Elision of consonants
is discussed in the respective chapters on elision.
3.4 Elision within the Word

The pronunciation and phonemic make-up of words depends greatly on the
tempo of speech and style. When the tempo increases and/or when the speaker"s style
is informal, the number of sounds in words tends to decrease. This phenomenon is
called elision. Elision (omission/ellipsis/deletion) is "A term used in phonetics and
phonology to refer to the omission of sounds in connected speech. Both consonants
and vowels may be affected, and sometimes whole syllables may be elided" (Crystal
1991: 119). When a phoneme is elided, we say that it is realized as zero, or that it has
zero realization. Elision within the word is also termed intraword elision.
We distinguish between two types of elision # historical and contextual.
Historical (established) elision.
In this type of elision, the omission of a sound occurred some time in the past
and is now fully established in English. The omitted sound is never used in speech.
Here are some examples of historical elision:
(A) Vowels
The loss of word-medial vowels in words like evening, chimney, Salisbury

/ ªslzbrΙ /, etc.
The loss of word-final vowels in words like time, name, written, cousin, etc.
(B) Consonants
The loss of / r / in words like arm, horse, church, more, other, etc.
The loss of / l / in words like half, calf, walk, talk, palm, etc.
135
The Word
The loss of / t / in words like listen, fasten, often, thistle, castle, etc.
The loss of / b / in words like limb, lamb, tomb, bomb, womb, etc.
The loss of / n / in words like autumn, column, damn, hymn, solemn, etc.
The loss of / p / in cupboard, etc.
Contextual (present colloquial) elision.
In this type of elision, the omission of a particular sound occurs under the
influence of increased tempo, the sound"s quality, its position in the word, the quality
of the neighbouring sounds in a given context, and the speaker"s style. The elision of
this kind is not usually present in slow and careful speech. Here are the basic rules:
(A) Vowels
1. If the penultimate syllable of simple and derived words contains / / or / ± /, and it

is preceded by a syllable with the primary stress, / / and / ± / tend to be elided.
possible / ªpÂs bl / → / ªpÂsbl /

university / ©j n ªvs t / → / ©j n ªvst /
difficult / ªd f klt / → / ªd fklt /
parliament / ªpCl±m±nt / → / ªpClm±nt /
grappling / ªgr p±l Ì / → / ªgr pl Ì /
national / ªn ±nl / → / ªn nl /
If / ± / or / / is preceded by / j / or / r / there is no elision, e.g. insulate / ª nsj±le t /, soluble

Note
/ ªsÂlj±bl /, intricate / ª ntr k±t /etc.
2. If a pre-penultimate syllable of simple and derived words contains / / or / ± /, and

it is preceded by a syllable with the primary stress, / / and / ± / tend to be elided.
difficulty / ªd f klt / → / ªd fklt /

delicacy / ªdel k±s / → / ªdelk±s /
reasonable / ªriz±n±bl / → / ªrizn±bl /
honorary / ªÂn±r±r / → / ªÂnr±r /
If / ± / or / / is preceded by / j / or / r / there is no elision, e.g. insulating / ª nsj±le t Ì /,

Note
intricacy / ª ntr k±s /etc.
136
The Word
3. Words ending in / h±r / have a tendency to drop either / h /, / ± /, or / h± / in rapid

speech.
/ h±r / → / ±r / sanctuary / ªs Ìkjh±r / → / ªs Ìkj±r /

February / ªfebjh±r / → / ªfebj±r /
obituary / ±'b tjh±r / → / ±'b tj±r /
estuary / ªesjh±r / → / ªesj±r /
/ h±r / → / hr / January / ªdµ njh±r / → / ªdµ njhr /

February / ªfebrh±r / → / ªfebrhr /
estuary / ªesjh±r / → / ªesjhr /
statuary / ªst jh±r / → / ªst jhr /
/ h±r / → / r / sanctuary / ªs Ìkjh±r / → / ªs Ìjr /

mortuary / ªmjh±r / → / ªmjr /
4. The initial sequences / C + ±ª + l / and / C + ±ª + r / tend to drop the vowel / ± / in

rapid speech, and the primary stress is moved to the beginning of the word (C =
Consonant).
police / p±ªlis / → [ pªis ] → / plis /6

parade / p±ªre d / → [ pªe d ] → [ pre d ]6
terrific / t±ªr f k / → [ tª f k ] → / ªtr f k /6
collision / k±ªl µn / → [ kª µ ] → / ªkl µn /6
believe / b±ªliv / → [ bªiv ] → [ ªbliv ]
5. The word-initial sequence / Cc + ±ª + C / tends to drop the vowel / ± / in rapid

speech, and the primary stress is moved to the beginning of the word (C = Consonant,
Cc = Continuant Consonant).
phonetics / f±ªnet ks / → / ªfnet ks /

supporter / s±ªpt± / → / ªspt± /7
satiric / s±ªt r k / → / ªst r k /7
6
In words where the elision of / ± / results in the sequence / p, t, k / plus / r, l /, the devoicing of / r /
and / l / is rare, in comparison with similar sequences without elision, e.g. cologne - clone, terrain -
train, etc.
7
In words where the elision of / ± / results in the sequence / s / plus / p, t, k /, the aspiration of the
/ p, t, k / is not full, but it is usually audible. In such cases, this partial aspiration may be the only
thing (if we disregard the context) differentiating the meaning of such pairs of words as succumb -
scum, support - sport, etc.
137
The Word
veranda / v±ªr nd± / → / ªvr nd± /

thermometer / *±ªmÂm t± / → / ª*mÂm t± /
6. The central element or both the central and the last element of English triphthongs
tend to be elided. When both the central and the last elements are elided, the
remaining vowel is lengthened.
layer / le ± / → / le± / → / le /

joyous / dµ ±s / → / dµ ±s /8
fire / fa ± / → / fa± / → / fa /
power / pah± / → / pa± / → / pa /
mower / m±h± / → / m±± / → / m /
The elision of both the central and the last element of a triphthong is typical of
Conservative RP.
7. Other vowel elisions.
perhaps / p±ªh ps / → / ph ps /, / pr ps /
geography / dµ ªÂgr±f / → / ªdµÂgr±f /
geometry / dµ ªÂm±tr / → / ªdµÂm±tr /
Note
All words containing syllabic consonants are examples of elision, since syllabic consonants
are produced by elision of the preceding / ± /.
(B) Consonants
1. The word-initial sequence / l + C / tends to drop the consonant / l / in rapid

speech ( C = Consonant ).
/ l + C / → / + C /
already / lªred / → / (l)ªred / → / ªred /

alright / ©lªra t / → / ©(l)ªra t / → / ©ªra t /
although / lªs±h / → / (l)ªs±h / → / ªs±h /
always / ªlw±z / → / ª(l)w±z / → / ªw±z /
8
Complete elision of / / in / ± / is rare.
138
The Word
2. Word-medial and word-final consonant clusters / nt + C /, / nd + C /, / st + C / and

/ ft + C / tend to drop the consonants / t / and / d / in rapid speech ( C = Consonant ).
/ nt + C / → / n + C /
front-page / ©frCntªpe dµ / → / ©frCn(t)ªpe dµ / → / ©frCnªpe dµ / → / ©frCmªpe dµ /9

countdown / ªkahntdahn / → / ªkahn(t)dahn / → / ªkahndahn /
eventful / ªventfl / → / ªven(t)fl / → / ªvenfl /
point-blank / ©p ntªbl Ìk / → / ©p n(t)ªbl Ìk / → / ©p nªbl Ìk / →
/ ©p mªbl Ìk /8
Note
In the sequence / nt + r /, the / t / is elided only in compounds where the first element ends
with the sequence / nt / and the second element starts with the phoneme / r /, e.g. front-
runner / ªfrCnt©rCn± / → / ªfrCn©rCn± /. All other words containing the sequence / nt + r / are
pronounced without the elision of / t /, e.g. country / ªkCntr /, control / k±nªtr±hl /,
downtrodden / ªdahn©trÂdn /, etc.
/ nd + C / → / n + C /
windmill / ªw ndm l / → / ªw n(d)m l / → / ªw nm l / → / ªw (m)m l / or / ªw m l /

landlord / ªl ndld / → / ªl n(d)ld / or / ªl nld /
standstill / ªst ndst l / → / ªst n(d)st l / or / ªst nst l /
landscape / ªl ndske p / → / ªl n(d)ske p / or / ªl nske p /
Note
In the sequence / nd + r /, the / d / is elided only in compounds where the first element ends
with the sequence / nd / and the second element starts with the phoneme / r /, e.g. fundraiser
/ ªfCnd©re z± / → / ªfCn©re z± /. In simple and derived words, the elision of / d / in the
unstressed sequence / nd + r / is rare, e.g. hundred / ªhCndr±d / → / hCnr±d /, foundry
/ fahndr / → / fahnr /, etc. All other words containing the sequence / nd + r / are
pronounced without the elision of / d /, e.g. raindrop / ªre ndrÂp /, sundress / ªsCndres /,
etc.
/ st + C / → / s + C / mostly / ªm±hstl / → / ªm±hs(t)l / → / ªm±hsl /

restless / ªrestl±s / → / ªres(t)l±s / → / ªresl±s /
wristwatch / ªr stwÂj / → / ªr s(t)wÂj / → / ªr swÂj /
dustman / ªdCstm±n / → / ªdCs(t)m±n / → / ªdCsm±n /
9
When the sequence / nd / or / nt / is followed by any of the consonants / p, m, b, k, g /,
simultaneous elision and assimilation usually takes place. For more detail, see 3.8.2.3.
139
The Word
Word-final sequence / sts / is reduced to / s() /. The length of the remaining / s() /
depends on the tempo of speech, and it can vary from / s / through / s; / to / s /.
/ sts / → / s() / or / s(s) / → / s /
guests / gests / → / ges() / or / ge(s)s / → / ges /

lists / l sts / → / l s() / or / l (s)s / → / l s /
adjusts / ±ªdµCsts / → / ±ªdµCs() / or / ±ªdµC(s)s / → / ±ªdµCs /
texts / teksts / → / teks() / or / tek(s)s / → / teks /
Note
In the sequence / st + r /, the / t / is elided only in compounds where the first element ends
with the sequence / st / and the second element starts with the phoneme / r /, e.g. rest room
/ ªrestrum / → / ªresrum /. All other words containing the sequence / st + r / are
pronounced without the elision of / t /, e.g. abstract / ª bstr kt /, keystroke / ªkistr±hk /,
pastry / ªpe str /, etc.
/ ft + C / → / f + C / softball / ªsÂftbl / → / ªsÂf(t)bl / → / ªsÂfbl /

thriftshop / ª*r ftÂp / → / ª*r f(t)Âp / → / ª*r fÂp /
softly / ªsÂftl / → / ªsÂf(t)l / → / ªsÂfl /
swiftness / ªsw ftn±s / → / ªsw f(t)n±s / → / ªsw fn±s /
Note
In the sequence / ft + r /, the / t / is elided only in compounds where the first element ends
with the sequence / ft / and the second element starts with the phoneme / r /, e.g. soft rot
/ ªsÂftrÂt / → / ªsÂfrÂt /. All other words containing the sequence / ft + r / are pronounced
without the elision of / t /, e.g. half-truth / ©hfªtru* /, AFTRA / ª ftr± /, etc.
3. Word-final consonant clusters (groups of two or more consonants) tend to be

reduced in rapid speech. Here are some examples.
/ *s / → / s / months / mCn*s / → / mCns /

twelfths / twelf*s / → / twelfs /
fifths / f f*s / → / f fs /
cloths / klÂ*s / → / klÂs /
/ sz / → / z / clothes / kl±hsz / → / kl±hz /

bathes / be sz / → / be z /
loathes / l±hsz / → / l±hz /
breathes / brisz / → / briz /
140
The Word
/ kts / → / ks / facts / f kts / → / f ks /

acts / kts / → / ks /
mulcts / mClkts / → / mClks /
products / ªprÂdCkts / → / ªprÂdCks /
/ pts / → / ps / prompts / prÂmpts / → / prÂmps /

attempts / ±ªtempts / → / ±ªtemps /
corrupts / k±ªrCpts / → / k±ªrCps /
scripts / skr pts / → / skr ps /
/ ktl / → / kl / exactly / gªz ktl / → / gªz kl /

tactless / ªt ktl±s / → / ªt kl±s /
perfectly / ªpf ktl / → / ªpf kl /
directly / d ªrektl / → / d ªrekl /
/ skt / → / st / asked / skt / → / st /

masked / mskt / → / mst /
basked / bskt / → / bst /
frisked / fr skt / → / fr st /
/ mps / → / ms / glimpsed / gl mpst / → / gl mst /

bumps / bCmps / → / bCms /
mumps / mCmps / → / mCms /
dumps / dCmps / → / dCms /
/ mpt / → / mt / kempt / kempt / → / kemt /

attempt / ±ªtempt / → / ±ªtemt /
prompt / prÂmpt / → / prÂmt /
empty / ªempt / → / ªemt /
/ ks* / → / ks / sixth / s ks* / → / s ks /

sixths / s ks*s / → / s ks() / or / s k(s)s /
sixthly / s ks*l / → / ªs ksl /
/ lf* / → / l* / twelfth / twelf* / → / twel* /
(C) Elision of both vowels and consonants.
1. The word-final sequences / r±r / and / ±r±r / tend to be reduced to / r / in rapid

speech.
141
The Word
/ r±r / → / r / arbitrary / ªb tr±r / → / ªb tr /

contrary / ªkÂntr±r / → / ªkÂntr /
library / ªla br±r / → / ªla br /
/ ±r±r / → / r / literary / ªl t±r±r / → / ªl tr /

temporary / ªtemp±r±r / → / ªtempr /
itinerary / a ªt n±r±r / → / a ªt nr /
honorary / ªÂn±r±r / → / ªÂnr /
Note
Function words drop sounds in unstressed positions in sentences. For more information, see
4.1.
3.5 Pronunciation of Suffixes

3.5.1 Inflectional Suffixes
Inflectional suffixes express certain grammatical relationships, such as the
plural, the present participle, the past tense, the third person singular present tense,
the possessive case ( -s genitive ), etc. They do not change the word class of the stems
to which they are attached, e.g. work, works, working, worked, etc. Stem is "... the
part of the word-form which remains when all inflectional affixes have been removed"
(Bauer 1983: 20). In the word !chairs", for example, the stem is !chair". In the word
!wheelchairs", however, the stem is !wheelchair", even though it contains two roots.
The pronunciation of inflectional suffixes in English words follows certain
rules. In most cases, it is the final sound of the stem that influences the pronunciation
of the inflectional suffix. Here are the rules for the individual suffixes (Gimson &
Cruttenden, 1994; Quirk et al, 1992):
1. Inflectional suffix -ed.

This suffix is added to regular verbs to form the past tense. The pronunciation of
this suffix follows these rules:
(a) If the stem ends in / t / or / d /, we add / d /.10

Examples: bleat / blit / # bleated / ªblit d /
fade / fe d / # faded / ªfe d d /
10
In fast informal English, there is a tendency towards obscuration of the suffix / d / into / ±d /.
142
The Word
(b) If the stem ends in a vowel or a voiced consonant (except / d /), we add / d /.
Examples: free / fri / # freed / frid /
judge / dµCdµ / # judged / dµCdµd /
(c) If the stem ends in a voiceless consonant (except / t /), we add / t /.

Examples: hatch / h j / # hatched / h jt /
lick / l k / # licked / l kt /
2. Inflectional suffixes -s, or -!s.

The suffix -s is found in regular plurals and the third person singular present tense,
whereas the suffix -!s forms the possessive case (the -s genitive). The
pronunciation of these suffixes follows these rules:
(a) If the stem ends in a sibilant # / s, z, , µ, j, dµ /, we add / z /.11

Examples: place / ple s / # places / ªple s z /
George / dµdµ / # George"s / ªdµdµ z /
(b) If the stem ends in a vowel or any non-sibilant voiced consonant, we add / z /.
Examples: flow / fl±h / # flows / fl±hz /
John / dµÂn / # John"s / dµÂnz /
(c) If the stem ends in any non-sibilant voiceless consonant, we add / s /.

Examples: cough / kÂf / # coughs / kÂfs /
earth / * / # earth"s / *s /
(d) Some nouns which in the singular end in the final voiceless fricatives / * / and
/ f /, follow special rules in the plural.
1. Nouns ending in -th / * /.
The following nouns ending in -th can form the plural with the sequence / *s / or
/ sz /.
oath / ±h* / # oaths / ±h*s /, / ±hsz /

sheath / i* / # sheath / i*s /, / isz /
truth / tru* / # truths / tru*s /, / trusz /
wreath / ri* / # wreaths / ri*s /, / risz /
lath / l* / # laths / l*s /, / lsz /
11
In fast informal English, there is a tendency towards obscuration of the suffix / z / into / ±z /.
143
The Word
The following nouns ending in -th form the plural with the sequence / sz / only.
bath / b* / # baths / bsz / path / p* / # paths / psz /

mouth / mah* / # mouths / mahsz / youth / ju* / # youths / jusz /
2. Nouns ending in -f(e) / f /.
The following nouns ending in -f(e) can form the plural with the sequence / fs /
or / vz /.
dwarf / dwf / handkerchief / ªh Ìk±j f / hoof / huf /

dwarfs / dwfs / handkerchiefs / ªh Ìk±j fs / hoofs / hufs /
dwarves / dwvz / handkerchieves / ªh Ìk±j vz/ hooves / huvz /
scarf / skf / wharf / wf /

scarfs / skfs / wharfs / wfs /
scarves / skvz / wharves / wvz /
The following nouns ending in -f(e) form plural with the sequence / vz / only.
calf / kf / # calves / kvz / self / self / # selves / selvz /

elf / elf / # elves / elvz / sheaf / if / # sheaves / ivz /
half / hf / # halves / hvz / shelf / elf / # shelves / elvz /
knife / na f / # knives / na vz / thief / *if / # thieves / *ivz /
leaf / lif / # leaves / livz / wife / wa f / # wives / wa vz /
life / la f / # lives / la vz / wolf / whlf / # wolves / whlvz /
loaf / l±hf / # loaves / l±hvz /
(e) With regular plurals and some words ending / s / or / z /, the genitive is
expressed by apostrophe only. In such cases the pronunciation follows these
rules:
1. Fixed expressions with #for ... sake! are pronounced with zero suffix.
Examples: for goodness" sake / f±©ghdn±sªse k /
for conscience" sake / f±©kÂnnsªse k /
2. Regular plurals in the genitive are pronounced with zero suffix.

Examples: boys" / b z /
horses" / ªhs z /
144
The Word
3. Greek names of more than one syllable in the genitive are pronounced with
zero suffix.
Examples: Socrates" / ªsÂkr±tiz /
Xerxes" / ªzksiz /
4. Some names ending in the voiced sibilant / z / take / z / in the genitive.

Examples: Burns" / ªbnz z /
Dickens" / ªd k nz z /
3. Inflectional suffix -ing.

This suffix is added to English verbs to create continuous forms of verbs, verbal
nouns, or adjectives. The pronunciation of this suffix follows these rules:
(a) If the stem ends in a consonant, or any of the vowels / i, u, 12 /, or
diphthongs other than / ±, e±, h± /, we add / Ì /.
Examples: break / bre k / # breaking / ªbre k Ì /
draw / dr / # drawing / ªdr Ì /
(b) If the stem ends in any of the following / , ,13 , ±, ±, e±, h± /, we add
/ r Ì /.
Examples: bar / b / # barring / ªbr Ì /
roar / r / # roaring / ªrr Ì /
4. Inflectional suffixes -er and -est.

These suffixes express comparative and superlative degrees of some adjectives and
adverbs. Their pronunciation follows these rules:
(a) If the stem ends in any sound except / Ì /, we add / ± / or / st / (or / ±st /)
respectively.
Examples: fast / fst / # faster / ªfst± / # fastest / ªfst±st /
great / gre t / # greater / ªgre t± / # greatest / ªgre t±st /
(b) If the stem ends in / Ì /, we add / g± / or / g st / ( or / g±st / ) respectively.

Examples: strong / strÂÌ / # stronger / ªstrÂÌg± / # strongest / ªstrÂÌg±st /
long / lÂÌ / # longer / ªlÂÌg± / # longest / ªlÂÌg±st /
12
This only applies to the stems not ending in !-r" or !-re".
13
This only applies to the stems ending in !-r" or !-re".
145
The Word
3.5.2 Derivational Suffixes

Derivational suffixes are morphemes which change the word class of the roots
they are attached to. By adding a derivational suffix to a root, we create a new lexical
word (lexeme), e.g. love $ lovely, lonely $ loneliness, hope $ hopeful, etc. The
pronunciation of derivational suffixes does not follow the rules comparable with those
for inflectional suffixes. Nevertheless, since some derivational suffixes may pose
problems for English learners, in the following section we present the pronunciation
of the most problematic ones.
-able / ±bl / break / bre k / # breakable / \bre k±bl /

size / sa z / # sizeable / \sa z±bl /
-age / dµ / break / bre k / # breakage / \bre k dµ /

seep / sip / # seepage / \sip dµ /
-en / n / loose / lus / # loosen / \lusn /

/ ±n / fall / fl / # fallen / \fl±n /14
-ful / fl / faith / fe * / # faithful / \fe *fl / } adjectives

boast / b±hst / # boastful / \b±hstfl /
/ fhl / room / rum / # roomful / \rumfhl / } nouns

spoon / spun / # spoonful / \spunfhl /
-less / l±s / help / help / # helpless / \helpl±s /

harm / hm / # harmless / \hml±s /
-ment / m±nt / pay / pe / # payment / \pe m±nt /

treat / trit / # treatment / \tritm±nt /
-ness / n±s / lonely / ªl±hnl / # loneliness / \l±hnl n±s /

bold / b±hld / # boldness / \b±hldn±s /
14
rare (see also 3.2.3 # Syllabic [ ])
146
The Word
3.6 Word Stress

In the following sections we will attempt to define and explain the term !stress",
which has already been mentioned several times in this book. The phenomenon of
!stress" is in some literature referred to as !accent", and it basically carries the same
meaning. However, the term !accent" also refers to a particular way of pronunciation
(RP accent, American accent, Yorkshire accent, etc.). That is why we will not use the
term !accent" to denote the phenomenon described in the following sections.
3.6.1 The Nature and Definition of Stress

When we listen to English words, we are not only able to determine the number
of syllables in those words, but we are also able to determine which of the syllables
stand out, as it were, more clearly from the rest of the syllables making up particular
words. Syllables of this sort are called strong, and the other ones are called weak.
Thus, the first syllable of the word !letter" is strong, whereas the last syllable of this
word is said to be weak. In other words, the first syllable of the word !letter" carries
more sonority than the last one. In addition to the fact that the syllable !le" of the
word !letter" is more sonorous, it can also be made more distinct by applying such
factors as loudness, pitch and length. The combination of the four factors # quality,
quantity, loudness, and pitch will be called prominence. The syllable or syllables
which carry more prominence than others are called stressed syllables. Stress can
therefore be defined as "The force employed in uttering a syllable, giving it a certain
degree of prominence" (Kingdon 1965: ix). The stress occurring within a word is
called word stress. Word (lexical) stress is defined as "... the relative degree of force
used in pronouncing the different syllables of a word ..." (Kingdon 1965: 1). Instead
of the word force, the words strength, or emphasis, are sometimes used.
We must distinguish between stress and prominence. Stress is something we
consciously apply to particular syllables, whereas prominence is the result of stress.
Stress is an articulatory phenomenon, prominence is an acoustic one. When we stress
a syllable, we usually increase the loudness and the length of the vowel(s), or we
change the pitch of the vowel(s) in some way. Sonority of the sound(s) is something
which is not directly influenced by stress, because we cannot say that by applying
stress to the vowel / C / in the word adult for instance, we make that vowel more
sonorous, since sonority is an inherent quality of sounds and it does not depend on
stress. However, the syllable carrying stress always has a clear vowel as a nucleus,
and as we know, clear vowels usually have higher sonority than obscured vowels, so
the prominence of the syllable increases. In other words, stress seems to influence the
choice of the vowel carrying it, but not the vowel quality. Thus, stress can be
147
The Word
defined as a conscious prominence-increasing change of length, loudness, and

pitch of syllable(s), fulfilling certain lexical and grammatical functions.
The degree of prominence of syllables depends of four factors: pitch, loudness,

quality, and quantity. Stress is the process of the conscious change of the first three:
1. Pitch. The change of the pitch and the pitch height are the most important
factors in distinguishing the stressed syllables from the unstressed ones. If the
syllables of a word are uttered with equal loudness and one of them has either higher
pitch or a pitch change, this syllable will generally be heard as stressed. The stress
based on pitch movement is called kinetic stress, and the stress without pitch
movement is called static stress. As a rule, in words containing both primary and
secondary stress, the secondary stress is static and the primary stress is kinetic, e.g.
afternoon / ©ft±ªnun /:
afternoon / © − − ª /
For the sake of simplicity, we will not analyze words in this book in terms of kinetic
and static stress.
2. Loudness. Stressed syllables are often perceived to be louder, and in many cases
this may be so. The loudness depends on the type of the sound; voiced sounds, for
instance, have greater amplitude of vibration of the vocal folds and therefore higher
intensity. Loudness also depends on the sonority of sounds. The more sonorous a
sound is, the more likely it is to be heard as louder, even though, in absolute acoustic
terms, it may not be of higher intensity at all. The change of loudness is often
connected with the change of pitch, quality, and quantity.
3. Quantity. The length of syllables also plays its role in prominence. "Long
vowels and diphthongs are generally more prominent than short vowels, ..." (Gimson
& Cruttenden 1994: 203).
4. Quality. The quality of vowels is important for the prominence of syllables. By

looking at the sonority scale we can see that open vowels are more sonorous and
therefore more prominent than close vowels, because sonorous sounds increase the
subjective impression of loudness. Therefore the syllables containing close vowels
and the mid / ± / will be less prominent than the syllables with open vowels, although
pitch, loudness, and length may increase the prominence of close vowels.
148
The Word
3.6.2 Types of Word Stress

It is evident that a combination of the four factors (pitch, loudness, quantity,
and quality) produces a whole range of different degrees of syllable prominence.
From the practical point of view, however, it is sufficient to distinguish between:
(a) syllables with a high degree of prominence # syllables carrying primary stress
(b) syllables with a medium degree of prominence # syllables carrying secondary
stress
(c) syllables with a low degree of prominence # unstressed syllables
!
/d ©l b±ªre n/
Fig. 43. The size of the black marks expresses the degree of prominence. The smallest ones
stand for unstressed syllables, the medium one indicates secondary stress, and the biggest
one shows the primary stress.
Thus, we have established two new terms which will be used frequently from
now on # primary stress and secondary stress.
Primary (strong/main/principal) stress is the increase of loudness, length, and
variation of pitch of a syllable, which, together with the inherent sonority of the
syllable, gives the syllable a high degree of prominence. We indicate it by means of a
small vertical mark placed high up before the syllable it refers to # / bet± /, / ±bCv /.
If a word consists of one syllable only, we usually do not indicate it # crew / kru /.
Primary stress can be placed on any syllable in English words, and it is present in
every content (lexical) word.15 A word can only contain one primary stress.
Secondary (half-strong/medium) stress is the increase of loudness, length, and
variation of pitch of syllable(s), which, together with the inherent sonority of the
syllable(s), gives the syllable(s) a medium degree of prominence. In RP, the
secondary stress can either precede or follow the primary stress. We can find it in
words consisting of two and more syllables, and there are usually not more than two
secondary stresses in one word. It can be placed on any syllable. We indicate it by
means of a small vertical mark © placed low down before the syllable it refers to, e.g.
/ ©def ªn n /, / ©e v ªe n /.
15
Form (grammatical) words are stressed only under certain conditions, or when in their dictionary
form. For more information on the subject see 4.1.
149
The Word
The distribution of primary and secondary stress follows these rules:
1. Simple and derived words have no secondary stress if the primary stress is on
the first syllable, e.g. buttering / ªbCt±r Ì /, dangerous / ªde ndµ±r±s /.
2. Simple words and derived words beginning with a syllable not containing the
vowel / i / have no secondary stress if the primary stress is on the second syllable,
e.g. above / ±ªbCv /, content / k±nªtent /.
3. Simple and derived words with the primary stress on the second syllable have
the secondary stress on the first syllable, if the first syllable contains the vowel / i /,
e.g. redo / ©riªd /, premeditate /©priªmed±te t /.
4. Derived words that have the primary stress on the third syllable, always have
the secondary stress on the first syllable, e. g. possibility / ©pÂs±ªb l±t /,
artificial / ©t ªf l /.
5. Compounds that have the primary stress on the second element, always have the
secondary stress on the first element,16 e.g. free-range / ©friªre ndµ /, bad-tempered
/ ©b dªtemp±d /.
6. Compounds have no secondary stress if the primary stress is on the first
syllable, and the second element consists of one syllable only, e.g. airmail / ªe±me l /,
freephone / ªfrif±hn /.
7. Compounds and compound derivatives that have the primary stress on the first
syllable can have the secondary stress on the second element if the second element
consists of more than one syllable, e.g. freemason / ªfri©me sn /, typewriter
/ ªta p©ra t± /.
The stress pattern in English words can be considered fixed in the sense that it
always falls on the same syllable in a particular word (every English word has its own
fixed stress pattern),17 and it can also be considered free in the sense that the English
words can have stress put on any syllable # first, second, etc. (In Slovak, for instance,
the word stress always falls on the first syllable.) The freedom of the stress placement
in English words can be demonstrated in the following analysis.
The primary stress on the first syllable:
one syllable: top / tÂp /, bake / be k /, feel / fil /

two syllables: fighting / ªfa t Ì /, other / ªCs± /
16
Except when the stress is used for contrastive purposes, e.g. He was ªbad-©tempered, not ªgood-
©tempered.
17
The fixed stress pattern of some words can sometimes change under the influence of the
following word with the strong primary stress placed on the first or second syllable. This
phenomenon is called !shift of stress" and we will discuss it in later chapters.
150
The Word
three syllables: gathering / ªg s±r Ì /, family / ªf m±l /

four syllables: favourably / ªfe v±r±bl /, terminated / ªtm ne t d /
The primary stress on the second syllable:
two syllables: above / ±ªbCv /, inside / nªsa d /

three syllables: reporter / r ªpt± /, convulsion / k±nªvCln /
four syllables: atrocity / ±ªtrÂs±t /, auditioning / ªd n Ì /
five syllables: deliberating / d ªl b±re t Ì /, invalidated / nªv l±de t d /
The primary stress on the second syllable, the secondary stress on the first syllable:
two syllables: outwit / ©ahtªw t /, three-day / ©*riªde /

three syllables: half-hearted / ©hfªht d /, four-cornered / ©fªkn±d /
four syllables: half-heartedly / ©hfªht dl /, outdistancing / ©ahtªd st±ns Ì /
The primary stress on the third syllable, the secondary stress on the first syllable:
three syllables: overgrow / ©±hv±ªgr±h /, underpin / ©Cnd±ªp n /

four syllables: photographic / ©f±ht±ªgr f k /, definition / ©def ªn n /
five syllables: preternatural / ©pr¬t±ªn j±r±l /, representative / ©repr ªzent±t v /
six syllables: representatively / ©repr ªzent±t vl /
photographically / ©f±ht±ªgr f k±l /
The primary stress on the fourth syllable, the secondary stress on the first syllable:
five syllables: coeducation / ©k±hedµhªke n /, reincarnation / ©ri kªne n /

six syllables: unrepresentative / ©Cnrepr ªzent±t v /
disequilibrium / ©d sekw ªl br ±m /
seven syllables: unrepresentatively / ©Cnrepr ªzent±t vl /
photosynthetically / ©f±ht±s nª*et k±l /
The primary stress on the fourth syllable, the secondary stress on the second syllable:
five syllables: deliberation / d ©l b±ªre n /, congratulation / k±©gr j±ªle n /

six syllables: unpopularity / Cn©pÂpj±ªl r±t /, investigatory / n©vest ªge t±r /
seven syllables: uneconomically / Cn©ik±ªnÂm k±l /
ecclesiasticism / ©kliz ª st s zm /
151
The Word
The primary stress on the fifth syllable, the secondary stress on the second syllable:
six syllables: demobilization / di©m±hb±la ªze n /

demagnetization / di©m gn±ta ªze n /
seven syllables: invariability / n©ve±r ±ªb l±t /
impracticability / m©pr kt k±ªb l±t /
The primary stress on the fifth syllable, the secondary stresses on the first and third
syllable:
six syllables: decontamination / ©dik±n©t m ªne n /

maladministration / ©m l±d©m n ªstre n /
seven syllables: palaeontological / ©pe l ©Ânt±ªlÂdµ kl /
The primary stress on the sixth syllable, the secondary stress on the second syllable:
seven syllables: conceptualization / k±n©sepjh±la ªze n /
The primary stress on the sixth syllable, the secondary stresses on the first and fourth
syllable:
seven syllables: telecommunication / ©tel k±©mjun ªke n /

eight syllables: palaeoanthropology / ©pe l ±h© n*r±ªpÂl±dµ /
The primary stress on the seventh syllable, the secondary stress on the first syllable:
eight syllables: overcapitalization / ©±hv±k p tla ªze n /
3.6.3 The Influence of Stress on the Clarity of Vowels

The stress influences the clarity of vowels both directly and indirectly.
The direct influence of stress can be seen on the syllables carrying stress,
either primary or secondary. Such syllables always contain a clear, non-obscured
vowel, that is, any vowel other than / ± /, e.g. / , C, h, /, etc. The vowel / ± / can
therefore never form the nucleus of syllables carrying primary or secondary stress in
RP accent.
Examples: automatic / ©t±ªm t k /
capacity / k±ªp s±t /
egocentric / ©ig±hªsentr k /
152
The Word
The indirect influence of stress can be seen on the syllables preceding or

following the syllables carrying the primary stress. Here are two basic rules:
1. The syllable final / / and / h / followed by a syllable carrying the primary stress
tend to be obscured to / ± /, e.g. mistake / m ªste k / → / m±ªste k /, simulation
/ ©s mjhªle n / → / ©s mj±ªle n /, etc.
2. The vowels / / and / h / of the penultimate syllable following the syllable with
the primary stress tend to be obscured to / ± /, e.g. clarity / ªkl r t / → / ªkl r±t /,
deputy / ªdepjht / → / ªdepj±t /, etc.
3.6.4 Stress Pattern of the Word

One of the main problems that the learners of English are faced with, whether
at the beginning of their studies or at an advanced level, is the question of the
placement of stress within the word. We often ask ourselves which syllable or
syllables of a particular word should be stressed. As we have already mentioned and
shown, the stress pattern in English is free in the sense that the primary stress can be
placed on any syllable, unlike in some other languages, where it is usually placed on a
particular syllable of the word. In French, for instance, the last syllable is usually
stressed, in Polish, the penultimate syllable is stressed, and in Slovak, the first syllable
is stressed.
There have been several attempts to make up a set of rules by which the
placement of stress in English is governed. Unfortunately, almost every rule is subject
to numerous exceptions, which makes the matter extremely complicated. The
following rules should therefore be viewed as general tendencies rather than absolute
norms.
The rules governing the stress pattern in English words can be based on several
factors:
1. Whether the word is a simple (root) word, or complex (derived or compound)

word.
2. The grammatical category of the word.
3. The number of syllables in the word.
4. The quality of vowels in the syllables.
3.6.4.1 Stress in Simple Words

The term !simple word" is used to refer to words composed of a morphological
unit called !root". The root forms a part of every lexical word (lexeme). "A root is a
form which is not further analysable, either in terms of derivational or inflectional
153
The Word
morphology. It is that part of a word-form that remains when all inflectional and
derivational affixes have been removed. A root is the basic part always present in a
lexeme" (Bauer 1983: 20). In the word !waiters", for example, the root is !wait",
whereas !-er" is a derivational affix and !-s" is an inflectional affix. It is not always
easy to decide whether a particular word is a root word or a derived word. For the
sake of simplicity we will now disregard this fact, but we will return to the problem in
the next chapter.
In the following sections, the word !stress" is used to cover the term !primary
stress".
Two-syllable verbs
1. "... if the second syllable of the verb contains a long vowel or diphthong, or if it
ends with more than one consonant, ...the second syllable is stressed" (Roach 1991:
89).
apply / ±ªpla / accuse / ±ªkjuz /
react / r ª kt / decay / d ªke /
2. "If the final syllable contains a short vowel and one (or no) final consonant, the
first syllable is stressed" (Roach 1991: 89).
badger / ªb dµ± / carry / ªk r /

offer / ªÂf± / open / ª±hp±n /
3. Two-syllable verbs ending in / ±h / have the primary stress on the first syllable.
borrow / ªbÂr±h / follow / ªfÂl±h /

billow / ªb l±h / shadow / ª d±h /
Two-syllable adjectives
Two-syllable adjectives have the same rules as two-syllable verbs.
1. divine / d ªva n / robust / r±hªbCst /

direct / da ªrekt / serene / s±ªrin /
2. lonely / ªl±hnl / equal / ªikw±l /

lucky / ªlCk / certain / ¥stn /
3. shallow / ª l±h / hollow / ªhÂl±h /

mellow / ªmel±h / yellow / ªjel±h /
154
The Word
Two-syllable nouns
1. "... if the second syllable contains a short vowel the stress will usually come on
the first syllable" (Roach 1991: 89).
honey / ªhCn / table / ªte bl /

sofa / ªs±hf± / product / ªprÂdCkt /
2. If the second syllable contains a long vowel or a diphthong the stress will
usually come on the second syllable.
estate / ªste t / domain / d±ªme n /

baboon / b ªbun / design / d ªza n /
Three-syllable verbs
1. "... if the last syllable contains a short vowel and ends with not more than one
consonant, that syllable will be unstressed, and stress will be placed on the preceding
(penultimate) syllable" (Roach 1991: 90).
encounter / Ìªkahnt± / determine / d ªtm n /
2. "If the final syllable contains a long vowel or diphthong, or ends with more
than one consonant,. that final syllable will be stressed" (Roach 1991: 90).
entertain / ©ent±ªte n / resurrect / ©rez±ªrekt /
Three-syllable nouns
1. "... if the final syllable contains a short vowel or ±h, it is unstressed; if the
syllable preceding this final syllable contains a long vowel or diphthong, or if it ends
with more than one consonant, that middle syllable will be stressed" (Roach 1991:
90).
potato / p±ªte t±h / disaster / d ªzst± /
2. "If the final syllable contains a short vowel and the middle syllable contains a
short vowel and ends with not more than one consonant, both final and middle
syllables are unstressed and the first syllable is stressed" (Roach 1991: 90).
quantity / ªkwÂnt±t / emperor / ªemp±r± /
155
The Word
3. If the final syllable contains a long vowel or diphthong, and/or ends with more
than one consonant, the stress will usually be placed on the first syllable.
intellect / ª nt±lekt / stalactite / ªst l±kta t /
Three-syllable adjectives
If the final syllable contains a long vowel or diphthong, and/or ends with more than
one consonant, the stress will usually be placed on the first syllable.
opportune / ªÂp±tjun / insolent / ª ns±l±nt /
The above rules should serve as a very brief and general outline of stress
patterns of English root words. No exceptions are given for the simple reason that
there are so many of them, that their enumerating would serve no practical purpose.
3.6.4.2 Stress in Derived Words

We will now return to the discussion about root and derived words, which we
have already mentioned in the previous chapter. Let us first define the term !derived
word". Derived words are words consisting of one root in combination with one or
more derivational affixes. This definition works well with the native English words
like care → careless, happy → happily, etc. However, some words, mainly of Latin
and Greek origin, pose a bit of a problem for the aforementioned definition. Let us
take, for instance, the words permit, submit, and commit. It can easily be seen, that all
these words contain a prefix : per-, sub-, com-. The remaining part !-mit" is something
that in original English words would be called root. The unit !-mit", however, does
not function as a separate word in English, and therefore it would be unreasonable to
treat such words as derived words, because we would first have to study the
morphology of the language that these particular words came from, which would be
very impractical. Put differently, from the synchronic point of view, the words permit,
submit, commit, and the like are no longer considered as derived.
As we have already said, derived words consist of a root plus an affix. Affixes
are of two kinds in English: prefixes and suffixes. Prefixes are morphemes standing
before the root, e.g. possible # impossible, act # react, etc. Suffixes are morphemes
placed after the root, e.g. ill # illness, play # playful, etc.
156
The Word
Affixes can influence the stress of English words in these three ways:
1. The affix itself carries the primary stress, e.g. refuge / ªrefjudµ / → refugee
/ ©refjhªdµi /
2. The affix is neutral, it does not influence primary stress in any way, e.g. surprise
/ s±ªpra z / → surprising / s±ªpra z Ì /
3. The primary stress remains on the root, but it is moved to a different syllable, e.g.
alcohol / ª lk±hÂl / → alcoholic / © lk±ªhÂl k /
PREFIXES
Prefixes applied to roots in English can be either neutral, or both neutral and
stress-carrying. The same prefix can be neutral in some words, but stress-carrying in
others. Here are some examples:
Neutral prefixes
!de-" : code / k±hd / # decode / ©diªk±hd /

