Iionyiueuo MUHUCTepCT80M 06pa308aHUR 8blCUle20 U cpeoneeo Cne4UaAbH020 y~e6H02o


8 Ka~eCT8e



U ¢aKYllbTeWB



MoCK "BwwlAwBJI wJ(OJIa"


55K 81.2 AHrn-1


P e U. e H 3 e H T bl: KacjleApa pOMaHo-repMaHCKO~
CKOro nenaroruxecxoro '1eCKI1X HayK



I:1HCTI:1TYTa I:1HOCTpaHHblX R3blKOB 14 K3H,II,I:1,11,aT :ll:1nOnorMQ

Ilununentco 0. CP.




n.B., MeTJlIOK A.A. Teoperusecxaa cjloHeTI1Ka






1I1H-TOB1:1 cpaK. MHOCTP_R3-1 -



1980. -

144 c.,

B nep.: 60


...... paccsaarp A3bfKa:






1(11 8HrllHiitcKoro 101 cnoroaeneaae,

¢oHollonl~eCKI1i-1 np"'poAB

aH81lH3 3syKOS, III xapaKTepHblB

cnoroo6pa30SBHHB oc;o6eHHOCT.., 8Hrn..,M-

c;po ...eTM~eCKaA YA8petMR, R3b1Ka.



Io1HTOH8UHA. eapH8T11SHOCTb SBeAeHbl 3ey KOB.

n POH3HOc;.., TBlTbHOi'i np..,HL\l1nbl aKY-




CTI1~eCKOM CKOM pe~l1, cnoco6bl

Kl1aC:CI1c;pHKau,HM aortpccei

C;POHOCTI1I1..,CTWleCKl1e I1HTepc;pepeHUHIII I1HTOH81.1,1II11. BHrm"MCKoro npill


aHrnl1M8 TBK>KB

c;poHeTI1~eCKOM ..,306pa>KeHIllA CTYAeHTBM





70104-133 6 ----------146--80. M304(051 -80


66K 81.2 AHrn-1 4H(AHrnl
"Sblw3i'iwaA uncona",

© V13AaTenbcHlo


FOREWORD The book undertaken at Pedagogical The authors on -theoretical acquaint foraiqn is designed Institutes for -students and Faculties of English of Foreign who have Languages. to information and

the study of a theoretical

course in English phonetics in view: firstly,

of the book had two objects teachers problems of English phonetics

give the prospective them with linguists. linguists

of English necessary

and, secondly, to the opinions of

some of the diverse

views of Soviet

The authors

have put forward

the foremost own attitude reference readers' forefront

on the points at issue and expressed their

to them. An endeavour has been made to treat the findings phonetic in this field, and to draw the phenomena which are in the

main problems of theoret ical and exper imenta I phonetics with due to the latest attention to

of modern lingu ist ics, such as phonet ic styles and proAn attempt has also been made to show the studies in solving practical problems con-

sodic interference.

relevance of theoretical Every chapter

nected with language teach ing. is followed by exercises. Their purpose is not as to encourage further which stimulate to be wh ich are read ing. or questions views independent refuted much to control understanding contain contraversial and assignments

The exescises decisions, supported,

which requ ire reference to pronouncfeatures of speech. it may

ing dictionaries be of interest phonetics.

and analysis of the phonetic to teachers

Although this book is intended primarily for students,

and to all those concerned with English



Baryshnicova who have inspired and encouraged this work. chapters 3. Prof. wr. 7 w.A. 6.V. are also expressed to those Chapters " 2. 9. Artyomov. 5. Vassilyev and the late Prof. Thanks for !"aluable suggestions who have rev iewed the MS.K. Borisova. by L. K.The authors make grateful acknowledgement to Prof. 4. V. 10 .V. 8.itten by A. V. A.