!dis-" : able / ªe bl / # disable / d sªe bl /
!il-" : legal / ªligl / # illegal / ªligl /
!im-" : possible / ªpÂs±bl / # impossible / mªpÂs±bl /
!in-" : formal / ªfml / # informal / nªfml /
!ir-" : regular / ªregj±l± / # irregular / ªregj±l± /
!mis-": lead / lid / # mislead / m sªlid /
!non-": verbal / ªvbl / # non-verbal / ©nÂnªvbl /
!pre-": arrange / ±ªre ndµ / # prearrange / ©pri±ªre ndµ /
!re-" : consider / k±nªs d± / # reconsider / ©rik±nªs d± /
!un-" : kind / ka nd / # unkind / Cnªka nd /
Both neutral and primary stress carrying prefixes
!anti-" : neutral : Catholic / ªk *l k / # anti-Catholic / © nt ªk *l k /

stress-carrying : hero / ªh ±r±h / # antihero / ª nt ©h ±r±h /
!auto-" : neutral : destruct / d ªstrCkt / # autodestruct / ©t±hd ªstrCkt /
stress-carrying : pilot / ªpa l±t / # auto-pilot / ªt±h©pa l±t /
!over-" : neutral : weight / we t / # overweight / ©±hv±ªwe t /
stress-carrying : time / ta m / # overtime / ª±hv±ta m /
157
The Word
!post-" : neutral : modern / ªmÂdn / # post-modern / ©p±hstªmÂdn /

stress-carrying : script / skr pt / # postscript / ªp±hstskr pt /
!ultra-" : neutral : sonic / ªsÂn k / # ultrasonic / ©Cltr±ªsÂn k /
stress-carrying : sound / sahnd / # ultrasound / ªCltr±sahnd /
SUFFIXES
From the vast number of suffixes found in English, we will only examine a
few, concentrating on those which are common and productive (capable of applying
to words and thus creating new English words). From the point of view of stress,
suffixes can be divided into neutral, and stress carrying.
Neutral suffixes
This type of suffixes does not affect the primary stress placement. Here are a few
examples:
!-able" : honour / ªÂn± / # honourable / ªÂn±r±bl /

!-acy" : celibate / ªsel±b±t / # celibacy / ªsel±b±s /
!-al" : deny / d ªna / # denial / d ªna ±l /
!-en" : deep / dip / # deepen / ªdip±n /
!-ful" : play / ple / # playful / ªple f /
!-ing" : sing / s Ì / # singing / ªs Ì Ì /
!-ish" : child / ja ld / # childish / ªja ld /
!-ism" : social / ªs±hl / # socialism / ªs±hl zm /
!-ist" : separate / ªsepr±t / # separatist / ªsepr±t st /
!-ize" : economy / ªkÂn±m / # economize / ªkÂn±ma z /
!-like" : snake / sne k / # snakelike / ªsne kla k /
!-less" : help / help / # helpless / ªhelpl±s /
!-ly" : love / lCv / # lovely / ªlCvl /
!-ment" : punish / ªpCn / # punishment / ªpCn m±nt /
!-ness" : lonely / ªl±hnl / # loneliness / ªl±hnl n±s /
!-ous" : fame / fe m / # famous / ªfe m±s /
!-fy" : false / fls, fÂls / # falsify / ªfls fa , ªfÂls fa /
!-ty" : difficult / ªd f klt / # difficulty / ªd f klt /
!-wise" : clock / klÂk / # clockwise/ ªklÂkwa z /
!-y" : dirt / dt / # dirty / ªdt /
158
The Word
Primary stress carrying suffixes
This type of suffix always carries the primary stress:
!-ain" (for verbs only) : certain / ªstn / # ascertain / © s±ªte n /

!-ee" : refuge / ªrefjudµ / # refugee / ©refjhªdµi/
!-eer" : mountain / ªmahnt n / # mountaineer / ©mahnt ªn ± /
!-ese" : Japan / dµ±ªp n / # Japanese / ©dµ p±ªniz /
!-ette" : usher / ªC± / # usherette / ©C±ªret /
!-esque" : picture / ªp kj± / # picturesque / ©p kj±ªresk /
!-ique" : critical / ªkr t kl / # critique / kr ªtik /
Stress-fixing suffixes
These suffixes influence the placement of the primary stress within the root.
1. Primary stress on the last syllable of the root:
!-eous" : advantage / ±dªvnt dµ / # advantageous / © dv±nªte dµ±s /

!-graphy" : photo / ªf±ht±h / # photography / f±ªtÂgr±f /
!-ial" : proverb / ªprÂvb / # proverbial / pr±ªvb ±l /
!-ic" : climate / ªkla m±t / # climatic / kla ªm t k /
!-ian" : music / ªmjuz k / # musician / mjuªz n /
!-ion" : perfect / ªpf kt / # perfection / p±ªfekn /
!-ious" : injure / ª ndµ± / # injurious / nªdµh±r ±s /
!-ity" : curious / ªkjh±r ±s / # curiosity / ©kjh±r ªÂs±t /
!-ive" : reflex / ªrifleks / # reflexive / r ªfleks v /
Words ending in -ic, -ian, and -ion, always have the primary stress on the penultimate
syllable.
2. Primary stress on the penultimate syllable of the root:
!-ate" : facility / f±ªs l±t / # facilitate / f±ªs l±te t /
Words ending in -ate usually have the primary stress on the pre-penultimate
syllable.
159
The Word
3.6.4.3 Stress in Compound Words

Compounds are lexemes composed of more than one root and they function as
a single word. They may be written as separate units, e.g. flower pot, they can be
hyphenated, e.g. tax-free, or they can be written as one word, e.g. bedroom. However,
there are no strict rules concerning this matter, so some compounds can be found
written in all three forms in different dictionaries. The placement of the primary and
secondary stress depends to a large extent on the word class that the compound
belongs to. The question is whether to put the primary stress on the first or on the
second element (root) of the compound. We will now attempt to present some basic
rules, which, however, should only be taken as tendencies, because there are a lot of
exceptions to most of the following rules.
1. Compounds functioning as nouns usually have the primary stress placed on the first
element.
car-ferry / ªk©fer / suitcase / ªsutke s /

ghetto blaster / ªget±h©blst± / pickpocket / ªp k©pÂk t /
showdown / ª±hdahn / set-up / ªsetCp /
Exceptions: apple pie / © plªpa /, cotton wool / ©kÂtnªwhl /, etc.
2. Compounds functioning as adjectives which have an -ed morpheme at the end

usually have the primary stress placed on the second element.
deep-seated / ©dipªsit d / bad-tempered / ©b dªtemp±d /

ham-fisted / ©h mªf st d / long-winded / ©lÂÌªw nd d /
half-timbered / ©hfªt mb±d / heavy-handed / ©hev ªh nd d /
Exceptions: time-honoured / ªta m©Ân±d /, hen-pecked / ªhenpekt /, etc.
3. Compounds functioning as adverbs usually have the primary stress placed on the
second element.
head-first / ©hedªfst / north-east / ©n*ªist /

downstream / ©dahnªstrim / head-on / ©hedªÂn /
4. Compounds functioning as verbs which have an adverbial first element usually

have the primary stress placed on the second element.
160
The Word
downgrade / ©dahnªgre d / ill-treat / © lªtrit /

back-fire / ©b kªfa ± / outwit / ©ahtªw t /
undergo / ©Cnd±ªg±h / download / ©dahnªl±hd /
5. Compounds in which the first element is a number in some form, usually have the
primary stress placed on the second element (unless they undergo shift of stress).
three-wheeler / ©*riªwil± / two-way / ©tuªwe /

second-class / ©sek±ndªkls / one-off / ©wCnªÂf /
four-footed / ©fªfht d / half-size / ©hfªsa z /
Exceptions: one-piece / ªwCnpis /, fourpenny / ªfpn /, six-pack / ªs ksp k /, etc.
The mutual relation of primary and secondary stress is governed by specific

rules which we have already specified in 3.6.2.
With compounds written as two separate words, problems may arise not only as
to whether they should be considered one word or two words, but also as to where the
primary stress should go. If we, for the sake of simplicity, put aside the question of
whether we are dealing with one single and stable unit or a combination of two
juxtaposed items, we are certain to find cases where the placement of the primary
stress brings about a change of meaning:
ªblackbird (drozd) ©black ªbird (þierny vták)

ªblackboard ($kolská tabuĐa) ©black ªboard (þierna tabuĐa)
ªEnglish ©teacher (uþiteĐ angliþtiny) ©English ªteacher (anglický uþiteĐ)
ªtoy ©factory (továreĖ na výrobu hraþiek) ©toy ªfactory (továreĖ ako hraþka)
3.6.5 Variation of Word Stress

There is a number of words in English which can have more than one stress
pattern. These words can be divided into two basic groups. The first group comprises
that type of words, whose different stress patterns change neither word class nor
meaning of those words. The other group consists of the so called word class pairs,
that is, words with identical spelling but different stress pattern and, of course,
different word class, e.g. !desert": noun = / dez±t /, verb = / d zt /. In both cases,
there can be additional changes in the phonemic make-up of the words.
161
The Word
3.6.5.1 Word-Class-Maintaining Variation of Stress

The following is a list of some of the words with double stress pattern, where the
differences in stress do not affect the word class.
adult / \ dClt /, / ±\dClt /

aristocrat / \ r st±kr t /, / ±\r st±kr t /
centrifugal / ]sentr \fjugl /, / sen\tr fj±gl /
centripetal / ]sentr \pitl /, / sen\tr p tl /
controversy / \kÂntr±vs /, / k±n\trÂv±s /
despicable / d \sp k±bl /, / \desp k±bl /
exquisite / k\skw z t /, / \ekskw z t /
inexplicable / ] n k\spl k±bl /, / n\ekspl k±bl /
integral / \ nt gr±l /, / n\tegr±l /
kilometre / \k l±]mit± /, / k \lÂm t± /
mirage / \m rµ /, / m±\rµ /
precedence / \pres dns /, / pr \sidns /
premature /\prem±jh±, \prem±j±, \prim±jh±, \prim±j± /, / ]prem±\jh± /
sonorous / \sÂn±r±s /, / s±\nr±s /
television / \tel v µn /, / ]tel \v µn /
The first stress pattern is more common in modern English.
There are also several compounds which have a different stress pattern in British
(RP) and American (GA) English:
RP stress pattern GA stress pattern
Adam"s apple / © d±mzª pl / / ª d±mz© pl /

peanut butter / ©pinCtªbCt± / / ªpinCt©bCt± /
half-time / ©hfªta m / / ªh fta m /
vocal cords / ©v±hklªkdz / / ªvohklkrdz /
season ticket / ªsizn©t k t / / ©siznªt k t /
weekend / ©wikªend / / ªwikend /
162
The Word
3.6.5.2 Word-Class-Changing Variation of Stress

Word stress in English can have a grammatical function. This phenomena can
be seen in the aforementioned word class pairs, which are words with identical
spelling but different stress pattern and word class. Most of these words are two-
syllable words of French origin. The phonemic make-up of the words is liable to
change under the influence of the stress. Here are the examples:
Noun Verb
abstract / ª bstr kt / / ±bªstr kt /

accent / ª ksnt / / ±kªsent /
attribute / ª tr±bjut / / ±ªtr bjut /
combine / ªkÂmba n / / k±mªba n /
compact / ªkÂmp kt / / k±mªp kt /
complement / ªkÂmpl±m±nt / / ©kÂmpl±ªment / also / ªkÂmpl±ment /
compliment / ªkÂmpl±m±nt / / ©kÂmpl±ªment / also / ªkÂmpl±ment /
compound / ªkÂmpahnd / / k±mªpahnd /
compress / ªkÂmpres / / k±mªpres /
concert / ªkÂns±t / / k±nªst /
conduct / ªkÂndCkt / / k±nªdCkt /
conflict / ªkÂnfl kt / / k±nªfl kt /
conscript / ªkÂnskr pt / / k±nªskr pt /
consort / ªkÂnst / / k±nªst /
contract / ªkÂntr kt / / k±nªtr kt /
contrast / ªkÂntrst / / k±nªtrst /
convict / ªkÂnv kt / / k±nªv kt /
desert / ªdez±t / / d ªzt /
digest / ªda dµest / / da ªdµest, d ªdµest /
discard / ªd skd / / d ªskd /
discount / ªd skahnt / / d ªskahnt /
discourse / ªd sks / / d ªsks /
escort / ªeskt / / ªskt /
export / ªekspt / / kªspt /
extract / ªekstr kt / / kªstr kt /
ferment / ªfment / / f±ªment /
import / ª mpt / / mªpt /
impress / ª mpres / / mªpres /
163
The Word
incline / ª nkla n / / nªkla n /

increase / ª nkris / / nªkris /
insult / ª nsYlt / / nªsYlt /
interchange / ª nt±je ndµ / / © nt±ªje ndµ /
misprint / ªm spr nt / / m ªspr nt /
object / ªÂbdµ kt / / ±bªdµekt /
overflow / ª±hv±fl±h / / ©±hv±ªfl±h /
perfume / ªpfjum / / p±ªfjum / also / ªpfjum /
permit / ªpm t / / p±ªm t /
pervert / ªpvt / / p±ªvt /
present / ªpreznt / / pr ªzent /
produce / ªprÂdjus / / pr±ªdjus /
progress / ªpr±hgres / / pr±ªgres /
project / ªprÂdµekt / / pr±ªdµekt /
protest / ªpr±htest / / pr±ªtest /
rebel / ªrebl / / r ªbel /
record / ªrekd / / r ªkd /
refill / ªrif l / / ©riªf l /
refund / ªrifYnd / / r ªfYnd /
refuse / ªrefjus / / r ªfjuz /
regress / ªrigres / / r ªgres /
reject / ªridµekt / / r ªdµekt /
reprimand / ªrepr mnd / / ©repr ªmnd / also / ªrepr mnd /
resit / ªris t / / ©riªs t /
segment / ªsegm±nt / / segªment /
subject / ªsCbdµ kt / / s±bªdµekt /
supplement / ªsCpl m±nt / / ©sCpl ªment / also / ªsCpl ment /
survey / ªsve / / s±ªve / also / ªsve /
suspect / ªsYspekt / / s±ªspekt /
torment / ªtment / / tªment /
transfer / ªtr nsf, ªtrnsf / / tr nsªf, trnsªf /
transplant / ªtr nsplnt, ªtrnsplnt / / tr nªsplnt, trnªsplnt /
transport / ªtr nspt, ªtrnspt / / tr nªspt, trnªspt /
164
The Word
Adjective Verb
absent / ª bsnt / / ±bªsent, bªsent /

abstract / ª bstr kt / / ±bªstr kt /
alternate / lªtn±t, Âlªtn±t / / ªlt±ne t, ªÂlt±ne t /
compound / ªkÂmpahnd / / k±mªpahnd /
frequent / ªfrikw±nt / / fr ªkwent /
present / ªpreznt / / pr ªzent /
Noun Adjective
compact / ªkÂmp kt / / k±mªp kt /

minute / ªm n t / / ma ªnjut /
3.7 Variation of Phonemic Structure of Words

There are cases in English when one word can be pronounced in two (or even
more) different ways, without any change in the stress pattern. Sometimes the
different pronunciations represent two different word classes, sometimes they are
mere variations of the same word.
3.7.1 Word-Class-Maintaining Variation of Phonemic

Structure
The following are lists of words in which the variation of phonemic structure
does not bring about any changes in the word class of the words.
1. Vowels
/ i / vs. / e / economical / ©ik±ªnÂm kl / vs. / ©ek±ªnÂm kl /

premature / ©prim±ªjh± / vs. / ©prem±ªjh± /
/ i / vs. / a / Argentine / ªdµntin / vs. / ªdµnta n /

iodine / ªa ±din / vs. / ªa ±da n /
/ / vs. / e / orchestra / ªk str± / vs. / ªkestr± /

alphabet / ª lf±bet / vs. / ª lf±b t /
165
The Word
/ / vs. / a / privacy / ªpr v±s / vs. / ªpra v±s /

dynasty / ªd n±st / vs. / ªda n±st /
/ / vs. / e / magistrate / ªm dµ str t / vs. / ªm dµ stre t /

holiday / ªhÂl±d / vs. / ªhÂl±de /
/ e / vs. / e / maintain / menªte n / vs. / me nªte n /

again / ±ªgen / vs. / ±ªge n /
/ / vs. / / transfer / ªtr nsf / vs. / ªtrnsf /

graph / gr f / vs. / grf /
/ / vs. / e / patriot / ªp tr ±t / vs. / ªpe tr ±t /

amoral / ªmÂr±l / vs. / e ªmÂr±l /
/ / vs. / e± / larynx / ªl r Ìks / vs. / ªle±r Ìks /

pharynx / ªf r Ìks / vs. / ªfe±r Ìks /
/ C / vs. / Â / constable / ªkCnst±bl / vs. / ªkÂnst±bl /

combat / ªkCmb t / vs. / ªkÂmb t /
/ C / vs. / ± / bankrupt / ªb ÌkrCpt / vs. / ªb Ìkr±pt /

dandruff / ªd ndrCf / vs. / ªd ndr±f /
/ Â / vs. / / false / fÂls / vs. / fls /

Australia / Âªstre l ± / vs. / ªstre l ± /
/ / vs. / h± / sure / / vs. / h± /

Europe / ªjr±p / vs. / ªjh±r±p /
/ h / vs. / u / room / rhm / vs. / rum /

groom / grhm / vs. / grum /
/ / vs. / ju / suit / sut / vs. / sjut /

super / ªsup± / vs. / ªsjup± /
/ e / vs. / / data / ªde t± / vs. / ªdt± /

esplanade / ©espl±ªne d / vs. / ©espl±ªnd /
/ ±h / vs. / ± / phonetics / f±hªnet ks / vs. / f±ªnet ks /

dislocate / ªd sl±hke t / vs. / ªd sl±ke t /
166
The Word
2. Consonants
/ j / vs. / tj / Christian / ªkr sj±n / vs. / ªkr stj±n /

actual / ª kjh±l / vs. / ª ktjh±l /
/ dµ / vs. / dj / educate / ªedµhke t / vs. / ªedjhke t /

verdure / ªvdµ± / vs. / ªvdj± /
/ dµ / vs. / µ / garage / ªg rdµ / vs. / ªg rµ /
/ nj / vs. / n / French / frenj / vs. / fren /

branch / brnj / vs. / brn /
/ Ìk / vs. / Ì / anxious / ª Ìk±s / vs. / ª Ì±s /

strength / streÌk* / vs. / streÌ* /
/ s / vs. / / associate / ±ªs±hs e t / vs. / ±ªs±h e t /
/ sj / vs. / / issue / ª sju / vs. / ª u /

sexual / ªseksjh±l / vs. / ªsekh±l /
/ / vs. / µ / Asia / ªe ± / vs. / ªe µ± /

magnesia / m gªni± / vs. / m gªniµ± /
/ zj / vs. / µ / azure / ª zjh± / vs. / ª µh±, ª µ± /
/ s / vs. / z / usage / ªjus dµ / vs. / ªjuz dµ /

unison / ªjun sn / vs. / ªjun zn /
/ f / vs. / p / diphthong / ªd f*ÂÌ / vs. / ªd p*ÂÌ /

naphtha / ªn f*± / vs. / ªn p*± /
3.7.2 Word-Class-Changing Variation of Phonemic

Structure
Words belonging to this category are relatively rare, and they can be divided
into two groups. The first group consists of word pairs identical in spelling and stress
pattern, but different in their phonemic structure. It is a sub-category of the so called
!homographs" (homographs also include words with different stress pattern, e.g.
minute, desert, etc.). Here are some examples:
167
The Word
abuse n. / ±ªbjus /, v. / ±ªbjuz /

associate n. / ±ªs±h ±t /, v. / ±ªs±h e t /
bow n. / b±h /, v. / bah /
deliberate adj. / d ªl br±t /, v. / d ªl b±re t /
estimate n. / ªest±m±t /, v. / ªest±me t /
excuse n. / kªskjus /, v. / kªskjuz /
house n. / hahs /, v. / hahz /
intimate adj. / ª nt±m±t /, v. / ª nt±me t /
lead n. / led /, v. / lid /
row n. / rah /, v. + n. / r±h /
separate adj. / sepr±t /, v. / sep±re t /
sow n. / sah /, v. / s±h /
tear n. / t ± /, v. / te± /
use n. / jus /, v. / juz /
wind n. / w nd /, v. / wa nd /
The second group consists of words pairs which differ graphically only in one
or two letters, have identical stress pattern and different phonemic structure. They
form word class pairs, and they were coined by the process of sound interchange.
Here are some examples:
Noun Verb
advice / ±dva s / advise / ±dva z /

bath / b* / bathe / be s /
blood / blCd / bleed / blid /
breath / bre* / breathe / bris /
food / fud / feed / fid /
gold / g±hld / gild / g ld /
life / la f / live / l v /
prophecy / prÂf±s / prophesy / prÂf±sa /
3.8 Assimilation within the Word

The quality of the individual phonemes of the word depends to a certain extent
on the type of the surrounding sounds and the style of the speaker. In informal style
and with increased tempo of speech, one or more sounds can change their acoustic
qualities. Sometimes the change is just a variation within the same phoneme, and only
a new subsidiary member of the same phoneme is produced. Sometimes the change is
168
The Word
so dramatic that a completely different phoneme is formed. With respect to these

facts, we speak of two separate phenomena # allophonic assimilation (similitude) and
phonemic assimilation. Assimilation within the word may also be termed intraword
assimilation.
3.8.1 Allophonic Assimilation within the Word

The term allophonic assimilation refers to an acoustic realisation of a sound in a
concrete phonetic context. It is the use of a particular allophone in a particular word,
where the choice of the allophone is conditioned by the phonetic environment of the
word. This phenomenon is discussed in Jones (1960) who, however, prefers the term
!similitude". He defines it as follows: "It often happens that a particular sequence of
two phonemes involves the use of a certain subsidiary member of one of them which
has a greater resemblance to a neighbouring sound than the principal member has.
In this case there is said to be similitude between that subsidiary member and the
neighbouring sound" (Jones 1960: 217). Thus, any allophonic change occurring
within the word caused by the presence of the neighbouring sound(s) will fall into
this group. Here are the most important categories:
1. Place of articulation
post-alveolar [ t ] in try [ tÚa ](influence of [ Ú ] )

dental [ } ] in eighth [ e }* ] (influence of / * / )
labio-dental [ P ] instead of / m / and / n / when followed by / f / or / v /:
comfort [ ªkhnPf±t ], conference [ ªkhÂPfÚ±ns ]
advanced (pre-velar) [ h ] in key [ hi ] (influence of / i / )
retracted [ \ ] in result [ r ªz\1t ] (influence of [ 1 ] )
velarized (dark) [ 1 ] in build [ b 1d ] (influence of the following consonant)
2. Voicing
Devoicing of / l, r, w, j / after an aspirated voiceless consonant / p, t, k /.
please [ pFiz ] cry [ kGa ] twit [ tz t ]

clean [ kFin ] pew [ pIu ] try [ tGa ]
Devoicing of voiced plosives, affricates, and fricatives in the word-final positions

when standing alone.
pub [ p\? ] smog [ smÂA ]

judge [ dµ\@µ ] above [ ±ªb\B ]
169
The Word
3. Lip rounding
Lip-spread Lip-rounded
[p] pea [ pw ] pool
w
[t] tea [t ] two
w
[k] key [k ] cool
w
[ m ] mean [ m ] moon
4. Nasalization
Vowels preceded or followed by nasals are often nasalized.

man [ m~n ] sunny [ ¥smn ]
and [ ~nd ] singer [ ¥sÌ± ]
3.8.2 Phonemic Assimilation within the Word

Phonemic assimilation, unlike allophonic assimilation, involves a change of
one phoneme into another. It is defined as "...the process of replacing a sound by
another sound under the influence of a third sound which is near to it in the word or
sentence" (Jones 1960: 217#218). It is a change of one or more phonemes into a
different phoneme (or phonemes) under the influence of the neighbouring sounds.
Phonemic assimilation may vary in its extent; it mainly depends on the tempo of
speech and the individual style of the speaker. It is most likely to occur in rapid,
casual speech, and it mostly affects consonants. It can be divided into several
categories.
3.8.2.1 Historical and Contextual Phonemic Assimilation

The basic division of assimilation is into historical and contextual assimilation.
Historical assimilation is defined as "...an assimilation which has taken place in the
course of development of a language, ... where ... a word which was once pronounced
in a certain way came to be pronounced subsequently in another way" (Jones 1960:
218).
Examples: amte / mt / → ant / nt / # The consonant / m / changed into / n /

under the influence of the following / t / ( both
/ n / and / t / have identical place of
articulation # alveolar).
170
The Word
picture / ªp ktjhr / → / ªp kj± / # The consonants / t / and / j / coalesced

into one: / j /.
grandeur / ªgr ndjhr / → / ªgr ndµ± / # The consonants / d / and / j /

coalesced into / dµ /.
nation [ ªnwsj±n ] → [ ªne ] # The consonants / s / and / j / coalesced

into one: / /, [ w ] changed into [e ].
Contextual (juxtapositional/synchronic) phonemic assimilation is an

assimilation which occurs in rapid casual speech where one or more phonemes
change into another phoneme (or phonemes) under the influence of the neighbouring
sound(s). This type of assimilation can occur both within the word and in the
sentence.
Examples: input / ª npht / → / ª mpht / # The consonant / n / changed into / m /

under the influence of / p /.
increase / nªkris / → / Ìªkris / # The consonant / n / changed into
/ Ì / under the influence of / k /.
All the following types of assimilation can be considered subtypes of the

contextual assimilation, because they all occur in rapid casual speech, whether within
words or at word boundaries.
3.8.2.2 Progressive (Perseverative) Phonemic Assimilation

In this type of assimilation a phoneme (or phonemes) changes into a different
phoneme (or phonemes) under the influence of the preceding sound. This type of
assimilation is not very frequent. Within a word, it can be found in all cases of !-s"
inflection added to the stem which ends in a voiced sound, and in the cases of syllabic
[ ] and [ ].
Examples:
dogs / dÂgz / # The consonant / s / changed into / z / under the influence of the
preceding / g /, which is a voiced sound.
seas / siz / # The consonant / s / changed into / z / under the influence of the
preceding / i /, which is a voiced sound.
171
The Word
happen [ ªh p ] → [ ªh p ] # The consonant [ ] assimilated into [ ] under

the influence of the preceding / p /.
thicken [ ª* k ] → [ ª* k ] # The consonant [ ] assimilated into [ ] under the

influence of the preceding / k /.
3.8.2.3 Regressive (Anticipatory) Phonemic Assimilation

In this type of assimilation a phoneme (or phonemes) changes into a different
phoneme (or phonemes) under the influence of the following sound. It can be found
frequently, both within the word and at word boundaries. Here are the individual
cases of regressive assimilation within the word:
1. The consonant / t / followed by any of the following consonants / p, b, m /

assimilates to / p /. The consonant / t / followed by any of the following
consonants / k, g / assimilates to / k /.
/ tp / → / pp / → / p / : footprint / ªfhtpr nt / → / ªfhppr nt / → / ªfhpr nt / or

[ ªfh=pr nt ]
/ tb / → / pb / : heartbreak / ªhtbre k / → / ªhpbre k / or [ ªh=bre k ]
/ tm / → / pm / : treatment / ªtritm±nt / → / ªtripm±nt / or [ ªtri=m±nt ]
/ tk / → / kk / → / k / : fruitcake / ªfrutke k / → / ªfrukke k / → / ªfruke k / or
[ ªfru=ke k ]
/ tg / → / kg / : shotgun / ªÂtgCn / → / ªÂkgCn / or [ ªÂ=gCn ]
In all of these cases, / t / assimilated to / p /, and / t / assimilated to / k / can be

superseded by [ = ]. In the cases of / tp / → / pp / and / tk / → / kk /, the elision of
two identical consonants occurs frequently: / ªfhpr nt /, / ªfruke k /.
2. The consonant / d / followed by any of the following consonants / p, b, m /

assimilates to / b /. The consonant / d / followed by any of the following
consonants / k, g / assimilates to / g /.
/ dp / → / bp / : tadpole / ªt dp±hl / → / ªt bp±hl /

/ db / → / bb / → / b / : goodbye / ghdªba / → / ghbªba / → / ghªba /
/ dm / → / bm / : headmaster / ©hedªmst± / → / ©hebªmst± /
/ dk / → / gk / : hardcover / ªhdkCv± / → / ªhgkCv± /
/ dg / → / gg / → / g / : mudguard / ªmCdgd / → / ªmCggd / →
/ ªmCgd /
172
The Word
In the case of / db / → / bb / and / dg / → / gg /, the assimilation can be followed by

elision of the two identical consonants: / ©ghªba /, / ªmCgd /.
3. The consonant / n / followed by any of the following consonants / p, b, m /

assimilates to / m /. The consonant / n / followed by any of the following
consonants / k, g / assimilates to / Ì /.
/ np / → / mp / : gunpoint / ªgCnp nt / → / ªgCmp nt /

/ nb / → / mb / : spoonbill / ªspunb l / → / ªspumb l /
/ nm / → / mm / → / m / : gunman / ªgCnm±n / → / ªgCmm±n / → / ªgCm±n /
/ nk / → / Ìk / : incredible / nªkred±bl / → / Ìªkred±bl /
/ ng / → / Ìg / : ingredient / nªgrid ±nt / → / Ìªgrid ±nt /
In the case of / nm / → / mm /, the assimilation can be followed by elision of one of

the two identical consonants: / ªgCm±n /. The real length of continuant consonants
depends on the tempo of speech. Thus, the possible transcription could be
/ ªgCm()±n /, or / ªgC(m)m±n /.
4. The consonant / s / followed by any of the following consonants / , j, dµ, j /

assimilates to / ∫ /.
/ s / → / / → / / : misshape / m sªe p / → / m ªe p / → / m ()ªe p /

/ sj / → / j / : crosscheck / ªkrÂsjek / → / ªkrÂjek /
/ sdµ / → / dµ / : misjudge / m sªdµCdµ / → / m ªdµCdµ /
/ sj / → / (j) / : tissue / ªt sju / → / ªt (j)u /
In the case of / s / → / /, the assimilation is followed by the elision of one of the

continuants. The real length of continuant consonants depends on the tempo of
speech. Thus, the possible transcription could be / m ()ªe p /, or / m ()ªe p /. / s /
followed by / j /, as found in the word !tissue", can be realized / ∫j / or / ∫ /. With
increased tempo of speech, the assimilation to / / is more frequent. In that case, the
assimilation is coalescent. For more on coalescent assimilation, see the following
chapter.
Note
The assimilated pronunciation of the words tissue / ªt / and issue / ª u / is nowadays
considered established (historical), and it is given in dictionaries as the first or the only
pronunciation, although some speakers still use the older non-assimilated form.
5. The consonant / z / followed by any of the following consonants / , j, dµ, j /

assimilates to / µ /. The number of words containing these sequences is rare.
173
The Word
/ z / → / µ / : quiz-show / ªkw z±h / → / ªkw µ±h /

/ zj / → / µj / : sailor's-choice / ªse l±zj s / → / ªse l±µj s /
/ zdµ / → / µdµ / : scissors-jack / ªs z±zdµ k / → / ªs z±µdµ k /
/ zj / → / µ(j) / : visual / ªv zjh±l / → / ªv µ(j)h±l /
/ z / followed by / j /, as in the word visual, can be realized / µj / or / µ /. With

increased tempo of speech, the assimilation to / µ / is more frequent. In that case, the
assimilation is coalescent. For more on coalescent assimilation, see the following
chapter.
The assimilated pronunciation of the words visual / ªv µh±l /, casual / ªk µh±l /, azure
Note
/ ª µ± /, etc. is nowadays considered established (historical), and it is given in dictionaries

as the first or the only pronunciation, although some speakers still use the older, non-
assimilated form.
6. The sequence of consonants / nt / followed by any of the following consonants

/ p, b, m / assimilates to / mp /. The assimilation is then followed by elision of
/ p /, leaving only / m /. (except the case of / ntm / → / mpm /, where / p / is
retained or superseded by [ = ].
The sequence of consonants / nt / followed by any of the following consonants
/ k, g / assimilates to / Ìk /. The assimilation is then followed by elision of / k /,
leaving only / Ì /.
/ ntp / → / mpp / → / mp / : front-page / ©frCntªpe dµ / → / ©frCmpªpe dµ / →

/ ©frCmªpe dµ /
/ ntm / → / mpm /, / m=m / : stuntman / ªstCntm n / → / ªstCmpm n /
or [ ªstCm=m n ]
/ ntb / → / mpb / → / mb / : paintbrush / ªpe ntbrC / → / ªpe mpbrC / →
/ ªpe mbrC /
/ ntk / → / Ìkk / → / Ìk / : brilliant-cut / ©br l ±ntªkCt / → / ©br l ±ÌkªkCt / →
/ ©br l ±ÌªkCt /
/ ntg / → / Ìkg / → / Ìg / : Montgomery / mÂntªgÂmr / → / mÂÌkªgÂmr / →
/ mÂÌªgÂmr /
7. The sequence of consonants / nd / followed by any of the following consonants

/ p, b, m / assimilates to / mb /. The assimilation is then followed by elision of
/ b /, leaving only / m /. (In the case of / ndm / → / mbm /, both / m / and the
following / b / are elided, leaving only one / m /.
The sequence of consonants / nd / followed by any of the following consonants
174
The Word
/ k, g / assimilates to / Ìg /. The assimilation is then followed by elision of / g /,

/ ndp / → / mbp / → / mp / : soundproof / ªsahndpruf / → / ªsahmbpruf / →

/ ªsahmpruf /
/ ndm / → / mbm / → / m()/ : landmark / ªl ndmk / → / ªl mbmk / →
/ ªl mk /
/ ndb / → / mbb / → / mb / : handbook / ªh ndbhk / → / ªh mbbhk / →
/ ªh mbhk /
/ ndk / → / Ìgk / → / Ìk / : handcuff / ªh ndkCf / → / ªh ÌgkCf / → /ªh ÌkCf /
/ ndg / → / Ìgg / → / Ìg / : endgame / ªendge m / → / ªeÌgge m / →
/ ªeÌge m /
In the assimilation / ndm / → / mbm / → / m /, the length of the remaining / m /
depends greatly on the tempo of speech. Thus, the possible transcription could be
/ ªl m()k /, or / ªl (m)mk /.
3.8.2.4 Coalescent (Reciprocal) Phonemic Assimilation

Coalescent assimilation is an assimilation where "... there is mutual influence,
or fusion, of the sounds upon each other..." (Crystal 1991: 28). It is a process of
coalescing of two sounds into one. We distinguish between two types of coalescent
assimilation: assibilation and transsibilation.
Assibilation is "... a specific kind of assimilation which results in acquiring a

sibilant pronunciation by the sound or sound-group which originally did not contain
it" (Vachek 1973: 65#66). Within the word, it normally occurs in unstressed
syllables, but there are tendencies in modern English # Estuary accent (see 5.1.1.3) #
towards assibilation in stressed syllables also. All assibilations in English are
becoming increasingly established (non-assimilated forms fall into disuse).
1. The consonant sequence / dj / tends to coalesce into / dµ /.
unstressed syllable : assiduous / ±ªs djh±s / → / ±ªs dµh±s /

stressed syllable : due / dju / → / dµu /
2. The consonant sequence / tj / tends to coalesce into / j /.
unstressed syllable : actual / ª ktjh±l / → / ª kjh±l /

stressed syllable : tune / tjun / → / jun /
175
The Word
Transsibilation is a kind of coalescent assimilation where a consonant

sequence containing a sibilant sound, / sj / or / zj /, is replaced by a new sibilant
sound, / / or / µ /. In common with assibilations, transsibilations in English are
becoming increasingly established (non-assimilated forms fall into disuse).
1. The consonant sequence / sj / tends to coalesce into / /.
tissue / ªt sju / → / ªt u /

issue / ª sju / → / ª u /
2. The consonant sequence / zj / tends to coalesce into / µ /.
visual / ªv zjh±l / → / ªv µh±l /

azure / ª zjh± / → / ª µh± / → / ª µ± /
3.9 Linking within the Word

Linking is a process known to occur mainly in connected speech, but it can also
be found within the word. It is a process of inserting a linking sound between two
vowels within the word in particular environments, and it is done to increase the ease
of articulation. It can be found within the word when two vowels meet at a syllable
boundary, more specifically, when one syllable ends with a vowel and the following
syllable begins with a vowel.
Depending on the type of vowels involved, we insert one of these three linking
sounds: / r /, / j /, or / w /.
Linking / r /
When a morpheme (bound or free) ends with any of the vowels / , , , ± /,
or diphthongs / e±, ±, h± /, and the last or the penultimate letter of that morpheme is
the letter !r" (except for morphemes ending in !-ra"), e.g. bore / b /, hear / h ± /, etc.,
the / r / link is inserted between this morpheme and the following morpheme (bound
or free), if the following morpheme begins with a vowel. This link is called linking
/ r /. Linking / r / is optional with words formed by prefixation and compounding,
because its omission will not interfere with the meaning of these words (although they
might sound unnatural). In these categories of words, linking / r / can be replaced by a
glottal stop. Words formed by suffixation, however, must keep linking / r /, because
the omission of linking / r / or its replacement by a glottal stop would ruin the
176
The Word
meaning of those words, e.g. the word !boring" / ªbr Ì / would hardly be
comprehensible when pronounced as / ªb Ì / or [ ªb= Ì ].
Prefix (bound morpheme) + root (free morpheme) = prefixation:
fore- + arm forearm / ªfrm /

hyper- + active hyperactive / ©ha p±rª kt v /
Root (free morpheme) + suffix (bound morpheme) = suffixation:
bore + -ing boring / ªbr Ì / fever + -ish feverish / ªfiv(±)r /

star + -y starry / ªstr / jeer + -ing jeering / ªdµ ±r Ì /
fur + -y furry / ªfr / scare + -y scar(e)y / ªske±r /
Root (free morpheme) + root (free morpheme) = compounding:
car + alarm car alarm / ªkr±©lm /

more + over moreover / mrª±hv± /
cover + up cover-up / ªkCv±rCp /
cure + all cure-all / ªkjh±rl, ªkjrl /
clear + out clearout / ªkl ±raht /
"By analogy, this / r / linking usage is extended to all / , , ± / endings, even
when there is no spelling justification" (Gimson & Cruttenden 1994: 262). Thus we
can formulate the following rule.
When a morpheme ends with any of the vowels / , , ± /, and the last or
the penultimate letter of that morpheme is not the letter !r" (except for morphemes
ending in !-ra"), e.g. draw / dr /, schwa / w /, contra / \kntr± /, etc., the / r / link
is sometimes inserted between this morpheme and the following morpheme, if the
morpheme begins with a vowel. This link is called intrusive / r /.
schwa + -ish schwaish / \wr /

draw + -ing drawing / drr Ì /
ultra + -ism ultraism / \Cltr±r zm /
contra- + indicate contraindicate / kntrr\ nd ke t /
The insertion of intrusive / r / before a suffix within the word is considered

incorrect and not acceptable in RP, although it may occasionally be heard with some
RP speakers.
177
The Word
Linking / w /
When a syllable within the word ends with the vowel / u /, or any of the
diphthongs / ±h / or / ah /, a slight / w / link is inserted between this syllable and the
following syllable, if the following syllable begins with a vowel.
Root (free morpheme):
ruin / ruw n /
suet / s(j)uw t /
Root + root (free morph.+free morph.) or root + affix (free morph.+bound morph.):
blue + -ish bluish / ªbluw / blue + eyed blue-eyed / ©bluwªa d /

snow + -y snowy / ªsn±hw / go + ahead go-ahead / ªg±hw±hed /
plough + ing ploughing / ªplahw Ì / now + days nowadays / ªnahw±de z /
co- + option co-option / k±hwªÂpn /
The insertion of linking / w / is something that happens automatically, when the

word is pronounced fluently. The fact that it is written as an indexed sound indicates
that it does not have the value of a fully pronounced / w /.
Linking / j /
When a syllable within the word ends with the vowels / / or / i /, or any of the
diphthongs / e /, / a / or / /, a slight / j / link is inserted between this syllable and
the following syllable, if the following syllable begins with a vowel.
Root (free morpheme):
aeon / ªij±n / CIA / ©sija jªe /

iodine / ªa j±din / kiosk / ªkijÂsk /
Root + root (free morph.+free morph.) or root + affix (free morph.+bound morph.):
re- + affirm reaffirm / ©rij±ªfm / agree + -able agreeable / ±ªgrij±bl /

pre- + occupy preoccupy / ©prijªÂkj±pa / fly + over flyover / ªfla j±hv± /
tri- + angle triangle / ªtra j Ìgl / lay + out layout / ªle jaht /
Linking / j /, just as linking / w

/, is inserted automatically, to ease the
pronunciation effort.
178
Connected Speech
4 Connected Speech
Connected speech is a term used to describe an utterance consisting of more

than one word. It is a continuous sequence of individual units (words), which we have
briefly analyzed in the previous chapters. The realization of words in connected
speech differs considerably from the realization of words as individual units. In the
following chapters we will attempt to describe the most important phenomena that
take place in connected speech.
4.1 Strong and Weak Forms

The phenomenon of strong and weak forms is closely connected with stress and
prominence. We have mentioned earlier that every syllable has its particular
prominence, that is, its relative distinctiveness, which depends on four factors:
quality, quantity, loudness, and pitch. The prominence of syllables in most English
words is relatively stable, as it is given by the stress pattern and phonemic make-up of
particular words. However, there are some words in English whose prominence can
differ depending on whether the word is pronounced as a separate unit, or whether it
is a part of connected speech. Almost all these words are the so-called !function
words".
Function (form/structural/grammatical/empty/synsemantic) words are mostly
monosyllabic words "... whose role is largely or wholly grammatical ..." (Crystal
1991: 142). They carry little or no lexical meaning. They include prepositions,
conjunctions, determiners, pronouns, auxiliary verbs, and some adverbs. When
pronounced separately and without relation to the rest of the elements of the sentence,
these words are stressed (they carry a medium or a high degree of prominence), and
they are said to be in their strong form. Such cases, however, are extremely rare,
because these words are usually used in sentences where they express certain
grammatical relations. When the function words are used in connected speech, they
are unstressed (they carry a low degree of prominence), and they are said to be in
their weak form.
The fact that the function words are reduced to their weak forms in connected
speech is a very important one, and everyone trying to learn English pronunciation
should be aware of the existence of this phenomena. There are at least two reasons
why. First of all, the correct use of weak forms in one"s pronunciation reduces the
unnatural foreign accent, which is something that most learners of English wish to
avoid. Secondly, ignorance of and unfamiliarity with the weak forms in English may
lead to difficulties in understanding the speakers who use them.
179
Connected Speech
During the change of the function word from its strong form to its weak form,
any of the following three phenomena, or a combination of them, can take place:
1. Reduction of length: to / tu / → / tu /

2. Obscuration of vowels: at / t / → / ±t /
3. Elision of sounds: him / h m / → / m /
There are a number of words which, although phonetically separate when said
in isolation, may be neutralized (they sound alike) in their weak form (Gimson &
Cruttenden, 1994):
of, her, are, a : / ± / The plays are poor. / s± ©ple z ± ªp /

He plays a poor man. / hi() ©ple z ± ªp ªm n /
She wants her dog. / i() ©wÂnts ± ªdÂg /
A friend of mine. / ± ªfrend ± ma n /
of, have : / ±v / The boys of Eton fish. / s± ªb z ±v ªitn ªf /

The boys have eaten fish. / s± ªb z ±v ªitn ªf /
or, are : / ± / Ten are under. / ©ten ±r ªCnd± /

Ten or under. / ©ten ±r ªCnd± /
The following is a list of the most common function words in their strong and weak
forms.
strong form weak form
a /e / /±/
am / m/ / ±m, /
an / n/ / ±n, /
and / nd / / ±nd, nd, ±n, /
are / / + consonant / ± / + consonant
/ r / + vowel / ±r, r / + vowel
as / z/ / ±z /
at / t/ / ±t /
be / bi / / bi() /
been / bin / / bi()n /
but / bCt / / b±t /
can (aux.) /k n/ / k±n, k, kÌ /
180
Connected Speech
could / khd / / k±d, kd /

do (aux.) / du / / d±, d / + consonant
/ du() / + vowel
does (aux.) / dCz / / d±z /
for / f / + consonant / f± / + consonant
/ fr / + vowel / f±r, f / + vowel
from / frÂm / / fr±m, fr /
had (aux.) 1 /h d/ / h±d, ±d, d /
has (aux.) 1 /h z/ / h±z, ±z, z, s /
have (aux.) 1 /h v/ / h±v, ±v, v /
he / hi / / hi(), i() /
her / h / / h±, , ± / + consonant
/ hr / / h±r, r, ±r / + vowel
him /hm/ / m/
his /hz/ / z/
is / z/ / s, z /
just / d(Cst / / d(±s / + consonant
/ d(±st / + vowel
me / mi / / mi() /
must / mCst / / m±s / + consonant
/ m±st / + vowel
my2 / ma / / m± / + consonant
/ mi / + vowel
of / Âv / / ±v, v, ± /
or3 / / + consonant / ± / + consonant
/ r / + vowel / ±r / + vowel
per / p / + consonant / p± /+ consonant
/ pr / + vowel / p±r / + vowel
Saint / se nt / / s±nt, snt, s±n, s /
shall / l/ / ±l, /
she / i / / i() /
should / hd / / ±d, d /
Sir / s / + consonant / s± / + consonant
/ sr / + vowel / s±r / + vowel
so 4 / s±h / / s± / + consonant
some (det.) 5 / sCm / / s±m, s /
than /' n/ / '±n, ' /
that (conj. and rel. pron.)6 /' t/ / '±t /
the / 'i / / '± / + consonant
/ 'i() / + vowel
them / 'em / / '±m, ±m, /
181
Connected Speech
there (indef. adv.)7 / 'e± / + consonant / '± / + consonant

/ 'e±r / + vowel / '±r / + vowel
this8 /ss/ / s±s /
to / tu / / t± / + consonant
/ tu() / + vowel
us / Cs / / ±s, s /
was / wÂz / / w±z /
we / wi / / wi() /
were / w / + consonant / w± / + consonant
/ wr / + vowel / w±r / + vowel
who / hu / / hu(), u() /
will /wl/ / ±l /, / /
would / whd / / w±d, ±d, d /
you9 / ju / / j± / + consonant
/ ju() / + vowel
your / j / + consonant / j± / + consonant
/ jr / + vowel / j±r / + vowel
1. There is a distinction between the auxiliary verb have (and all its forms), and the
main verb have. Here are some examples:
auxiliary verb main verb

I've seen it / a v ªsin t / I have a headache / a ©h v ± ªhede k /
He has come. / hi() ±z ªkCm / He has a brother. / hi() ©h z ± ªbrCs± /
She had left. / i() ±d ªleft / She had many friends / i() ©h d ªmen ªfrendz /
2. The word my can occasionally be used in its weak form. British English speakers
may have the pronunciation / mi / before a vowel, and / m± / before a consonant.
American English speakers also use / m± / before a consonant, but / mi / is not
acceptable. Here are some examples:
/ mi / + vowel / m± / + consonant
On my own. / ©Ân mi ª±hn / Yes, my dear? / ªjes m± ©d ± /
3. The weak form of !or" is relatively rare, occasionally found in phrases such as:
Two or three pounds / ©tu ± ©*ri ªpahndz /
4. The weak form of the word so is used only rarely, and only in casual speech before
adjectives and adverbs, if they begin with a consonant.
182
Connected Speech
Not so bad / ©nÂt s± ªb d /

Don't go so fast / ªd±hnt ©g±h s± ªfst /
So long# / s± ªlÂÌ /
5. When the word some functions as a determiner (it specifies the following noun), it
is usually reduced to / s±m / or [ s ]. It may usually be omitted without destroying
the structure of the sentence. When the word some functions as a pronoun (it is used
instead of a noun, which it represents), it is in its strong form / sCm /. The omission of
this type of some changes the meaning or destroys the structure of the sentence. When
some stands at the end of the sentence, it is always in its strong form $ / sCm /. Here
are some examples:
Some / s±m, sm / as a determiner ( trochu, zopár)
I need some apples. / a ªnid s±m ª plz /

Have some more wine. / ªh v s±m ©m ªwa n /
There's some ice in the fridge. / se±z s±m ªa s n s± ©fr dµ /
Some / sCm / as a pronoun (nejaký, niektorý)
Some say it was an accident. / ªsCm ©se t w±z ±n ª ks dnt /

Many houses were damaged and some were totally destroyed.
/ ªmen ©hahz z w± ªd m dµd ±n ªsCm w± ªt±htl d ªstr d /
Some of his jokes were very rude. / ªsCm ±v z ©dµ±hks w± ªver ªrud /
6. When the word that functions as a demonstrative pronoun or an adverb, the strong
form / s t / is used. If it functions as a relative pronoun or a conjunction, the weak
form / s±t / is used. Here are some examples:
Strong form / s t /
demonstrative pronoun : Who's that man? / ªhuz s t ªm n /

Where did that noise come from?
/ ªwe± d d s t ªn z ©kCm frÂm /
adverb : I can't walk that far. / a ªknt ©wk ªs t f /

It's about that long. / ts ±©baht ªs t ©lÂÌ /
183
Connected Speech
Weak form / s±t / (ktorý, þo, %e)
relative pronoun : The letter that came this morning is mine.

/ s± ªlet± s±t ªke m s±s ªmn Ì z ªma n /
All that I have is yours. / ªl s±t a ªh v z ªjz /
conjunction: He told me that he would come. / hi() ªt±hld mi() s±t i() w±d ªkCm /
It's possible that he didn't do it. / ts ªpÂs±bl s±t i() ªd dnt ©du() t /
7. When the word there functions as an adverb or an interjection, the strong form
/ se± / is used. When it functions as a pronoun, the weak form / s±, s±r / is used.
Strong form / se± /
adverb : The chance was there, but I didn't take it.

/ s± ªjns w±z ªse± b±t a ªd dn(t) ªte k t /
There you are. / ªse± ju() /
interjection : There# I've done it# / ªse±a v ªdCn t /
There, there, don't get so upset# / ªse± ©se± ªd±hnt ©get s±h Cpªset /
Weak form / s±, s±r /
pronoun : There must be some explanation# / s± ©mCs bi() ªsCm ©ekspl±ªne n /

There are two books on the table. / s±r± ªtu ªbhks Ân s± ©te bl /
This form can be found in relatively stable constructions of the following type (also in
the negative) :
there are / s±r± / there seems / s±ªsimz /
there is / s±r z / there must / s±ªmCst /
there'll be / s(±)lbi() / there appears / s±r±ªp ±z /
there was / s±w±z / there were / s±w± /
8. Some people use a weak form of the word this / s±s / in the constructions of the
following type:
this morning / s±s ªmn Ì /

this afternoon / s±s©ft±ªnun /
184
Connected Speech
9. In British English, the weak form / ju() / is used before vowels and the weak form
/ j± / is used before consonants. In American English, the weak form / j± / can be used
in both environments.
If you can. / © f j± ªk n /
You ought to go. / ju() ©(t) t± ªg±h /
FURTHER NOTES ON THE TRANSCRIPTION OF WEAK FORMS
With some form words there are several possible degrees of reduction, and we
often ask ourselves which of the possible forms should be used. In the transcription of
connected speech these rules apply:
1. The faster the tempo of speech, the greater the degree of reduction.
Examples: Boys and girls: slow speech $ / ªb z ±nd ªglz /
medium tempo $ / ªb z ±n ªglz /
fast speech $ / ªb z ªglz /
2. Regular alternation of vowels and consonants.

Our speech is made up of a combination of vowels and consonants. This combination
is not haphazard. On the contrary. There is a tendency towards regular alternation of
vowels and consonants. It is unnatural to find several vowels or consonants following
each other. Even if there are such cases, for instance a succession of several
consonants in the words like glimpse / gl mps /, asked / skt /, etc., some of the
consonants always tend to be elided, e.g. / gl ms / and / st / respectively. The least
articulatory effort is required when there is a regular alternation of one vowel and one
consonant, which can graphically be shown as:
... C + V + C + V ...
Therefore, the choice of the degree of reduction of some form words depends also on
the preceding or following sound:
Examples:
I want her to go. / a ªwÂnt ± t± ªg±h / $ the function word her has been reduced to
/ ± / because it is preceded by / t /. Thus, the regular alternation of consonant + vowel
+ consonant is ensured $ / ªwÂnt ± t± /.
Most of them. / ªm±hst ±v ±m / $ the form word them has been reduced to / ±m /
because it is preceded by / v /. The regular alternation of consonant + vowel +
consonant is ensured $ / ªm±hst ±v ±m /. Or, there can be another possibility:
185
Connected Speech
/ ªm±hst ± s±m / $ the form word of has been reduced to / ± / because it is followed
by / s /. The regular alternation of vowel + consonant + vowel is ensured / ªm±hst ±
s±m /
CASES OF NON-REDUCED AND PARTIALLY REDUCED FUNCTION

WORDS
As we have already mentioned, function words tend to occur in their reduced

(weak) form in connected speech. However, this rule is subject to several exceptions.
Here are the most important ones:
1. Most function words standing in the final position in the sentence keep their strong
form, even though they are not necessarily stressed.
Where do you come from? / ªwe± d± j± ªkCm frÂm /

What are you laughing at? / ªwÂt ± j± ªlf Ì t /
Yes, he does. / ªjes i() ªdCz /
There are a few exceptions. Function words her, him, he, them, and us can have a
weak form even when standing in the final position in the sentence:
Tell him. / ªtel m / Call her. / ªkl ± / Is he? / ª z i() /

Take them. / ªte k (s)±m / All of us. / ªl ±v ±s /
Function words be, do, me, she, to, who, and you retain their strong form at the end of
the sentence, but their length may be reduced:
I do. / a ¥du() / Tell me. / ªtel mi() / Where to? / ªwe± tu() /
You will be. / ju() ªw l bi() / Is she? / ª z i() / Says who? / ©sez ªhu() /
Do you? / ªdu ju() /
2. When we contrast or compare two or more pronouns, or two or more prepositions,

these pronouns and prepositions retain their strong form.
They invited me, not her. / se nªva t d ªmi nÂt ªh /

He travels to and from London a lot.
/ hi() ©tr vlz ªtu ±n ªfrÂm ©lCnd±n ± ªlÂt /
The letter was from him, not to him. / s± ªlet± w±z ªfrÂm m nÂt ªtu m /
A work of and about literature. / ± ©wk ªÂv ±n ±ªbaht ªl tr j± /
186
Connected Speech
3. Auxiliary verbs shall and will retain their strong form when standing at the
beginning of the sentence. The rest of auxiliary verbs may retain their strong form
when standing at the beginning of the sentence, although the weak form is more
common.
Shall we wait for him? / ª l wi() ªwe t f± h m /
Have you done it? / ©h v ju() ªdCn t /
Can I come too? / ©k n ªa ©kCm ªtu /
4. When a function word is being cited or quoted, the strong form is used:
You shouldn't put *and* at the end of a sentence.

/ ju() ªhdn(t) pht ª nd ±t si() ªend ±v ± ªsent±ns /
5. Function words do, into, my, so, the, to, and you followed by a word beginning
with a vowel retain a strong form, even though their length may be reduced.
Strong form Weak form
so do I / ªs±h du() ªa / so do they / ªs±h d± ªse /

get into a fight / ªget ntu() ± ªfa t / into the sky / © nt± s± ªska /
my aunt / ma ªnt / he's my friend / hi()z m± ªfrend /
so important / ©s±h mªptnt / so long / s± ªlÂÌ /
the apple / si() ª pl / the pear / s± ªpe± /
you ought / ju() ªt / if you can / © f j± ªk n /
6. When a function word is given stress for the purpose of emphasis, the strong form
is used.
You must give me more money. / ju() ªmCst ©g v mi() ©m ©mCn /
It was him. / t ªwÂz h m /
You should go there. / ju() ªhd ©g±h se± /
John and his mother went. (not just one of them) / ©dµÂn ª nd z ©mYs± ©went /
187
Connected Speech
4.2 Sentence Stress

Sentence stress (phrasal stress) is "... the relative degree of force given to the
different words in a sentence" (Kingdon 1965: 1). In common with word stress, it is a
conscious prominence-increasing change of loudness, length and pitch. The
overall degree of prominence is also influenced by the sonority of the sounds in the
syllable. Sentence stress is a stress distributed on different words within the sentence,
and it differs from the word stress, which does not concern units larger than words.
The placement of stress within the sentence depends on the relative importance
of the words of that sentence. The basic rule is that "The more important a word is,
the stronger is its stress" (Jones 1960: 262). The most important words are usually
content words. Content (lexical/full/notional/autosemantic) words are the "... words
which have stateable lexical meaning ..." (Crystal 1991: 78), and they are the
counterpart of function words, which we have described in the previous chapter. In
content words, lexical meaning prevails over their grammatical meaning. They
include nouns, verbs, adjectives, numerals, and some adverbs. The dividing line
between function and content words is not always clear-cut.
This leads us to the formulation of the basic principle of sentence stress:
Sentence stress is the stress usually placed on content words in the sentence.
The price is the thing that annoys me / s± ªpraΙs Ιz s± ª*ΙÌ s±t ±ªnΙz mi() /
We distinguish between primary sentence stress and secondary sentence stress.

Primary sentence stress falls on the content words which carry most importance. In
special cases, it may also fall on function words, and in that case the function words
are in their strong form. Secondary sentence stress usually falls on content words of
less importance, and it may also fall on function words in strong form. The words
carrying neither primary nor secondary stress are considered unstressed (mainly
function words).
Can I have a coffee and a cup of tea, please?

/ ©k n aΙ ©h v ± ªkÂfΙ ±n ± ©kCp ±(v) ªti ©pliz /
I"ve never been to a car rally. / ©aΙv ©nev± ªbin tu() ± ©k ©r lΙ /
Sometimes all content words in the sentence are equally important. In such a
case, all content words carry primary stress. This phenomenon is sometimes called
level stress or even stress.
John has just bought two large brown dogs.

/ ªdµÂn ±z ªdµCs ªbt ªtu ªldµ ªbrahn ªdÂgz /
188
Connected Speech
Here are some additional rules to placement of primary and secondary

stress within a sentence, as well as some exceptions to the basic principle of
sentence stress:
1. When we feel the need to emphasize a function word for the purpose of contrast,
we give the word the primary stress. This type of stress is sometimes called
overstress.
It is my book. / Ιt Ιz maΙ ªbhk /

It is my book. / Ιt Ιz ªmaΙ bhk /
It is my book. / Ιt ªΙz maΙ bhk /
Note
This phenomenon overlaps with the stress function of intonation. For more detail, see
4.9.3.2.
2. If we use a word which has already been used in the sentence (or in the previous
sentence), this word takes secondary stress, or no stress at all.
How many times have you been there? / ªhah menΙ ªtaΙmz ±v ju() ªbin se± /
Three times. / ª*ri ©taΙmz /
We think of that as a child thinks. / wi() ª*ΙÌk ±v ©s t ±z ± ªjaΙld ©*ΙÌks /

The boys shouted to the other boys. / s± ªbΙz ªahtΙd t± si() ªCs± ©bΙz /
Note
This phenomenon overlaps with the discourse function of intonation. For more detail, see
the respective chapter on the discourse function of intonation.
3. !Wh"-words standing at the beginning of a question carry primary or secondary

stress. They are seldom unstressed.
What color is the new carpet? / ªwÂt ªkCl±r Ιz s± ©nju ªkpΙt /

Which did he choose? / ªwΙj dΙdi() ªjuz /
Where do you come from? / ªwe± d± j± ªkCm frÂm /
4. The exclamatory what standing at the beginning of a sentence is not stressed.
What beautiful weather# / wÂt ªbjutfl ªwes± /

What a stupid idea# / wÂt ± ªstjupΙd aΙªdΙ± /
What a dreadful thing# / wÂt ± ªdredfl ª*ΙÌ /
189
Connected Speech
5. Auxiliary verbs in the negative usually carry primary stress.
I can!t tell you anything. / aΙ ªknt ©tel ju() ªenΙ*ΙÌ /

He won!t be able to come. / hi() ªw±hnt bi() ©eΙbl t± ªkCm /
Don!t you worry. / ªd±hnt ju() ªwCrΙ /
6. Auxiliary verbs in question tags always carry primary stress.

She told you, didn!t she? / i() ªt±hld ju() ªdΙdn(t) i() /
Yes it would, wouldn!t it? / ªjes Ιt ªwhd ªwhdnt Ιt /
He"s your brother, isn!t he? / hi()z j ªbrYs±r ªΙznt i() /
7. When the subject follows the verb, the verb generally carries secondary stress.
!Yes", said his father. / ªjes ©sed Ιz ªfs± /
After a storm comes a calm. / ªft±r ± ªstm ©kCmz ± ªkm /
8. Monosyllabic prepositions at the beginning of a sentence may carry primary or

secondary stress.
On your own. / ©Ân j±r ª±hn /

In the room they found a dog. / ©Ιn s± ªrum seΙ ©fahnd ± ªdÂg /
9. When a noun is preceded by a word which defines or describes the noun more
closely, the noun takes the primary stress and the preceding word usually takes the
secondary stress.
That"s a nice dog. / ªs ts ± ©naΙs ªdÂg /
His own car. / hΙz ©±hn ªkC /
10. When an intensifier, such as absolutely, horribly, completely etc., precedes

another word, both words usually take primary stress.
I"m terribly sorry. / aΙm ªter±blΙ ªsÂrΙ /
You're absolutely correct. / jr ª bs±lutlΙ k±ªrekt /
11. In phrasal verbs, the verbal element usually takes the secondary stress and the
adverbial element takes the primary stress.
Would you like to come in? / ªwhd ju() ªlaΙk t± ©kCm ªΙn /
Take it off# / ©teΙk Ιt ªÂf /
190
Connected Speech
12. In a casual conversational style, the number of unstressed syllables and syllables
with secondary stress tends to exceed the number of the syllables with primary stress.
Can I have a coffee and a cup of tea, please?

/ ©k n aΙ ©h v ± ªkÂfΙ ±n ± ©kCp ±(v) ªti ©pliz /
I"ve never been to a car rally. / ©aΙv ©nev± ªbin tu() ± ©k ©r lΙ /
Would you like another one? / whd ju() ©laΙk ±ªnCs± wCn /

It is important to realize one thing. Sentence stress depends greatly on the

context of the situation, which is why the same word can have different stress in
different contexts. Let us take a look at the phrasal verb take off in different contexts.
Take it off# / ©teΙk Ιt ªÂf / $ In short sentences containing only form words (except
for the phrasal verb itself), the phrasal verb keeps its
general stress pattern: / © ... ª ... /.
The plane took off. / s± ªpleΙn ©thk ªÂf / $ When the phrasal verb stands in the final
position in a sentence, it keeps its general stress pattern:
/ © ... ª ... /.
Take your clothes off. / ªteΙk j() ªkl±hsz ©Âf / $ When the phrasal verb is divided
by another important lexical word, its stress pattern
changes to / ª ... © ... /.
He ordered me to take my clothes off. / hi() ªd±d mi() t± ©teΙk maΙ ªkl±hsz ©Âf /
When the phrasal verb is preceded and followed by
important lexical words, its stress pattern may change to
/ © ... © ... /.
This simple demonstration should suffice to show that the rules given
concerning the placement of sentence stress are only generalizations of the much
more complex reality, where one must take into consideration such factors as the
number of words in the sentence, the type and position of words in the sentence, the
overall context of the situation, etc.
191
Connected Speech
4.3 Shift of Stress

We have shown in earlier chapters that stress in English is fixed, that is, every
word has its own stable stress pattern. With some words, however, this fixed stress
pattern is relative. It is mainly words where the primary stress is preceded either by
secondary stress, e.g. afternoon / ©ft±ªnun /, or by an unstressed syllable, e.g.
unknown / Cnªn±hn /. Two rules can be specified for these two categories:
1. When a word where the primary stress is preceded by the secondary stress, e.g.
afternoon / ©ft±ªnun /, is followed by a word with the primary stress on the first or
on the second syllable, the primary stress of the first word is shifted to the first
syllable and is reduced to secondary stress, e.g. afternoon tea / ©ft±nun ªti /.
2. When a word where the primary stress is preceded by an unstressed syllable,

e.g. / Cnªn±hn /, is followed by a word with the primary stress on the first or on the
second syllable, the primary stress of the first word is shifted to the first syllable and
is reduced to secondary stress, e.g. an unknown land / ±n ©Cn()±hn ªl nd /.
Here are some more examples of both types:
Normal stress Stress shift
close to Piccadilly / ªkl±hs t± ©p k±ªd l / Piccadilly Circus / ©p k±d l ªsk±s /

he's bad-tempered / hi()z ©b dªtemp±d / bad-tempered man / ©b dtemp±d ªm n /
vanilla ice cream / v±ªn l±r ©a sªkrim / ice cream soda / ©a skrim ªs±hd± /
she's Chinese / i()z ja ªniz / Chinese restaurant / ©ja niz ªrestrÂnt /
a royal princess / ± ªr ±l pr nªses / Princess Victoria / ©pr nses v kªtr ± /
Some dictionaries indicate the shift of stress, either by means of a backward-

pointed black arrow, or simply by stating stress shift, and sometimes giving an
example (Longman Dictionary of Contemporary English, English Pronouncing
Dictionary by D. Jones 15th edition, etc.).
192
Connected Speech
4.4 Rhythm
English speech is said to be rhythmical. It is often explained by the fact that
English belongs to the so called stress-timed languages. "In stress-timed languages,
it is claimed that the stressed syllables recur at regular intervals of time, regardless
of the number of intervening unstressed syllables ..." (Crystal 1991: 329). Thus, "... a
general rule of English rhythm is that we take an equal amount of time from one
stressed syllable to the next, i.e. that English rhythm has an isochrony based on
stresses" (Cruttenden 1991: 24). Let us consider the following example:
You ªput your ªbrakes on when the ªlight ªturns to ªred.
We have indicated the stressed syllables in words put, brakes, light, turns, and
red. According to the stress-timed theory of English rhythm, we should take an equal
amount of time from one stressed syllable to the next. Thus, in the following example,
we get five groups of words (divided by vertical lines) which should be pronounced
in equal amounts of time:
You put yourbrakes on when thelightturns tored.
It is evident, that the second block of words $ breaks on when the, which
contains four syllables, will have to be pronounced with much greater speed than the
third block $ light, which only contains one syllable. It would be absurd to believe
that in cases like this an exactly equal period of time separates the stressed syllables.
It is therefore safer to say that there is a tendency towards taking an approximately
equal period of time between one stressed syllable and the next.
Some linguists have further developed the theory concerning English rhythm
by inventing the term !rhythm unit". A rhythm unit (rhythmic group/foot/beat) is a
unit containing a stressed syllable plus all the following unstressed syllables up to
(but not including) the next stressed syllable.
Walkdown thepath to theend of the canal.
If the sentence starts with an unstressed syllable, this syllable is included with the first
rhythm unit.
You put yourbreaks on when thelightturns tored.
As we have noted, the regularity of the English rhythm is a very problematic

matter and we can find many instances contradicting the rhythm rule. At the same
time, there are cases where the strict rhythm is kept, even if that means breaking the
established stress pattern of the word. This happens, for instance, in counting, where
193
Connected Speech
the placement of stress and rhythm pattern thus created depends on the stress patterns
of the initial or preceding item(s):
ªone ªtwo ªthree ... ªthirteen, ªfourteen, ªfifteen ...
The numerals thirteen, fourteen, fifteen, etc. are normally stressed on the second
syllable,1 but here they are influenced by the stress pattern of the preceding items.
The strict regular rhythm can also be introduced in cases when " ... one is
implying repetition of something which ought to be accepted without argument, and
... perhaps when the speaker is expressing irritation or sarcasm" (Quirk et al 1992:
1043):
You should ªnever ªmove the ªpapers ªon my ªdesk
The influence of the overall rhythm of the utterance caused the preposition
!on", which is normally unstressed, to take the primary stress, just like the stressed
syllables in all preceding content words.
Another rule of English rhythm is the so-called borrowing rule. The

borrowing rule states that "... a syllable with a reduced vowel 'borrows time' from
any immediately preceding syllable containing a full vowel" (Gimson & Cruttenden
1994: 227). Reduced vowels are / ± /, which is always unstressed, and unstressed / /
and / h /. Stressed / / and / h / and all remaining vowels are considered full. Thus the
syllable of the word speed will be shortened in the word speedy, because the final
unstressed / /, which is a reduced vowel, borrowed time from the stressed initial
vowel / i /. Therefore, approximately the same time is taken to pronounce the words
speed and speedy. If we apply this rule to a sentence, we will get the following
results:
The plane was approaching the runway at high speed.

R F- R R F- R R F F- R F F
where F is a full vowel and lasts the longest

F- is a shortened full vowel
R is a reduced vowel and lasts the shortest
In this way we can show how the type of vowels in syllables influences the
length, and how length, in turn, influences the rhythm of the English sentence.
1
Numerals of this type, however, are subject to shift of stress.
194
Connected Speech
4.5 Linking in Connected Speech

Linking (liaison) is "A term used in phonology to refer to a sound which is
introduced between linguistic units, usually for ease of pronunciation" (Crystal 1991:
205). It is a process of joining two words together by means of a linking element
(sound). Linking occurs both within the word and in connected speech.
In connected speech, linking occurs where two vowels meet at word
boundaries. Let us examine the following scheme.
/ word ending with a vowel / + linking sound + / word beginning with a vowel /
Depending on the type of vowels involved, we insert one of these three linking
sounds $ / r /, / j /, or / w /.
Linking / r /
When a word ends with any of the vowels / , , , ± / or diphthongs / ±,
e±, h± /, and the last or the penultimate letter of that word is the letter !r" (except for
words ending in !-ra"), e.g. far / f /, here / h ± /, etc., the / r / link is inserted between
this word and the following word, if the following word begins with a vowel. This
link is called linking / r /. Whereas within the word it usually functions as an element
determining the meaning of the word, at word boundaries it can be dropped and
superseded by [ = ] without any change in meaning.
far away / ]fr ±ªwe / fir area / ªfr ]e±r ± /

other ideas / ªCs±r a ªd ±z / four hours / ]fr ªah±z /
"By analogy, this / r / linking usage is extended to all / , , ± / endings, even
when there is no spelling justification" (Gimson & Cruttenden 1994: 262). Thus we
can formulate the following rule:
When a word ends with any of the vowels / , , ± / or diphthong / ± /, and
the last or the penultimate letter of that word is not the letter !r" (except for words
ending in !-ra"), e.g. idea / a ªd ± /, raw / r /, Barbara / \bbr /, etc., the / r / link is
inserted between this word and the following word, if the following word begins with
a vowel. This link is called intrusive / r /.
the spa at Bath / s± ©spr ±t ªb* / Russia and China / ªrC±r ±n ªja n± /
raw onion / ©rr ªCnj±n / the idea is / si() a ªd ±r z /
Barbara is here / \bbrr z ]h /
195
Connected Speech
The usage of intrusive / r / after / ± / is widespread. The usage of intrusive / r /

after / / and / / is common too, although English speakers try to avoid using it.
Some English speakers and teachers still regard intrusive / r / as incorrect or
substandard, but no one can deny its frequent usage in rapid casual connected speech.
Linking / w /
When a word ends with the vowel / u / or any of the diphthongs / ±h / or / ah /,

a slight / w / link is inserted between this word and the following word, if the
following word begins with a vowel.
two eyes / ªtuwªa z /

so important / ©s±hw mªptnt /
how interesting / ©hahwª ntr±st Ì /
Linking / w / is inserted automatically, providing the two words are pronounced

fluently and without breaks, so no special drilling of this sound is necessary.
Linking / j /
When a word ends with any of the vowels / / or / i /, or any of the diphthongs
/ e /, / a /, or / /, a slight / j / link is inserted between this word and the following
word, if the following word begins with a vowel.
pretty eyes / ©pr t jªa z /

three hours / ©*rijªah±z /
day and night / ªde j±nªna t /
my arm / ©ma jªm /
boy and girl / ªb j±nªgl /
As in the previous case, linking / j / is produced automatically and requires no

extra practising.
196
Connected Speech
4.6 Elision in Connected Speech

In the sub-chapter 3.4, we have already talked about elision occurring within
the word, and now we will look at elision in connected speech. We have defined
elision as omitting (leaving out) one or more sounds to facilitate the ease and fluency
of pronunciation. Elisions are usually brought about by increased tempo of
pronunciation, specific sound environment and the style of the speaker. Elision in
connected speech is to be understood as elision occurring at word boundaries, that is,
interword elision. We distinguish between two types of elision $ historical and
contextual.
Historical elision only concerns the environment of the separate word, and it is
the omission of one or more sounds, which happened in the course of time and is now
firmly established, so the word is always pronounced with the elision, regardless of
the tempo and style of speech. We have covered this case in 3.4.
Contextual elision, on the other hand, only takes place in rapid casual
connected speech, and it is usually not present in slow and careful pronunciation. This
type of elision can be found both within the word, and in units bigger than one word.
Contextual elision happening within the word has been covered in 3.4, and we will
not concern ourselves with this type of elision here. Instead, we will focus on
contextual elision in connected speech.
ELISION OF VOWELS
1. When a word ending with any of the closing diphthongs / e , a , , ±h, ah / is

followed by a word beginning with / ± /, the final vowel of the diphthong tends to be
elided. The remaining vowel sequence ending in / ± / may further drop the / ± / and
the remaining vowel may be lengthened.
may as well / ©me ±zªwel / → / ©me±zªwel / → / ©me()zªwel /

try again / ©tra ±ªgen / → / ©tra±ªgen / → / ©tra()ªgen /
boy and girl / ªb ±nªgl / → / ªb nªgl / → / ªb()nªgl /
go away / ©g±h±ªwe / → / ©g±±ªwe / → / ©g{()ªwe /
how about / ªhah±©baht / → / ªha±©baht / → / ªha()©baht /
It is a process identical to that happening to triphthongs within words.
2. When a word ending with a consonant is followed by a word starting with a

sequence / ± / plus a continuant consonant, the / ± / in the sequence tends to be elided.
In such cases the continuant consonant often becomes syllabic.
197
Connected Speech
not alone / ©nÂt±ªl±hn / → / ©nÂtª±hn / → / ©nÂtªl±hn /

get another / ©get±ªnCs± / → / ©getªCs± / → / ©getªnCs± /
live among / ªl v±©mCÌ / → / ªl v©CÌ / → / ªl v©mCÌ /
will arrive / ©w l±ªra v / → / ©w lªa v / → / ©w lªra v /
3. Word-final / ± / may be elided if it is followed by linking [ r ]. If the word-final / ± /

is a part of a diphthong, the remaining vowel may be slightly lengthened. If the word-
final / ± / is a part of a triphthong, the triphthong can be monophthongized according
to the rules specified in earlier chapters.
father-in-law / ªfs±r nl / → / ªfsr nl /
for instance / f±rª nstns / → / ªfr nstns /2
here is the news / ªh ±r zs±ªnjuz / → / ªh ()r zs±ªnjuz /
fire away / ]fa ±r±\we / → / ]fa r±\we / → / ]far±\we /
ELISION OF CONSONANTS
1. When two identical consonants occur at word boundaries, one of them is usually
elided. The occurrence of two identical consonants standing together is called
gemination.
take care / ©te kªke± / → / ©te (k)ªke± / or / ©te ªke± /
hot tea / ©hÂtªti / → / ©hÂ(t)ªti / or / ©hÂªti / or [ ©hÂ=ªti ]
much cheese / ©mCjªjiz / → / ©mC(j)ªjiz / or / ©mCªjiz /
When these identical consonants occurring at word boundaries are continuants,

the elision may not be complete, and they may fuse into one longer continuant. The
length of the resulting continuant depends on the tempo of speech.
ten nights / ©tenªna ts / → / ©ten()ªa ts / or / ©te(n)ªna ts / or / ©teªna ts /
some men / s±mªmen / → / s±m()ªen / or / s±(m)ªmen / or / s±ªmen /
tough fight / ©tCfªfa t / → / ©tCf()ªa t / or / ©tC(f)ªfa t / or / ©tCªfa t /
2. When a word ending with a sequence: consonant plus / t / or / d / is followed by a

word beginning with a consonant (other than / j / or / h /), / t / and / d / in these
sequences tend to be elided. Graphically, this rule can be represented as follows:
/ C + t / + / C/ / → / C + C/ / or / C + d / + / C/ / → / C + C/ /
(C = consonant, C/ = any consonant except / j / and / h /)
2
The primary stress may be moved to the beginning of the word
198
Connected Speech
next day / ©nekstªde / → / ©neks(t)ªde / or / ©neksªde /

cold lunch / ©k±hldªlCnj / → / ©k±hlªlCnj / or / ©k±hl()ªCnj / or / ©k±hªlCnj /3
robbed both / ©rÂbdªb±h* / → / ©rÂbªb±h* / or / ©rÂ(b)ªb±h* / or / ©rÂªb±h* /3
helped me / ªhelptmi() / → / ªhelp(t)mi() / → / ªhelpmi() /
felt bad / ©feltªb d / → /©felpªb d / or /©fel=ªb d /4
left hand / ©leftªh nd /5
blond hair / ©blÂndªhe± /5
3. When two homorganic plosives or affricates differing only in the presence or

absence of voicing ( / t / $ / d /, / b / $ / p /, etc.) meet at word boundaries, the plosion
of the first one is incomplete, and the consonant may sometimes be completely elided.
The incomplete plosion may be phonetically expressed as [ _ ].
sit down / ©s tªdahn / → [ ©s t_ ªdahn ] or / ©s (t)ªdahn / or / ©s ªdahn /

good time / ©ghdªta m / → [ ©ghd_ ªta m ] or / ©gh(d)ªta m / or / ©ghªta m /
jump bail / ©dµCmpªbe l / → [ ©dµCmp_ ªbe l ] or / ©dµCm(p)ªbe l /
or / ©dµCmªbe l /
much jam / ©mCjªdµ m / → [ ©mCj_ ªdµ m ] or / ©mC(j) ªdµ m /
or / ©mCªdµ m /
4. All cases of contracted forms are examples of contextual elisions which came to be
used in written form.
he had $ he"d / hi()d /
they are $ they"re / se± /
5. The elision of sounds in the following collocations is frequent in colloquial speech,

but it is usually characterized as vulgar.
want to / ªwÂn± /
give me / ªg m /
let me / ªlem /
get me / ªgem /
I"m going to BrE / ©a mg±n±, ©a Ì±n± /,
AmE / ©amg±n±, ©am±n±, ©amohn±6 /
3
In cases like these, the simultaneous elision of both word final consonants may occur. This is a
combination of the rules 1. and 2.
4
The sequence / lt / + / C / is usually realized as / lp / + / C / or [ l= ] + / C /.
5
In the sequences / C + t / + / h / and / C + d / + / h /, the / t / and / d / are not elided.
6
Southern American accent
199
Connected Speech
4.7 Assimilation in Connected Speech

Assimilation is a term which "... refers to the influence exercised by one sound
segment upon the articulation of another, so that the sounds become more alike, or
identical" (Crystal 1991: 28). In the same way as phonemic and allophonic
assimilations may occur within a single word, these phenomena exist in
connected speech. Assimilation in connected speech is an assimilation occurring at
word boundaries, and it may also be termed interword assimilation.
4.7.1 Allophonic Assimilation in Connected Speech

In connected speech, allophonic assimilation involves a change of a phoneme
into one of its variants at word boundaries. It is a process identical to that occurring
within the word, but the sound environments may differ slightly. We distinguish these
categories:
1. Place of articulation
dental [ } ] in that thing [ ©s }ª* Ì ] (influence of / * / )

come for me [ ªkh\Pf±mi() ]
run fast[ ©Ú\Pªfst ]
advanced (pre-velar) [ h ] in take it [ ªte h t ] (influence of / / )
retracted [ s ] in this road [ ©s sªÚ±hd ] (influence of [ Ú ] )
2. Voicing
Devoicing of / l, r, w, j / after an aspirated voiceless consonant / p, t, k /. This

type of allophonic variation is "... restricted to cases where a sequence of words
forms a close-knit entity" (Gimson & Cruttenden 1994: 256).
at last [ ±ªtFst ] at once [ ±ªtZ\ns ]

at rest [ ±ªtGest ] thank you [ ª* ÌkIu() ]
Devoicing of voiced plosives, affricates, and fricatives when followed by a

voiceless consonant at word boundaries.
of course [ ±Bªkhs ] child support [ ©ja l@ s±ªpht ]

big shoe [ ©b Aªu ] his friend [ ©h Dªfrend ]
200
Connected Speech
3. Lip rounding
Any word-final consonant except / p, b, m / can be labialized, if it is followed

by an unstressed word beginning with / w /.
thick one [ ª* kww\n ] this way [ ªs swwe ]

long one [ ªlÂÌww\n ] that way [ ªs twwe ]
4. Nasalization
Vowels preceded or followed by nasals are often nasalized at word boundaries.
bring another [ ©bÚÌ ªnms± ] come along [ ©khmm ªlÌ ]

come in [ ©khmm ªn ] every man [ ©evÚ ªm~n ]
4.7.2 Phonemic Assimilation in Connected Speech