sound. Both the oral and written speech forms have a material substance. It is a process of . each of which is a specially organ ized language system with a certain number of its un its. The grammatical form of a verb or a noun can be changed on Iy by chang ing the sounds which compose them. For example. is an exclamat ion. communication by means of language. The phonet ic system of a language is a set of phonet ic un its arranged in an orderly way to replace each other in a given framework. In written speech the substance is graphic. . intonahon groups. accentual (rhythmic) units. Segmental and prosodic units serve to form and differentiate units at ather subsystems of language. tempo. expressing the speaker's evaluation. vowels and oonsonants which form the vocalic and the Consonantal subsystems. expressing the speaker's uneertalntv and desire to get further information.pwvT"/ (fo:ne:) *The terms "phonic". The sound substance gives shape to a spoken message in communicat ion. In fact it contains two systems (or levels) ."the immediate actualitv of thought" [46]' and "the most important means of human intercourse" (47] exists in two main speech forms: oral and written. the lexical end grammatical units. wide pitch range and greater loudness. <. but it is what forms patterns of language. The modifications of words and their combination into utterances (sentences) are first of all sound phenomena. stress. utterances. '!Well'done!". is a question. THE ROLE OF SOUND PHENOMENA IN COMMUNICATION THE PHONETIC SYSTEM OF A LANGUAGE Language . 5 . pronounced with the falling tone. It is clearly seen from the utterances of identical lexical and I grammatical structures. "phonetics" rome from the Greek word . The rising-falling tone in the utterance "Well'~one" may express a challenging or quizzical attitude on the part of the speaker. The sound substance is a med iurn in wh ich the whole system of language is embodied. "phonettc". "WeIl8one?" pronounced with the riSing tone.segmental and suprasegmental. By changing the prosodic structure (intonation) of an utterance one changes the meaning of the utterance. pauses. A substance is not in itself language. Segmental un its are elementary sounds. rhythm. it is the sound substance or the sound matter. which form the subsystems of pitch. it forms un its of the phonetic system of a concrete language.or prosodic. Speech is a man itestation of ··Ianguage. Prosodic units are syllables.C·R APT E R t. In oral speech the substance is phonic".

1))/1 . Speech sounds are products of human organs of speech. Sound product ion is impossib Ie without resp irat ion. Speech sounds are based ell iefly on expi ration. which are closely interoonnected: the art icu tatorv aspect. educat ion.))) Ar/leu/afIO/) I +tHl 'i))} IIIHtIIl' sasa wav~s 1111'1111 \ P~rcepllOIl Fig. Moreover. Every act of speech presupposes the presence of a person who speaks and a person who listens. ASPECTS OF SOUND PHENOMENA The ability to form language units is not the only property of the sound medium.inspiration and expiration. sound phenomena have different aspects. ~ ((HII)). the listener hears and interprets them. Consequently. Expiration. the sound medium has its own independent properties as a physical phenomenon. Gleason notes that to speak any language a person must know nearly all the 100. geograph ical or ig in. They result from the activities of the diaphragm. Being created by the speaker. the sound medium indicates the speaker's personality (sex. social status and so on.To underline the importance of the sound medium of language H. In phonic expiration the air co6 . though in some African languages there are sounds produced py inspirat ion. Commun icat ion is possible onlv because the speaker and the listener interpret the sounds as units of the same language. The articulatory (sound-production) aspect. the larvnx with the vocal cords in it. the sounds travel through the air to the listener in the form of complex combinations of sound waves. 1. while only 50-900/0 of its grammar and 1%ofthe\locabulary maybe sufficient (24). wh ich cons ists of two alternating phases . the auditory and the lingUistic aspect.10 of its phonetics. the pharynx. the acoust ic. Stages in the passing of a spoken ~ssage. the lungs. the mouth cav itv with the speech organs situated in it and the nasal cavity. individual features) and reveals his physiological and emotional state. age. it is a product of human activity. The speaker produces sounds. during which speech sounds are produced is called phonic expiration as distinct from quiet breathing. In addition to it. the trachea. the bronchi.

Organs of speech. expiration lasts much longer than inspira~)on. whereas in quiet breathing inspiration and expiration each take about the same period of time. Diagrams of the vocal cords: (a) tightly closed 85 for [?]. Fig. (Q) (A) (c) Fig. The air-stream coming from the lunqs undergoes important modifications in them. though mainly regular or periodic. vocal cords. For example. 3. thus producinq variations in the intensity of speech sounds. . the air-pressure below the vocal cords becomes very high and the air comes from the lungs in regular puffs making the vocal cords vibrate. because during speech the air'passage is periodically blocked by the speech organs. Sound product ion actually takes place in the larynx. One part of sound product ion is phonation.mes from the lungs not freely but in spurts. or VOice-production. The regular vibrations of the vocal cords are transmitted to the air-stream and the acoustic effect perceived by the human ear is that of a vocal tone. the air pressure is greater on the peak of the syllable and it is less on its margins. (e) Ioosly together and vibrating as for voice. The lungs supply the necessary air-pressure and regu late its force. This is what we call vo ice. the pharynx and the oral and nasal cavities. When the 7 . are tensed and brought loosely together. Their vibrations are complex. situated in the larynx. (b) wide open as for br. Therefore in speech.

Thus there are two main sources of vibration in the production of speech sounds . organs necessary to pronounce a speech sound.owel sound of a certain Quality is produced. Fif. The simultaneous vibrations of each part of t\"le vocal cords produce cart ial tones (over tones. The fundamental frequency determines the pitch of the voice and forms an acoustic 8 . When in the supralaryngear cavities there is an obstruction to the airstream. 4.(\ The number of vibrations per second is called frequency. When an obstruction is created and the vocal cords vibrate. . intensitv. As a result.frequency. depends on tnt type of obstruction (a constriction or a complete closure) and determines the particular quality at a consonant. a no ise is produced. the mouth and the nasal cavity) thus modifying the voice which comes from the lungs. A sound wave is created by a vibration which may ~periadic or nonperiodic. Frequency of the basic vibrations of the vocal oords is the fundamental frequency (marked by Fa). ) VlJ\TV ~ F i{I. a voiced consonant is produced. Like any other sound of nature speech sounds exist in the form of sound waves and have the same physical properties . duration and spectrum. The character:. The movements of speech organs modify the shape.'\?jbnltiql'!s(l) of entire cord. a t. When the vocal cords do not vibrate.the vocal cords arid various kinds of obstruction.11\/11\. 5. The acoustic aspect. (b) of its perts. of the noise (frict ion or plosion). Complllx vibrations.or harmon ics}. The vocal cords v ibrate in such a way that they produce various kinds of waves simultaneously. the result is a voiceless consonant.. Frequency is measured in hertz or cycles per second (cis). simple or complex. The basic-vibrations of the vocal cords over their whole length produce the fundamental tone of voice. size and volume of the supralervnqeal cavities (the pharynx.The other part at sound-production is articulation which comprises all the movements and positions of the s~ech.

. inventor of the magnetit telephone. F 1 of the 'lowe r la:/ is equal to 800 Hz and F2 . Intensity is measured in decibels (dB"·).to 1100 Hz. The duration of a sound is the quantity of time during which the same vibrations continue. Changes in intensity are associated with stress in those languages which have dynamic stress. Due to the particular changes in their shape. "Each half of the cord vibrates at a frequency twice as great . sonants and voiced oonconants because these sounds are formed 'with vo ice and it is absent in the spectra of voiceless consonants. 9 . The formants of li:1 are equal to 380 Hz and 2500 Hz. FO is present in the spectra of vowe ls. F2 . Vowel sounds have at least two formants markedrby F.four times faster. Thus the specific qualities of vowel sounds are achieved. and so on .as that of the cord as a whole. Thus. each quarter . and F2.. The complex range of intensified frequencies wh ich form the quality of a sound is called the acoust ic spectrum of the sound. F1 is characterized by lower frequencies. Intensity of speech sounds depends on the amplitude of v ibration.· and partial vibrations The complex tone is modified in the oral and nasal resonators. For this reason the duration of a sound is often referred to as its quantity.The decibel is one tenth of a bel. The frequencies of the overtones are much higher. The intensified frequencies in the spectrum which characterize the quality of a sound and distinguish it from other sounds of different quality are called formants. of the fundamental low. which can be viewed in two ways. size and volume the oral and nasal cavities function as an acoustic filter: they intensify certain frequencies contained in the complex tone and weaken the others. Any sound has a certain duration. The perception of speech sounds involves the activity of our hearing mechanism. respectively. The auditory (sound-perception) higher "frequencies. der . The ber is so named after AlexanGraham Bell (1847-1922}. The formant of the fundamental tone (Fo) is irrelevant to vowel differention. Like any other form of matter. It is relatively superposition tone. Each third vibrates three times faster. sound ex ists and moves in time. The duration of sounds is measured in mifliseoonds (ms). Speech sounds may also be analysed from the point of view of perception. which are together responsible for the particular quality of eacb vowel type.basis of range oi The complex speech melody. about 40-400 cis in the total results in a votce..

the higher is the pitch of the voice and vice versa. formant frequency. as the difference between them is not singificant in distinguishing words or grammatical forms in Russian. Thus. The greater is the intensity of a sound. Our perception of the pitch of the voice depends largely (but not solely) on the fundamental frequency generated by vocal cord vibration. But . So a listener's reactions are conditioned by his experience of handling his own language. Thus /a:/ is more powerful than /i:/. are louder or more sonorous than others. a 10 . The human ear does not percei ve all the acoust ic featu res present in a sound wave. frequency. intensi tv and duration in terms of four perceptible categories of pitch. the hinger limit is 20 000 Hz. In this way. Moreover.On the one hand. On the other hand. In ideal conditions we can perceive even a difference in 3 Hz. But the perceived pitch variat ion may also be affected by var iat ions of intens ity on the same frequency. The human brain interprets aeaust ic phenomena in terms of a given language system. Our judgements relating to loudness are not as fine as those relating to either quality or pitch. and vowels generally have more carrying power than consonants.our percept ion of loudness does not depend on intens ity alone. the louder is the sound. in comparison with neighbouring sounds or syllables. it is a physiological mechanism which reacts to acoustic stimuli: the human ear transforms mechanical vibrations of the air into nervous stimuli and transmits them to the brain. the lower limit of frequency which we can perceive is 16 Hz. The greater is the. but such small distinctions are not usable in speech. Changes in intensity are perceived by ou r ear as var iatalons in the loudness of a sound. A sound or a syllable may be perceived as louder. because of a marked pitch change on it or because it is longer than the others. it is also a psychological mechanism which selects from the great amount of acoustic information only that which is linguistically Significant. some sounds. By reducing the intensity of a sound we can achieve a high pitch in perception. different acoustic stimuli may be interpreted as be jng the same sound un it. quality. Thus for Russian the open / e I as in "uncr" and the mid-open I e I as in "nera" are one and the same sound un it. loudness and length. Our physiolog ical capabilities in perception are limited. owing to their nature. The listener hears the acoustic features of fundamental frequency. It shou Id be remembered that percept ib Ie featu res of sounds are not fu!ly conditioned by the related articulatory and acoustic ones. An Englishman would consider these sounds as different sound units since in the English language the former resembles the open / <lie las in "had" and the latter is very much like the mid-open I e las in "head" which serve to differentiate words.