Phonemic assimilation is "...the process of replacing a sound by another sound
under the influence of a third sound which is near to it in the word or sentence"
(Jones 1960: 217$218). It is a change of one or more phonemes into a different
phoneme (or phonemes) under the influence of the neighbouring sounds. It mainly
depends on the tempo of speech and the individual style of the speaker. Phonemic
assimilation occurring at word boundaries can be divided into these types:
progressive, regressive and coalescent assimilation.
4.7.2.1 Progressive (Perseverative) Phonemic Assimilation

Progressive phonemic assimilation at word boundaries occurs when the initial
sound of a word changes to become like the final sound of the preceding word. It
involves a change of one phoneme into another. This type of assimilation is not very
frequent. It can be found in cases where a word ending in a voiceless consonant is
followed by auxiliaries has, does, and is.
it is / t z / → / ts /
Jack has / ªdµ k h±z / → / ªdµ ks /
What does it look like? / ©wÂt d±z t ªlhk la k / → / ©wÂts t ªlhk la k /
In all of these cases, the auxiliaries were reduced to their final phoneme / z /
which, in turn, assimilated into / s / under the influence of the preceding voiceless
201
Connected Speech
consonant. This phenomenon is also reflected orthographically by contracting the

auxiliaries in question: it is = it"s, Jack has = Jack"s, What does it look like? = What"s
it look like?
4.7.2.2 Regressive (Anticipatory) Phonemic Assimilation

Regressive phonemic assimilation at word boundaries occurs when the final
sound (or sounds) of a word change to become like the initial sound of the following
word. It involves a change of one phoneme into another. It can be found in these
environments.
1. The consonant / t / followed by any of the following consonants / p, b, m /

assimilates to / p /.
/ tp / → / pp / → / p / : that pen / ©s tªpen / → / ©s pªpen / → / ©s ªpen / or

[ ©s =ªpen ]
/ tb / → / pb / : that boy / ©s tªb / → / ©s pªb / or [ ©s =ªb ]
/ tm / → / pm / : that man / ©s tªm n / → / ©s pªm n / or [ ©s =ªm n]
2. The consonant / t / followed by any of the following consonants / k, g /

assimilates to / k /.
/ tk / → / kk / → / k / : that cup / ©s tªkCp / → / ©s kªkCp / → / ©s ªkCp / or

[ ©s =ªkCp ]
/ tg / → / kg / : that girl / ©s tªgl / → / ©s kªgl / or [ ©s =ªgl ]
In all of these cases, / t / assimilated to / p /, and / t / assimilated to / k / can be

superseded by [ = ]. In the cases of / tp / → / pp / and / tk / → / kk /, the elision of
two identical consonants occurs frequently: / ©s ªpen /, / ©s ªkCp /.
3. The consonant / d / followed by any of the following consonants / p, b, m /

assimilates to / b /.
/ dp / → / bp / : good pen / ©ghdªpen / → / ©ghbªpen /

/ db / → / bb / → / b / : good boy / ©ghdªb / → / ©ghbªb / → / ©ghªb /
/ dm / → / bm / : good man / ©ghdªm n / → / ©ghbªm n /
202
Connected Speech
4. The consonant / d / followed by any of the following consonants / k, g /

assimilates to / g /.
/ dk / → / gk / : good cup / ©ghdªkCp / → / ©ghgªkCp /

/ dg / → / gg / → / g / : good girl / ©ghdªgl / → / ©ghgªgl / → / ©ghªgl /
In the case of / db / → / bb / and / dg / → / gg /, the assimilation can be

followed by elision of the two identical consonants: / ©ghªb /, / ©ghªgl /.
5. The consonant / n / followed by any of the following consonants / p, b, m /

assimilates to / m /.
/ np / → / mp / : ten pens / ©tenªpenz / → / ©temªpenz /

/ nb / → / mb / : ten boys / ©tenªb z / → / ©temªb z /
/ nm / → / mm / → / m / : ten men / ©tenªmen / → / ©temªmen / → / ©teªmen /
In the case of / nm / → / mm /, the assimilation can be followed by elision of

the two identical consonants: / temen /. The real length of continuant consonants
depends on the tempo of speech. Thus, the possible transcription could be
/ ©tem()ªen /, or / ©te(m)ªmen /.
6. The consonant / n / followed by any of the following consonants / k, g /

assimilates to / Ì /.
/ nk / → / Ìk / : ten cups / ©tenªkCps / → / ©teÌªkCps /

/ ng / → / Ìg / : ten girls / ©tenªglz / → / ©teÌªglz /
7. The consonant / s / followed by any of the following consonants / , µ, j, dµ, j /

tends to assimilate to / /.
/ s / → / / → / / : this shop / ©s sªÂp / → / ©s ªÂp / → / ©s ªÂp /

/ sµ / → / µ / : this genre / ©s sªµÂnr / → / ©s ªµÂnr /
/ sj / → / j / : this change / ©s sªje ndµ / → / ©s ªje ndµ /
/ sdµ / → / dµ / : this jug / ©s sªdµCg / → / ©s ∫ªdµCg /
/ sj / → / (j) / : this year / ©s sª(j) ± / → / ©s ª(j) ± /
In the case of / s / → / /, the assimilation is followed by elision of one of

the continuants: / ©s ªÂp /. The real length of continuant consonants depends on the
tempo of speech. Thus, a possible transcription could be / ©s ()ªÂp /, or / ©s ()ªÂp /.
203
Connected Speech
/ s / followed by / j / can be realized / j / or / /. With increased tempo of speech, the

assimilation to / / is more frequent. In that case the assimilation is coalescent. For
more on coalescent assimilation, see the following chapter.
8. The consonant / z / followed by any of the following consonants / , µ, j, dµ, j /

tends to assimilate to / µ /.
/ z / → / µ / : his shop / h zªÂp / → / h µªÂp /

/ zµ / → / µµ / → / µ / : his genre / h zªµÂnr / → / h µªµÂnr / → / h ªµÂnr /
/ zj / → / µj / : his chin / h zªj n / → / h µªj n /
/ zdµ / → / µdµ / : his jug / h zªdµCg / → / h µªdµCg /
/ zj / → / µ(j) / : his yacht / h zªjÂt / → / h µ ª(j)Ât /
/ z / followed by / j / can be realized as / µj / or / µ /. With increased tempo of

speech, the assimilation to / µ / is more frequent. In that case the assimilation is
coalescent. For more on coalescent assimilation, see the following chapter.

/ p, b, m / assimilates to / mp /. The assimilation is then followed by elision of
/ p /, leaving only / m /.
/ ntp / → / mpp / → / mp / or [ m=p ] : don"t pass / ©d±hntªps /

→ / ©d±hmpªps / → / ©d±hmªps / or [©d±hm=ªps ]
/ ntm / → / mpm / or [ m=m ] : don"t miss / ©d±hntªm s / → / ©d±hmpªm s /
→ / ©d±h(m)ªm s / or [ ©d±hm=ªm s ]
/ ntb / → / mpb / → / mb / or [ n=b ] : don"t be / ªd±hntbi() / → / ªd±hmpbi() /
→ / ªd±hmbi() / or [ ªd±hm=bi() ]

/ k, g / assimilates to / Ìk /. The assimilation is then followed by elision of / k /,
/ ntk / → / Ìkk / → / Ìk / : don't come / ©d±hntªkCm / → / ©d±hÌkªkCm / →

/ ©d±hÌªkCm /
/ ntg / → / Ìkg / → / Ìg / : don't go / ©d±hntªg±h / → / ©d±hÌkªg±h / →
/ ©d±hÌªg±h /
204
Connected Speech

/ p, b, m / assimilates to / mb /. The assimilation is then followed by elision of
/ b /, leaving only / m /. (In the case of / ndm / → / mbm /, both / m / and the
following / b / are elided, leaving only one / m /.
/ ndp / → / mbp / → / mp / : found peace / ©fahndªpis / → / ©fahmbªpis / →

/ ©fahmªpis /
/ ndb / → / mbb / → / mb / : found both / ©fahndªb±h* / → / ©fahmbªb±h* / →
/ ©fahmªb±h* /
/ ndm / → / mbm / → / m / : found me / ªfahndmi() / → / ªfahmbmi() / →
/ ªfahmi() /
In case of assimilation / ndm / → / mbm / → / m /, the length of the remaining

/ m / depends greatly on the tempo of speech. Thus, the possible transcription could
be / ªfah(m)mi() / or / ªfahm()i() /.

/ k, g / assimilates to / Ìg /. The assimilation is then followed by elision of / g /,
/ ndk / → / Ìgk / → / Ìk / : found Kim / ©fahndªk m / → / ©fahÌgªk m / →

/ ©fahÌªk m /
/ ndg / → / Ìgg / → / Ìg / : found gold / ©fahndªg±hld / → / ©fahÌgªg±hld / →
/ ©fahÌªg±hld /
4.7.2.3 Coalescent (Reciprocal) Assimilation

Coalescent assimilation is an assimilation where "...there is mutual influence,
or fusion, of the sounds upon each other..." (Crystal 1991: 28). At word boundaries, it
is a process of coalescing of the final and the initial consonants of two successive
words. We distinguish between two types of coalescent assimilation: assibilation and
transsibilation.
Assibilation is "... a specific kind of assimilation which results in acquiring a

sibilant pronunciation by the sound or sound-group which originally did not contain
it" (Vachek 1973: 65$66).
205
Connected Speech
1. The consonant sequence / dj / tends to coalesce into / dµ /.
did you / ªd dju() / → / ªd dµu() /

bad youth / ©b dªju* / → / ©b ªdµu* / or / ©b dµªu* /
2. The consonant sequence / tj / tends to coalesce into / j /.
get you / ªgetju() / → / ªgeju() /

not yet / ©nÂtªjet / → / ©nÂªjet / or / ©nÂjªet /
In all these word groups it is possible to supersede the phoneme / t / by [ = ]. In

such case there is no assimilation.
Transsibilation is a kind of coalescent assimilation where a consonant

sequence containing a sibilant sound, / sj / or / zj /, is replaced by a new sibilant
sound, / / or / µ /.
1. The consonant sequence / sj / tends to coalesce into / /.
miss you / ªm sju() / → / ªm u() /

this year / ªs sj ± / → / ªs ± /
2. The consonant sequence / zj / tends to coalesce into / µ /.
as yet / ±zªjet / → / ±ªµet / or / ±µªet /

donkey's years / ªdÂÌk z©j ±z / → / ªdÂÌk ©µ ±z / or / ªdÂÌk µ© ±z /
Note
The complete elision of / j / in all cases of coalescent assimilation happens in very rapid
speech. Sequences of the above type usually retain a certain trace of / j /, even in fast
pronunciation, e.g. get you / ªgej(j)u() /, this year / ªs (j) ± /, etc. In such cases, the
assimilation is regressive.
206
Connected Speech
4.8 Juncture
Juncture is "A term used in phonology to refer to the phonetic boundary
features which may demarcate grammatical units such as morpheme, word or clause"
(Crystal 1991: 188). Juncture is a way in which two morphemes, words, or clauses
are connected. In this chapter, we will focus only on junctures between morphemes
and words. Linking and intrusive / r /, for instance, are a type of juncture. There are
two basic types of juncture:
Open juncture refers to connecting two morphemes or words by means of

silence,
e.g. co-opt [ k±h=ªÂpt ], my turn / ma x ªtn /, etc.
Close juncture refers to connecting two morphemes or words without silence, e.g.
co-opt [ k±hwªÂpt ], my turn / ma ªtn /, etc.
A common way of transcribing the open juncture is the sign [ + ], any of the pause
symbols [ x ], [ f ], or a glottal stop [ = ].
The close juncture, on the other hand, is transcribed as [ $ ], or by any of the linking
or intrusive sounds.
Sometimes there may be problems with the analysis of close junctures. For
instance, how do we distinguish the word-group #night rate! from #nitrate! ? The
phonological analysis gives us identical pattern $ / ªna tre t /, so, phonologically,
these two word-groups are indistinguishable. However, phonetic analysis will reveal
some differences:
night rate [ ªna t_Úe t ] or [ ªna =Úe t ]

nitrate [ ªna t{e t ]
In the word-group night rate, the / t / is released with little audible plosion, or it
can be realized as a glottal stop, in which case we can speak of open juncture. The
phoneme / r / in the word-group night rate is realized as a voiced approximant [ Ú ].
In the word nitrate the phoneme / t / is realized as a plosive, and / r / is

represented by its fricative variant $ [ { ].
Here are some other examples of similar cases:
I scream / a ªskrim / / a / is unclipped and it is unstressed.

/ s / is stressed.
/ r / is voiced.
207
Connected Speech
ice cream / ©a s ªkrim / / a / is clipped by the following voiceless

consonant and it carries secondary stress.
/ s / is not stressed.
/ r / is devoiced
might rain / ma t ªre n / / a / is clipped by the following voiceless

consonant.
/ t / is released with little audible plosion.
/ r / is voiced and stressed.
my train / ma ªtre n / / a / is unclipped

/ r / is devoiced
he lies / hi ªla z / / i / is unstressed, / l / is a clear [ lj ]

heel eyes / ©hil ªa z / / i / carries secondary stress, / l / is a dark [ 1 ]
Finally, we must point out that these contrastive phonetic differences, which
work well in slow speech, are usually neutralized in rapid connected speech, that is,
their contrastive value is only potential, not absolute.
4.9 Intonation
In the following sections, we will focus our attention on even larger parts of
connected speech than we have already examined $ on tone-units and utterances. We
will mainly discuss the phenomena connected with the variation of pitch, and we will
look at it from the point of form and function.
4.9.1 The Definition of Intonation

Intonation is a very complex phenomenon for which it is rather difficult to find
a satisfactory definition. The majority of the definitions proposed so far, however,
agree that the most important role in intonation is played by the pitch of the voice.
"Intonation may be defined as the variations which take place in the pitch of the voice
in connected speech, i.e. the variations in the pitch of the musical note produced by
the vibration of the vocal cords" (Jones 1960: 275).
208
Connected Speech
Intonation is "A term used in the study of suprasegmental phonology, referring to the
distinctive use of patterns of pitch, or melody. The study of intonation is sometimes
called intonology" (Crystal 1991: 182).
Intonation is "The pattern of pitch changes that occur during a phrase, which may be
a complete sentence" (Ladefoged 1975: 279).
Intonation is "... pitch variations within a phrase" (Calvert 1986: 179).
"Intonation is the variation given to the pitch of the voice in speaking" (Kingdon
1965: 1).

It is important to make a distinction between stress and intonation. What these

two phenomena have in common is the fact that both stress (kinetic stress) and
intonation are based on the variation of the pitch of the voice. Nevertheless, stress and
intonation usually work at different levels and fulfil different functions.
The primary function of word stress is to determine the meaning and acoustic
shape of words. Sentence stress is built on the stress of individual words, and
although it creates a certain pattern in the sentence (rhythm), it mainly influences the
sentence from the point of view of form, not content. Intonation, on the other hand,
functions at a level of content, and usually comprises units larger than a single word.
It can express such aspects as the mood and attitude of the speaker, it can influence
the meaning of the tone-unit or utterance, it performs a role similar to punctuation in
writing, etc. Needless to say, both stress and intonation are mutually intertwined and
interdependent.
When people speak, their pitch varies from high to low. The span between the
lowest and the highest pitch a speaker can produce is called pitch range. The pitch
range can be different with different speakers, and it depends on the physical structure
of the articulators.
A particular pitch or a variation of the pitch placed on a syllable is called tone.
In many languages, tone is essential for the lexical meaning of the word. In Beijing
Mandarin Chinese, for instance, the word ma has four different meanings, depending
on the type of tone used:
ma in a level tone means !mother" ma in a falling tone means !scold"

ma in a rising tone means !hemp" ma in a falling-rising tone means !horse"
"Such languages, where word meanings or grammatical categories (such as

tense) are dependent on pitch level, are known as tone languages" (Crystal 1991:
353). In English, the tone does not influence the lexical meaning of words, thus,
209
Connected Speech
English is not a tone language. However, the tone or pitch differences in English are
used for other purposes. English is therefore called an intonation language.
4.9.2 The Forms of Intonation

In the following sections we will attempt to describe the physical (acoustic)
nature of intonation, that is, how it manifests itself on the level of form.
4.9.2.1 The Tone and the Tone-unit

Tone is "... the distinctive pitch level of a syllable" (Crystal 1991: 353). We
recognize five different tones in English:
1. falling (fall, glide-down) %yes

2. rising (rise, glide-up) $yes
3. level _yes
4. falling-rising (fall-rise, dive) 'yes
5. rising-falling (rise-fall) &yes
Of course, each of these five tones can be pronounced in the higher or in the
lower part of the pitch range:
High level: yes

Low level: yes
In English, the meaning of the words pronounced on different pitch levels is the
same, unlike in some other languages, where the level of the pitch actually influences
the lexical meaning of words (e.g. African language Kono).
Tones "... are in practice always associated with stressed syllables, though they
may not always be confined to such a syllable" (Kingdon 1958: xxvii). Each syllable
of a word, word-group, or a sentence is pronounced at a certain pitch (changing or
sustained), but only a small number of prominent syllables carry a tone. It is
important to distinguish between pitch as a physical phenomenon and tone as a
linguistic means. Tone is a distinctive pitch level, that is, a pitch level which means
something in terms of intonation ( it can express the mood or attitude of the speaker,
influences the meaning, etc.). The syllable which carries a tone is called a tonic
syllable (nucleus). Tonic syllables have a high degree of prominence and they carry a
stress called tonic (nuclear) stress. The tonic syllable may be accompanied by other
components forming a sequence in which it (the tonic syllable) stands out as the most
prominent element. A sequence of this type is called the tone-unit. The tone-unit
210
Connected Speech
(tone-group / sense-group / breath-group / intonation group / phonological phrase /

intonational phrase/phrase) is rather a difficult unit to define. The problem lies in the
fact that the tone-unit can be treated from various points of view: grammatical,
phonetic, semantic and physiological. How, therefore, do we ascertain the boundaries
between tone-units?
If we take grammatical considerations into account, the tone-unit can be
defined as a sequence of elements corresponding to a clause. "It would certainly be
most convenient if there were a normal co-occurrence of tone-unit with clause, as is
frequently suggested, but while the evidence does show a certain relationship of this
kind, it can hardly be called a norm in a very strong sense" (Crystal 1969: 257).
From the phonetic point of view, there are two principles: "... it is possible in
most cases to detect some sudden change from the pitch level at the end of one tone-
unit to the pitch level that starts the following tone-unit, and recognition of the start
of the following tone-unit is made easier by the fact that speakers tend to "return
home" to a particular pitch level at the beginning of a tone-unit" (Roach 1991: 159).
The second principle is based on rhythm: "... within the tone-unit, speech has a
regular rhythm, but that rhythm is broken or interrupted at the tone-unit boundary"
(Roach 1991: 159).
A semantic criterion for delimiting the tone-unit was proposed by D. Jones
who used the term sense-group. He argued that "Sentences are usually divisible into
smaller sequences between which pauses may be made, though they are not
essential" (Jones 1960: 274).
The tone-unit can also be defined from the physiological point of view, that is,
taking into account some extra-lingual phenomenon. One such phenomenon is the
employment of breath during communication. The places where we pause to take a
breath during speaking are sometimes equated with tone-unit boundaries. The term
breath-group has been introduced to indicate a sequence of elements delimited by
breathing.
It is obvious that none of the adduced criteria can objectively be used to define
the tone-unit. The combination of the four criteria is probably the most reasonable
way to arrive at an objective delimitation of the tone-unit boundaries, although some
linguists argue that the tone-unit definition should be based primarily on phonetic and
phonological considerations.
The tone-unit, in its smallest form, may consist of only one syllable. There are
two types of tone-units: simple and compound. In this book, we will only examine
the structure of the simple tone-unit.
The simple tone-unit contains only one tonic syllable, and the tonic syllable is
an obligatory component of the tone-unit. The simple tone-unit can consist of four
components: pre-head, head, tonic syllable, and tail.
211
Connected Speech
The tonic syllable

The tonic syllable is the central component of the tone-unit. It is the syllable
carrying the tonic stress.
%hour
The head
The head is "... all that part of a tone-unit that extends from the first stressed
syllable up to (but not including) the tonic syllable" (Roach 1991: 146).
ªless than an %hour ªlittle ªless than an %hour
less than an = head little less than an = head

hour = tonic syllable hour = tonic syllable
If there is no stressed syllable preceding the tonic syllable, there cannot be a head.
Depending on the pitch of the head, we distinguish between high head and low
head. "In the case of the high head, the stressed syllable which begins the head is
high in pitch; usually it is higher than the beginning pitch of the tone on the tonic
syllable" (Roach 1991: 154).
The ªbus was %late

___________________
___________________
"In the low head the stressed syllable which begins the head is low in pitch;
usually it is lower than the beginning pitch of the tone on the tonic syllable" (Roach
1991: 155).
The ©bus was %late

___________________
___________________
The pre-head
The pre-head consists of all unstressed syllables preceding the tonic syllable or
the head.
212
Connected Speech
in an %hour in a little less than an %hour

in an = pre-head in a = pre-head
hour = tonic syllable little less than an = head
hour = tonic syllable
The tail
The tail consists of the syllables between the tonic syllable and the end of the
tone-unit.
%look at it $what did you ªsay
look = tonic syllable what = tonic syllable

at it = tail did you say = tail
Thus, a simple tone-unit can contain all four components:
On my ªway to the %station.
On my = pre-head
way to the = head
sta- = tonic syllable
-tion = tail
Tone-units can be combined together to form an utterance. The utterance is

"... a continuous piece of speech beginning and ending with a clear pause ..." (Roach
1991: 134). Utterance can consist of one tone-unit only.
By now we can see that the tone-unit and the utterance have their place in the
set of phonological units with which phonology operates. These units are in
hierarchical relationship. The basic phonological units are phonemes. One or more
phonemes constitute a syllable, one or more syllables make up a tone-unit, and one
or more tone-units represent an utterance. Thus, an utterance can consist of one
phoneme only, e.g. Ah$ / /.
4.9.2.2 Pitch Behaviour within the Tone-unit

The behaviour of the pitch of the voice within the tone-unit abides by certain
rules. We have already stated that there are five tones in English. The tone-unit must
therefore contain one of these five tones, which are placed on the tonic syllable. The
distribution of the tone within the tone-unit depends on the number of elements
constituting the tone-unit and the position of the tonic syllable. From the practical
213
Connected Speech
point of view, we will now be interested in those tone-units which contain the tail.
Here we can formulate this basic rule:
When there is a tail following the tonic syllable, the movement of the tone
is not completed on the tonic syllable, but it is carried over the syllables of the
tail.
Here are the specific cases for the individual tones:
Falling tone (Fall)
One syllable tone-unit Multi-syllable tone-unit with a tail

(it only consists of the tonic syllable)
%Why? %Why did you go?
_____________________ ______________________
_____________________ ______________________
Rising tone (Rise)

$What? $What did you say?
__________________ _______________________
__________________ _______________________
Falling-rising tone (Fall-rise)

'Some. 'Some chairs.
_______________ ___________________
_______________ ___________________
"If there is a tail of two or more syllables, the normal pitch movement is for the
pitch to fall on the tonic syllable and to remain low until the last stressed syllable in
the tail. The pitch then rises from that point up to the end of the tone-unit" (Roach
1991: 153).
214
Connected Speech
I 'might ©buy it. I 'might have ©thought of ªbuying it.

________________ ___________________________________
________________ ___________________________________
"If there is no stressed syllable in the tail, the rise happens on the final
syllable" (Roach 1991: 153).
'Most of them. 'Most of it was for them.

________________ __________________________
________________ __________________________
Rising-falling tone (Rise-fall)

& No# & No sir.
________ ___________
________ ___________
"When there are two or more syllables in the tail, the syllable immediately
following the tonic syllable is always higher and any following syllables are low"
(Roach 1991: 154).
& Beautiful. & All of them ©went.

_____________ _________________
_____________ _________________
4.9.3 The Functions of Intonation

In the previous sections we have shown that intonation has its specific forms
manifested in the five different tones. Now we will take a look at the functions these
215
Connected Speech
tones fulfil in the English language. In this respect we speak of four different
functions of intonation, some of which, of course, can overlap. The four functions are
attitudinal, stress, grammatical, and discourse function.
4.9.3.1 The Attitudinal Function of Intonation

When we communicate, we often need or want to express our feelings and
attitudes, not just the pure facts of the message we are delivering. The reasons for this
need or want are simple. We may be happy, angry, bad-tempered, grateful, bored, etc.
However, the process of expressing feelings, emotions, or attitudes is a very complex
one. Firstly, the expression of feelings, emotions, or attitudes may be genuine (we
really feel happy, angry, bored, etc.), or it may be artificial (we pretend to be happy,
angry, bored, etc.). Secondly, we express feelings, emotions, or attitudes differently
with different people. For instance, an angry boss speaking to his employee might
express his anger in a different way than an angry employee speaking to his boss. The
employee, fearing for his job, will probably try to control himself, while the boss will
give full vent to his anger. The intonation (among other things) will clearly reflect
this. Thirdly, the intonation is not the only means by which we can express feelings,
emotions, or attitudes. If we pronounce the same sentence in different ways, trying to
express happiness, anger, sadness, etc., we will be able to hear that the sentences have
different loudness, tempo, voice quality, etc. All of these factors are of great
importance in conveying attitudes and emotions, and they can be divided into two
feature groups: prosodic features and paralinguistic features. These terms are defined
in 3.1.
Intonation is viewed "... not as a single system of contours, levels, etc., but as a
complex of features from different prosodic systems" (Crystal 1969: 195). The tone,
pitch range, key, pause, and loudness form the basis of intonation in English,
with rhythmicality and tempo closely related.
It is clear that when expressing our feelings, emotions and attitudes, a lot of
features come into play. For the sake of simplicity and clarity, however, we have to
make certain generalizations. Most books agree on these tones and categories:
1.Fall
Finality, definiteness: ªThat"s the ªend of the %news.

I"m ªabsolutely %certain.
He ªstopped %talking.
Strong commands: ªTake your ªfeet off the %chair.

ªShut your %face.
ªHave some %cheese.
216
Connected Speech
Exclamations: ªGood %Heavens#

What a ªvery ªpretty %dress#
%Nonsense#
2. Rise
Encouraging, soothing: It ©won"t $hurt.

ªShe"ll be ©here $soon.
I ªwon"t ©drive ªtoo $fast. (So don't worry.)
Grumble: I ©didn"t $hurt you. (So why make all that fuss?)
You ªcan"t ªpossibly ©do $that. (You should know better.)
I $did. (grumbling contradiction)
Tag-questions after commands: %Come over ©here, $will you?

ªLet"s ©have some %music, $shall we?
%Hold ©this for me, $would you?
(Some of the above cases with a rising tone have also grammatical function)
3. Fall-rise
Uncertainty, doubt: It"s 'possible.

He"s 'generous, but I ©don"t ªtrust him.
He"s my %brother. Your 'brother? You ©don"t %have
a ©brother.
Requesting: ©Can I 'buy it?

©Will you 'lend it to me?
Polite or pleading command: ' Shut the ªwindow.

' Don't ©make me ªangry.
' Do ©have some ©more ªtea.
Note
It is important to remember that the movement of the tone is not completed on the tonic
syllable, but it is carried over the syllables of the tail:
'Shut the ªwindow. 'Don"t ©make me ªangry. ' Do ©have some ©more ªtea.
− _ _ − − _ _ _ − − _ _ _ −
217
Connected Speech
4. Rise-fall
Surprise, being impressed: You were & first#

&All of them#
Have you heard about Jane? She"s &pregnant#
I enjoyed the whole concert, but what a fi & nale#
Challenging: (I don"t like to keep reminding him.) Well, you damn well &ought to#
(They"re emptying the dustbins.) They &always empty them on Fridays.
(I couldn"t find any adverbs.) But there are &lots of them in the text.
5. Level
Boredom, disinterest, routine: A: Have you ever been in prison?

B: (No.
A: Do you suffer from any serious illness?
B: (No.
A: Is your eyesight defective?
B: (No.
(Answering to a series of questions for an insurance
policy.)
Listing, enumerating: ( One, ( two, ( three, ( four, etc.

We ©have ( milk, ( bread, ( rolls, ( butter, etc.
(The last item on the list will have a falling tone.)
4.9.3.2 The Stress Function of Intonation

Many linguists refer to this type of intonation as !accentual", which comes from
the word !accent". However, the word !accent" is used in a different meaning in this
course, so we will avoid using it here.
The title suggests that the placement of stress is determined by intonation. This
may not be absolutely true. Word stress is something independent of intonation,
working on a different level and fulfilling different functions. Tonic stress, however,
could be regarded as a part of intonation, since it is placed on the tonic syllable,
which is the basis of the tone-unit. Thus, we can claim that intonation brought
about by placing tonic stress on certain syllables in an utterance, with the
intention to emphasize or contrast a particular word, has a stress function.
However, since most types of intonation are based on tonic stress, it must be
218
Connected Speech
emphasized that we should consider them to be closely related and often overlapping
in function.
The tonic stress (focus) is usually placed on the last content word in the
utterance. This happens under the so-called normal circumstances, where the tonic
stress is neutral. "The neutral position of focus is what we may call END-FOCUS, that
is (generally speaking) chief prominence on the last open-class (lexical) item or
proper noun in the clause" (Quirk et al 1992: 938). Here are some examples:
He was ¥born in % London. I"ve ¥broken a % plate.
There are, however, several exceptions to this rule:
(a) !Event" sentences

"Event sentences (sometimes called #presentation! sentences) typically involve an
intransitive verb which denotes (dis)appearance or misfortune" (Cruttenden, 1991:
83). In such sentences it is commonly the subject which receives focus, i.e. the tonic
stress:
(Watch out#) That %chimney"s falling ¤down#
(What happened while I was away?) The %milkman ¤called.
(What"s all the fuss about?) A %criminal has e¤scaped#
(b) Final adverbials

Final adverbials (most commonly adverbials of time) do not usually receive end-
focus, i.e. tonic stress:
I ¤went to %London on ¤Thursday.

I"m ¤seeing %John this ¤morning.
We ¤had ¤roast %beef for ¤dinner.
(c) Adjectival wh-objects

In wh-questions where an adjectival wh-word functions as the object of the verb, the
tonic stress falls on the object noun following the adjective:
What %seeds did you ¤use?

Which %course did you ¤take?
Whose ad%vice will you ¤accept?
219
Connected Speech
(d) Cleft sentences

A cleft sentence is a "... special construction which gives both thematic and focal
prominence to a particular element of the clause..." (Quirk et al 1972: 951). Most
cleft sentences begin with the pronoun it plus the verb be followed by the element
carrying tonic stress (focus):
It was %John who ]wore his ]best \suit to the $dance ]last \night.
It was his best %suit that ]John \wore to the $dance ]last \night.
It was ]last %night that ]John \wore his ]best \suit to the $dance.
Tonic stress can also be used for contrastive purposes, and such tonic stress is
called contrastive. It is used when we need to contrast or emphasise certain words,
whether lexical or grammatical. Consider the following examples.
ªThis is my %book. Normal placement of tonic stress (on a content word).

ªThis is %my ©book (not yours) Contrastive placement of tonic stress (on a form word).
It was ©very %boring. Non-emphatic use of tonic stress.

It was %very ©boring. Emphatic use of tonic stress.
From what has been said follows that contrastive focus (contrastive tonic
stress) can be placed on different elements of the clause:
Focus on the subject:

(Who was born in London?) %John was ¤born in ¤London.
Focus on the verb:

(John was married in London, wasn"t he?) %No, he was %born in ¤London.
Focus on the object:

(I hear you"re painting the bathroom blue.) %No, I"m ¤painting the %kitchen ¤blue.
etc.
There are cases where the placement of the tonic stress is not determined by
contrast or emphasis:
I ©have ªplans to %leave. I am planning to leave.

I ©have % plans to ©leave. I have some plans/diagrams/drawings that I have to leave.
The tonic stress in the second sentence is clearly not placed for contrastive or
emphatic purposes. Here, the different position of the tonic stress actually changes the
220
Connected Speech
grammatical meaning of the sentence. In this example, the stress and grammatical
functions overlap.
4.9.3.3 The Grammatical Function of Intonation

In the context of intonation, the word !grammatical" tends to be used in a very
loose sense. In spoken language, intonation can play the same role as some
grammatical means in written language. It is therefore said that intonation can have a
grammatical function. Here are the main categories:
1. Intonation fulfilling the function of the comma.
The function of the comma in written language is reflected in two ways in

speech. Firstly, the comma may be indicated in speech as a short pause, or, secondly,
it may be indicated by the preceding tone-unit.
ªThose who 'sold , x ©quickly ©made a %profit.

ªThose who ªsold 'quickly , x ©made a %profit.
A short pause between the individual tone-units separated by a comma in

writing, is signalled by the a single vertical line x in transcription. Longer pauses
between utterances are signalled by a double vertical line f .
The Conªservatives who 'like the pro©posal , x are %pleased.

The Con'servatives , x who 'like the pro©posal , x are %pleased.
From the intonation pattern it is clear that the first sentence is a restrictive
clause, while the second sentence is a non-restrictive one. The first sentence implies
that only some Conservatives liked the proposal, whereas the second sentence implies
that all Conservatives liked it.
In many cases, the tone-unit followed by a comma in writing contains the

falling-rising tone (as in all above examples).
2. Intonation signalling the declarative, interrogative, and imperative

sentence types.
The choice of the tone within the tone-unit can signal whether the sentence is a
statement, question, or a command. Of course, the individual sentence types do not
necessarily use only one kind of tone. Here are the individual sentence types:
221
Connected Speech
(a) Declarative sentences

Most declarative sentences expressing only facts, without expressing the attitude of
the speaker, use the falling tone.
He"s a ]university %lecturer.

A©bout a ªquarter of an %hour.
There ªaren"t e©nough %chairs.
If a declarative sentence has two parts, of which the first is more important than the
second, we usually use the fall-rise.
I ©went to 'London on ©Monday.

You can 'keep it if you ©really ©want it.
He was ªvery 'well when I ©last ©saw him.
(b) Interrogative sentences
General questions usually have rising tone in English.
Do you $like it?

$Must we ©do as he ©says?
Is it $you?
Yes / no questions are questions which can be answered with a simple yes or no. We
use a rising tone for this kind of questions.
©Have you got any $money?

Do you ©have $time?
Are you $married?
Alternative questions are questions in which we offer or ask about two different or
related things. They have the rising-falling tone.
&Coffee or ªtea?
Was it &you or ªhim?
To&day or toªmorrow?
When the question serves to list or enumerate, we use the rising tone for all listed
items except the last one, which has the falling tone.
222
Connected Speech
Was it $brown, $yellow, $black, or %blue?

Would you ©like $milk, $coffee, $tea, or %wine?
Is your ©name $Bill, $John, $Jack, or %Tom?
Echo questions are questions repeating a word from the previous utterance, usually
because the listener wants to make sure that the word is really what he heard, or he
doesn"t understand why the speaker used that particular word (it makes no sense to
him semantically), or he simply doesn't know the word. Echo questions have a rising
tone.
A: He"s ªgone to %school.

B: To $school?
A: %Yes. He ©wants to ªtalk to the %teacher.
B: He ©wants to $talk to him?
Wh-questions are questions starting with any of the wh-pronouns, including the
pronoun 'how'. Wh-questions usually have a falling tone.
%Why did you ªchange your ªmind?

%Where?
ªWhat are you %doing ©here?
Tag-questions are question structures "... usually consisting of an auxiliary verb plus
pronoun, attached to the end of a statement in order to convey a negative or positive
orientation" (Crystal 1991: 345). Depending on the nature of assumption and
expectation underlying them, we distinguish these four types of tag-questions:
1. Positive assumption + neutral expectation

He ]likes his %job, $doesn"t he?
2. Negative assumption + neutral expectation
He ]doesn"t like his %job, $does he?
3. Positive assumption + positive expectation
He ]likes his %job, %doesn"t he?
4. Negative assumption + negative expectation
He ]doesn"t like his %job, %does he?
"The tag with the falling tone, it will be noted, invites confirmation of the statement,
and has the force of an exclamation rather than a genuine question" (Quirk et al
1972: 391).
223
Connected Speech
(c) Imperative sentences
Imperative sentences typically express commands, therefore this particular case of the
grammatical function of intonation overlaps with the attitudinal function of
intonation. For more information, see the section concerned with falling and falling-
rising tones in 4.9.3.1.
4.9.3.4 The Discourse Function of Intonation

So far we have been discussing the individual functions of intonation in terms
of single tone-units or utterances. Recent studies, however, have shown that the
choice of the tone and the choice of its placement is also dependent on the importance
and information content of the elements of a larger unit $ discourse. Discourse is "A
term used in linguistics to refer to a continuous stretch of (especially spoken)
language larger than a sentence ..." (Crystal 1991: 106). From the phonological point
of view, it is "... a set of utterances which constitute any recognisable speech event
..." (Crystal 1991: 106).
It has been discovered that separate tone-units and utterances may have the
tonic stress placed on different syllables in discourse. The most important (or one of
the most important) factor influencing the placement of the tonic stress in discourse is
the information content of the words out of which the discourse is composed. Look
at the following extract from a conversation.
A: I"ve ©lost my %hat.

B: What %kind of ©hat?
A: It was a %sun hat.
B: What %colour ©sun hat?
A: It was %white. ©White with %stripes.
B: There was a ªwhite ©hat with ©stripes in the %car.
A: %Which ©car?
B: The ©one I %sold.
(Rogerson & Gilbert 1990: 46)
In the first sentence, the tonic stress is placed on the word hat. This placement
is consistent with the general rule, that tonic stress is usually placed on the last noun
(or lexical word) in a tone-unit or an utterance (end-focus). In the second sentence,
however, the word hat is no longer important because it has already been mentioned,
that is, it does not carry any new information $ it has low information content. The
224
Connected Speech
word kind is the word which is most important in this sentence, because it helps the
conversation to continue $ it has high information content. In the third sentence, the
new information is carried by the adjectival (first) element of the compound sun hat.
Therefore the tonic stress is placed on this element. In the fourth sentence, the tonic
stress is placed on the word colour, more precisely on its first syllable, because it is
the word colour that brings new information into the conversation. The rest of the
conversation is stressed in a similar manner $ words containing new information take
tonic stress.
It can easily be seen that the situational context plays an important role in
intonation, a fact that has not previously been recognised.
225
Geographical Varieties and Accents of English
5 Geographical Varieties and Accents

of English
The term geographical variety refers to a type of English spoken in a

particular geographical area, and having certain grammatical, lexical, and
pronunciation peculiarities by which it can be distinguished from other types of
English. Grammatical and lexical peculiarities of a variety of English are called a
dialect, while the pronunciation peculiarities of a variety of English are referred to as
accent. Thus, a geographical variety of English covers both accent and dialect of a
particular area.
Another term which needs to be defined with respect to the matters discussed in
the following chapters is the term Standard English. Standard English is "... the
variety of English language which is normally employed in writing and normally
spoken by 'educated' speakers of the language" (Trudgill, Hannah 1985: 1). Since
Standard English refers only to grammar and vocabulary, and not to accent, it can be
called !the dialect of educated people in the English-speaking countries".
In this course, we will only be interested in accents, not dialects. More
precisely, accents found in different geographical areas of the world. Accent is a
particular way of pronunciation. We distinguish between regional accents and social
accents. Regional accent is a way of pronunciation in certain part (region) of the
country. Social (class) accent is a way of pronunciation of a certain social group
characterized by equal social status, or common educational and cultural background.
In the following chapters, we will discuss these geographical varieties and their
accents:
1. British English (a) English English # RP accent

# London accent
# Estuary English
(b) Scottish English # Scottish accent
(c) Welsh English # Welsh accent
2. Irish English Northern Ireland # Belfast accent
3. North American English (a) U.S. English # General American accent

# Southern American accent
# Eastern American accent
(b) Canadian English # Canadian accent
4. Australian English # Australian accent
226
5.1 British English

The term British English in this course is used to refer to the pronunciation
found in Great Britain. Great Britain is an island of NW Europe, separated from
the mainland by the English Channel and the North Sea. It comprises England,
Scotland, and Wales. In this respect we can speak of the English spoken in England,
which we will term English English, then, the English spoken in Scotland # Scottish
English, and the English spoken in Wales # Welsh English.
Scotland
Edinburg
England
Wales
London
Fig. 44. Great Britain and the main accent areas.
5.1.1 English English

There are a number of accents found in England, and we have chosen only
some of them. Most of the accents in England are regional accents, although we will
also describe a very important social accent # RP.
227
5.1.1.1 RP Accent
Whenever and wherever a type of British English (as opposed to American) is
taught, "... the accent presented as a model for the learner will almost always be
!received pronunciation#, or !RP#. Received here is to be understood in its
nineteenth-century sense of !accepted in the best society# " (Hughes, Trudgill 1979:
2). Although British society has changed a lot since that time, the RP accent still
remains an indicator of social superiority and better education. It is the accent taught
and required at public schools # schools which are in fact private and very expensive.
Originally, the RP accent was a regional accent coming from the speech of
London and the surrounding area, but today it is a (predominantly) social accent. This
means that when a speaker uses the RP accent, it is impossible to say which part of
the country he comes from.
It has been estimated that only about 3 to 5 per cent of the population of
England speak RP. Why should we then teach this accent to learners of English?
There are several reasons:
1. British people consider the RP accent the most correct accent, and some might
even say that it is the most beautiful accent.
2. The RP accent is used on BBC television and radio.
3. The RP accent is the form of pronunciation most often described in the books on
English phonetics and phonology.
4. All dictionaries transcribing words in British English use the RP accent as a basis
for the selection of the transcription symbols.
5. All course-books teaching British English use the RP accent as a basis for
pronunciation of words.
6. The RP accent is readily understood in all English speaking countries, and often
serves as a model of correct pronunciation for those countries.
Many members of British society with a different accent often strive to change
their accent so that it approximates the RP, because they associate it with correctness
and high social status. On the other hand, "... some members of the present younger
generation reject RP because of its association with the !Establishment# in the same
way that they question the validity of other forms of traditional authority" (Gimson &
Cruttenden 1994: 79). In the eyes of the younger generation, regionally marked
accents are more desirable.
Within RP, we distinguish three different types of pronunciation, which are
largely the result of age-group differences:
1. Conservative RP
This type of accent is sometimes called Refined RP, and it is used mainly by
older speakers. The number of speakers using this accent is increasingly
declining.
228
2. General RP
This type of RP is often considered to be the most accepted and suitable type
of RP accent (although, as in the case of the Conservative RP accent, the
number of people using it seems to be decreasing), and it can be found in
most books describing British English. All the distinctive phonemes used in
this course come from this accent. From now on, the RP General will be
used as a basis for all comparisons made with other accents discussed in
the following sections.
3. Advanced RP
This type of accent is sometimes referred to as Regional RP, which some
regard as contradiction in terms, because RP is a social accent, which means
that it should not tell you where the speaker comes from. The term !regional"
describes the "... type of speech which is basically RP except for the presence
of a few regional characteristics which go unnoticed even by other speakers
of RP" (Gimson & Cruttenden 1994: 80).
1 2 3
past pore Burn, fire off Back letter city go here there poor
after war fir, tyre lost that other petty so fear fair sure
bath boar early mire froth glad actor lazy flow beer lair lure
etc. etc. etc. etc. etc. etc. etc. etc. etc. etc. etc. etc.
C. RP [ ] [3±] [ \ ] [Q±] [ 3] [w±] [ \ ] [ ] [4` ] [ j] [4] [ 6c ]
G. RP [ 7] [3] [ ] [Qb±] [ 9 ] [ ] [ ] [ i ] [:` ] [ c ] [ wc ] [6c ][3]
A. RP [ ;] [ o] [ ] [ Q ] [ 9 ] [ a ] [ ] [ i ] [`] [ < ] [ w ] [ o ]
Table 8. Main vowel and diphthong differences among Conservative, General, and
Advanced RP.
1. / / followed by any of the fricatives / f, v, *, s, s /, and sequences / nt, nd, ns, nj, mp /.
2. / Â / followed by any of the fricatives / f, *, s /.
3. Words of two or more syllables ending in !-y" usually preceded by a consonant.
1 2 3
constitute, educate, super, lute, network,
statue, deciduous, Sue, luminous, not now,
etc. etc. etc. etc. etc.
C. RP / tj / / dj / / sj / / lj / /t/
G. RP / j /, / tj / / dµ /, / dj / / s /, / sj / / l /, / lj / [ = ], / t /
A. RP /j/ / dµ / /s/ /l/ [=]
Table 9. Main consonant distribution differences among Conservative, General, and Advanced RP.
1. / tj / in unstressed positions within the morpheme.
2. / dj / in unstressed positions within the morpheme.
3. / t / followed by a consonant between morphemes.
229
In the description of the individual accents, we will try to focus only on those
features by which the described accents differ from the RP General accent.
RP General RP General
vowel and vowel and Example words
diphthong diphthong
phonemes allophones
/ / [ ]1 it, city, kill
[ i ]2 happy, marry, only
/ i / [ i] bee, seedy, meat
/e/ [>] pet, bed, said
/ / [ ] hat, back, bad
/C/ [\] putt, hurry, but
/ / [ 7] dance, daft, bard
/Â/ [9] cot, pot, doll
/ / [ 3] awe, pause, for
/h/ [h] pull, put, full
/ u / [ hu]1, [ Þ ]2 ooze, cool, threw
/±/ [ ± ]1 above, famous, forget (non-final)
[ ]2 China, sofa, butter (final)
/ / [ ] bird, fern, learn
/e / [ >a ] eight, plate, weigh
/a / [ Qb ] eye, flight, my
/ / [ Lb ] boy, adroit, alloy
/ ±h / [ :`] pole, boat, nose
/ ah / [ 7`] bout, house, now
/ ±/ [ c] ear, fear, here
/ e± / [ wc] air, share, care
/ h± / [ 6c ] sure, poor, tour
Table 10. RP General vowel and diphthong phonemes and allophones. See also Fig. 20.
230
RP RP
General General Example words
consonant consonant
phonemes allophones
[ ph ] part, appeal (stressed positions)
/p/ [p] upper, opium (unstressed medial positions)
[ p_ ] kept, hop (followed by a plosive or affricate, and word-finally)
[ ?Ð ] rib cage, cab (followed by a voiceless consonant and word-finally)
/b/ [ bÐ ] robbed, cab (followed by a plosive or affricate, and word-finally)
[?] bar, bog (initially preceded by silence or a voiceless consonant)
[b] able, fob off (intervocalically)
[ kh ] car, decay (stressed positions)
/k/ [k] acre, weekly (unstressed medial positions)
[ k_ ] locked, take (followed by a plosive or affricate, and word-finally)
[ AÐ ] dog kennel (followed by a voiceless consonant and word-finally)
/g/ [ gÐ ] logged, fog (followed by a plosive or affricate, and word-finally)
[A] game, go (initially preceded by silence or a voiceless consonant)
[g] ago, beggar (intervocalically)
[ th ] time, atone (stressed positions)
/t/ [t] later, attic (unstressed medial positions)
[ t_ ] bit, that time (followed by a plosive or affricate, and word-finally)
[ @Ð ] good chance (followed by a voiceless consonant and word-finally)
/d/ [ dÐ ] good day, rod (followed by a plosive or affricate, and word-finally)
[@] day, dog (initially preceded by silence or a voiceless consonant)
[d] udder, fed up (intervocalically)
/f/ [f] fine, after, life
[v] oval (intervocalically)
/v/ [B] veal, love (word-initially preceded by silence or a voiceless consonant,
/*/ [*]
and word-finally followed by silence or a voiceless consonant )
thick, lethal, bath
[s]
/s/
other, either (intervocalically)
[C] this, lathe (word-initially preceded by silence or a voiceless consonant,
/s/ [s] stone, visa, less
[z] ozone (intervocalically)
/z/ [D] these pens (word-initially preceded by silence or a voiceless consonant,
// [] shake, usher, rash
[µ]
/µ/
measure (intervocalically)
[E] rouge powder (word-initially preceded by silence or a voiceless
consonant, and word-finally followed by silence or a voiceless consonant )
231
/h/ [h] how, he, hoot (word-initially, not followed by / j /)

[] behind, anyhow (intervocalically)
/j/ [j] chin, macho, rich
[ dµ ] ajar (intervocalically)
/ dµ / [ @µ ] grudge someone, jaw (word-initially preceded by silence or a voiceless
consonant, and word-finally followed by silence or a voiceless consonant )
[m] mother, amiable, sum (not followed by / f / or / v /)
/m/ [P] comfort, some vote (followed by / f / or / v /)
[] rhythm, chasm (syllabic)
[n] no, venal, run (not followed by / f / or / v /)
/n/ [0] dance, avalanche, pins (followed by / s /, / z /, / /, / /)
[P] inferior, lean veal (followed by / f / or / v /)
[] cotton, nation (syllabic)
[Ì]
/Ì/
sing, ringworm
[] thicken, second class (assimilated word-finally, especially when
preceded and followed by / k / or / g /)
j j
[l ][F ] leap, clear (followed by a vowel or / j /, devoiced by preceding / p, t, k /)
/l/ [1] felt, kill (followed by a consonant other than / j /, or word-finally)
[2] cattle, racial (syllabic)
[Ú] rat, error
/r/ [G] pray, tree, cry (devoiced by the preceding aspiration of / p, t, k /)
[|] bakery, Hungary, flavouring (syllabic)
/j/ [j] yes, spaniel
[I] queue, pew (/ j / preceded by an aspirated / p, t, k /)
/w/ [w] wine, aware
[Z] quite, twice (/ w / preceded by an aspirated / k, t /)
Table 11. RP General consonant phonemes and some of their frequently used allophones.
In recent years, the term BBC English has been used to describe the type of
pronunciation considered correct, or socially acceptable. The reasons for the
emergence of this term are ascribed to the fact that most BBC announcers in the past
used the RP pronunciation, and therefore RP pronunciation often became identified
with !BBC English", that is, the accent heard on the BBC radio and television. At
present, however, the scope of the accents heard on the BBC radio and television is
much broader than it used to be, and the term itself becomes very vague. When
teaching pronunciation, we need to specify the number of phonemes and describe the
particular allophones of an accent. This is not possible with so-called !BBC English",
because the term covers more than one accent. For this reason, we will not use the
term !BBC English" in this book.
232
5.1.1.2 London Accent

The typical London accent is Cockney. Cockney is as much a social accent as a
regional one. It is mainly spread through the working class of London, but traces of
Cockney are often present in most middle-class speech of that area (Hughes &
Trudgill, 1979; Gimson & Cruttenden, 1994).
Cockney differs from the General RP in these respects:
1. / C / in RP General is realized as [ a ] in Cockney, e.g. blood [ blad ], cut [ kat ],

etc.
2. / / in RP General is realized as [ w ] in Cockney, e.g. bag [ bwg ], hack [ hwk ],
etc.
3. In Cockney, RP / / is realized as [ o ] in words without an inflectional
morpheme, e.g. pause [ poz ], paw [ po ], etc.
In words with an inflectional morpheme, RP / / is realized as [ ± ], e.g. bored
[ b±d ], paws [ p±z ], etc.
4. The phoneme / e /, which is pronounced as [ > ] in the RP General accent, is
closer to [ e ] in Cockney, e.g. bed [ bed ], set [ set ].
5. Word-final / / is in Cockney realized as [ i ], e.g. city [ ªs ti ], any [ ªeni ], etc.
6. / i / in RP is realized as [ ±i ] in Cockney, e.g. tea [ t±i ], free [ fr±i ], etc.
7. / u / in RP is realized as [ ±ü ] in Cockney, e.g. two [ t±ü ], do [ d±ü ], etc.
8. Diphthong / e / is in Cockney pronounced as [ ], e.g. day [ d ], may
[ m ], etc.
9. Diphthong / a / is in Cockney pronounced as [ ¥ ], e.g. die [ d¥ ], fly [ fl¥ ],
etc.
10. Diphthong / ±h / is in Cockney pronounced as [ h ], e.g. go [ g h ], no
[ n4h ], etc.
11. Diphthong / ah / is in Cockney pronounced as [ ] (or [ a ]), e.g. now
[ n ], allow [ ±ªl ], etc.
12. / h / is often dropped in Cockney, whether in stressed or in unstressed positions,
e.g. hand [ wnd ], heavy [ ªevi ], etc.
13. In Cockney, RP consonant [ 1 ] is pronounced as [ ö ] after vowels, e.g. milk
[ m ök ], kill [ k ö ], etc., and as [ ±ö ] after consonants, e.g. bottle [ ªbÂ=±ö ],
label [ ªl b±ö ], etc.
14. / * / in RP is substituted by / f / in Cockney, e.g. think / f Ìk /, bath / bf /, etc.
15. / s / in RP is substituted by / v / in Cockney, e.g. bathe / b v /, father / ªfv± /,
etc.
16. / t / is realized as a glottal stop, in all except word-initial positions, e.g. butter
[ ªba=± ], water [ ªw=± ], etc.
233
17. / p / and / k / are realized as a glottal stop providing they are followed by a
consonant and they are not in the word-initial position, e.g. soapbox
[ ªs h=bÂks ], technical [ ªten k±ö ], etc.
18. Word-initial / p, t, k / are heavily aspirated in Cockney. With the phoneme / t /,
the aspiration results in affrication, e.g. tea [ ts±i ], two [ ts±ü ], etc.
5.1.1.3 Estuary English

The term Estuary English was coined in 1983 by David Rosewarne.1 It first
appeared in print in the Times Educational Supplement in 1984. Estuary English is
defined as a variety of regional speech, which is typical of the Thames estuary. It is a
mixture of Received Pronunciation and local accents (mostly Cockney). The term
Estuary English suggests a reference not only to pronunciation, but also to grammar
and lexis (although Estuary English uses standard grammar). Since we are only
interested in pronunciation in this book, it would seem more appropriate to coin a
term referring to pronunciation only, for instance, Estuary accent. Compared to RP
terminology, the Estuary accent can be seen as an even more advanced form of
Advanced RP. The Estuary accent is both a regional and social accent. It is regional
because it is mainly found on the banks of the Thames and its estuary (although it
shows signs of spreading), and social in the sense that it is a result of interaction of
different social groups.
The process of acquisition of Estuary accent is based on accommodation
resulting in accent convergence. In comprehensive schools in England, a minority of
children starts with RP, and the majority starts with a local accent. The two accents
converge and most of the pupils will leave school with Estuary accent. In public
schools in England (which are, of course, private and very expensive), the majority of
children starts off with RP accent and the rest have a local accent. Due to accent
convergence, most will end up with Estuary accent. In other words, the RP speakers
accommodate !downwards" because they want to increase their street credibility and
appear tough, whereas the local accent speakers accommodate !upwards" as they want
their accent to be more sophisticated. The popularity of Estuary accent is becoming
more and more significant for the future. Some speak of Estuary accent as the !RP of
the future".
Here are the basic features by which Estuary accent differs from the General
RP accent (Rosewarne, 1984; Maidment, 1994; Wells, 1994a/1994b):
1. Dark [ 1 ] vocalization
In common with Cockney, Estuary accent exhibits dark [ 1 ] vocalization.
Thus, General RP [ 1 ] is pronounced as [ ö ] or [ ±ö ] in Estuary accent, e.g.
milk [ m ök ], kill [ k ö ], etc.
1
/ de v d rzwn /
234
2. Diphthong allophony
The quality of the diphthong in words like goat, roll, flow, etc. depends on
the following sound. If the following sound is a dark [ 1 ] or its Estuary
variant [ ö ], the diphthong is realized as [ Âh ], e.g. roll [ rÂhö ], soul [ sÂhö ],
etc. If the following sound is any other sound, the diphthong is realized as
[ h ], e.g. go [ g h ], smoke [ sm hk ], etc.
3. Diphthong shift
The RP diphthong / e / is in Estuary accent realized as [ ], e.g. day [ d ],
may [ m ], etc.
The RP diphthong / a / is in Estuary accent realized as [ ¥ ], e.g. die [ d¥ ],
fly [ fl¥ ], etc.
The RP diphthong / ±h / is in Estuary accent realized as [ h ] or [ Âh ],
depending on the following sound (see 2. above), e.g. go [ g h ], no [ n h ],
etc.
4. Word-final / / tensing
In Estuary accent, the word-final RP / / is pronounced, as in Cockney, as a
tense [ i ], e.g. city [ ªs ti ], happy [ ªh pi ], etc.
5. Glottal replacement
RP / t / is pronounced as [ = ] when it is followed by a consonant (except / r /),
and not preceded by / s /, or when standing in the word-final position, e.g.
football [ ªfh=bö ], bit [ b = ], etc.
Note
mattress [ ªm tr±s ] # / t / followed by / r /
best [ best ] # / t / preceded by / s /
6. Yod coalescence
In RP, the sequences / tj / and / dj / may coalesce into / j / and / dµ / in
unstressed positions. In Estuary accent, this type of coalescent assimilation
occurs also in stressed positions, e.g. tune [ jun ], reduce [ r ªdµus ], etc.
5.1.2 Scottish English

Educated Scottish people speak and write a form of Standard English which, in
terms of grammar and lexis, is not very different from English used in England.
However, they speak with a typical Scottish accent. The Scottish accent differs from
235
RP mainly in that it has a radically different system of vowels (Trudgill & Hannah,
1985; Gimson & Cruttenden, 1994).
RP vowel Scottish
and allophones Example words
diphthong representing
phonemes RP phonemes
/ / [ ] pit, city, kill
/ i / [ i()] bee, seedy, meat
/e/ [w] pet, bed, said
/4/ [ä] hat, back, bad
/C/ [{] putt, hurry, but
/ / [ä] dance, daft, bard
/Â/ [] cot, pot, doll
/ / [] caught, pause, fork
/h/ [K] pull, put, full
/ u / [K] boot, cool, fool
/±/ [:] butter, wanted, above
[ ] bird, first, fir
/ / [ w ] fern, learn, divert
[ { ] hurt, curt, burn
/e / [ e;] bay, plate, weight
/a / [\ ] buy, fly, my
/ / [M ] boy, coy, alloy
/ ±h / [ o; ] pole, boat, nose
/ ah / [ {h ] bout, house, now
/ ±/ [ i ] beer, fear, here
/ e± / [ w ] there, share, care
/ h± / [ K ] sure, poor, tour
Table 12. The comparison of RP and Scottish vowel and diphthong sounds. It is evident
from the table, that the Scottish accent has not only a different phoneme realization, but also
a different phonemic inventory.
The previous table shows the way the individual sounds are realized in the
Scottish accent. It also indicates differences in the vowel and diphthong phonemic
236
inventories of RP and Scottish accents. The following is a table of a typical Scottish

inventory of vowel and diphthong phonemes.
/i/ bee, peer

/e/ bay, pair
/w/ bed, merry, fern
/ / bid, bird, butter, wanted
/C/ putt, hurry, fur, sofa
/a/ bad, marry, bard, path, father, calm
/u/ put, boot, poor
/o/ boat
// pot, long, cough, fork, paw
/ ai / buy
/ au / bout
/ i / boy
Table 13. Vowel and diphthong phonemic inventory of the Scottish accent (Trudgill,
Hannah, 1985: 83).
The differences between an RP accent and a Scottish accent can be found in all
categories of sounds. Here are the main features of the Scottish accent:
1. The Scottish accent is a rhotic accent (r-full accent). This means that all r-sounds
in writing are pronounced in speech. Therefore, RP vowels / ± / and / /, standing
for !r" in spelling, and diphthongs / ±, e±, h± / do not exist in Scottish accent in the
traditional sense. Thus, word pairs like sawed # soared, caught $ court, etc. are
clearly different in Scottish accent, unlike in RP, where these pairs are homophones
(they sound alike).
2. The RP distinction between / / and / / does not exist in Scottish accent. The
sound [ ä ] is used for both / / and / /, e.g. cat [ kät ], dance [ däns ], etc.
3. The RP distinction between / u / and / h / does not exist in Scottish accent. The
sound [ K ] is used for both, / u / and / h /, e.g. put [ pKt ], boot [ bKt ], etc.
4. The RP distinction between / / and / Â / does not exist in Scottish accent. The
sound [ ] is used for both / / and / Â /, e.g. cot [ kt ], court [ kt ], etc.
237
5. Vowels and diphthongs standing in the word-final position or followed by an

inflection are longer than other vowels, e.g. booze [ bKz ] vs. boos [bKz ], greed
[ grid ] vs. agreed [ ±ªgrid ], etc.
6. The Scottish accent preserves a distinction between [ w ] and [ Z ], e.g. witch

[ w j ] vs. which [ Z j ], wine [ w\ n ] vs. whine [ Z\ n ],etc.
7. The consonant / r / used by Scots is most usually a flap [ ], although some

speakers use a frictionless continuant [ Ú ].
8. The initial / p, t, k / are often unaspirated in the Scottish accent.
9. Intervocalic RP / t / is often realized as a glottal stop in the Scottish accent, e.g.

butter [ ªb{ Ú ], etc.
10. The phoneme / l / is most commonly a dark [ 1 ] in Scottish accent in all positions,
e.g. little [ ¥1 t2 ], etc.
11. The velar fricative [ x ], which is very similar to Slovak !ch", occurs in a number
of specifically Scottish words, e.g. loch [ lx ] (lake), dreich [ dix ] (dull), etc.
5.1.3 Welsh English

In most areas of Wales, English was, until recently, a second language learnt in
school. This is no longer true, and nowadays a majority of the Welsh population are
native speakers of English. The Welsh English of educated people is not much
different from Standard English spoken in England. The main differences are in the
accent.
The principal differences between the Welsh accent and the General RP accent
are the following (Trudgill & Hannah, 1985):
1. The RP / / is realized as [ a ] in words like chance, pass, last, etc.
2. There is no / C / # / ± / contrast in Welsh accent, e.g. mud [ m±d ], rubber [ ª±b± ],

etc.
3. RP word-final / / represented by -y in writing is pronounced as [ i ] in Welsh, e.g.

happy [ ªhapi ], city [ ªs ti ], etc.
4. Unstressed orthographic a tends to be pronounced as [ a ], e.g. sofa [ ªsofa ], etc.
238
RP vowel Welsh
[i] lazy, city, happy
/ i / [ i ] bee, seedy, meat
/ / [a] hat, back, bad
/C/ [±] putt, rubber, bud
/ / [ a] calm, cart, bard
[a] chance, last, dance
/Â/ [] cot, pot, doll
/ / [] caught, pause, fork
/ u / [ u ] boot, cool, fool
/±/ [±] butter, famous, actor
[a] sofa, above, China
[] condemn, corrupt, tomato
/ / [ R] bird, fern, hurt
/e / [[] bay, bait, maid (words spelled with ai or ay)
[ e] made, bake, late
/a / [± ] buy, fly, my
/ / [ ] boy, coy, alloy
/ ±h / [ 3h ] blow, knows
[ o] boat, nose, so
/ ah / [ ±h ] house, now
/ ±/ [ ±] beer, fear, here
/ e± / [ e± ] there, share, care
/ h± / [ h± ] sure, poor, tour
Table 14. The comparison of RP and Welsh vowel and diphthong sounds.
5. Unstressed orthographic o tends to be pronounced as [ ] rather than [ ± ], e.g.

condemn [ knªdwm ], etc.
239
6. In some areas, RP diphthong / e / can be pronounced in two different ways # [ e ]

or [ [ ]. These sounds are capable of differentiating meaning:
made [ med ] maid [ m[ d ]
7. In some areas, RP diphthong / ±h / can be pronounced in two different ways # [ o ]

or [ 3h ]. These sounds are capable of differentiating meaning:
nose [ noz ] knows [ n3hz ]
8. Words such as tune, music are pronounced as [ t hn ], [ ªm hz k ].
9. Most speakers make a distinction between pairs such as:
blew [ bl h ] blue [ blu ]

threw [ * h ] through [ *u ]
10. / h / in Welsh accent is usually absent, but it may be present in stressed positions.
11. The RP vowel / / is pronounced with rounded lips # [ R ].
12. The Welsh accent is mainly non-rhotic, with the occurrence of linking and
intrusive / r /.
13. Welsh / r / is often a tapped [ ].
14. Voiceless plosives tend to be strongly aspirated, and in word-final positions are
generally not glottalized, e.g. pit [ ph th ].
15. The phoneme / l / is a clear [ lj ] in all positions.
5.2 Irish English
There are two distinct Irish accents, northern and southern. In this book, we
will only examine the one found in the north, used mainly by the people of Northern
Ireland. The English spoken in Northern Ireland has its roots in Scotland. That is why
the Irish accent resembles the Scottish accent in many ways.
Let us now take a closer look at the vowel and diphthong differences between
the General RP and the Irish accent (Vachek, 1960; Trudgill & Hannah, 1985).
240
RP vowel Irish
/ / [ ] hat, back, bad
[w] bag, bang, flag ( before / g / and / Ì / )
/C/ [O] putt, rubber, bud
/ / [ P] daft, last, dance
/Â/ [] cop, pot, dock ( before / p, t, k / )
/ / [ ] caught, hawk, fork ( before / p, t, k / )
/h/ [K] pull, put, full
/ u / [K] boot, cool, fool
/±/ [±] famous, sofa, above,
/ / [ ],[ ± ] bird, fern, hurt
/e / [ w ] bay, say, days
[ i± ] gate, daze, late
/a / [w ] buy, fly, my
/ / [ ] boy, coy, alloy
/ ±h / [ o] blow, knows, boat
/ ah / [ K] house, now
/ ±/ [ i ] beer, fear, here
/ e± / [ w ] there, share, care
/ h± / [ g ] sure, poor, tour
Table 15. The comparison of RP and Irish vowels and diphthongs.
The pronunciation of educated people in Northern Ireland differs from the

General accent in there respects:
1. Vowels followed by / p, t, k, j / are short.
2. General RP / / is pronounced as [ w ] before / g / and / Ì /, bag [ bwg ], bang

[ bwÌ ], etc.
241
3. / Â / and / / contrast before / p, t, k /, e.g. caught [ kt ] # cot [ kt ], but they
are of the same length in other environments, e.g. awful [ ªf ] # offal [ ªf ].
4. There is no distinction between / h / and / u /; they are both pronounced [ K ], e.g.

full [ fKl ] # fool [ fKl ].
5. General RP / / is pronounced [ ] or [ ± ], e.g. bird [ b±d ], fur [ f± ], etc.
6. Word-final / e /, and / e / followed by an inflectional morpheme, is pronounced as

[ w ], e.g. day [ dw ], days [ dwz ]. / e / followed by a consonant which is not an
inflectional morpheme, is pronounced as [ i± ], e.g. gate [ gi±t ], lake [ li±k ], etc.
Thus, there is contrast between word-pairs like:
days [ dwz ] vs. daze [ di±z ]

rays [ wz ] vs. raise [ i±z ]
7. RP diphthong / a / is variable, but it is often pronounced as [ w ], die [ dw ], fly

[ flw ], etc.
8. General RP / ±h / is monophthongized to [ o ], go [ go ], so [ so ], etc.
9. General RP / ah / is pronounced as [ K ], e.g. now [ n K ], how [ h K ], etc.
10. General RP centring diphthongs do not exist in Irish accent. The combinations
[ i ], [ w ], and [ K ] are used instead.
11. / r / in the northern Irish accent is usually a frictionless continuant, very similar to
the one found in General American accent. The phonetic symbol representing this
sound is [ ].
12. The phoneme / l / is most often a clear [ lj ].
13. Intervocalic / t / is often pronounced as a voiced tap [ ].
5.3 North American English

The English spoken in North America is, in terms of pronunciation, rather
different from the rest of the English-speaking world. Apart from the fact that there is
considerable regional variation in pronunciation, there seems to be no universally
accepted standard accent, like the one we can find in England. Therefore, the correct
242
or accepted accent is often claimed to be the one used by educated speakers on

national television and radio.
Geographically, North American English can be divided into US English and
Canadian English. Let us take a closer look at these geographical varieties.
5.3.1 US English
The present-day pronunciation of English found in the United States reflects
the strongly mixed character of the US population. It has been estimated that before
World War II, only about 69 per cent of the US population were of American descent.
The result of this fact was that the observation of the pronunciation norms in the US
became impossible to impose, unlike in Britain, where social pressure gave rise to the
Received Pronunciation. Moreover, the new national and cultural consciousness,
which arose in the US in the nineteenth century, brought about a distinctly American
way of pronunciation, which had been, until then, strongly influenced by British
standards (Vachek, 1960).
EA Accent
GA Accent
SA Accent
Fig. 45. The United States and the main accent areas.
The English spoken in the US nowadays can be roughly divided, in terms of

pronunciation, into three distinct accents. These accents coincide with specific
243
geographical areas, so they are all regional accents. They are the General American
accent, the Southern American accent, and the Eastern American accent.
5.3.1.1 General American Accent

The General American accent (GA accent), which is sometimes called the
Northern Standard, is the most typical accent of the American people. It is used by an
estimated 100 million speakers, and it is spread throughout the country, with the
exception of the southern and eastern parts (see Fig. 45). This accent is the most
representative of all US accents, and it also exhibits the biggest differences from the
General RP accent.
A phonological analysis of General American speech sounds reveals that there
are fewer phonemes than in General RP accent. Here are the General American
phonemes:
Vowels Consonants
/ / is, sit, hit /p/ pan, open, stop

/ i / ease, feel, see /b/ bin, able, hub
/h/ oops, look, took /k/ can, acre, thick
/ u / ooze, two, fool /g/ guy, again, gig
/ / or, soft, paw /t/ tack, later, bat
/e/ ebb, said, well /d/ day, udder, raid
/ / apple, dance, half /f/ fall, often, laugh
/C/ /v/ vase, oval, dove
/*/
up, utter, mud
thick, ether, fifth
/s/
/ / almond, father, calm
/±/ about, famous, drama that, either, bathe
/s/ sin, mason, this
/z/ zip, razor, these
// shop, usher, rush
Diphthongs /µ/ genre, vision, beige
/h/ how, hit, behind
/e / late, eight, day /j/ chin, nature, much
/a / ice, fine, lie / dµ / jaw, region, grudge
/ / oil, coil, boy /m/ me, summer, some
/n/ no, canal, fun
/Ì/
/ oh / oval, coat, slow
/ ah / out, foul, now singer, bring, bang
/l/ leak, column, kill
/r/ rat, oral, car, here, burn
/j/ yes, yacht, yet
/w/ well, swine, quite
244
As one can see, the main differences on the phonological level are between
vowels and diphthongs. There are twelve vowel phonemes in General RP accent,
whereas in the General American accent, there are ten vowel phonemes. As far as
diphthongs are concerned, the General American accent has only five diphthong
phonemes compared to eight diphthong phonemes found in the General RP accent.
There is no difference between the type and number of the consonant phonemes in the
General RP and General American accents.
There are several important allophones found in the GA accent, and they are
illustrated in the following tables.
GA vowel
and GA Example words
diphthong allophones
phonemes
/ / [ <] pit, city, kill
[i] happy, daddy, silly
[ i;] feel, heel, deal (/ i / followed by / l /)
/h/ [ 6] look, book, foot
[ 6] full, bull, pull (/ h / followed by / l /)
/ u / [ hu], [u] rude, through, do
[ u;±] cool, fool, rule (/ u / followed by / l /)
/ / [ ;]1 soft, long, talk
[ o ]2 oar, boring, core (/ / followed by / r /)
/e/ [ [] bed, let, ten
[ w] act, bat, flag
/ / [ w;] dance, half, ask (/ / followed by any
of the / n, f, v, *, s /)
/C/ [ ] cut, run, want
[9] gull, scull, dull (/ C / followed by / l /)
[ a] cop, hot, bother ( 'o' in writing)
/ ¥ / [ T] father, calm, balm
[ ¥] mark, are, dark (/ ¥ / followed by / r /)
/±/ [±] above, sofa, famous
/e / [ ea ] late, aim, day
/a / [ aa ] ice, five, high
/ / [ La ] oil, boil, coy
/ oh / [ 3` ] omen, goat, so
/ ah / [a` ] out, doubt, now
Table 16. The General American vowel and diphthong allophones.
245
GA GA Example words
consonant allophones
phonemes
[ th ] tea, attack (in a stressed syllable not preceded by / s /)
[t] late, stop, hefty (preceded by a voiceless consonant within
root, root finally followed by silence or a consonant)
/t/ [] better, eater, later (unstressed / t / followed by / r /)
[S] mighty, consulted, battle (intervocalically, or preceded
or followed by a voiced consonant)
[ t_ ] late, night time (root finally followed by silence or
a plosive or affricate)
/w/ [w] we, well, await
[ Z ], [ z ] what, which, where
[] rat, outrage, rib
/r/ [ H ], [ ] teacher, pervade, hurry
[ H], [] burn, fur, urge
/l/ [1] leap, hello, kill
Table 17. Some of the General American allophones of the phonemes / t /, / w /, / r /, and / l /.
Most consonant sounds of the General American accent are very similar to the
ones found in the General RP accent. Therefore we have focused only on those
consonant allophones, which show considerable differences or undergo some
important variation.
The main differences between the General RP and the General American are
the following:
Inventory
The phonemic inventories of the RP accent and the GA accent are different. The RP
accent has forty-four phonemes, the GA accent has thirty-nine phonemes.
Vowels
1. In the GA accent, there are ten vowel phonemes, while there are twelve vowel
phonemes in the RP accent.
246
2. GA vowel sounds are different from RP vowel sounds. For concrete allophones,
see Table 16.
3. There is a strong tendency towards diphthongization of vowels in the GA accent.

The diphthongization is symbolized by a small raised [ ± ], which is a reduced form of
schwa. The extent of diphthongization may vary according to the type of the
preceding and following sounds. For a detailed analysis of the diphthongized sounds,
see Table 16.
4. The RP phoneme / Â / can be represented in the GA accent either by the phoneme

/ /, e.g. soft / sft /, dog / dg /, etc., or by the phoneme / /, e.g. hot / ht /,
doctor / ªdktr /, etc. The distribution of / / and / / may differ from speaker to
speaker, e.g. dog / dg / # / dg /, sorry / ªsr / # / ªsr /, etc. In a small number
of cases, RP / Â / is in GA represented by / C /, e.g. want / wCnt /, what / wCt /,etc.
5. The RP phoneme / / in words like sore, or, boar, etc., is represented in the GA
accent by / r /, e.g. sore / sr /, boar / br /, etc. In words like talk, walk, etc. it
may be pronounced either / / or / /, e.g. talk / tk / # / tk /, walk / wk / #
/ wk /, etc., although the variant with / / seems to be more frequent.
6. The RP phoneme / / is represented in the GA accent either by / /, e.g. balm

/ bm /, father / ªfsr /, etc., or by / /, e.g. dance / d ns /, last / l st /, etc.
7. The RP phoneme / / in words like bar, are, harm, etc., is represented in the GA
accent by / r /, e.g. bar / br /, harm / hrm /, etc.
8. The RP phoneme / ± / is represented in the GA accent either by / ± /, when it does

not stand for an !r" in the spelling, e.g. sofa / ªsohf± /, famous / ªfe m±s /, etc., or by
/ r /, when it does stand for an !r" in the spelling, e.g. teacher / ªtijr /, per cent
/ prªsent /, etc. / r / in these cases is a short syllabic [ H ], which can also be
symbolized by [ ].
9. The RP phoneme / / is represented in the GA accent by / r /, e.g. burn / brn /, fur

/ fr /, etc. / r / in these cases is a long syllabic [ H ], which can also be symbolized by
[ ].
10. The RP / C / is represented in the GA accent either by / C /, e.g. cut / kCt /, up

/ Cp /, etc., or, in the words like hurry, worry, furry, etc., by / r /, e.g. hurry / ¥hrr /,
worry / ¥wrr /, etc. / r / in these cases is a short syllabic [ H ], which can also be
symbolized by [ ].
247
11. In the GA accent, vowels / / and / h / in unstressed positions are almost always
pronounced as / ± /, e.g. mistake / m±ªste k /, deputy / ªdepj±t /, etc.
Diphthongs
1. There are eight RP diphthong phonemes compared to five GA diphthong

phonemes.
2. The quality of the diphthongs found in the GA accent differs from RP diphthongs.
For concrete allophones, see Table 16.
3. There are no centring diphthongs in the GA accent. The sequences / r /, / er /, / hr /

are used instead of RP / ± /, / e± /, / h± /.
4. The second element of the GA diphthongs tends to be strongly reduced, even more
than in the RP accent.
5. The RP diphthong / a / in words like organization, civilization, colonization, etc.,

is pronounced as / ± / in the GA accent, e.g. organization / ©rg±n±ªze n /,
civilization / ©s v±l±ªze n /, etc.
Consonants
1. There are twenty four consonant phonemes in both RP and GA accents.
2. In intervocalic positions, the RP consonant phoneme / t / is realized as a voiceless

plosive consonant [ t ], whereas in the GA accent, it is often realized as a voiced tap
[ S ] or [ ] (sometimes also expressed by [ ]), which are very similar to Slovak [ ].
Thus, words like better, battle, eighty, etc., can be phonetically transcribed as
[ªb[± ] or [ªb[±H ], battle [ ªbw;±S2 ], etc.
3. The RP phoneme / w / is represented in the GA accent by the phoneme / w /, which

can be realized in two distinct ways. In words beginning with 'wh' the phoneme / w /
is realized as a voiceless bilabial approximant [ Z ], also symbolized by the allophone
[ z ]. In other environments, it is realized as a voiced bilabial approximant [ w ].
Thus, word-pairs like which $ witch, whine $ wine, whey $ weigh, etc., can be
distinguished by the quality of the allophone, in which case these two allophones
become two separate phonemes, since they can differentiate meaning. However, the
contrast between [ w ] and [ Z ] is becoming obsolete, and most GA speakers use
only the voiced variant [ w ].
248
4. The GA phoneme / r / differs from its RP counterpart in the place of distribution.