they perform certain linguistic functions. word-forms. word-forms and utterances. consequently. The relat ions between the art ieu latory. rhythm. because of the role which sound matter plays in the functioning of language as a social phenomenon. They constitute meaningful units . disturbances in the production of speech sounds are likely to appear. The better we hear the differences between the sounds. Represent ing language un its in actual speech. Most of the mean ingfu 1distinctions of the language are based on d ist inctions in sound. word-forms or utterances. the length of rhythm ic units in an English utterance is considered to be approximately the same since it is a characteristic feature of English rhythm that stressed syllables occur at more or less equal intervals of time. they are characterized by certain pitch-andstress patterns. the sound phenomena enable the listener to identify them as concrete words. The process of communication would be impossible if the speaker himself did not hear the sounds he pronounces. aud itory and lingu ist ie aspects of speech sounds can be presented roughly in the following way: . the better we pronounce them. acoust ic. and. This is an example of how our brain interprets from the acoustic material only that which is linguistically significant. Sounds and pressod ic featu res serve to differentiate the un its they form since communication by means of language is possible only because sound phenornena can be opposed 1:0 one another for purposes of d ifferentiat ing words. temporal features.morphemes. of sounds. If the link between listening and pronouncing is disturbed. utterances. The linguistic aspect. distinctive and identificatory functions. For example. All the utterances consist of words. Therefore in learn ing to pronounce the sounds of a foreign language one shou ld bear in mind the importance of ear-train ing. Our hearing mechanism acts as a monitor of what we ourselves are saying. All the words of a language consist of speech sounds which are grouped and arranged in the way specific for the language and which are unified by stress. But our perception of length does not always correspond to the actual duration of speech sounds or other un its. words.Different duration of speech sounds is perceived as a difference in their length. Segmental sounds and' prosodic features are Iingu ist ic phenomena. Thus. The Iingu lst ic aspect of speech sounds is also called the funct ional or social aspect. Simultaneously. But the actual duration of rhythmic units is far from being equal. segmental sounds and prosodic features of speech perform constitutive.

"to"'. the radio. pauses). Phonetics occupies itself with the study of the ways in wh ich the sounds are organ ized into a system of un its and the variation of the units in all types and styles of spoken language. deaf aids raises a great number of primarily phonetic problems. rhythm) The study of the sound phenomena of language. the physiological basis of sound production and the sound phenomena that reveal the individual peculiarities of the speaker. Thus the sound medium has a special science all to itself because of the exclusive importance of oral speech as compared with written speech. rhythm.Articu latory characteristics Acoustic properties Auditory (perceptible) qualities Linguistic phenomena ------~----~----r-----I I I different positions and movements of speech organs formant I frequencies vibrations of the voca I cords I fundamental I pitch I frequency I I I I I I I prosody stress)· (melody. Being a science in its own right. further development of such technical means of sound transmission as the telephone. recording and speaking machines. tempo. Oral speech is primary. Phonetics as a science is a branch of linguistics. stress. speech recognizers. whereas written speech is secondary (it is constructed on the basis of oral speech). prosody (stress) -----T---T----~----the quantity of time I duration I length I prosody dur inq which the sound is pronounced the amplitude vibrations Of-/. The sound medium needs special attention also because of the complex character of its product ion and percept ion. Phonet ics as a branch of Iingu ist ics stud ies sounds in the broad sense. phoneme. constitutes the subject of the phonetic science. It also studies the acoustic properties of sounds. I (timbre) quality . Besides. it is at the same time cbsely connected with other lingu istic sciences12 .ty-t ioudn~--II I '~~OdY (stress)- I I I PHONETICS I I AS A SCIENCE I I I (tempo. in all their aspects and varieties. comprising segmental sounds (vowels and consonants) and prosodic phenomena (pitch.