The GA accent is a rhotic accent, that is, every written !r" letter is pronounced in the
speech. Therefore, in the GA accent, words like teacher, burn, perfume, etc., are
pronounced with / r / sound rather than / ± / or / / sounds. In addition to that, the
quality of the GA / r / is distinctly different from RP / r /. GA / r / is a retroflex palatal
voiced sonorant, symbolized by [ ], whereas RP / r / is a post-alveolar voiced
sonorant, symbolized by [ Ú ].
There are three important realizations of the GA phoneme / r /.
The first one is found in the word-initial position, e.g. rat, road, etc., intervocalically,
e.g. irritate, oral, etc., and in the word-final positions if preceded by a vowel, e.g. far,
beer, etc. This sound always functions as a consonant and the phonetic symbol for
this sound is [ ].
The second variant is found in the word-medial positions if preceded and
followed by a consonant, e.g. permit, perfume, etc., and in the word-final positions if
preceded by a consonant, e.g. teacher, later, etc. This sound always functions as a
vowel and the phonetic symbol for this sound is the syllabic [ H ] or [ ].
The third variant occurs in all positions in words and it functions as a vowel. It
can be found in words like urge, burn, fur, etc., and the phonetic symbol for this
sound is the syllabic [ H ] or [ ].
5. The RP accent has two distinct allophones of the phoneme / l /: clear [ lj ], and dark
[ 1 ]. The GA accent, however, uses dark [ 1 ] in all positions, e.g. leap [ 1ip ], kill
[ kh<±1 ], etc.
6. General RP unstressed sequences / nt / or / nt± / are in the GA accent realized as

/ n / or / nr /, e.g. plenty / ªplen / [ ªp[±ni ] , winter / ªw nr / [ ªw<±n±· ], etc.
7. General RP stressed sequences / tju /, / dju /, sju /, and / lju / are realized in the
GA accent without / j /, e.g. student / ªstudnt /, due / du /, Susan / ªsuzn /, lewd
/ lud /.
Stress
1. In the GA accent, polysyllabic words (with the primary stress on the fourth syllable
from the end of the word) ending in -ary, -ory, and -mony, have a secondary stress
placed on these endings. In the General RP accent, these endings are unstressed:
RP accent GA accent
dictionary / ªd knr / / ªd k±©ner /

lavatory / ªl v±tr / / ªl v±©tr /
testimony / ªtest m±n / / ªtest±©mohn /
249
Note
British dictionaries usually do not indicate the secondary stress in these words.
2. Some words have a different stress pattern in the RP accent and in the GA accent:
Simple words
ballet / ªb le / / b ªle /
baton / ªb tn /, / b ªtÂn / / b±ªtn /
beret / ªbere / / b±ªre /
café / ªk fe / / k ªfe /, / k±ªfe /
cigarette / ©s g±ªret / / ªs g±ret /, / ©s g±ªret /
debris / ªde br¬ /, / ªdebri / / d±ªbri /, / ªde bri /
fillet / ªf l t / / f ªle /
chalet / ª∫ le /, / ª∫ l / / ∫ ªle /
magazine / ©m g±ªzin / / ªm g±zin /
garage / ªg rµ /, / ªg r dµ / / g±ªrµ /
Note
Different stress patterns usually cause some changes in the phonemic structure of the words.
Compound words
Adam's apple / © d±mzª pl / / ª d±mz© pl /

ice cream / ©a sªkrim / / ªa skrim /
peanut butter / ©pinCtªbCt± / / ªpinCt©bCtr /
police state / p±©lisªste t / / p±ªlisste t /
season ticket / ªsizn©t k t / / ©siznªt k t /
stage manager / ©ste dµªm n dµ± / / ªste dµ©m n dµr /
vocal cords / ©v±hklªkdz / / ªv±hklkrdz /
weekend / ©wikªend / / ªwikend /
250
Phonemic structure
1. Word-final unstressed sequence -ile is pronounced as / a l / in the RP accent, and

usually as / ±l / or / l / in the GA accent. Here are some examples:
RP accent GA accent
docile / ªd±hsa l / / ªdsl /

fertile / ªfta l / / ªfrtl / [ ªfS2 ]
fragile / ªfr dµa l / / ªfr dµl /
hostile / ªhÂsta l / / ªhstl /
missile / ªm sa l / / ªm sl /
mobile / ªm±hba l / / ªmohbl /
2. There is a number of words with differing phonemic structure in RP accent and GA

accent. Here are some of them:
RP accent GA accent
advertisement / ±dªvt sm±nt / / © dvrªta zm±nt /

brazier / ªbre zj± / / ªbre µr /
clerk / klk / / klrk / [ kk ]
either / ªa s± / / ªisr /
glacier / ªgl sj± / / ªgle r /
herb / hb / / rb / [ =b ]
illustrative / ª l±stre t v / / ªlCstr±t v /
laboratory / l±ªbÂr±tr / / ªl br±©tr /
neither / ªna s± / / ªnisr /
route / rut / / raht /, / rut /
schedule / ªedjul / / ªskedµul /, / ªskedµ(±)l /
tomato / t±ªmt±h / / t±ªme toh /
vase / vz / / ve z /
Note
Many of the above words exhibit not only different phonemic structure, but also different
stress patterns.
251
5.3.1.2 Southern American Accent

The second distinct accent type found on the US territory is the Southern
American accent (SA accent), which is used by approximately 30 million people. The
SA accent is spoken in the southern half of Maryland, Virginia, West Virginia, the
southern parts of Ohio, Indiana and Illinois, southeastern Missouri and Oklahoma,
Arkansas, the eastern quarter of Texas, North Carolina, South Carolina, Kentucky,
Tennessee, Florida, Georgia, Alabama, Mississippi, and Louisiana (see Fig. 45).
The phonemic inventory of the Southern American accent is the same as the
inventory of the General RP accent, except for the phoneme / Â /, which does not
exist in the SA accent, and the presence of the phoneme / /. However, there are
some important vowel and diphthong allophones in the SA accent (Calvert, 1986).
SA vowel
and SA allophones Example words
diphthong
phonemes
/ / [ <j±] pit, city, kill
[i] happy, daddy, silly
[ i;±] feel, heel, deal (/ i / followed by / l /)
/h/ [6±], [6w±] look, book, foot
[ 6] full, bull, pull (/ h / followed by / l /)
/ u / [ hu ] rude, through, do
[ u;±] cool, fool, rule (/ u / followed by / l /)
/ / [ ;`], [w±] long, talk, hoarse, mourning, four
/e/ [ ej ± ] bed, let, ten
/ / [ wj±] act, bat, flag
[ wj±] dance, half, ask (/ / followed by
any of the / n, f, v, *, s /)
/C/ [ ]±
cut, run, want
[9] gull, scull, dull (/ C / followed by / l /)
/ / [ a] cop, hot, bother ('o' in writing)
[ T] father, calm, balm
[ ] mark, are, dark (/ / followed by !r"
/±/ [±]
in spelling)
above, sofa, teacher
/ / [ ] urge, first, fir
252
/e / [ a] late, aim, day

/a / [ a ] ice, five, high
/ / [ a ] oil, boil, coy
/ oh / [ \` ] omen, goat, so
/ ah / [ `] out, doubt, now
/ ±/ [ <c] here, ear, fear
/ e± / [ Uc ] air, there, snare
/ h± / [ 6c ] sure, tour, lure
Table 18. The vowel and diphthong phonemes and allophones of the SA accent.
Below are the main characteristic features of the SA accent:
1. The SA accent has forty-three phonemes in its inventory, that is, all RP General
phonemes minus / Â /.
2. The Southern American accent is a non-rhotic accent, although in the areas

bordering with the General American accent this may not be so. In such cases the
sounds [ ], [ e ], and [ h ] are not considered phonemes, and [ ] and [ ] fall
into one phoneme # / r /.
3. The diphthongization of some vowels is even stronger than in the GA accent. This,
among other things, creates the phenomenon called !the Southern drawl". For
concrete allophones, see Table 18.
4. The SA accent / / represents both RP / Â / and / /, e.g. talk / tk /, dog / dg /,
horse / ªhs /, etc.
5. RP phoneme / Â / can be in the SA accent represented either by the phoneme / /,

e.g. soft / sft /, dog / dg /, etc., or by the phoneme / /, e.g. hot / ht /, doctor
/ ªdkt± /, etc. In a small number of cases, RP / Â / is in GA represented by / C /, e.g.
want / wCnt /, what / wCt /, etc.
6. The RP phoneme / / is represented in the SA accent either by / /, e.g. balm

/ bm /, father / ªfs± /, bar / b /, etc., or by / /, e.g. dance / d ns /, last / l st /,
etc.
7. In the SA accent, vowel / / in unstressed positions is almost always pronounced as

/ /, unlike in the GA accent, e.g. mistake / m ªste k /, kitchen / ªk j n /, etc.
253
8. The second element of the SA diphthongs tends to be strongly reduced, even more
than in the RP accent. For concrete allophones, see Table 18.
9. RP diphthong / a / in words like organization, civilization, colonization, etc., is

pronounced as / / or / ± / in SA accent, e.g. organization / ©g±n±ªze n /,
civilization / ©s vl±ªze n /, etc.
10. In intervocalic positions, RP consonant phoneme / t / is realized as a voiceless

plosive consonant [ t ], whereas in SA accent, it is often realized as a voiced tap [ S ]
or [ ], which are very similar to Slovak [ r ]. Thus, words like better, battle, eighty,
etc., can be phonetically transcribed as [ ªbej±± ], battle [ ªbw;j±S2 ], etc.
11. SA phoneme / r / differs from its RP counterpart in quality. SA / r / is a retroflex

palatal voiced sonorant, symbolized by [ ], whereas RP / r / is a post-alveolar voiced
sonorant, symbolized by [ Ú ].
12. RP accent has two distinct allophones of the phoneme / l /: clear [ lj ], and dark [ 1 ].
SA accent, however, uses dark [ 1 ] in all positions, e.g. leap [ 1ip ], kill [ kh<j±1 ], etc.
13. General RP unstressed sequences / nt / or / nt± / are in SA accent realized as / n /

or / n± /, e.g. plenty / ªplen /, winter / ªw n± /, etc.
14. The GA sequence / er / in words like dairy, Mary, etc. tends to be pronounced as
/ e r / in SA accent, e.g. dairy / ªde r /, Mary / ªme r /, etc.
15. In SA accent, the word greasy is pronounced as / ªgriz /, whereas in RP and in

the rest of the US, it is pronounced as / ªgris /.
16. The use of linking [ ] is common in SA accent.
"Industrialization, migrations, and easy travel, as well as pervasive public

media, are influencing Southern American-English dialect (and accent) toward
General American pronunciation. In such major cities as Atlanta, Birmingham,
Charleston, and Houston, one is likely to hear General American almost as
commonly as Southern dialect (and accent). This is especially true in colleges,
universities, and the headquarters of national businesses. However, because of the
greater portion of rural area in the South, its dialect (and accent) is likely to change
more slowly than in that of New England" (Calvert 1986: 204).
254
5.3.1.3 Eastern American Accent

The Eastern American accent (EA accent) is spoken in the states of Rhode
Island, New Hampshire, Maine, and the eastern parts of Vermont, Massachusetts, and
Connecticut, and also in New York City and its environs. The number of speakers
using this type of accent is estimated at approximately 11 million.
This area was colonized mainly by settlers from the south of England, many of
whom were educated and emigrated for religious reasons. This part of the US has
always been in close contact with England, and the pronunciation clearly reflects this
fact. The similarities between the RP and EA accents are so striking, that this area
was aptly named !New England".
The most typical accent of this area is the accent used in Boston. Here are the
main characteristic features of this accent (Calvert, 1986):
1. There are forty-four phonemes in the EA accent, just like in the RP accent.
2. The EA accent is a non-rhotic accent.
3. Long RP / / in words like four, door, etc. is pronounced in the EA accent as

[ o± ], e.g. four [ fo± ], door [ do± ], etc.
4. The phoneme / / is pronounced as [ a ] in the EA accent, whereas in RP it is

close to [ ], e.g. car [ ka ], dance [ dans ], etc.
5. The EA phoneme / Â / in words like top, hot, pop, etc. is sometimes pronounced as
[ Â ], e.g. in Boston, or as [ ¥; ], e.g. in New York City.
6. All other features are similar to the General RP.
Like the SA accent, the EA accent is strongly influenced by the GA accent.

This influence can mostly be seen in the usage of / / instead of / Â / in words like
hot, top, etc., the usage of / / instead of / / in words like past, dance, etc., and in
pronouncing all !r" letters in writing, that is, rhotacization of the EA accent.
5.3.2 Canadian English

Although it is possible to speak of a Canadian accent in theory, there are so few
differences between the GA accent and the Canadian accent, that some might
consider the term useless. However, for the sake of simplicity, clarity and
255
consistency, we will use the term Canadian accent to refer to those few particular
features by which the Canadian pronunciation differs from the GA accent.
Instead of comparing the Canadian accent to RP, we will now compare it to the
GA accent, since it only differs from it by a few features (Trudgill & Hannah, 1985):
1. In most of Canada, the distinction between / / and / / is absent or obsolescent.

The sound used is phonetically close to [ ;± ], which, compared to GA [ a ] or [ T ]
is best transcribed phonologically as / /. Thus, the following pairs of words may be
neutralized:
cot # caught : / kt / [ kh;±t ] not # nought : / nt / [ n;±t ]
2. The Canadian phoneme / / is usually realized from a lower position than the
GA / /, which could phonetically be represented as [ ;± ], e.g. bad [ b;±d ], sad
[ s;±d ], etc.
3. A typical feature of the Canadian accent is the so called !Canadian raising". It

involves the occurrence of different allophones of / aΙ / and / ah / diphthongs,
depending on whether these are followed by a voiceless consonant or not. The rule is
as follows:
If the diphthongs / aΙ / and / ah / are followed by a voiceless consonant, the
first element of the diphthongs # / a / # changes to / ± /. Thus:
/ aΙ / ice / ±Ιs / eyes / aΙz /

/ ah / lout / l±ht / loud / lahd /
4. In the Canadian accent, the word been is occasionally / bin /, unlike in GA, where
it is usually / bΙn /.
5. Words like corollary, capillary, etc., which have the primary stress on the first
syllable in GA accent, have the primary stress on the second syllable in the Canadian
accent:
GA accent Canadian accent
corollary / ªkr±©lerΙ / / k±ªrl±rΙ /

capillary / ªk p±©lerΙ / / k±ªpΙl±rΙ /
6. In the Canadian accent, the word shone is pronounced as / Cn /, unlike in GA,
where it is / ohn /.
256
5.4 Australian English

The accent used in Australia exhibits little regional variation. The variation can
mainly be seen on the social level. In this respect, the pronunciation in Australia
ranges from broad (strong) Australian accent all the way up to RP. The broad
Australian accent shares many features with Cockney accent of London, but, of
course, it has some specific Australian features.
There are no phonemic differences between the Australian accent and the RP
accent (Trudgill & Hannah, 1985; Gimson & Cruttenden, 1994).
Australian Australian
vowel and allophones Example words
diphthong
phonemes
/ / [i ] pit, city, kill
[±] horses, rely, village (/ / in unstressed syllables)
[ i; ] happy, daddy, money (word-finally)
/ i / [ {;a ] bee, seedy, meat
/e/ [U] pet, bed, said
/ / [w] hat, back, bad
/C/ [V] putt, rubber, bud
/ / [ a ] daft, balm, bard (before fricatives, and 'l' and 'r')
[ (; ] dance, sample, plant (before nasals)
/Â/ [] cop, pot, dock
/ / [o], [o;\] caught, hawk, fork
/ / [ XJ ] boot, cool, fool
/±/ [±] famous, above, correct
[\] sofa, teacher, ever (word-final / ± /)
/ / [ ] bird, fern, hurt
/e / [ ;a ] bay, say, days
/a / [ ;a ] buy, fly, my
/ / [ o;a ] boy, coy, alloy
/ ±h / [ ¥;] blow, knows, boat
/ ah / [ ; ] house, now, cow
/ ±/ [ Z ] beer, fear, here
/ e± / [ [ ] there, share, care
/ h± / [ o ] sure, poor, tour
Table 19. The phonemes and allophones of broad Australian accent.
257
The following are the main features of the broad Australian accent:
1. The realizations of the individual phonemes in Australian accent differs from the
RP accent.
2. The Australian phoneme / / is realized in three ways:

(a) In stressed positions it is realized as [ i ], e.g. bit [ bit ], is [ iz ], etc.
(b) In unstressed non-final positions it is reduced to [ ± ], e.g. mistake [ m±ªst ;ak ],
horses [ ªhos±z ], etc.
(c) In word-final position, representing the letter !y", / / is pronounced as [ i; ], e.g.
happy [ ªwpi; ], money [ ªmVni; ], etc.
3. The RP / / is pronounced either as [ a ] if followed by !r" or !l" in spelling, or a

fricative in pronunciation, e.g. path [ pa* ], calm [ kam ], bar [ ba ], etc. or it can
be pronounced as [ w; ] if followed by a nasal in pronunciation, e.g. dance [ dw;ns ],
sample [ ªsw;mp2 ], etc.
4. The RP phoneme / ± / is realized as [ \ ] if standing in the word-final position, e.g.

sofa [ ªs¥;f\ ], teacher [ ªth{; j\ ], etc.
5. In the Australian accent, the closing diphthongs are pronounced as in Cockney, and
the first element tends to have extra length, e.g. late [ l ;at ], fight [ f;at ], etc.
6. Australian centring diphthongs are monophthongized, e.g. beer [ bZ ], there [ s[ ],
etc.
7. The Australian phoneme / e / is close in its quality to [ ], and it is transcribed

phonetically as [ U ], e.g. bed [ bUd ], met [ mUt ], etc.
8. The Australian accent is a non-rhotic accent and it has both linking and
intrusive / r /. Australian / r / is often more strongly retroflexed than in RP, however,
it does not quite reach the quality of the GA [ ].
9. The intervocalic / t / often becomes a voiced flap [ ] in Australian accent.

Nevertheless, the voiceless plosive [ t ] is also frequent in this environment.
10. The glottal stop may substitute / t / in the Australian accent, but not
intervocalically, e.g. lately [ ªl ;a=li; ], not there [ ©n=ªs[ ], etc.
11. Words that can have either / Â / or / / in RP, e.g. Australia, auction, salt, etc.,
have only / Â / in the Australian accent.
258
12. In the Australian accent, days of the week tend to be pronounced with / e /
instead of RP / /, e.g. Monday / ªmCnde /, Tuesday / ªjuzde /, etc.
13. Australian sequences / tj / and / dj / may coalesce into / j / and / dµ / in all

positions, e.g. tune / jun /, due / dµu /, etc.
14. The phoneme / h / may be dropped in the Australian accent, e.g. hand / nd /,
happy / ª p /, etc.
259
A Confrontational Analysis of English and Slovak Pronunciation Systems
6 A Confrontational Analysis of English and

Slovak Pronunciation Systems
In this chapter we will attempt to characterize briefly the main differences

between English and Slovak pronunciation, both on segmental and suprasegmental
levels.
When comparing sound systems of different languages, it is important to use
systematic transcription with symbols and diacritics referring to certain absolute
qualities. So far, we have been using the International Phonetic Alphabet (IPA) for
this purpose. Unfortunately, the phonetic transcription of many Slovak sounds is not
based on the IPA. Thus, to avoid possible confusion, we will use the IPA
transcription symbols for both English (RP accent) and Slovak. Most data on Slovak
speech sounds are taken from Dvonþová, Jenþa & KráĐ (1969), KráĐ (1988), and KráĐ
& Sabol (1989), but some of the symbols have been changed to be compatible with
the IPA.
6.1 Vowels
A. PHONEMIC INVENTORY
English: There are twelve distinctive vowels in RP English:

/ /, / i /, / e /, / /, / Y /, / /, / Â /, / /, / h /, / u /, / ± /, / /
Slovak: There are eleven distinctive vowels in Slovak (The symbols are adjusted to
suit our comparison):
/ i /, / i /, / e /, / e /, / a /, / a /, / /, / /, / /, / u /, / u /
Slovak does not have phonemes which would be similar in nature to the
English / ± / and / /. Slovak / /, although similar in nature to English / /, is a
marginal phoneme and it is used only in literary style, unlike the English / / which
is frequently used. The Slovak long phoneme / e / has no counterpart in English,
where only short / e / can be found. While the rest of the phonemic inventories of the
two languages is similar in nature, there are, of course, differences in phonetic
quality.
B. QUALITY
The concrete allophones of English and Slovak vowel phonemes show considerable
differences:
260
English:
front front-central central central-back back
close
i i J u

close-mid
3

mid
>
open-mid
\
7 9
open
Fig. 46. The diagram of General RP vowel allophones.
Slovak:
front front-central central central-back back
close
Z, Z ], ]
close-mid
[, [
mid
L, L
open-mid
^_
open
T, T
Fig. 47. The diagram of Slovak vowel allophones.
261
English vowel phonemes i e Y Â h u ±

Slovak vowel phonemes i i e e a a u u
English allophones ,i i > \ 7 9 3 h hu,J ±,
Slovak allophones Z Z [ [ ^_ T T L L ] ]
Table 20. A comparison of English and Slovak vowel phonemes and allophones.
ENGLISH front front-central central central-back back

close / i / / u /
close-mid / / /h/ / /
mid / ± /, / /
open-mid /e/ /C/
open / / / / /Â/
Table 21. An analysis of English vowel phonemes in terms of vertical (open-close) and
horizontal (front-back) tongue position. The positions are approximate.
SLOVAK front front-central central central-back back

close / i /, / i / / u /, / u /
close-mid
mid / e /, / e / / /, / /
open-mid
open / / / a /, / a /
Table 22. An analysis of Slovak vowel phonemes in terms of vertical (open-close) and
horizontal (front-back) tongue position. The positions are approximate.
The different quality of English and Slovak short vowels is due to bigger
distances between the tongue positions of the Slovak vowels. More specifically, the
distances between Slovak / i /, / e /, / a /, / /, / /, and / u /, are greater than in
English, and the vowels / i /, / a /, and / u / are also purer in the sense that they are
produced at the near-maximum points of the close-open and front-back positions.
Another thing contributing to the difference is the fact that long vowels in Slovak
have practically the same quality and are produced in the same place as short vowels,
thus creating only six different tongue positions (long and short vowels plus / /),
whereas in English we have twelve different tongue positions. An English speaker is
therefore capable of making a finer analysis of sounds, because his ear distinguishes
among twelve different phoneme qualities, while a Slovak speaker normally
distinguishes and uses only five (/ / is not normally used). That is why English
speakers sometimes analyze Slovak / a / in stressed positions as / /, or / C /, and
262
Slovak / e / in unstressed positions as / e / or / ± /. On the other hand, an untrained

Slovak ear tends to analyze the English / / as either / a / or / e /.
Furthermore, English vowels tend to be reduced into a more obscure form ! / ± /
! in unstressed non-final positions, e.g. possibility / ©pÂs ªb l t / → / ©pÂs±ªb l±t /,
deputy / depjht / → / depj±t /, etc. This quality reduction of vowels does not
occur in Slovak.
C. QUANTITY
English: The real acoustic length of English vowels depends on the type of the
following sound. English vowels followed by a voiceless consonant are shortened,
those followed by a voiced sound or standing in the word-final position are
uninfluenced. Thus, in English we have four phonetic lengths, but only two
phonological lengths:
Phonological lengths: Phonetic lengths:
1. short vowels bit / b t /, bid / b d / 1. shortened short vowels bit [ b t ]

2. long vowels beat / bit /, bead / bid / 2. short vowels bid [ b ;d ]
3. shortened long vowels beat [ bi;t ]
4. long vowels bead [ bid ]
Slovak: In Slovak we distinguish between two lengths, whether phonologically or

phonetically. The type of the following consonant does not dramatically influence the
phonetic length of Slovak vowels. Long vowels are approximately twice the length of
short vowels:
short vowels mak / mak /
long vowels mám / mam /
6.2 Diphthongs
English: There are eight distinctive diphthongs in RP English:

Five closing diphthongs / e /, / a /, / /, / ±h /, / ah /
Three centring diphthongs / ± /, / e± /, / h± /
Slovak: There are four distinctive diphthongs in Slovak:
/ ia /, / ie /, / iu /, / uo /
Slovak and English diphthongs are so different in nature, that no parallels can be
drawn between them.
263
B. QUALITY
The quality of the individual elements of diphthongs in English and Slovak is

different. The differences lie mainly in the starting and ending tongue positions (see
Tables 24, 25), and in the prominence of the individual diphthong elements:
English: As far as the prominence of the individual elements of English diphthongs is

concerned, we distinguish two categories:
1. falling diphthongs, where the first element is more prominent
2. rising diphthongs, where the second element is more prominent
Slovak: All Slovak distinctive diphthongs are rising, that is, the second element is
more prominent. There are no falling diphthongs in Slovak.

e

a a
Fig. 48. English closing diphthongs.

e
Fig. 49. English centring diphthongs.
264
iu
ie uo
ia
Fig. 50. Slovak diphthongs.
English diphthongs Slovak diphthongs

Phonemes e a a e ia ie iu u
Allophones >a Qb Lb :` 7` c wc 6c e f[ f g3
Table 23. A comparison of English and Slovak diphthong phonemes and allophones.
English diphthongs starting position ending position

/e / front, open-mid front-central, close-mid
/a / front-central, open front-central, close-mid
/ / back, open-mid front-central, close-mid
/ ±h / central, mid central-back, close-mid
/ ah / central-back, open central-back, close-mid
/ ±/ front-central, close-mid central, mid
/ e± / front, open-mid central, mid
/ h± / central-back, close-mid central, mid
Table 24. An analysis of English diphthong phonemes in terms of vertical (open-close) and
horizontal (front-back) tongue positions. The positions are approximate.
Slovak diphthongs starting position ending position

/ ia / front, close-mid front-central, open
/ ie / front, close front, mid
/ iu / front, close central-back, close-mid
/ uo / back, close back, close-mid
Table 25. An analysis of Slovak diphthong phonemes in terms of vertical (open-close) and
horizontal (front-back) tongue positions. The positions are approximate.
265
In Slovak we have several non-distinctive two-vowel sequences. These are, for

instance, the sequences "ao# or "eo#, e.g. video, kakao, etc. These two-vowel
combinations can be found only in words of foreign origin, and they are interesting
from the point of view of language acquisition. When children acquire language, and,
of course, pronunciation, they form a system of distinctive sounds ! phonemes ! in
their brains, and they analyze all sounds of their language in terms of this system. The
system of Slovak phonemes does not contain "ao# or "eo# in its inventory, and there
are no such sequences found in native Slovak words. Therefore, when a child hears
these two-vowel sequences, (s)he automatically separates them into individual vowel
phonemes and inserts a consonant between the individual elements, e.g. kakao
/ kakav /, video / videj /, etc., because (s)he intuitively knows that these
sequences do not form one syllable. These mistakes are perfectly consistent with the
Slovak phonemic system, and this example shows that language acquisition is not just
imitation, but a creative process based on application of a certain system of
functionally relevant units.
C. QUANTITY
English: The length of English diphthongs equals the length of English long vowels.
However, the real acoustic length of English diphthongs, like the length of vowels,
depends on the type of the following consonant. If an English diphthong is followed
by a voiceless consonant, it is shortened, if it is followed by a voiced sound or
standing in the word-final position, it is uninfluenced.
There is also internal differentiation between the lengths of the individual
elements of a diphthong. In falling diphthongs, the first element is longer and more
prominent, in rising diphthongs, it is the second element which is longer and more
prominent. However, the overall lengths of falling and rising diphthongs are equal.
[ long vowel + voiceless sound ] = [ diphthong + voiceless sound ]

[ long vowel + voiced sound ] = [ diphthong + voiced sound ]
Slovak: The real acoustic length of Slovak diphthongs does not depend on the type of
the following consonant. Slovak diphthongs are approximately of the same length as
Slovak long vowels.
[ Diphthong ] = [ Long vowel ]
The length of the individual elements is not equal. The first element is shorter,
the second element is longer.
266
6.3 Consonants
English: There are twenty-four distinctive consonants in English:

/ p /, / b /, / t /, / d /, / k /, / g /, / j /, / dµ /, / f /, / v /, / * /, / s /,
/ s /, / z /, / /, / µ /, / h /, / m /, / n /, / Ì /, / l /, / r /, / j /, / w /
Slovak: There are twenty-seven distinctive consonants in Slovak. The symbols are
taken from IPA:
/ p /, / b /, / t /, / d /, / c /, / I /, / k /, / g /, / f /, / v /, / s /, / z /, / /, / µ /,
/ ts /, / dz /, / j /, / dµ /, / x /, / /, / m /, / n /, / /, / l /, / \ /, / r /, / j /
Since some of the symbols may be unfamiliar or confusing to Slovak learners,

we offer an explanatory set of words. Each symbol represents the sound of the
underlined letter of the word standing next to it:
/ c / Ģava / / $iĢ / ts / cap / j / þas / x / chata

/ I / ćateĐ / µ / %iĢ / dz / medza / dµ / d%bán / / hora
/ / vaĖa / l / lano / \ / Đad
By looking at the above sets of phonemes, we can see that there are differences
not only in the number of phonemes, but also in the type of phonemes.
In Slovak, there are no such phonemes as / * / or / s /. The sounds [ Ì ] and [ w ]
do exist in Slovak, but not as separate phonemes. The phoneme / r / exists in Slovak,
but the sound most often representing this phoneme in Slovak is different from that
found in English. The rest of the English phonemes are similar in nature to Slovak
phonemes.
English, on the other hand, does not have the phonemes / c /, / I /, / x /, / /,

/ \ /, / ts / and / dz / in its inventory. The Slovak phonemes / c /, / I /, / x /, / /, / \ /
could all be considered variant realizations of the English sequences / tj - /, / dj - /,
/ hj - /, / nj - /, / lj - /, although the absolute quality differs. The Slovak phonemes / ts /
and / dz / are broken down into separate elements in English ! / t + s / and / d + z /.
B. QUALITY
Very few consonants of Slovak and English can be said to be completely

identical in quality, because they behave differently in the two languages. The
problem lies partly in the nature of the particular consonants, and partly in the
267
possible sound environments, which can be different in Slovak and English. If we

were to look for such identical consonants, we would have to focus on those which
behave very similarly in all possible positions and environments. The phonemes
closest to meeting these criteria are / b /, / d /, / g /, / f /, / s /, / z /, / m /, / n /, / j /. Of
course, if we take into consideration the fact that English voiceless consonants
influence the preceding vowels and Slovak voiceless consonants do not, the number
of such theoretically identical consonants would be still smaller.
Nasal Oral
Plosive Affricate Fricative Lateral Central
Bilabial p b m w
Labiodental f v
Dental * s
Alveolar t d s z n l
Post-alveolar r
Palato-alveolar t dµ µ
Palatal j
Velar k g Ì
Glottal h
Table 26. A chart of English distinctive consonants.
Nasal Oral
Lateral Central
Plosive Affricate Fricative Trill Glide
Bilabial p b m
Labiodental f v
Alveolar t d ts dz s z n l r
Palato-alveolar j dµ µ
Palatal c I \ j
Velar k g x
Glottal
Table 27. A chart of Slovak distinctive consonants.
There are certain important allophonic variations of English and Slovak

consonants, and we will now attempt to characterize them.
268
English:
The phonemes / p /, / t /, and / k / have two main variants. The first one is a
voiceless plosive consonant [ p ], [ t ], and [ k ] standing in the unstressed positions in
words, e.g. open [ ª±hp±n ], hot [ hÂt ], lake [ le k ], etc. The second allophone is an
aspirated voiceless plosive [ ph ], [ th ], and [ kh ], which can be found in stressed
positions not preceded by / s /, e.g. part [ pht ], tea [ thi ], car [ kh ], etc.
The English phoneme / h / can be realized in three different ways. The
voiceless glottal fricative [ h ] can be found in stressed positions when followed by a
vowel, e.g. how [ hah ], the voiced glottal fricative [ ] stands in intervocalic
positions, e.g. behind [ b ªa nd ], and the voiceless palatal fricative [ I ] occurring,
for instance, in the word huge [ Iud ].
The phoneme / m / can be realized as [ m ] as in the word mud [ mCd ], or as a
syllabic [ ] found in the assimilated word happen [ ªh p ], or, finally, as a
labiodental [ P ], in cases where / m / is followed by / f / or / v /, e.g. comfort
[ ªkCPf±t ], etc.
The phoneme / n / has five variants: [ n ] as in not [ nÂt ], [ ] as in cotton
[ ªkÂt ], a labiobental [ P ], in cases where / n / is followed by / f / or / v /, e.g. envy
[ ªePv ], the allophone [ 0 ], which is found when followed by / s, z, , µ /, e.g.
dance [ ¥d0s ], etc., and the allophone [ Ì ] (also functioning as a separate
phoneme) when followed by / k / or /g /, e.g. income [ \ Ìkm ].
The English phoneme / Ì / has two main variants. [ Ì ] is found in words
like sing [ s Ì ], etc., and the syllabic [ ] found only in assimilated words like
thicken [ ª* k ].
There are three main variants of the General RP English / r /. The post-alveolar
approximant [ Ú ] is the most frequent allophone of / r / in General RP. The syllabic
[ | ] is found in words like literal [ ªl t|±l ]. The flapped variant [ ] is found (mainly
in the intervocalic position) in the speech of some older RP speakers and it is not very
frequent.
The phoneme / l / has three distinct variants: clear [ lj ] pronounced when
followed by a vowel or / j /, e.g. last [ ljst ], leap [ ljip ], etc., dark [ 1 ] pronounced
when followed by the consonant (except / j /), or when standing in the word-final
position, e.g. felt [ fe1t ], kill [ k 1 ], etc., and the syllabic dark [ 2 ] variant, e.g. bottle
[ ªbÂt2 ].
Phonemes p t k h m n Ì r l
Allophones p p t th k k h I m Pn P 0 Ì Ú |
h h j
l 1 2
Table 28. The most common allophones of some English consonant phonemes.
269
Slovak:
The Slovak phoneme / v / has four distinct allophones. The first one is a voiced
labiodental fricative [ v ], and we can find it in the non-final position when followed
by a voiced consonant except / r, j, l, , n, /, or when the preposition "v# is followed
by a voiced consonant except / r, j, l, , n, /, vdova [ ¥vda ], v zime [ ¥v zime ].
The second variant is a bilabial approximant [ w ], pronounced in the syllable-final
position preceded by a vowel or / r / or / l /, e.g. pravda [ ªprawda ], krv [ krw ]. The
third variant of / v / is the voiceless labiodental fricative [ f ], found when the
preposition "v# or a root-initial "v# is followed by a voiceless obstruent, e.g. vtedy
[ ªftedi ], v tme [ f tme ]. The fourth allophone is the labiodental approximant [ ],
which is found non-finally followed by a vowel or / r, j, l, , n, /, and when the
preposition "v# is followed by a vowel or / r, j, l, , n, /, e.g. vata [ ¥ata ], vlk [ k ],
v roku [ ¥ ku ].
The phoneme / x / has two allophones: a voiceless velar fricative [ x ] found in
unassimilated words, e.g. hrach [ hrax ], and a voiced velar fricative [ ] found at
word boundaries as a result of assimilation, when followed by a voiced sound, e.g.
hrach bol [ ¥hra ¤bl ].
The phoneme / / can be realized either as a voiced glottal fricative [ ], e.g.
hore [ e ], noha [ na ], or as a voiced velar fricative, found at word boundaries
as a result of assimilation when followed by a voiced sound, e.g. roh ulice [ r
uitse ].
The Slovak phoneme / m / has two distinct variants. A bilabial nasal [ m ] is
found in all positions except when followed by / f / or / v /, e.g. mama [ ªmama ]. A
labiodental nasal [ P ] is found only when followed by / f / or / v /, e.g. nymfa
[ ªniPfa ].
There are three main allophones of the phoneme / n /: the alveolar nasal [ n ],
e.g. strana [ ªstrana ], the velar nasal [ Ì ] found when followed by / k / or / g /, e.g.
lanko [ ªlaÌk ], cengaĢ [ ªtseÌgac ], and the allophone [ 0 ], found when followed
by / s, z, , µ /, e.g. penzia [ ¥pe0zia ].
The phoneme / r / in Slovak has four variants. The first one is a flapped
alveolar consonant [ ], which can sometimes be interchangeable with the second
variant ! a trilled alveolar consonant [ r ]. The single-flap alveolar [ ] is used mostly
in intervocalic position, e.g. varí [ ªai ], whereas the trilled alveolar consonant [ r ]
is found mainly when preceded or followed by another consonant, e.g. krása [ ªkrasa ],
norma [ ªnrma ], etc., although the two allophones may be interchanged. The third
variant is a short syllabic trilled [ ] found in words like brloh [ ªblx ], krt [ kt ],
etc. The fourth variant is a long syllabic trilled consonant [ ], found in words like
pĚhliĢ [ ªph\ic ], vĚba [ ªba ], etc.
270
The Slovak phoneme / l / has three distinct allophones. The first variant is a
lateral alveolar consonant [ l ]. It is found in all positions, e.g. láva [ ªlaa ], málo
[ ªmal ], dal [ dal ], etc. The second variant is a syllabic short lateral alveolar
consonant [ ], found in words like tlkot [ ªtkt ], vlk [ k ], etc. The third variant is a
long lateral alveolar consonant [ ] found in words like vĎþa [ ªja ], tĎk [ tk ], etc.
Phonemes v x m n r l
Allophones v w f x m P n Ì 0 r l
Table 29. The most common allophones of some Slovak consonant phonemes.
Finally, it is important to remember that the quality of English consonants

influences the real acoustic length of preceding vowels, whereas in Slovak, no
dramatic influence of this type can be heard.
C. QUANTITY
English: From the phonological point of view, no consonants in English have the
status of long sounds, although from the phonetic point of view we can speak of long
consonants, or lengthened consonants, in some environments. However, not all
English consonants can be lengthened. In general, only continuant consonants can
change their length. These are the main cases in which consonants can change their
length:
1. All continuant consonants standing in the word-final position are lengthened

when preceded by a short vowel, e.g.
come [ kYm; ] ! calm [ km ] Chris [ kr s; ] ! crease [ kris ]

sin [ s n; ] ! scene [ sin ] buzz [ bYz; ] ! bars [ bz ]
2. Lateral and nasal consonants can often be lengthened for the sake of emphasis
(usually emotional reasons), when they are preceded by a short vowel and
followed by a voiced consonant, e.g.
splendid [ ªsplen;d d ] hungry [ ªhYn;gr ]

numbers [ ªnYm;b±z ] seldom [ ªsel;d±m ]
3. When two identical continuant consonants meet at the word boundary, they
merge into a single longer one, e.g.
271
some men [ ©sYm;ªen ] this site [ ©s s;ªa t ]

tough fight [ ©tYf;ªa t ] ten nights [ ©ten;ªa ts ]
Slovak: Phonologically, there are no long consonants in Slovak. Phonetically, there

are two official long allophones [ ] and [ ]. These allophones have the status of
long consonants and their length equals the length of long vowels. They can be found
both within morphemes and at morpheme boundaries. In English, long consonants
can only occur at morpheme boundaries.
As in English, all Slovak continuant consonants can be lengthened, both within
the word and at word boundaries, in cases of gemination:
vy$$ý [ ªi;i ] denný [ ªIen;i ]

predám mak [ ªpredam;ªak ] pes sedel [ ªpes;ªeIel ]
The real length depends to a great extent on the tempo of speech.
6.4 Stress
A. PROMINENCE
English: There are three degrees of syllable prominence brought about by stress in
English ! unstressed syllables (low degree of prominence), syllables with secondary
stress (medium degree of prominence), and syllables with primary stress (high degree
of prominence). The syllables carrying secondary stress are clearly distinguishable
from the syllables carrying no stress, which can also be seen in the choice of the
vowel in the unstressed syllable, where we often find an obscured vowel / ± /.
Slovak: In Slovak, there are also three degrees of syllable prominence brought about
by stress ! unstressed syllables (low degree of prominence), syllables with secondary
stress (medium degree of prominence), and syllables with primary stress (high degree
of prominence). In general, the primary stress in English has more prominence than
the primary stress in Slovak, and the secondary stress in English usually has more
prominence than the secondary stress in Slovak. In Slovak, the choice of vowels does
not seem to correspond to the amount of the stress imparted to the syllables. Compare
the levels of prominence of the syllables carrying primary and secondary stress in
English and Slovak:
−
typewriter / ª © _ / þiernobiely / ª − _ © − _ /
272
B. DISTRIBUTION
WORD STRESS
English: The placement of stress in English is free in the sense that the primary stress
can be put on any syllable in English words, that is, there is not a universal stress
pattern compulsory for all English words. This flexibility can also be seen in the fact
that the secondary stress can either precede or follow the main stress in English. The
placement of secondary stress does not depend on the number of syllables in English
words. For a complete set of rules on the placement of primary and secondary stress,
see 3.6.
Slovak: Slovak is one of the languages where a certain stable stress pattern is
compulsory for all words, or, at least, when these words are pronounced in isolation.
The primary stress is always placed on the first syllable and, unlike in English, the
secondary stress never precedes the primary stress. Here are some basic rules:
1. Simple and derived words (with or without inflections) have the primary stress
placed on the first syllable.
dobre / ªdbre / robotníci / ªrbtitsi /
2. Compound words and compound derivatives (with or without inflections) have the
primary stress placed on the first syllable, and the secondary stress may be placed on
the second element.
tmavomodrý / ªtmav¤mdri / poĐnohospodár / ªp\n¤spdar /
SENTENCE STRESS
English: In general, the stress in an English sentence is placed on content words, that
is, words carrying lexical meaning. The function words are stressed only on special
occasions.
Slovak: The sentence stress in Slovak is not necessarily placed on content words,
although there are certain categories of function words which are usually unstressed
(som, si, je, sme, mi, ma, ti, Ģa, etc.). Unlike in English, Slovak monosyllabic
prepositions take primary stress from the following content word (in certain contexts),
e.g. na, vo, za, etc.
na dvore / ªna ©dvre / vo dverách / ªv ©dverax / za rohom / ªza ©rm /

do $koly / ªd ©kli / bez obetí / ªbez ©beci / k otcovi / ªg tsvi /
273
C. FUNCTION
English: The stress pattern in English can distinguish one word from another,
sometimes without any phonemic changes, although such changes often accompany
such word-pairs. To put it in another way, one word class changes into another word
class mainly by means of stress. Here are some examples:
permit n. / ªpm t / v. / p±ªm t /

import n. / ¥ mpt / v. / m¥pt /
For a complete list of word class pairs see 3.6.5.2.
Slovak: The stress pattern in Slovak is stable, and it can never distinguish between
word classes.
6.5 Assimilation
Assimilation in both English and Slovak is most noticeable in the category of
consonant sounds, therefore in the following sections we will focus solely on
consonants. The analysis is done from the point of view of rapid connected speech.
A. ALLOPHONIC ASSIMILATION WITHIN THE WORD AND AT THE

WORD BOUNDARY
The consonants assimilated allophonically can be divided into different groups.

We will attempt to analyze them in terms of place of articulation and voicing:
English:
1. Assimilation of the place of articulation
post-alveolar [ t ] in try [ tÚa ] (influence of [ Ú ])

dental [ } ] in eighth [ e }* ], that thing [ ©s }ª* Ì ] (influence of / * /)
comfort [ ªk\Pf±t ], conference [ ªkÂPfÚ±ns ]
come for me [ ªk\Pf±mi() ], run fast [ ©Ú\Pªfst ]
advanced (pre-velar) [ h ] in key [ hi ], take it [ ªte h t ] (influence of / i / and / /)
velarized (dark) [ 1 ] in build [ b 1d ] (influence of the following consonant)
retracted [ s ] in this road [ ©s sªÚ±hd ] (influence of [ Ú ])
palatal fricative [ I ] in queue [ kIu ] (influence of advanced aspirated / k /)
274
2. Voicing
Devoicing of / l, r, w, j / after an aspirated voiceless consonant / p, t, k /.
please [ pFiz ] at last [ ±ªtFst ]

cry [ kGa ] at once [ ±ªtz\ns ]
twit [ tz t ] at rest [ ±ªtGest ]
pew [ pIu ] thank you [ ª* ÌkIu() ]
Devoicing of voiced plosives, affricates, and fricatives in the word-final positions,

and when followed by a voiceless consonant at word boundaries.
pub [ p\? ] of course [ ±Bªkhs ]

smog [ smÂA ] child support [ ©ja l@ s±ªpht ]
judge [ dµ\@µ ] big shoe [ ©b Aª∫u ]
above [ ±ªb\B ] his hand [ ©h Dª h n@ ]
Slovak:
1. Assimilation of the place of articulation
Labio-dental [ P ] instead of / m / when followed by / f / or / v /:
amfiteáter [ ªaPfiteater ] dám veĐa [ ©daP ªe\a ]
Velar nasal [ Ì ] instead of / n / when followed by / k / or / g /:
skrinka [ ªskriÌka ] Kongo [ ªkÌg ]

syn kriþal [ ªsiÌ ªkrijal ]* ten graf [ ªteÌ ªgraf ]*
Bilabial approximant [ w ] instead of / v /:
pravda [ ªprawda ] dav [ daw ]

dav burácal [ ªdaw ªburatsal ]
Note
The assimilations marked with an asterisk are considered substandard, and should be
avoided in the codified Slovak pronunciation.
275
2. Voicing
The voiceless velar fricative [ x ] changes to a voiced velar fricative [ ] when

followed by a voiced sound at the word boundary:
hrach je [ ªra je ] sneh bol [ ªse bl ]
Devoicing of the voiced plosives, affricates, and fricatives in the word-final

positions, and when followed by a voiceless consonant at word boundaries is
complete in Slovak, so a new phoneme is produced. Therefore, these cases, unlike in
English where the devoicing is not necessarily complete, are part of phonemic
assimilation.
B. PHONEMIC ASSIMILATION WITHIN THE WORD AND AT THE

WORD BOUNDARY
English: According to the direction and extent of assimilation, we distinguish three

major types of assimilation in English: progressive, regressive, and coalescent.
1. Progressive phonemic assimilation occurs when the initial consonant of a word

changes to become like the final consonant of the preceding word, e.g. happen
[ ªh p ].
2. Regressive phonemic assimilation occurs when the final consonant (or consonants)
of a word change to become like the initial consonant of the following word, e.g.
ten cups / ©teÌªkCps /.
3. Coalescent assimilation occurs when two sound fuse into a new one, e.g. get you
/ ªgeju() /.
Complete lists of the individual categories and cases of these three types of
assimilation can be found in the respective chapters on assimilation.
Slovak: In Slovak, progressive phonemic assimilation does not exist. There is only
regressive and coalescent assimilation in Slovak (the following rules are based on
KráĐ, 1988).
REGRESSIVE ASSIMILATION
1. All voiced plosives, affricates, and fricatives / z /, / µ / and / h /, or a combination

of them, followed by a voiceless consonant, assimilate to voiceless consonants with
the same place of articulation (VLC = Voiceless consonant).
/ b, d, I, g, dz, dµ, z, µ, h / + / VLC / → / p, t, c, k, ts, j, s, , x / + / VLC /
276
bozkaĢ / ªbskac / dá%ć pr$í / ªdacªpri /

obþan / ªpjan / dub spadol / ªdupªspadl /
rehtaĢ / ªrextac / plod práce / ªpltªpratse /
In many cases when this assimilation produces two plosive consonants, or two
affricate consonants, or a combination of a plosive and an affricate consonant in
sequence, the first consonant is pronounced without audible release, or, in fast speech,
it may sometimes be completely dropped (this, however, is considered substandard,
and the students of Slovak are recommended to use the standard form, as given in
KráĐ, 1988):
odtade [ ªt_taIe ] chyĢte [ ªxic_ce ]

podĢaĢ [ ªpt_cac ] robĢe [ ªrp_ce ]
oddych [ ªd_dix ] oddiel [ ªd_Iiel ]
2. Consonants / t, d, c, I / plus / s, z, ts, dz / may assimilate to / tsts / or / dzdz /.

Consonants / t, d, c, I / plus / , µ, j, dµ / may assimilate to / jj / or / dµdµ /.
In both cases, the first of the two assimilated consonants is pronounced without
audible release, or, in very fast speech, it may be completely dropped. This
phenomenon overlaps with coalescent assimilation:
otca [ ªts_tsa ] dobytþí [ ªdbij_ji ]

vedca [ ªets_tsa ] pod%ieraĢ [ ªpdµ_dµierac ]
Note
A combination of two identical consonants in Slovak pronunciation is termed "doubled
consonants# = zdvojené spoluhlásky (for further details see KráĐ, 1975a, f, g, h).
3. All voiced plosives, affricates, and fricatives / z /, / µ / and / /, or a combination

of them, standing in the word-final position and followed by silence, assimilate to
voiceless consonants with the same place of articulation (except / /, which
assimilates to / x /, that is, the place of articulation is different):
/ b, d, I, g, dz, dµ, z, µ, / → / p, t, c, k, ts, j, s, , x /
chlieb / x\iep / gulag / ªgulak / raz / ras /

chlad / xlat / jedz / jets / krí% / kri /
keć / kec / imid% / ªimij / stoh / stx /
277
4. Voiceless consonants or a combination of them occurring at the morpheme

boundary, assimilate to voiced consonants with the same place of articulation, when
followed by a voiced sound (mainly a voiced obstruent).
(VLC = Voiceless consonant, VCD = Voiced consonant, VS = Voiced sound).
/ VLC / + / VS / → / VCD / + / VS /
kresba / ªkrezba / vlak huþí / ªvlagªuji /

ná$ho / ªnaµ / les vonia / ª\ezªvia /
kliatba / ªk\iadba / viac ráz / ªviadzªras /
Exceptions:
(a) The assimilation of this type does not occur when the prepositions "s#, "so# and
"k#, "ku# precede inflected personal pronouns:
s nami / ªs nami / k nemu / ªk emu / so mnou / ªs mnu /

s vami / ªs vami / k nej / ªk ej / ku mne / ªku me /
s ním / ªs im / k nám / ªk nam /
s Ėou / ªs u / k nim / ªk im /
(b) The sequence / v / + / VLC / only assimilates to / f / + / VLC / when the

sequence is in the word-initial position followed by a voiceless consonant, or when
/ v / represents the preposition "v# and is followed by a voiceless consonant:
vþela / ªfjela / v práci / ªf pratsi /

vták / ªftak / v tomto / ªf tmt /
v$iĢ / ªfic / v þistote / ªf jistce /
5. The consonants / t, d, n, l / followed by / i, i, e, ia, ie, iu / within words,

assimilate to / c, I, , \ /, but usually not in words of foreign origin:
teplo / ªcepl / pani / ªpai /

dielo / ªIiel / ale / ªa\e /
Exceptions:
Assimilation of this sort does not take place in inflected forms of some adjectives,
numerals and participles:
pekní, milí, matkini, matkinej, stáleho, stálemu, zaviatej, jednej, %iadni, etc.
278
6. The alveolar nasal / n / followed by / p /, / b / or / m / assimilates to / m /.
klenba / ªk\emba / on poþul / mªpjul /

hanba / ªamba / zvon bije / ªzvmªbije /
bonbón / ªbmbn / ten mal / ªtem mal /
Note
The assimilation of / n / to / m / occurring at the word boundary is considered substandard,
and it is not recommended in the codified Slovak pronunciation.
7. The sound / s / followed / t, / at the morpheme boundary may assimilate to / /.

The sound / z / followed / dµ, µ / at the morpheme boundary may assimilate to
/ µ /.
rozþúliĢ / ¥rztu\ic / → / ¥rstu\ic / → / ¥rtu\ic /*

roz$afný / ¥rzafni / → / ¥rsafni / → / ¥rafni /*
roz%aĢ / ¥rzµac / → / ¥rµµac /*
Note
COALESCENT ASSIMILATION
Consonants / t, d, c, I / plus / s, z / may assimilate to / ts / or / dz /.

Consonants / t, d, c, I / plus / , µ / may assimilate to / t / or / dµ /.
In both cases, the first of the two assimilated consonants is pronounced without
audible release, or, in very fast speech, it may be completely dropped. In such cases
the assimilation is coalescent, because two consonants fuse into a different one:
poć sem / ªpIsem / → / ªpcsem / → / ªptsem /* / I + s / → / ts /

dvadsaĢ / ªdvadsac / → / ªdvatsac / / d + s / → / ts /
pod%ieraĢ / ªpdµierac / → / ªpdµierac / / d + µ / → / dµ /
rad$ej / ªradej / → [ ªrattej ] → / ªratej /* / d + ∫ / → / t /
daĢ znamenie / ©dacªznameie / → / ©daªdznameie /* / c + z / → / dz /
Note
279
To conclude the matters regarding assimilation, let us sum up the main

differences and similarities between English and Slovak assimilation.
Differences:
1. Both allophonic and phonemic assimilations produce different results in the two
languages, which is mostly due to the possible sound environments and sequencing.
2. The presence of aspiration in English causes allophonic assimilation (devoicing) of
some sounds. This is not possible in Slovak.
3. The devoicing of word-final voiced consonants is usually partial in English
(producing allophones), but complete in Slovak (producing phonemes).
4. In Slovak, regressive assimilation involving voicing within the word also occurs in
slow speech, which is something not found in English. In this respect, the assimilation
of voice in Slovak could be regarded as established.
5. There is no progressive phonemic assimilation in Slovak.
Similarities:
1. Some allophonic assimilations occur in identical sound environment and give
identical results: / m / followed by / f, v / → [ P ].
/ n / followed by / k, g / → [ Ì ] or, in English, the phoneme / Ì /.
etc.
2. Coalescent assimilation in both English and Slovak occurs usually in informal
colloquial speech.
6.6 Elision
VOWELS
English: Contextual elision of vowels in English can be found both within the word
and at the word boundary. The particulars have been discussed in 3.4 and 4.6. The
majority of cases of vowel elision results in the reduction of the number of syllables:
police: / p±ªlis / 2. syllables → / plis / 1. syllable

library: / ªla br±r / 3. syllables → / ªla br / 2. syllables
possible: / ªpÂs±bl / 3. syllables → / ªpÂsbl / 2. syllables
for instance: / f±rª nstns / 3. syllables → / ªfr nstns / 2. syllables
The important thing about vowel elisions in English is that they can occur both within
the morpheme and at the morpheme boundary.
280
Slovak: Contextual elision of vowels in Slovak is much less frequent than in English,
and it is confined mainly to the word boundary. It occurs when two vowels of
identical quality (the length can be different) meet at the word boundary, and the
second word of the sequence is either unstressed or carries secondary stress:
staþí iba / ªstaji iba / 4. syllables → / ªstajiba /* 3. syllables

málo ostalo / ªmal ©stal / 5. syllables → / ªmal©stal /* 4. syllables
prácu u% dostal / ªpratsu uµ ªdstal / 5. syll. → / ªpratsuµªdstal /* 4. syll.
mama asi nechce / ªmama asi ªextse / 6. syll. → / ªmamasiªextse /* 5. syll.
When the second word of the sequence carries primary stress, the elision
usually does not occur:
auto odi$lo / ]aut ªdil / my isto prídeme / mi ªist ©priIeme /
Within the word, the vowel elision can be found only in compounds whose
morphemes have identical vowels in the place of the morpheme boundary:
sivooký / ªsiv©ki / 4. syllables → / ªsivki /* 3. syllables

maloobchod / ªmal©pxt / 4. syllables → / ªmalpxt /* 3. syllables
All vowel elisions in Slovak occur only at the morpheme boundary and they result in
the reduction of the number of syllables.
Note
The elisions marked with an asterisk are considered substandard, and should be avoided in
the codified Slovak pronunciation.
CONSONANTS
English: The contextual elision of consonants in English is a frequent phenomenon,

and it mainly involves cases where several consonants follow in succession. All
details concerning the elision of consonants within the word and at the word
boundary can be found in the respective chapters on elision.
Slovak: The contextual elision of consonants, in common with English, occurs

mainly in consonants clusters. Even though most elisions might be considered
substandard by some Slovak phoneticians, no one can deny their widespread use in
the speech of educated people. Here are the major categories:
281
1. When two identical consonants in sequence occur within the word or at the word
boundary, the first one is usually elided in rapid informal speech.
chyĢte / ªxice /* byĢ ticho / ªbiªcix /*

mäkký / ªmeki /* chlap povedal / ªxlaªpvedal /*
poddaĢ / ªpdac /* chlieb bol / ªx\ieªbl /*
This elision also applies to those phonemes, which have undergone the process of
assimilation:
odtade / ªdtaIe / → / ªttaIe / → / ªtaIe /*

subpolárny / ªsubplarni / → / ªsupplarni / → / ªsuplarni /*
rozsúdiĢ / ªrzsuIic / → / ªrssuIic / → / ªrsuIic /*
chlieb padol / ªx\iebªpadl / → / ªx\iepªpadl / → / ªx\ieªpadl /*
voz stojí / ªvzªstji / → / ªvsªstji / → / ªvªstji /*
smäd trápil / ªsmedªtrapil / → / ªsmetªtrapil / → / ªsmeªtrapil /*
When two identical consonants are continuant consonants, they may merge into a
single longer one. The length of such a consonant depends on the tempo of speech:
vy$$í / ªvi()i /* mám málo / ªmam()ªal /*

zlomme / ªzlm()e /* má$ $aty / ªma()ªati /*
zákonne / ªzak()e /* mu% %il / ªmuµ()ªil /*
Note
2. The following consonant clusters tend to be reduced in rapid informal speech:
/ t / → / t / väþ$mi / ªvetmi / → / ªvetmi /

väþ$í / ªveti / → [ ªvetti ] → / ªveti /*
hráþ $kúlil / ªhratªku\il / → / ªhratªku\il /*
/ zs / → / s / rozseknúĢ / ªrzseknuc / → / ªrsseknuc / → / ªrseknuc /*

rozsah / ªrzsax / → / ªrssax / → / ªrsax /*
voz sa / ªvzsa / → / ªvssa / → / ªvsa /*
282
/ z / → / / ráz$tep / ªrazcep / → / ªracep /

perz$tina / ªperzcina / → / ªpercina /
voz $iel / ªvzªiel / → / ªvsªiel / → / ªvªiel /*
/ scs / → / s() / $esĢsto / ªescst / → / ªest /

$esĢ synov / ªescªsinov / → / ªeªsinov /*
nájsĢ si / ªnajscsi / → / ªnajsi /*
/ stn / → / sn / þestný / ªjestni / → / ªjesni /*

vlastný / ªvlastni / → / ªvasni /*
list na$iel / ª\istªnaiel / → / ª\izdªnaiel / → / ª\izªnaiel /*1
/ st / → / s / námestník / ªnamestik / → / ªnamesik /*

vlastne / ªvlaste / → / ªvlase /*
list nemám / ª\istªemam / → / ª\izdªemam /
→ / ª\izªemam /*1
/ zd / → / z / prázdniny / ªprazdini / → / ªprazini /*

hvizdne / ªhvizde / → / ªhvize /*
zjazd nebol / ªzjazdªebl / → / ªzjazªebl /*
/ tn / → / n / zvlá$tny / ªzvlatni / → / ªzvlani /*

zi$tný / ªzitni / → / ªzini /*
/ t / → / / zvlá$tne / ªzvlate / → / ªzvlae /*

zi$tník / ªzitik / → / ªziik /*
/ µd / → / µ / tý%dne / ªtiµde / → / ªtiµe /*

dá%dnik / ªdaµdik / → / ªdaµik /*
/ stk / → / sk / dvanástka / ªdvanastka / → / ªdvanaska /*

þiastka / ªjiastka / → / ªjiaska /*
list klamal / ª\istªkalmal / → / ª\isªkalmal /*
1
Simultaneous elision and assimilation
283
/ ptsk / → / psk / egyptský / ªegiptski / → / ªegipski /

koptský / ªkptski / → / ªkpski /
lept sklamal / ª\eptªsklamal / → / ª\epªsklamal /*
/ stsk / → / s()k / mestský / ªmestski / → / ªmeski /

list skrþil / ª\istªskrjil / → / ª\iªskrjil /*
Note
Similarities:
As we can see, certain parallels can be drawn between English and Slovak
elision. More specifically, two identical consonants or vowels following each other
are elided, or, in case of continuant consonants, lengthened in both languages. Also,
consonants / t / and / d / are frequently elided in consonant clusters in both languages.
Differences:
The main difference between vowel elision in English and Slovak lies in the
fact that vowel elisions in English can be found both within the morpheme and at the
morpheme boundary. In Slovak they are only found at the morpheme boundary.
Contracted forms in English, e.g. won#t / w±hnt /, he#s / hi()z /, etc., result in elision
of both vowels and consonants. Nothing like this exists in Slovak.
The differences in the elision of consonants found in English and Slovak are
mainly due to different clustering (sequencing) of the consonant phonemes in the two
languages.
6.7 Rhythm
English: English speech has regular rhythm. This rhythm is brought about by
alternating stressed and unstressed syllables, where there is a tendency towards
keeping an equal amount of time from one stressed syllable to the next. English
therefore belongs to the so called stress-timed languages.
Slovak: The stressed syllables in Slovak do not occur at regular intervals of time,
which is why the rhythm of a Slovak sentence differs from the rhythm of an English
sentence. The rhythm in Slovak is based on the number and the type of syllables
284
within words. The rule is that syllables with long vowels should not follow each other
within one word. A diphthong in Slovak counts as a long vowel (for the purpose of
rhythm). This rule is called "the rhythmical law# (rytmický zákon/rytmické krátenie).
Here are some examples:
pekný ! krásny bodiek ! þiarok nosím ! súdim
As one can see, these pairs of words are inflected in the same way, yet the
length of the vowels in the inflections is not identical. This is due to the fact that the
roots of the pairs contain vowels of different length, and according to the rhythmical
law, a long vowel cannot be followed by another long vowel in the same word.
Nevertheless, there are exceptions, e.g. lístie, hrádzí, páví, zmúdrieĢ, chvália,
kĚmievaĢ, múþiar, tisíckrát, niekým, nádielka, etc.
Since English makes use of only a handful of inflections, no direct parallels can
be drawn between the two languages. Nonetheless, in phonological terms, long
vowels in English root words are not usually followed by another long vowel or a
diphthong within the same word, e.g. over, important, nation, etc. The exceptions to
the rule (for root words) seem to be mostly words of foreign origin, e.g. vacation,
phoneme, migraine, etc. On the other hand, derived and compound English words can
frequently contain two long vowels in succession, e.g. pre-war, streetwise, freewheel,
etc. However, as we mentioned earlier, the length of syllables in English is ruled by
stress, whereas in Slovak no such influence can be detected. Slovak is therefore called
a syllable-timed language.
285
6.8 Intonation
Since intonation is a very complicated suprasegmental phenomenon, both in
English and in Slovak, we will try to focus only on the most salient similarities and
differences between the two languages. It is also important to bear in mind, that the
approaches to intonation analysis are different in English and in Slovak, which, as a
consequence, results in different terminology. For the sake of clarity and simplicity,
we will use the terminology introduced earlier in this book in the chapter on English
intonation.
Similarities:
1. All tones found in English (rise, fall, rise-fall, fall-rise, level) can be found in
Slovak too, albeit with some limitations:
Fall: Som %hladný.

Rise: Si $hladný?
Fall-rise: 'Adam najprv rozmý$Đal, %potom sa pustil do roboty.
Rise-fall: 'Voda na þaj u% vrie, no &zeliniek niet.
Level: (jeden, (dva, (tri, ...
2. When there is a tail following the tonic syllable, the movement of the tone is not
completed on the tonic syllable, but it is carried over the syllables of the tail:
%When did you come back? %Kedy ste sa vrátili?
3. The general function of some tones is very similar in the two languages:
The rising tone is found in general questions:
Are you $hungry? Si $hladný?
The rising tone is found in echo questions:
I'm going to %school. To $school? Idem %do $koly. $Do $koly?
The rising tone in requests is considered to be a sign of politeness:
Could you pass me the $salt? Podali by ste mi $soĐ?
286
The falling tone is found in general statements, commands, etc.:
I am %hungry. Som %hladný.

Go a%way; Choć %preþ;
The falling tone is placed on words to which we want to impart special emphasis:
%This is my book. %Toto je moja kniha.

This is %my book. Toto je %moja kniha.
The falling tone is placed on words carrying high information content ! discourse
function.
Wh-questions (in Slovak, the questions with the respective wh-pronouns) have a
falling tone:
%Who is it? %Kto je to?
The level tone expresses boredom, mechanical enumeration, etc.:
(one, (two, (three, ... (jeden, (dva, (tri, ...
Rising and falling tone in alternative questions:
$Coffee or %tea? $Kávu alebo %þaj?
Differences:
1. Fall-rise and rise-fall are usually not found within one syllable in Slovak. In
English, on the contrary, one syllable tone-units containing rise-fall or fall-rise are
quite frequent.
2. The individual tones may fulfil different functions in the two languages. Fall-rise,
for instance, is never used in Slovak questions, unlike in English, where fall-rise is
used frequently. Similarly, rise-fall, expressing surprise or big impression in English,
does not have the same function in Slovak.
3. Slovak sentences usually begin with a stressed syllable, because words in Slovak
have stress placed on the first syllable. In English, on the contrary, sentences often
start with one or more unstressed syllables, and this influences not only rhythm, but
also intonation. That is why Slovak sentences (compared to their English
287
counterparts) often start with a high pitch which is placed on the stressed syllable,
unlike in English, where the pitch of the initial unstressed syllables is usually low:
I need money. Potrebujem peniaze.
I#ve got a headache. Bolí ma hlava.
4. The intonation of tag questions may differ in English and Slovak:
(a) Tag questions expressing certainty in Slovak can have both rising and falling
intonation, depending on the lexical item used:
He#s your brother, %isn#t he? Je to tvoj brat, $v$ak?

Je to tvoj brat, %v$ak%e (%v$akáno/%v$akhej)?
(b) Tag questions expressing uncertainty in Slovak have only rising intonation:
He#s your brother, $isn#t he? Je to tvoj brat, $nie (alebo (azda)$nie)?
It is clear from the above examples that certainty and uncertainty in Slovak is
expressed not only by intonation but also by lexical means, i.e. v!ak/v!ak"e/v!akhej
express certainty (regardless of intonation) and nie/alebo nie expresses uncertainty.
5. The overall pitch range in English is broader than the overall pitch range in Slovak.
288
A Concise English-Slovak Dictionary of Phonetic and Phonological Terms
A Concise English-Slovak Dictionary of Phonetic and

Phonological Terms
A amplitude / ª mpl tjud / n. amplitúda,

rozkmit
accent / ª ksnt / n. prízvuk, analysis / ±ªn l±s s / n. analýza, rozbor
charakteristický spôsob výslovnosti anterior / nªt ±r ± / adj. predný
acoustic / ±ªkust k / adj. akustický, anticipatory / n©t s ªpe tr / =
zvukový regressive
~ phonetics / f±ªnet ks / akustická ~ assimilation / ±©s m ªle n / =
fonetika regressive assimilation
~ spectrum / ªspektr±m / akustické anvil / ª nv l / = incus
spektrum apparatus / © p±ªre t±s / n. aparát,
acoustics / ±ªkust ks / n. akustika ústrojenstvo
adjective / ª dµ kt v / n. adjektívum, approximant / ±ªprÂks m±nt / n.
prídavné meno aproximant
adverb / ª dvb / n. adverbium, approximation / ±©prÂks ªme n / n.
príslovka aproximácia, priblí!enie
affix / ª f ks / n. afix (predpona al. articulate / ªt kj±le t / v. artikulovaĢ,
prípona) vyslovovaĢ
affricate / ª fr k±t / n. afrikáta, articulation / ©t kj±ªle n / n.
polosykavka; adj. afrikovaný, artikulácia, hláskovanie
polozáverový ~ proper / ¥prÂp± / n. tenzia, stredová
airflow / ªe±fl±h / n. prúd vzduchu (hlavná) þasĢ artikulácie hlásky
airstream / ªe±strim / = airflow articulator / ªt kj±le t± / = organ of
allophone / ª l±f±hn / n. alofóna, articulation
variant fonémy articulatory / ªt kj±l±tr / adj.
allophonic / © l±ªfÂn k / adj. alofónny artikulaþný
~ transcription / ©tr nªskr pn / ~ phonetics / f±ªnet ks / artikulaþná
alofónna transkripcia fonetika
alphabet / ª lf±bet / n. abeceda arytenoid / © r ªtin d, ªr tn d / adj.
alveolar / © lv ª±hl± / n. alveolára, arytenoidný
ćasnová spoluhláska; adj. alveolárny, ~ cartilage / ªktl dµ / arytenoidná
ćasnový chrupka, krhlovitá chrupka
~ ridge / r dµ / alveolárny výstupok, ascending / ±ªsend Ì / = rising
ćasno aspirated / ª sp±re t d / adj.
American / ±ªmer k±n / adj. americký aspirovaný, sprevádzaný prídychom
289
aspiration / © sp±ªre n / n. aspirácia, body language / ªbÂd ©l Ìgw dµ / n.

prídych neverbálna komunikácia
assibilation / ±©s b ªle n / n. asibilácia, prostredníctvom gestikulácie a mimiky.
vznik trecieho "umu bound / bahnd / adj. viazaný
assimilation / ±©s m ªle n / n. ~ morpheme / ªmfim / viazaná
asimilácia, spodobovanie morféma
audible / ªd±bl / adj. poþuteĐný brackets / ªbr k ts / n. zátvorky
auditive / ªd±t v / adj. auditívny, breath / bre* / n. dych
auditórny, sluchový British / ªbr t / adj. britský
auditory / ªd tr / = auditive broad transcription / ©brd
~ nerve / ªnv / sluchový nerv tr nsªkr pn / n. "iroká transkripcia
~ ossicles / ªs klz / sluchové kostiþky
phonetics / f±ªnet ks / auditórna C
Canadian / k±ªne d ±n / adj. kanadský

fonetika
auricle / ªr kl / = pinna
Australian / Âªstre l ±n / adj. austrálsky cardinal / ªkd nl / adj. kardinálny
autosemantic word / ]t±hs \m nt k ~ vowel / vah±l / kardinálna samohláska
]wd / = content word cartilage / ªktl dµ / n. chrupka
auxiliary / gªz l ±r / adj. auxiliárny, central / ªsentrl / adj. centrálny,
pomocný stredový
~ vowel / vah±l / stredová samohláska
centring / ªsentr Ì / adj. dostredivý
~ verb / vb / pomocné sloveso
B ~ diphthong / ªd f*ÂÌ / dostredivý

diftong
back / b k / adj. zadný checked / jekt / adj. zatvorený
~ vowel / vah±l / zadná samohláska ~ syllable / ªs l±bl / = closed syllable
basilar / \b z l± / adj. bazálny, clear / kl ± / adj. jasný, mäkký
základný ~ / l / jasné / l /
~ membrane / \membre n / základná click / kl k / n. mĐask
(bazálna) membrána close / kl±hs / adj. zatvorený
bilabial / ba ªle b ±l / adj. bilabiálny, ~ vowel / vah±l / zatvorená samohláska
obojperný, pernoperný closed / kl±hzd / adj. zatvorený
bilateral / ba ªl tr±l / adj. bilaterálny ~ syllable / ªs l±bl / zatvorená slabika
bilingual / ba ªl Ìgwl / adj. bilingválny, close-mid / ©kl±hsªm d / adj.
dvojjazyþný polozatvorený
binary / ¥ba n±r / adj. dvojþlenný, ~ vowel / vah±l / polozatvorená
samohláska
closing / ªkl±hz Ì / adj. zatváravý
binárny
blade / ble d / n. þepeĐ jazyka
290
~ diphthong / ªd f*ÂÌ / zatváravý concave / kÂÌªke v / adj. konkávny,

diftong dutý, vyhĎbený
closure / ªkl±hµ± / n. záver ~ / r / konkávne / r /
cluster / ªklYst± / n. zhluk, zoskupenie, conjunction / k±nªdµYÌkn / n. spojka
skupina consonant / ªkÂns±n±nt / n. konsonant,
~ of consonants / ±v ªkÂns±n±nts / spoluhláska
zhluk spoluhlások consonantal / ©kÂns±ªn ntl / adj.
coalescent / ©k±h±ªlesnt / adj. konsonantický, spoluhláskový
koalescentný, spájavý, splývavý constant / ¥kÂnst±nt / adj. kon"tantný,
~ assimilation / ±©s m ªle n / stály
koalescentná asimilácia constrictive / k±n\str kt v / n.
coarticulation / ©k±ht kj±ªle n / n. konstriktíva, ú!inová hláska; adj.
koartikulácia, kombinácia dvoch konstrikþný, ú!inový
odli"ných miest alebo spôsobov content word / ªkÂntent ]wd / n.
artikulácie plnovýznamové slovo
cochlea / ªkÂkl ± / n. kochlea, slimák contextual / k±nªteksjh±l / adj.
cockney / ªkÂkn / n. londýnsky prízvuk kontextový, ovplyvnený hláskovým
okolím
~ assimilation / ±©s m ªle n /
ni!"ích sociálnych vrstiev
colloquial / k±ªl±hkw ±l / adj. hovorový
combinatory / ªkÂmb n±tr / adj. kontextová asimilácia
kombinatórny continuant / k±nªt njh±nt / n.
~ variant / ªve±r ±nt / kombinatórny kontinuanta
(poziþný/závislý) variant contoid / ªkÂnt d / n. kontoid
comma / ªkÂm± / n. þiarka contracted form / k±nªtr kt d ªfm /
comparative / k±mªp r±t v / adj. n. stiahnutý tvar
komparatívny, porovnávací
compound / ªkÂmpahnd / n. zlo!ené D
slovo
dark / dk / adj. tmavý, tvrdý,
complementary / ©kÂmpl ªmentr / adj.
velarizovaný
komplementárny, doplnkový
~ / l / tmavé / l /
~ distribution / ©d str ªbjun /
dB = decibel
komplementárna distribúcia decibel / ªdes bel / n. decibel
compression / k±mªpren / n.
declarative / d ªkl r±t v / adj.
kompresia, stlaþenie
oznamovací
~ stage / ste dµ / kompresné "tádium
~ sentence / ªsent±ns / oznamovacia
commutation / ©kÂmjhªte n / n.
veta
komutácia, substitúcia dental / ªdentl / n. dentála, zubná
~ test / test / komutaþný test,
spoluhláska; adj. dentálny, zubný
substituþná metóda
291
descending / d ªsend Ì / = falling elide / ªla d / v. vypustiĢ, vynechaĢ,

determiner / d ªtm n± / n. determinant nevysloviĢ
devoice / ©diªv s / v. zbaviĢ znelosti, elision / ªl µn / n. elízia, vypustenie,
desonorizovaĢ vynechanie
devoicing / ©diªv s Ì / n. zbavenie ellipsis / ªl ps s / = elision
znelosti, desonorizácia emphasis / ªemf±s s / n. emfáza, dôraz
diacritics / ©da ±ªkr t ks / n. diakritické empty word / ªempt ]wd / = form
znamienka word
dialect / ªda ±lekt / n. dialekt, náreþie English / ª Ìgl / n. angliþtina
diaphragm / ªda ±fr m / n. bránica epiglottis / ©ep ªglÂt s / n. hrtanová
difference limen / \d fr±ns ]la men / príchlopka
diferenþný prah equipollent / ©ikw ªpÂl±nt / adj.
diphthong / ªd f*ÂÌ / n. diftong, ekvipolentný
dvojhláska, kĎzavá samohláska Eustachian tube / ju]ste n \tjub / n.
discourse / ªd sks / n. diskurz, Eustachova trubica
jazykový prejav exhalation / ©eks±ªle n / n. exhalácia,
distinctive / d ªst nkt v / adj. výdych
di"tinktívny, rozli"ovací exhale / eksªhe l / v. exhalovaĢ,
~ feature / ªfij± / di"tinktívny príznak, vydýchnuĢ
rozli"ovacia vlasnosĢ
~ speech sound / ªspijsahnd / F
di"tinktívna hláska, fonéma
distribution / ©d str ªbjun / n. facultative / ¥f klt±t v / adj.
distribúcia, rozmiestnenie, mo!ný výskyt fakultatívny
drawl / ªdrl / n. pomalá Ģahavá reþ ~ variant = free variant
duration / djhªre n / n. trvanie falling / ªfl Ì / adj. klesavý
~ diphthong / ªd f*ÂÌ / klesavý diftong
E ~ tone / t±hn / klesavý tón
fall-rise tone / ªflra z ©t±hn / n.
ear canal / ª ± k±n l / = meatus klesavostúpavý tón
eardrum / ª ±drm / n. (u"ný) bubienok flap / fl p / n. jednokmitová alveolárna
hláska
echo / ªek±h / n. ozvena, opakovanie;
foot / fht / = rhythm unit
adj. opakovací
form word / ªfm ]wd / n.
~ question / ªkwesjn / opakovacia
neplnovýznamové slovo
otázka
formant / ªfm±nt / n. formant
egressive / ªgres v / adj. egresívny,
fortis / ªft s / adj. fortisový, napätý,
produkovaný výdychom
ejective / ªdµekt v / n. ejektíva; adj. silný
free / fri / adj. voĐný
ejektívny, produkovaný nad glottisom
292
~ morpheme / ªmfim / voĐná glottal / ªglÂtl / adj. glotálny, hrtanový,

morféma hlasivkový
~ syllable / ªs l±bl / = open syllable ~ stop / stÂp / (hrtanový) ráz,
~ variant / ªve±r ±nt / voĐný hlasivková explozíva, glotický pulz
(fakultatívny/nezávislý) variant ~ tone / t±hn / glotálny tón
~ variation / ©ve±r ªe n / voĐná variácia glottalic / glÂªt l k / adj. glotalický,
frequency / ªfrikw±ns / n. frekvencia, glotisový
glottalized / ªglÂt±la zd / adj.
kmitoþet
fricative / ªfr k±t v / n. frikatíva, glotalizovaný, abruptívny
glottis / ªglÂt s / n. glottis, hlasivková
spiranta, trená hláska; adj. frikatívny,
trený, spirantný "trbina
friction / ªfr kn / n. frikcia, trenie gradual / ¥gr dµh±l / adj. graduálny
frictionless / ªfr knl±s / adj. grammatical word / gr±ªm t kl ]wd /
nefrikatívny, netrený = form word
front / frYnt / n. predok, predná þasĢ;
adj. predný H
~ vowel / vah±l / predná samohláska
full word / ªfhl ]wd / = content word hair cell / \he±sel / vlásková nervová
function word / ªfYÌkn ]wd / = form bunka
half-close / ©hfªkl±hs / = close-mid
word
fundamental / ©fYnd±ªmentl / adj. half-open / ©hfª±hp±n / = open-mid
základný, hlavný, fundamentálny hammer / ªh m± / = malleus
~ frequency / ªfrikw±ns / základná hard / hd / adj. tvrdý
frekvencia ~ palate / ªp l±t / tvrdé podnebie,
predné podnebie
G harmonics / hªmÂn ks / n. parciálne
(harmonické) tóny, harmonické
gemination / ©dµem±ªne n / n. frekvencie
geminácia, zdvojenie spoluhlásky head / hed / n. hlava tónickej jednotky
Germanic / dµ±ªm n k / adj. height / ha t / n. vý"ka
germánsky Hertz / hts / n. hertz
~ language / ªl Ìgw dµ / germánsky Hz = Hertz
high / ha / adj. vysoký
jazyk
glide / gla d / n. 1. kĎzavá hláska 2. ~ vowel / vah±l / vysoká samohláska
kĎzavý pohyb homophone / ªhÂm±f±hn / n. homofón,
gliding / ªgla d Ì / adj. kĎzavý, rovnozvuþné slovo
s kvalitatívnymi zmenami homophonous / h±ªmÂf±n±s / adj.
~ vowel / vah±l / = diphthong homofonický, homofónny, rovnozvuþný
293
homophony / h±ªmÂf±n / n. ~ sentence / ªsent±ns / opytovacia veta

homofónia, rovnozvuþnosĢ intonation / © nt±ªne n / n. intonácia,
homorganic / ©hÂmªg n k / adj. melódia
homorganický, s rovnakým miestom intrinsic / n\tr ns k / adj. vlastný,
artikulácie vnútorný, prirodzený
~ duration / dj\re n / prirodzená dĎ!ka
I (dĎ!ka neovplyvnená zvukovým okolím)
intrusive / nªtrus v / adj. intruzívny,
imperative / mªper±t v / adj. prenikavý, vlo!ený
rozkazovací ~ / r / intruzívne / r /
~ sentence / ªsent±ns / rozkazovacia inventory / ª nv±ntr / n. inventár,
veta mno!ina hlások/foném
incomplete / © Ìk±mªplit / adj. neúplný, IPA / ©a piªe / = International Phonetic
nedokonalý, þiastoþný Alphabet
~ plosion / ªpl±hµn / neúplná explózia Irish / ªa r / adj. írsky
incus / ª Ìk±s / n. nákovka irregular / ªregj±l± / adj. nepravidelný
ingressive / Ìªgres v / adj. ingresívny, isolated / ªa s±le t d / adj. izolovaný
vdychový
inhalation / © nh±ªle n / n. inhalácia, J
vdýchnutie, vdych
inhale / nªhe l / v. inhalovaĢ, vdýchnuĢ jaw / dµ / n. þeĐusĢ, sánka
inner / ª n± / adj. vnútorný juncture / ªdµYÌkj± / n. spojenie,
~ ear / ± / vnútorné ucho prechod medzi dvoma hláskami
intelligibility / n©tel dµ±ªb l±t / n. juxtapositional / ©dµYkst±p±ªz nl / =
contextual
~ assimilation / ±©s m ªle n / =
zrozumiteĐnosĢ, pochopiteĐnosĢ
intelligible / nªtel dµ±bl / adj.
zrozumiteĐný, pochopiteĐný contextual assimilation
intensity / nªtens±t / n. intenzita, sila
intermittent / ] nt±\m tnt / adj. K
preru"ovaný
~ closure / \kl / preru"ovaný záver key / ki / register, þasĢ hlasového
International Phonetic Alphabet / rozsahu
© nt±ªn nl f±ªnet k ª lf±bet / n. kinesics / k ªnis ks / n. neverbálna
Medzinárodná fonetická abeceda komunikácia zalo!ená na pohybe rúk,
International Phonetic Association / hlavy, tela, oþí, atć.
© nt±ªn nl f±ªnet k ±©s±hs ªe n / n. kinetic / k ªnet k / adj. kinetický,
Medzinárodná fonetická asociácia pohyblivý
interrogative / © nt±ªrÂg±t v / adj. ~ stress / stres / kinetický prízvuk
opytovací
294
L ~ / r / spojovacie / r /
~ / w / spojovacie / w /
labial / ªle b ±l / n. labiála, pernica; adj. lip / l p / n. pera
labiálny, perný ~ rounding / ªrahnd Ì / zaokrúhlenie
labialization / ©le b ±la ªze n / n. pier, labializácia
labializácia, artikulácia s výraznej"ou loudness / ªlahdn±s / n. hlasitosĢ
úþasĢou pier lower / ªl±h± / adj. spodný
labiodental / ©le b ±hªdentl / n. ~ jaw / dµ / spodná þeĐusĢ
labiodentála, pernozubná spoluhláska; ~ lip / l p / spodná pera
adj. labiodentálny, pernozubný
larynx / ªl r Ìks / n. larynx, hrtan
lungs / lYÌz / n. pĐúca
lateral / ªl trl / n. laterála; adj. laterálny, M

bokový
~ plosion / ªpl±hµn / laterálna explózia mallet / ªm l t / = malleus
~ release / r ªlis / = lateral plosion malleus / ªm l ±s / n. maleus, kladivko
lateralization / ©l tr±la ªze n / n. mandible / ªm nd bl / = lower jaw
lateralizácia, "pecializácia mozgových manner / ªm n± / n. spôsob
hemisfér ~ of articulation / ±v ©t kj±ªle n /
lax / l ks / adj. laxný, voĐný, uvoĐnený, spôsob artikulácie
nenapätý meatus / m ªe t±s / n. zvukovod
~ vowel / vah±l / uvoĐnená samohláska mel / mel / n. mel, jednotka subjektívnej
length / leÌ* / n. dĎ!ka vý"ky tónu
length-mark / ªleÌ*mk / n. znak middle / ªm dl / adj. stredný
dĎ!ky ~ ear / ª ± / n. stredné ucho
lenis / ªlin s / adj. lenisový, slabý,
minimal pair / ©m n ml ªpe± / n.
nenapätý
minimálna dvojica (pár)
level tone / ªlevl t±hn / stabilný tón
mixed / m kst / adj. zmie"aný, neurþitý
lexical word / ªleks kl ]wd / = content ~ vowel / vah±l / = schwa
word modulating / ªmÂdj±le t Ì / adj.
lingual / ªl Ìgwl / adj. lingválny, modulaþný, obmieĖací, tónovací
jazykový
linguistic / l Ìªgw st k / adj. lingvistický,
~ organ / ªg±nz / modulaþný orgán
molar / ªm±hl± / n. molár, stoliþka
jazykovedný (zub); adj. molárny, stoliþkový
~ phonetics / f±ªnet ks / = phonology
linguistics / l Ìªgw st ks / n. lingvistika,
monophthong / ªmÂn±f*ÂÌ / n.
monoftong, jednoduchá (þistá)
jazykoveda
linking / ªl Ìk Ì / n. viazanie, spájanie;
samohláska, jednohláska
morpheme / ªmfim / n. morféma
adj. spojovací, spájajúci multilateral / ¤mYlt ¥l tr±l / adj.
~ / j / spojovacie / j / multilaterálny
295
muscle / ªmYsl / n. sval O