electromyography. we use a spectrograph to analyse the acoust ic spectra of sounds. pedagogy. and various kinds of techn ique to study sound-percept ion. subjective method . psychology. It is also closely interconnected with physioklgy. articulatory and perceptual phonetics are generally termed physiological phonetics. cinematography. Since sound production and sound perception are physiological processes. and prosodic phenomena. phonation (voire-production). It deals with our voice-producing mechanism and the way we produce sounds.grammar. It studies speech sounds with the helpof experimental (instrumental) methods. stvlist ics and the history of the language. 13 . biology. BRANCHES OF PHONETICS Depending on which of sound phenomena is studied. X-ray cinematography. This method involves observat ion of the movements of speech organs when pronouncing sounds and analysis of one's muscular' sensations during the articulation of speech sounds. since the phonetic system of a language. cybernetics. articulation and also the mental processes necessary for the mastery of a phonet ic system. Phonetics has a long history. loudness and duration. With the help of an electro-acoustic synthesizer synthetic speech is produced which is a good means of test ing the results of the electro-acoustic analysis. Besides these objective methods articulatory phonetics uses its oldest. pitch variation. X-ray photography. intensity and duration. phonetics is subdivided into four main branches. mathematics. But as a science in its own right it began to develop in RUssia and in Wes'tern Europe only in the second half of the 19th centurv. The methods used in perceptual phonetics are also experimental.the method of direct observation. They include various kinds of auditory tests. They involve palatography. photography. -Var ious kinds of apparatus are applied for analysing the acoustic structure of segmental sounds and prosodic phenomena. Articulatory phonetics is concerned with the study of sound as a result of the activities of speech organs. It was known to the ancient Greeks and H indus. Perceptual (auditory) phonetics occupies itself with the study of man's perception of segmental sounds. Methods employed in articulatory phonetics are experimental. It studies the ways in which sound perception is determined by the phonetic system of a language. an osci llog raph and an intonograph to analyse frequency. laryngoscopy. lexicology. For example. Acoustic phonetics is concerned with the acoustic aspect of speech sounds. physics. It studies respiration. its vocabulary and grammar constitute one indivisible ~hole.

Phonology was founded in Prague by a group of linguists (N. Special phonetics is concerned with the study of the phonetic system of a concrete language. Jakobson and others) . The methods employed by phonology are linguistic. such as special. applied. words. R. created in Russia by I. But not all linguists are of the opinion that phonology is an integral part of phonet ics. whereas phonology of intonat ion (prosody) is termed intonology (or prosodem ics). Phono logy sets out to discover those segmental and prosodic features that have a differential value in a language.Phonology. A great number of phoneticians abroad adhere to the same 'point of view. Kru shevsky and. N. Trubetz kov. Baudouin de Courtenay and developed by his pupils and followers l. p. a Swedish a close un itv of acoust ic. descriptive. and it establishes the system of phonemes and prosodernes. According to the Prague School. The basis of phonology is the phoneme theory. It deais with the funct ional aspect of sound phenomena. When the phonetic system is studied in its static form at a particular period (synchronically) . All the above branches of phonetics are closely connected since the object of their study .Hjelmslev who advocated total separation between phonetics and phonology. established by him. The two types of studies are interdependent and condition each other. there are other branches of phonetics. N. auditory and linguistic aspects. The distribution and groupin!. Consequently it seems preferable to grou p them together under the trad itional genera I head ing of phonet ics" [97. phonology is a lingu ist ic science and is concerned with the social function of phonet ic phenomena.l of phonemes in syllables and words in a particular language are dealt with in an area of phonology which is called phonotact ics. Trubetzkoy [40) claims that phonology should be separated from phonetics. phonetics and phonology are independent sciences: phonetics is a biological science and is concerned with physical and physiological characteristics of speech sounds.speech sounds . comparative. general.we deal with 14 . writes as follows: "It was a grave error on the part of the Prague School to want to establish a strict separation between phonetics and phonology". This point of view is supported by the Danish linguist L. later. It also sets out to determ ine the frequency of occurrence of these un its in syllables. Malmberg.Shcherba. Phono logy of segmental un its is often ca lied phonem ics. 971. or functional phonetics is a purely linguistic branch of phonetics. B. Besides the four branches given above. rhythmic un its and other sequences which form the utterance. But the vast majority of Soviet phone ticians do not consider it logical to separate function from form and to exclude phonetics from the linguistic sciences [80). For instance. art icu latory. historical. by other Sov iet and fore ign Iingu ists.