muscular / ªmYskj±l± / adj. svalový
~ tension / ªtenn / svalové napätie obscuration / ©Âbskjhªre n / n. redukcia
samohlások na / ± /, potláþanie þistoty
samohlások
obscure / ±bªskjh± / adj. nevýrazný,
N
narrow / ªn r±h / úzky, podrobný, zmie"aný

detailný ~ vowel / vah±l / nevýrazná (zmie"aná)
~ transcription / tr nsªkr pn / úzka samohláska
obstruent / ªÂbstrh±nt / n. ob"truenta,
transkripcia
"umová hláska
occlusion / ±\klu n / n. oklúzia, záver
nasal / ªne zl / n. nazála, nosová hláska,
occlusive / ±\klus v / n. okluzíva,
nosovka; adj. nazálny, nosový
~ cavity / ªk v±t / nosová dutina
záverová hláska
~ plosion / ªpl±hµn / nosová explózia
off-glide / ¥Âfgla d / n. detenzia,
~ release / r ªlis / = nasal plosion
~ twang / tw Ì / nosová výslovnosĢ
závereþná fáza artikulácie hlásky
omission / ±ªm n / = elision
nasalization / ©ne z±la ªze n / n.
on-glide / ¥ÂÌgla d / n. intenzia,
nazalizácia, úþasĢ nosovej dutiny pri poþiatoþná fáza artikulácie hlásky
artikulácii
open / ª±hp±n / adj. otvorený
nasalize / ªne z±la z / v. nazalizovaĢ
~ syllable / ªs l±bl / otvorená slabika
nasalized / ªne z±la zd / adj.
open-mid / ©±hp±nªm d / adj.
nazalizovaný
neuron / ªnjh±rn / n. neurón polootvorený
~ vowel / vah±l / polootvorená
neutral / ªnjutr±l / adj. neutrálny, samohláska
nevýrazný opennes / ª±hp±nn±s / n. otvorenosĢ
neutralizable / ªnjutr±la z±bl / adj.
opposition / ¤Âp±¥z n / n. protiklad
neutralizovateĐný
oral / ªr±l / adj. orálny, ústny
noise / n z / n. hluk, "um; adj. "umový
non-pulmonic / ©nÂnpYlªmÂn k / adj. ~ cavity / ªk v±t / ústna dutina
nepulmonálny, nepĐúcny ~ plosion / ªpl±hµn / ústna explózia
non-standard / ©nÂnªst nd±d / adj. ~ release / r ªlis / = oral plosion
organ / ªg±n / n. orgán
~ of articulation / ±v ©t kj±ªle n /
ne"tandardný
noun / nahn / n. podstatné meno
nucleus / ªnjukl ±s / n. 1. = peak 2. = artikulaþný orgán
tonic syllable ~ of Corti / ±v \kt / Cortiho orgán
~ of speech / ±v ªspij / = organ of
articulation
orthography / ª*Âgr±f / n. ortografia,
pravopis
296
ossicles / \s klz / n. kostiþky (see ~ apparatus / © p±ªre t±s / fonaþný

auditory) aparát (ústroj)
outer / ªaht± / adj. vonkaj"í phoneme / ªf±hnim / n. fonéma,
~ ear / ± / vonkaj"ie ucho základná fonologická jednotka
oval / ª±hvl / adj. oválny rozli"ujúca význam
~ window / \w nd±h / phonemic / f±hªnim k / adj. fonemický
overtones / ª±hv±t±hnz / = harmonics ~ transcription / tr nsªkr pn /
fonemická transkripcia
P phonemics / f±hªnim ks / = phonology
phonetic / f±ªnet k / adj. fonetický,
palatal / ªp l±tl / n. palatála, zvukový
tvrdopodnebná (predopodnebná) hláska; ~ transcription / tr nsªkr pn /
adj. palatálny, tvrdopodnebný fonetická transkripcia
palatalization / ©p l±t±la ªze n / n. phonetician / ©f±hn±ªt n / n. fonetik
palatalizácia, presunutie artikulácie do phonetics / f±ªnet ks / n. fonetika
oblasti tvrdého podnebia phonological / ©fÂn±ªlÂdµ kl / adj.
palate / ªp l±t / n. palátum, podnebie
fonologický
palato-alveolar / ©p l±t±h© lv ª±hl± / phonology / f±ªnÂl±dµ / n. fonológia
adj. palatoalveolárny, alveopalatálny, phonotactics / ©f±hn±ªt kt ks / n.
ćasnovopodnebný
paralinguistic / ªp r±l Ìªgw st k / adj.
fonotaktika, sekvenþné radenie foném
pinna / ªp n± / n. u"nica, aurikula
paralingvistický, mimojazykový
pitch / p j / n. vý"ka tónu
pause / pz / n. pauza, ticho; v.
zastaviĢ sa pitch-range / ªp jre ndµ / n. hlasový
peak / pik / n. jadro slabiky, hrebeĖ (tónový) rozsah
place / ple s / n. miesto
~ of articulation / ±v ©t kj±ªle n /
slabiky, vrchol slabiky
perseverative / p±ªsevr±t v / =
progressive miesto artikulácie
~ assimilation / ±©s m ªle n / = plosion / ªpl±hµn / n. explózia, fáza
progressive assimilation artikulácie okluzív
pharyngeal / ©f r nªdµi±l / adj. plosive / ªpl±hs v / n. explozíva; adj.
faryngálny, hltanový explozívny
pharynx / ªf r Ìks / n. farynx, hltan post-alveolar / ©p±hst lv ª±hl± / adj.
phon / fn / n. fón, jednotka hladiny postalveolárny, zaćasnový
hlasitosti pre-head / ªprihed / n. slabiky tónickej
phonation / f±hªne n / n. fonácia, jednotky nachádzajúce sa pred hlavou
tvorenie hlasu tónickej jednotky
phonatory / ªf±hn±tr / adj. fonaþný, prefix / ªprif ks / n. prefix, predpona
tvoriaci hlas preposition / ©prep±ªz n / n. predlo!ka
297
pressure / ªpre± / n. tlak Received Pronunciation / r ©sivd

primary / ªpra mr / adj. primárny, pr±©nYns ªe n / n. Prijatá výslovnosĢ
hlavný (výslovnostný úzus vo VeĐkej Británii)
~ stress / stres / primárny prízvuk reciprocal / r ªs pr±kl / adj. reciproþný,
privative / ªpr v±t v / adj. privatívny vzájomný
progressive / pr±ªgres v / adj. ~ assimilation / ±©s m ªle n / =
progresívny, postupujúci, perseveraþný coalescent assimilation
~ assimilation / ±©s m ªle n / reduced / r ªdjust / adj. redukovaný,
progresívna asimilácia oslabený
prominence / ªprÂm n±ns / n. ~ schwa / ∫w / redukovaná "va
prominencia regressive / r ªgres v / adj. regresívny,
pronoun / ªpr±hnahn / n. zámeno anticipaþný, spätný
pronunciation / pr±©nYns ªe n / n. ~ assimilation / ±©s m ªle n / regresívna
výslovnosĢ asimilácia
proportional / pr±¥pnl / adj. Reissner!s membrane / ]re sn±z
proporcionálny \membre n / Reissnerova membrána
prosodic / pr±ªsÂd k / adj. prozodický release / r ªlis / = plosion
~ features / ªfij±z / prozodické resonant / ªrez±n±nt / adj. rezonantný,
príznaky, prozódia ozvuþný
prosody / ªprÂs±d / n. prozódia ~ cavity / ªk v±t / rezonantná dutina
pulmonic / pYlªmÂn k / adj. respiration / ©resp±ªre n / n. respirácia,
pulmonálny, pĐúcny dýchanie
pure / pjh± / adj. þistý, bez respiratory / r ªsp r±tr / adj.
kvalitatívnych zmien respiratórny, týkajúci sa dýchania
~ vowel / vah±l / = monophthong retroflex / ªretr±fleks / adj. retroflexný
~ / r / retroflexné / r /
Q rhotacized / ªr±ht±sa zd / adj.
rotacizovaný, majúci kvalitu / r /
quality / ªkwÂl±t / n. kvalita, akosĢ ~ vowel / vah±l / rotacizovaná
zvuku, zafarbenie hlasu samohláska
quantity / ªkwÂnt±t / n. kvantita, rhotic / ªr±ht k / adj. rotický,
trvanie, dĎ!ka zvuku pou!ívajúci / r / presne podĐa pravopisu
rhoticity / r±hªt s±t / n. roticita
R rhythm / ªr sm / n. rytmus
~ unit / ªjun t / rytmický takt
rarefaction / ©re±r±ªf kn / n. rhythmicality / ©r sm ªk l±t / n.
rarefakcia, zrednutie rytmickosĢ
rise-fall tone / ªra zfl ©t±hn / n.
stúpavoklesavý tón
298
rising / ªra z Ì / n. stúpavý spectrograph / ªspektr±gr f / n.

~ diphthong / ªd f*ÂÌ / stúpavý diftong spektrograf
~ tone / t±hn / stúpavý tón spectrum / ªspektr±m / n. spektrum
roll / r±hl / = trill speech / spij / n. reþ, reþový prejav;
root / rut / n. koreĖ slova adj. reþový
rounded / ªrahnd d / adj. zaokrúhlený ~ organ / ªg±n / = organ of
~ vowel / vah±l / zaokrúhlená articulation
samohláska ~ sound / sahnd / hláska
RP / ©ªpi / = Received Pronunciation spelling / ªspel Ì / = orthography
stapes / ªste piz / n. strmienok
S stirrup / ªst r±p / = stapes
stop / stÂp / = plosive
schwa / w / n. "va, zmie"aná stress / stres / n. prízvuk, silové
(neurþitá) samohláska zvýraznenie slabiky
Scottish / ªskÂt / adj. "kótsky stressed / strest / adj. prízvuþný
secondary / ªsek±ndr / adj. ~ syllable / ªs l±bl / prízvuþná slabika
sekundárny, vedĐaj"í, druhotný stricture / ªstr kj± / n. zú!enie, ú!ina
~ stress / stres / sekundárny prízvuk strong / strÂÌ / adj. silný
segment / ªsegm±nt / n. segment ~ form / fm / silná forma,
semiocclusive / ]sem ±\klus v / n. neredukovaný tvar
semiokluzíva, polozáverová hláska suffix / ªsYf ks / n. sufix, prípona
semivowel / ªsem vah±l / n. supra-glottal / ©supr±ªglÂtl / adj.
polosamohláska supraglotálny, nadglotálny
sentence / ªsent±ns / n. veta suprasegmental / ©supr±segªmentl /
sibilant / ªs b l±nt / n. sibilanta, sykavka adj. suprasegmentálny
similitude / s ªm l tjud / = allophonic syllabic / s ªl b k / adj. slabikotvorný
assimilation syllabicity / ©s l±ªb s±t / n.
soft / sÂft / adj. mäkký slabikotvornosĢ
~ palate / ªp l±t / mäkké podnebie, syllable / ªs l±bl / n. slabika
velum synapsis / s \n ps s / n. synapsia
sonorant / ªsÂn±r±nt / n. sonóra, synsemantic word / ]s ns \m nt k
sonoranta, ne"umová hláska ]wd / = form word
sonority / s±ªnÂr±t / n. sonorita,
sonórnosĢ, zvuþnosĢ T
sound / sahnd / n. zvuk
~ wave / we v / zvuková vlna tail / te l / n. slabiky tónickej jednotky
spectrogram / ªspektr±gr m / n. nachádzajúce sa za hlavou tónickej
spektrogram jednotky
tamber / ªt mb± / = quality
299
tap / t p / = flap ~ lip / l p / horná pera

teeth / ti* / n. zuby utterance / ªYtr±ns / n. prehovor
tense / tens / adj. napätý uvula / ªjuvj±l± / n. uvula, þapík
~ vowel / vah±l / napätá samohláska uvular / ªjuvj±l± / n. uvulára, þapíková
ternary / ¥tn±r / adj. trojþlenný, spoluhláska; adj. uvulárny, þapíkový
ternárny
throat / *r±ht / n. hrdlo V
thyroid / ª*a r d / adj. "títny
~ cartilage / ªktl dµ / "títna chrupavka valley / ªv l / = trough
timbre / ªt mb± / = quality variant / ªve±r ±nt / n. variant
tone / t±hn / n. tón, tonéma velar / ªvil± / n. velára, mäkkopodnebná
tone-unit / ªt±hn©jun t / n. tónická spoluhláska; adj. velárny,
jednotka, melodéma zadopodnebný, mäkkopodnebný
tongue / tYÌ / n. jazyk velarization / ©vil±ra ªze n / n.
tonic / ªtÂn k / tónický velarizácia, presunutie artikulácie do
~ stress / stres / tónický prízvuk oblasti mäkkého podnebia
~ syllable / ªs l±bl / tónická slabika velum / ªvil±m / = soft palate
trachea / tr±ªki± / n. trachea, vibration / va ªbre n / n. vibrácia,
priedu"nica chvenie, kmitanie
transcription / tr nsªkr pn / n. vocal / ªv±hkl / adj. vokálny, hlasový,
transkripcia, prepis samohláskový
transsibilation / ©tr nss b±ªle n / n. ~ cords / kdz / = vocal folds
transsibilácia, zmena jednej sykavky na ~ folds / f±hldz / hlasivky
druhú vocalic / v±hªk l k / = vocal
trill / tr l / n. hrþivá (kmitavá) hláska vocoid / ªv±hk d / n. vokoid
trilled / tr ld / adj. hrþivý, kmitavý voice / v s / n. hlas
triphthong / ªtr f*ÂÌ / n. triftong, voiced / v st / adj. znelý
trojhláska voiceless / ªv sl±s / adj. neznelý
trough / trÂf / n. okraj slabiky, svah vowel / vah±l / n. samohláska, vokál
slabiky
W
U
waveform / ªwe vfm / n. graficky
unaspirated / Ynª sp±re t d / adj.
znázornená zvuková vlna
neaspirovaný, bez prídychu weak / wik / adj. slabý
unstressed / Ynªstrest / adj.
~ form / fm / slabá forma, redukovaný
neprízvuþný, bez prízvuku tvar
upper / ªYp± / adj. horný windpipe / ªw ndpa p / = trachea
~ jaw / dµ / horná þeĐusĢ word / wd / n. slovo
300
Appendix A
A Feature Matrix of RP General Vowels
English vowel phonemes

i u ;e
Length short + ! + ! + ! + ! + ! + +
long ! + ! + ! + ! + ! + ! !
Voicing voiced + + + + + + + + + + + +
voiceless ! ! ! ! ! ! ! ! ! ! ! !
Lip rounded ! ! + + + + ! ! ! ! ! !
rounding spread + + ! ! ! ! ! ! ! ! + !
neutral ! ! ! ! ! ! + + + + ! +
Horizontal front ! + ! ! ! ! ! ! ! ! + +
position of front-central + ! ! ! ! ! ! ! ! ! ! !
the raised central ! ! ! ! ! ! + ! + + ! !
part of the central-back ! ! + ! ! ! ! + ! ! ! !
tongue back ! ! ! + + + ! ! ! ! ! !
close ! + ! + ! ! ! ! ! ! ! !
Vertical close-mid + ! + ! ! + ! ! ! ! ! !
position ! mid ! ! ! ! ! ! ! ! + ! + !
openness open-mid ! ! ! ! ! ! + ! ! + ! !
open ! ! ! ! + ! ! + ! ! ! +
Muscular lax + ! + !
tension tense ! + ! +
301
Appendix B
A Feature Matrix of RP General Consonants
English consonant phonemes

p b k g t d f v * s s z µ h j dµ m n Ì l r j w
voiced ! + ! + ! + ! + ! + ! + ! + ! ! + + + + + + + +
voiceless + ! + ! + ! + ! + ! + ! + ! + + ! ! ! ! ! ! ! !
bilabial + + ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! + ! ! ! ! ! +
labiodental ! ! ! ! ! ! + + ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !
dental ! ! ! ! ! ! ! ! + + ! ! ! ! ! ! ! ! ! ! ! ! ! !
alveolar ! ! ! ! + + ! ! ! ! + + ! ! ! ! ! ! + ! + ! ! !
post-alveolar ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! + ! !
palato-alveolar ! ! ! ! ! ! ! ! ! ! ! ! + + ! + + ! ! ! ! ! ! !
palatal ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! + !
velar ! ! + + ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! + ! ! ! +
glottal ! ! ! ! ! ! ! ! ! ! ! ! ! ! + ! ! ! ! ! ! ! ! !
plosive + + + + + + ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !
affricate ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! + + ! ! ! ! ! ! !
fricative ! ! ! ! ! ! + + + + + + + + + ! ! ! ! ! ! ! ! !
nasal ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! + + + ! ! ! !
lateral ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! + ! ! !
approximant ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! + + +
obstruent + + + + + + + + + + + + + + + + + ! ! ! ! ! ! !
sonorant ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! + + + + + + +
fortis + ! + ! + ! + ! + ! + ! + ! + + ! ! ! ! ! ! ! !
lenis ! + ! + ! + ! + ! + ! + ! + ! ! + + + + + + + +
continuant ! ! ! ! ! ! + + + + + + + + + ! ! + + + + + ! !
sibilant ! ! ! ! ! ! ! ! ! ! + + + + ! + + ! ! ! ! ! ! !
302
References
References
Abercrombie, D. (1964). English Phonetic Texts. Faber and Faber Ltd.

Abercrombie, D. (1967). Elements of General Phonetics. Edinburgh University Press.
Akmajian, A. et al. (1995). Linguistics: An Introduction to Language and
Communication, 4th edition. Massachusetts: The MIT Press.
Arnold, G. F. and Tooley, O. M. (1972). Say it with Rhythm. Harlow: Longman.
Baart, J. (1999). A Field Manual of Acoustic Phonetics. SIL International.
Baker, A. (1982). Introducing English Pronunciation. Cambridge University Press.
Barber, C. (1993). The English Language: A Historical Introduction. Cambridge
University Press.
Bartek, H. (1944). Správna výslovnosĢ slovenská. Bratislava ! Turþiansky Sv. Martin:
Kníhtlaþiarsky úþ. spolok v Turþianskom Sv. Martine.
Bauer, L. (1983). English Word-formation. Cambridge University Press.
Bauer, L. (1984). "Linking /r/ in RP: some facts#, Journal of the International
Phonetic Association 14: 74!79.
Bradford, B. (1988). Intonation in Context. Cambridge University Press.
Brazil, D. (1985). The communicative Value of Intonation in English. Birmingham:
Bleak House Books, English Language Research.
Broshnan, L. and Malmberg, B. (1970). Introduction to Phonetics. Cambridge
University Press.
Bujalka, A., Balá$, P., Rýzková, A. (1998). Slovenský jazyk I: Zvuková stránka
jazyka, Náuka o slovnej zásobe. Univerzita Komenského Bratislava.
Calvert, D. (1986). Descriptive Phonetics, 2nd edition. New York: Thieme Inc.
Carrell, J. and Tiffany, W. R. (1977). Phonetics: Theory and Application, 2nd
edition. New York: McGraw-Hill Book Company.
Clark, J. and Yallop, C. (1995). An Introduction to Phonetics and Phonology, 2nd
edition. Oxford: Blackwell.
Cruttenden, A. (1986). Intonation. Cambridge University Press.
Crystal, D. (1991). A Dictionary of Linguistics and Phonetics, 3rd edition. Oxford:
Blackwell Publishers Ltd.
Crystal, D. (1969). Prosodic Systems and Intonation in English. Cambridge
University Press.
Crystal, D. (1975). The English Tone of Voice. Bristol: Edward Arnold.
Dalton, C. and Seidlhofer, B. (1994). Pronunciation. Oxford University Press.
Denes, P. and Pinson, E. (1963). The Speech Chain. New York: Bell Telephone
Laboratories.
Dvonþová, J. (1980). Fyziologická fonetika. Bratislava: Slovenské pedagogické
nakladateĐstvo.
Dvonþová, J., Jenþa, G. and KráĐ, Á. (1969) Atlas slovenských hlások. Bratislava:
VydavateĐstvo SAV.
303
References
Ellis, A. and Beattie, G. (1986). The Psychology of Language and Communication.

London: Weidenfeld and Nicolson.
Flanagan, J. (1972). Speech Analysis Synthesis and Perception, 2nd edition. Berlin:
Springer-Verlag.
Fox, A. (1978). "To "r# is human? Intrusive remarks on a recent controversy#, Journal
of the International Phonetic Association 8: 72!74.
Fry, D. B. (1979). The Physics of Speech. Cambridge University Press.
Fudge, E. C. (1973). Phonology. Harmondsworth: Penguin Books.
Fudge, E. C. (1977). "Long and short [n] in one Southern British speaker#s English#,
Journal of the International Phonetic Association 7: 55!65.
Gimson, A. C. and Cruttenden, A. (1994). Gimson!s Pronunciation of English, 5th
edition. London: Edward Arnold.
Graddol, D., Cheshire, J. and Swann, J. (1994). Describing Language, 2nd edition.
Buckingham: Open University Press.
Hála, B. (1929). Základy spisovné výslovnosti slovenské a srovnání s výslovností
þeskou. Praha: Státní tiskárna.
Hála, B. (1960). Fonetické obrazy hlásek. Praha: SPN.
Handbook of the International Phonetic Association (1999). Cambridge University
Press.
Holþík, J. (1980). "Skracovanie dlhých samohlások v spisovnej slovenþine#,
Jazykovedný þasopis 31: 147!151.
Holþík, J. (1982). "O skracovaní dlhých samohlások v spisovnej slovenþine#, Kultúra
slova 16: 41!43.
Hughes, A. and Trudgill, P. (1979). English Accents and Dialects: An Introduction
to Social and Regional Varieties of British English. London: Edward Arnold.
Isaþenko, A. V. (1968). Spektrografická analýza slovenských hlások. VydavateĐstvo
Slovenskej akadémie vied Bratislava.
Jones, D. (1960). An Outline of English Phonetics, 9th edition. Cambridge University
Press.
Jones, D. (1967). The Phoneme: its Nature and Use, 3rd edition. Cambridge: Heffer.
Jones, D. (1997). Cambridge English Pronouncing Dictionary, 15th edition. Edited
by Peter Roach and James Hartman. Cambridge University Press.
Keating, P. (1984). "Phonetic and phonological representation of stop consonant
voicing#, Language 60: 286!318.
Kenyon, J. and Knott, T. A. (1953). A Pronouncing Dictionary of American
English. Springfield, Mass: G & C Merriam Company.
Kim, C. W. (1970). "A theory of aspiration#, Phonetica 21: 107!116.
Kingdon, R. (1958). The Groundwork of English Intonation. London: Longmans,
Green and Co Ltd.
Kingdon, R. (1965). The Groundwork of English Stress, third impression (with
corrections). London: Longmans, Green and Co Ltd.
KráĐ, Á. (1965). "ýinnosĢ jazyka pri artikulácii slovenských perných spoluhlások
v súvislej reþi#, Jazykovedný þasopis 16: 3!19.
304
References
KráĐ, Á. (1966). "K otázke podnebnohltanového záveru pri výslovnosti samohlások

v slovenþine v porovnaní s þe%tinou a inými jazykmi#, Slavica slovaca 1: 26!37.
KráĐ, Á. (1968). "Fonetická systematika slovenských samohlások a dvojhlások#,
Slovenská reþ 33: 271!280.
KráĐ, Á. (1974a). "Fonetická povaha a systém slovenských samohlások a dvojhlások#,
Studia academica slovaca 3: 51!66.
KráĐ, Á. (1974b). "Samohláska ä#, Kultúra slova 8: 139!145.
KráĐ, Á. (1974c). "Slovenské samohlásky v konfrontácii so samohláskami iných
jazykov#, Studia academica slovaca 3: 67!81.
KráĐ, Á. (1974d). "Spoluhláska v a jej varianty I.#, Kultúra slova 8: 292!296.
KráĐ, Á. (1974e). "Spoluhláska v a jej varianty II.#, Kultúra slova 8: 338!344.
KráĐ, Á. (1974f). "VýslovnosĢ samohláskových skupín#, Kultúra slova 8: 245!250.
KráĐ, Á. (1975a). "E%te raz o zdvojených spoluhláskach#, Kultúra slova 9: 272!277.
KráĐ, Á. (1975b). "Spoluhlásky t, d, n, l ! Ģ, ć, Ė, Đ v domácich slovách#, Kultúra
slova 9: 46!51.
KráĐ, Á. (1975c). "Spoluhlásky t, d, n, l ! Ģ, ć, Ė, Đ v prevzatých slovách#, Kultúra
slova 9: 101!106.
KráĐ, Á. (1975d). "Tvrdé a mäkké spoluhlásky#, Kultúra slova 9: 9!15.
KráĐ, Á. (1975e). "Z problematiky intonácie#, Kultúra slova 9: 300!306.
KráĐ, Á. (1975f). "Zdvojené spoluhlásky I#, Kultúra slova 9: 148!154.
KráĐ, Á. (1975g). "Zdvojené spoluhlásky IIa#, Kultúra slova 9: 200!203.
KráĐ, Á. (1975h). "Zdvojené spoluhlásky IIb#, Kultúra slova 9: 238!242.
KráĐ, Á. (1975i). "Znelostná asimilácia#, Kultúra slova 9: 334!339.
KráĐ, Á. (1976a). "Preberanie cudzích slov a znelostná asimilácia#, Kultúra slova 10:
197!202.
KráĐ, Á. (1976b). "Spoluhláska n#, Kultúra slova 10: 306!312.
KráĐ, Á. (1976c). "Spoluhlásky [x ! , ! h]#, Kultúra slova 10: 145!149.
KráĐ, Á. (1976d). "Systematika znelostnej asimilácie I#, Kultúra slova 10: 14!17.
KráĐ, Á. (1976e). "Systematika znelostnej asimilácie II#, Kultúra slova 10: 48!52.
KráĐ, Á. (1976f). "Systematika znelostnej asimilácie III#, Kultúra slova 10: 74!78.
KráĐ, Á. (1976g). "Systematika znelostnej asimilácie IV#, Kultúra slova 10: 108!113.
KráĐ, Á. (1976h). "Znelostná asimilácia v prevzatých slovách I#, Kultúra slova 10:
241!244.
KráĐ, Á. (1976i). "Znelostná asimilácia v prevzatých slovách II#, Kultúra slova 10:
269!274.
KráĐ, Á. (1985a). "O fonetike a fonológii a ich vzĢahu k materiálnym a ideálnym
systémom#, Jazykovedný þasopis 36: 5!12.
KráĐ, Á. (1985b). "Predmet a definícia fonetiky a fonológie na podklade vzĢahov
materiálnych a ideálnych systémov#, Jazykovedný þasopis 36: 112!119.
KráĐ, Á. (1988). Pravidlá slovenskej výslovnosti, 2nd edition. Slovenské
pedagogické nakladateĐstvo Bratislava.
KráĐ, Á. and Sabol, J. (1989). Fonetika a fonológia. Slovenské pedagogické
nakladateĐstvo Bratislava.
305
References
Ladefoged, P. (1960). "The value of phonetic statements#, Language 36: 387!396.

Ladefoged, P. (1975). A Course in Phonetics. New York: Harcourt Brace
Jovanovich Inc.
Ladefoged, P. (1980). "What are linguistic sounds made of?#, Language 56: 485!502.
Ladefoged, P. (1996). Elements of Acoustic Phonetics, 2nd edition. Chicago: The
University of Chicago Press.
Ladefoged, P., Cochran, A. and Disner, S. (1977). "Laterals and trills#, Journal of the
International Phonetic Association 7: 46!54.
Laver, J. (1994). Principles of Phonetics. Cambridge University Press.
Lehiste, I. (1970). Suprasegmentals. Cambridge, Mass.: MIT Press.
Lenhardt, J. (1981). Konfrontaþná fonetika anglických a slovenských hlások.
Bratislava: Univerzita Komenského.
Lewis, J. Windsor (1975). "Linking /r/ in the General British pronunciation of
English#, Journal of the International Phonetic Association 5: 37!42.
Lewis, J. Windsor (1977). "The r-link business ! a reply#, Journal of the
International Phonetic Association 7: 28!31.
Lisker, L. (1957). "Closure duration and the intervocalic voiced-voiceless distinction
in English#, Language 33: 42!49.
Lisker, L. and Abramson, A. (1964). "A Cross-Language Study of Voicing in Initial
Stops: Acoustical Measurements#, Word 20: 384!422.
Li%ka, J. (1964). "O niektorých problémoch slovenskej spisovnej výslovnosti#,
Li%ka, J. (1971a). "Palatálna laterála v slovenþine#, Jazykovedný þasopis 22: 128!156.
Li%ka, J. (1971b). "Slovenské vibranty r, Ě#, Jazykovedné "túdie 11: 262!287.
Longman Dictionary of Contemporary English (1995). 3rd edition. Harlow.
Lyons, J. (1992). Language and Linguistics. Cambridge University Press.
Mackenzie, M. (1967). Modern English Pronunciation Practice. London: Longman.
Maidment, J.A. (1994). Estuary English: Hybrid or Hype? 4th New Zealand
Conference on Language & Society, Lincoln University, Christchurch, New
Zealand.
Malmberg, B. (1963). Phonetics. New York: Dover Publications, Inc.
Menhard, Z. (1991). A Workbook in English Phonetics. Praha: Univerzita Karlova.
O#Connor, J. D. (1973). Phonetics. Harmondsworth: Penguin Books.
O#Connor, J. D. (1980). Better English Pronunciation, 2nd edition. Cambridge
University Press.
O#Connor, J. D. and Arnold, G. F. (1963). Intonation of Colloquial English, 2nd
impression (with additions and minor corrections). London: Longman Group.
O#Connor, J. D. and Fletcher, C. (1989). Sounds English. Harlow: Longman.
Oxford Advanced Learner!s Dictionary (1995). 5th edition. Oxford.
Paþesová, J. (1992). Fonetika a fonologie souþasné angliþtiny. Brno: Masarykova
univerzita.
Pauliny, E. (1979) Slovenská fonológia. Slovenské pedagogické nakladateĐstvo
Bratislava.
306
References
Pierce, J.R. and David, E. (1958). Man!s World of Sound. New York: Doubleday &
Company, Inc.
Pike, K. (1943). Phonetics. The University of Michigan Press.
Pike, K. (1945). The Intonation of American English. The University of Michigan
Press.
Pike, K. (1947). Phonemics. The University of Michigan Press.
Potter, R., Kopp, G. and Green, H. (1947). Visible Speech. D. New York: Van
Nostrad Company, Inc.
Pulgram, E. (1964). Introduction to the Spectrography of Speech. The Hague:
Mouton.
Quirk, R. et al (1992). A Grammar of Contemporary English. Harlow: Longman.
Roach, P. (1991). English Phonetics and Phonology, 2nd edition. Cambridge
University Press.
Rogerson, P. and Gilbert, J. B. (1990). Speaking Clearly. Cambridge University
Press.
Romportl, M. (1985). Základy fonetiky, 2nd edition. Státní pedagogické
nakladatelství Praha.
Rosewarne, D. (1984). "Estuary English#. Times Educational Supplement, 19.
Sabol, J. (1968). "Poznámky o výslovnosti spoluhlásky v#, Kultúra slova 2: 298!300.
Sabol, J. (1972a). "O þasovej modulácii súvislej reþi#, Kultúra slova 6: 97!99.
Sabol, J. (1972b). "O silovej modulácii súvislej reþi#, Kultúra slova 6: 129!133.
Sabol, J. (1972c). "O tónovej modulácii súvislej reþi#, Kultúra slova 6: 193!195.
Sabol, J. (1974). "Typy neutralizácií v slovenskom fonologickom systéme#, Studia
academica slovaca 3: 405!417.
Sabol, J. (1975a). "Osobitosti pri znelostnej asimilácii v slovenþine#, Kultúra slova 9:
76!78.
Sabol, J. (1975b). "Spoluhláskové skupiny a fonologické protiklady#, Kultúra slova 9:
204!207.
Sabol, J. (1975c). "VýslovnosĢ spoluhlások ć, Ģ, Ė, Đ pred e, i, í, ia, ie, iu#, Kultúra
slova 9: 18!21.
Sabol, J. (1976a). "Kvantita v slovenských slovách a tvaroch#, Kultúra slova 10:
68!70.
Sabol, J. (1976b). "O tempe reþi#, Kultúra slova 10: 17!21.
Sabol, J. (1976c). "Prozodická stavba slovenského slova#, Kultúra slova 10: 141!144.
Sabol, J. (1976d). "Rozlo$enie prízvuþných a neprízvuþných slabík v spisovnej
slovenþine#, Kultúra slova 10: 206!209.
Sabol, J. (1976e). "Spolupôsobenie pauzy a tempa reþi#, Kultúra slova 10: 263!266.
Sabol, J. (1977a). "Krátke samohlásky, dlhé samohlásky a dvojhlásky v spisovnej
slovenþine#, Kultúra slova 11: 97!101.
Sabol, J. (1977b). "Kvantita a rytmický zákon v slovenþine#, Kultúra slova 11:
193!197.
Sabol, J. (1978). "O striedaní spoluhlások v slovenþine#, Kultúra slova 12: 35!39.
Sabol, J. (1979). "Rytmický ráz slovenþiny#, Kultúra slova 13: 193!196.
307
References
Sabol, J. (1986). Fonetika a fonológia, 2nd edition. Filozofická fakulta v Pre%ove.

Sabol, J. (1994). "Slovenská slabika (Náþrt problematiky)#, Studia academica slovaca
23: 214!224.
Sabol, J. (1996). "Fonetický a fonologický aspekt hodnotenia zvukových javov
slovenþiny#, Studia academica slovaca 25: 183!188.
Sabol, J. (1997). "Dynamika fonickej sústavy slovenþiny v ostatných desaĢroþiach#,
Studia academica slovaca 26: 137!146.
Sabol, J. (1998). "Syntetická fonologická teória a fonologická typológia#, Studia
academica slovaca 27: 141!148.
Sabol, J. and Zimmermann, J. (1986). Základy akustickej fonetiky, 3rd edition.
Ediþné stredisko UPJ' v Ko%iciach.
Sabol, J. and Zimmermann, J. (1994). "Z experimentálneho výskumu prízvuku
v spisovnej slovenþine#, Slovenská reþ 59: 328!339.
Saussure, F. de (1974). Course in General Linguistics. London: Fontana-Collins.
Skaliþková, A. (1982). Fonetika souþasné angliþtiny. Státní pedagogické
nakladatelství Praha.
Stanislav , J. (1953). Slovenská výslovnosĢ. Martin: Osveta.
'tekauer, P. (1995). An Encyclopaedia of English Linguistics. Pre%ov:
Slovacontact.
Tabaþeková, E. (1981). "Fonetická realizácia labiodentál v spisovnej slovenþine#,
Thompson, I. (1981). Intonation Practice. Oxford University Press.
Trudgill, P. and Hannah, J. (1985). International English: A Guide to Varieties of
Standard English, 2nd edition. London: Edward Arnold.
Urbanová, ď. (1989). An Introduction to English Phonetics and Phonology, 2nd
edition. Ko%ice: Ediþné stredisko UPJ'.
Urbanová, ď. (1998). A Handbook of English Phonetics and Phonology. Brno:
Masarykova univerzita.
Vachek, J. (1960). Some Geographical Varieties of Present-Day English. Státní
pedagogické nakladatelství Praha.
Vachek, J. (1970). U základĤ pra#ské jazykovČdné "koly. Praha: Academia.
Vachek, J. (1973). Chapters from Phonology of Modern English. Státní
Vachek, J. (1985). "Some remarks on English and Slovak vocalic systems#,
Philologica 35: 69!75.
Vachek, J. and Firbas, J. (1963). Zvukový rozbor souþasné angliþtiny. Státní
Ward, I. C. (1972). The Phonetics of English, 5th edition. Cambridge: Heffer.
Webster!s II New Riverside University Dictionary (1984). Boston: The Riverside
Publishing Company.
Wells, J.C. (1990). Longman Pronunciation Dictionary. Harlow.
308
References
Wells, J.C. (1994a). "The Cocneyfication of RP?# In: G. Melchers and N-L.
Johannesson (eds.), Nonstandard Varieties of Language. Papers from the
Stockholm Symposium 11!13 April 1991, 198!205. Stockholm Studies in English
LXXXIV. Stockholm: Almquist & Wiksell International.
Wells, J.C. (1994b). "Transcribing Estuary English#. Speech Hearing & Language:
UCL Work in Progress 8, 261!267.
Yule, G. (1985). The Study of Language. Cambridge University Press.
309
View publication stats

Pavlk 2003 English Phoneticsand Phonology

Uploaded by

Copyright:

Available Formats

Pavlk 2003 English Phoneticsand Phonology

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

Pavlk 2003 English Phoneticsand Phonology

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

A Theoretical Introduction to English Phonetics and Phonology

Book · January 2003

The user has requested enhancement of the downloaded file.

Publikáciu finanþne sponzorovalo

Recenzenti: prof. PhDr. Ábel KráĐ, DrSc.

Vydal: RETAAS s r.o.

1 Main Aspects of Communication 9

2 The Classification and Description of English Speech Sounds 71

3 The Word 116

3.6.4.2 Stress in Derived Words 156

4 Connected Speech 179

5 Geographical Varieties and Accents of English 226

6 A Confrontational Analysis of English and Slovak Pronunciation

A Concise English-Slovak Dictionary of Phonetic and Phonological Terms 289

Appendix A. A feature matrix of RP General vowels 301

Appendix B. A feature matrix of RP General consonants 302

1 Main Aspects of Communication

1.1 Language and Communication

(phoneme) if it is intended to participate in meaning. For instance, the first sound in

1. Historical priority ! there are no known communities without the capacity of

When humans communicate, a very complex and complicated process is

Auditory-vocal channel Visual channel

Table 1. Systems of human face-to-face communication

It should be noted that prosodic, paralinguistic, and kinesic features can be

1.2 Concrete and Abstract Sides of Language

(a) The position of the sound within a certain environment

(a) The number of positionally determined allophones (combinatory/positional/

[ hh ] in the word "key# (the tongue is advanced)

Allophones must be distinguished from phones.

A phoneme, in this threefold system, can be characterized as a purely abstract

phoneme /p/ second-degree abstraction

Table 2. shows the relationship between phonemes, allophones and phones.

1.3 Phonetics and Phonology

1. Articulatory (organic) phonetics is concerned with:

(a) the organs of articulation ! it names and describes organs participating in

(c) the description and classification of speech sounds ! it describes and

2. Acoustic phonetics is concerned with:

3. Auditory phonetics is concerned with:

Table 3. Sub-branches of phonetics and their categories of study.

(a) Descriptive phonetics

(a) Descriptive phonetics

(a) Diachronic phonetics

In terms of the applicability of the results of phonetic studies we distinguish between

(a) Theoretical phonetics

(a) General studies of articulation, acoustics, and reception of speech sounds.

This twofold character of phonetics creates a problem for linguists: is phonetics

In phonetic research, these types of methods can be found:

1. (a) Subjective methods

2. (a) Mathematical methods

(b) Non-mathematical methods

3. (a) Experimental methods

4. (a) Instrumental methods

phonology ! phonemes, and suprasegmental (non-segmental) phonology, which

1.3.2.2 Methods of Determining Phonemic Inventories

(a) minimal pair test

(a) Minimal pair test

cut / kYt / ! gut / gYt / fit / f t / ! feet / fit /

(b) Commutation test

cut / kYt / ! gut / gYt / fit / f t / ! feet / fit /

/ p / ! / b / : pour / p / ! bore / b /

very [ ¥ve ] ! [ ¥veÚ ]

slávny [ ¥slani ] ! [ ¥slawni ]

tok / tk / ! bok / bk / drak / drak / ! dral / dral /

leap [ ljip ] ! fool [ fu1 ]

banka [ ¥ba)ka ] ! nos [ ns ]

high ! / i / mid ! / e / low ! / /

In phonemic transcription, the quality of such vowels as / Ι / ! / i /, / h / !

/ i / sit / i / see / ai / eye / i± / here

/ / sit / i / see / a / eye / ± / here

tries / tra z / = [ kwGwQ;fD ]