The deaf people can. General phonet ics is concerned with the study of man's sound-preducing possibilities and the functioning of his speech mechanism. theor ies of stress. the way they are produced and the role they play when forming and expressing thoughts. in correcting speech defects and in curing patholog ical phenomena of speech.or applied phonetics we mean all the practical applications of phonetics.e. It studies written documents and compares the spelling and pronunciation of one and the same word in different periods of the history of the language. broadcasting. radio-announcers and other public speakers. Historical phonetics uses the philological method of investigation.e. teaching normal oral speech to deaf-mutes. General phonetics is based on the extensive material which is provided by the special phonet ics of a number of languages and on the materia Iof other sciences. In order to speak well and to teach others the proper way of speaking one must understand the mechan ism of art ieu lat ion and the mechan ism of phonation. To correct dey iat ions from the pronunciat ion norm. Phonetics is of considerable importance for other fields of language study. Phonet ic data is also made use of in teaching children to read and write their mother tongue. i. By practicaJ. Therefore. general phonetics has been able to make a number of general conclusions concerning the complex nature of speech sounds and to formulate a number of theories: the phoneme theory. in most cases. or evolutionary phonetics. etc. Young teachers should bear it in mind that to teach efficiently they must learn to speak efficiently. It establishes the types of speech sounds which exist in various languages.descriptive phonetics. the theory of syllable formation. Phonetics is applied in logopedics. All the branches of phonetics are of great use and importance in teaching the pronunciat ion of fore 19n languages. espscia lIy kindred ones. speech recognition. Comparative phonetics is concerned with the comparative study of the phonet ic systems of two or more languages. To build a microphone. to teach to speak one must know thorouqhlv the articulatory aspect of speech sounds. intonat ion. Phonetics is widely used in the field of sound transmission: in telephony. one must have a good knowledge of normal phonetics. a tape recorder. use only muscular sensation to control and guide their articulation. such as aphasia. i. singers. Theoret ical phonet ics of a part'icu lar language appl ies those theories to the language it analyses. As a result of this. a 15 . Phonetics is of great practical importance in the teaching of diction to actors. wh ich have made use of the approaches and the linguist ic methods worked out by phonetics. When the system is studied in its historical ~evelopment (diachronically) we speak about historical. in creating orthographies for unwritten languages. Phonetics is of great importance in surdo-pedagogics.

What un its does it include? What are the two main sources of sound? How does the frequency of the vocal cord vibrations change in the production of the falling tone? (the rising tone?) Why do the sounds of the violin. a speech synthesizer or any other apparatus capable of transmitting the spoken language. 2. 3. 8. Select some examples: 1) to illustrate articulatory and functional distinctt. 5. psychophonetics and other phonetic sciences which contributed oonsiderablv to the formation of speechology . B. Close interact ion and collaboration between phonetics and other sciences has given birth to new scientific branches such as technical acoustics.lns between English and Russian (or Byelorussian) sounds.spectrograph. the sound engineer must know acoustic phonetics and.the science of speech. EXERCISES 1. . 9. is a concrete man ifestation of the whole system of language at work. very important to train one's hearing abilities in foreign language Jearning? Why is phonetics placed amog linguistic sciences and not among physiological or physical? 6. 7. 4. A. the guitar and other musical instruments differ? What affects their specific qualities? Is there any analogy in the product ion of speech sounds? In what way does the listener perceive the acoust ic properties of speech sounds? What does percept ion of pitch and loudness depend on? How would you explain the difference between the physiological and psycho log ical aspects of our hearing mechan ism? Wh ich of them is responsible for "hear ing the message"? Which is responsible for "understand ing the message"? Wh ich of them is involved when we listen to a foreign language that we don't know? Why is . 3) to illust rate that an utterance. Think about the following questions for class discussion: What are the relat ions between language and speech? Def ine the phonet ic system of a language. 2) to illustrate the distinctions in sounds which express distinctions in meaning. an act of oral speech. so Ive the same problems as the Ilngu ist does.

Pike Il03]). There is no obstruction in their articulation.a feature characteristic of vowels. These are sonorants 1m. Voiced consonants are actually a combination of noise and tone. the force of the air stream coming from the lungs. lt.C HAP T E R 2. ri. the removal of which causes noise . That is why they are attributed to consonants. j. Thus. The articu latory boundary between vowe Is and consonants is not well marked. There exist speech sounds that occupy an intermediate position between vowels and consonants and have common features with both the vowels and consonants. The wide passage for the air stream in the articulation of sonorants means that the oral and nasal cavities are active. the distribution of museu lar tension. I. The muscular tension is concentrated at the place of obstruction. tel . And sonants are predominantly sounds of tone with an admixture of noise. Because of their strong vocalic characteristics sonorants iv«. n. ARTICULA TORY AND ACOUSTIC ANALYSIS OF ENGLISH SPEECH SOUNDS PRINCIPLES OF CLASSIFICATION OF SPEECH SOUNDS In all languages speech sounds are traditionally divided into two main types .ga: -~n). The air passage in their production is rather wide and the force of the air is weak as in the case of vowels.j.vowels and consonants. Vowels are speech sounds based on voice which is modified in the supra-« laryngeal cavities. Consonants are speech sounds in the articulation of which there is an obstruction. results in greater audibility (sonority. The air stream is strong. lJ . The force of the air stream is rather weak. From the articulatory point of view the main principles of the division are as follows: the presence or absence of obstruction. or perceptibility) of the sounds .6248 17 535153 . t} are often referred to as semivoweJs. Due to their great sonority some sonorants can be syllabic in some particular positions (e. Li ke vowels they are largely based on voice. w. But generally sonorants do not perform the function of syJJable formation. The muscular tension is spread evenly throughout the speech organs. They are combinations of the main tone and overtones intensified by the supralaryngeal cavities.noises. after K. consonants can be subdivided into sonorants and noise consonants (or contoids. carrying power.~I.plosion or friction.g. Consonants are non-periodic vibrations . There is an obstruction in the ir art ieu lat ion and the museu lar tension is concentrated at the place of obstruct ion as in the product ion of consonants. From the acoustic point of view vowels are complex periodic vibrationstones. 2 JaK. Voiceless consonants are pure noises.

The following quotation from Webster's New International Dictionary of the English language presents some marked peculiarities of the English articu latory bas is: "Every language has certain characterist ics of pronunciation which give it distinctive phonetic character. It is not made so tense as in many other languages. for instance. and the resulting different articulatory energy in vowel and consonant production. the French and German. The notion of the articulation basis of a language can be broadened if we analyse not enly the articulation part in sound production but phonation as well. THE ARTICULATION BASIS OF ENGLISH Due to the identical structure of speech organs of people of different races and nationalities. the lips do not protrude for I . the tongue is slightly drawn back. p.rhus. Numerous experiments show that the criterion which justifies the division of speech sounds into vowels and consonants is the physiological criterion put forward by V. Bogorod itz ky [13]. nor in the back vowels drawn so far back. all languages have soundsof identical types (e. the tongue. checked vowels in English which require a great force of utterance at the end of thei r art icu lat lon) . The phonat ional habits may also be dependent on the amount of consonants that occur in consonantal clusters. whereas voiceless consonants are much more energet ic.g. 18 . like a spoon. It is the d ist inct ion in the groups of muscles. as the French. In articulating the front vowels it is not pushed so far for~ard. The articulation basis of English differs from that of Russian: voiced consonants are less energetiC. The phonational habits of the native speakers of different languages may differ depending on the character of sounds (such as clicks and suction which are pronounced during inspiration. These articulatory habits characteristic of all the native speakers of a language are called the articulation basis of the language. and to draw it away from the teeth.articulatorily.. as v igorous and clear" [112. is characterized by laxity. the lips are inactive . the sounds are not identical . Russian IHI and English In/). u:1 as they do for /0. for exarnp Ie. hollow the front of it more or less. 'the acoustic boundary between vowels and consonants is not well mar ked either. be d'laracterized as sluggish or muffled. But being identical typologically. Their articulatory distinctions are explained by the fact that each language has its own tendencies and modes of art icu lat ion. and on the character of sound transitions in connected speech. vl. There is also a constant tendency to lower and flatten the tongue. English articulation may. In English.}:.. when articulating. in general. xlv ii]. which operate in vowel and consonant production.

comprises both the articulatory and phonational habits of the nat ive speakers of a language. b a c k-a d van c e d: I tr . according to the height of the tong14e.according to the position of the lips. Soviet phoneticians G. THE ARTICULATORY CLASSIFICATION OF ENGLISH SPEECH SOUNDS A.according to the length of a vowel. mix e d: I 3:. The position of the speech organs in the articulation of vowels may be kept for a variable period of I. e. English vowers have been traditionally subdivided into c los e (h i g h). According to the verticar movement of the tongue.. All these factors predetermine the principles according to which vowels are class if jed: .rmine the specific articulatory characteristics of its~ound system. . V.0:1 and the nuclei of the diphthongs lov • V'.Ll basis of English dpt .the tongue and the lips.) 1 and b a c k: {u:. . 1\ .accord ing to the force of art leu lat ion at the end of a vowel. . volume and shape. E:e . Vowe'ls The various qualities (timbres) of English vowels are determined by the oral resonator ."ll /. English vowels are classified into fro n t: Ii:. therefore. The peculiarities of the articulatilJ. to define the articulatory features of vowels in terms of these 3 degrees of opening of the mouth cavity. mid and 0 pen (low). open) into a n a r row and a b r 0 a d variation. Torsuyev (35L A. 1. . <lB I and the nuclei of the diphthongs leI.):./u:-u/.1 o:-l) {arenotdescribedfrom the point of view of their t r a cted:/Ilandthenucleusofthediphthong /1"" I. According to the horizontal movement of the tongue.The articulation basis.according to the degree of the muscular tension of the articulatory organs. the character of sound modifications in connected speech and the physiological mechanism of syllable formation. vowe Is can be classified as h jgh-narrow 2· 19 . Trakhterov [39}. the quality of a vowel depends on whether the speech organs are tense or lax and whether the force of articu lat ion weakens or is stab Ie.according to the horizontal movement of the tongue: .. The resonator is modified by the most movable speech organs .its size.according to the stability of articulation. mid. however. Thus. since functionally different vowels 1 1:-1 1. fro n t . .. 2. It is Jnsuff lcient. Moreover.accord ing to the vert ieal movement of the tongue. . Vassilyev (110J classify these sounds in a more precise manner subd iv iding each class (close. ~ I.

I\. D.:. 3 :. v i}. In the articulation of short I I I and I V" I these organs are relat ively re taxed. All the longyowels are believed to be tense. Not all phoneticians share this opinion. but may be usefu I in describing concreate realizations of the phonemes. i.e. because in the case of open vowels it is difficult to define whether there is any tenseness or not. This is due to the long period of time for which the speech organs are kept in a certain position and this. 4.Jones [84J only the long li:/ and lu:1 may be considered as tense. so these vowe Is are characterized as lax. u:l. t'r I. Jones applies the terms "tense" and "lax" only to close vowels. . whether they are rounded. I:. Eng Iish vowels are classed into r 0 U n d e d I :J :. This point should be clarified with the help of special electromyographic invest igat ions.Ie.o( v v].e. mid-broad I a. 5. v I. U: . English 1i:1 and lu i] are characterized as tense. ~ 2 [ Q Inucleu5 of the diphthongs [a!) .the nucleus of the d iphthoflg [0'" J. requ ires greater muscular tension of the speech organs. D . The subdivision of vowels into lipspread and lip-neutral is unnecessary for a phonological analysis. in its turn. Table of English vowels: 11 e ) .1?: a( I > v). 6. 3 :. According to the degree of muscular tension.!. high-broad I I. ~ I. a:. mid-narrow (Q )/. 3. low-broad lee> vowelS vowels vowels Fully fronl8iJCk.JI. According to the force of articulation at the end of the vowel (the character of the end). English vowels are subdivided into f r e e and c h e c ked.V" I and U n r 0 un d e d II :. According to the position of the lips. English vowels are classified into ten s e and I a x.ful1lJ froni re/raclea advanced biCk fran! Milel1 [jack (high) BrOd{j vowels v8f1i1fton Mid Narrow (mld'O{Ji'n) varriaflofl vowels sroea Close vam/lon IWrow vaneuon Open flaw) 'owels Broad v{Jnafion Fig. Thus. while short vowels are lax [35]. ). Iow-naHow I "'. 4 [v I .the nucleus of thediphthong [:>. Free vowels are pronounced in an opel! syllable with a weakening in the force 20 .nd (a. because the speech organs that participate in their formation (the tongue and the lips) are considerably tensed. [01' . Their classification reflects the distinctive differencies in the quality of the historically long and historically sh8'rt vowels. spread or neutral. I . According to D.b I. ~:I. Q:. for instance.the nucleus of the diphthong [c.

of articulation towards their end, i.e. they have a fading character. These are all the Eng fish long monophthongs and diphthongs and unstressed short vowels. Checked vowels are those in the articulation of which there is no weakening of the force of articulation. They are pronounced abruptly at the end, immediately followed by a consonant that checks them. These are historically short vowels-under stress. 6. According to the stability of articulation, English vowels are classed into monophthongsl I;. I ,e, ill! ,a:, fI ,;):,ZI ,u: ,vi, dip h tho n g s (e I ,a I ,D I , au, OV, 1~ ,e;9 , oa , 'Va I and dip h thongo ids,ordiphthongizedvQwels/1: ,u: l . The stability of articulation as in the case of monophthongs or its instability as in the case of diphthongs and diphthongoids is, actually, the stabitity (or lnsteb llltv] of the shape of the oral resonator. When the position of the tongue and the lips during the pronunciation of a vowel is altered to some extent, a new vowel quality is produced. In diphthongs two vowel elements are distinguished - the nucleus and the glide. The nucleus is stronger, more definite in timbre, more prominent and syllabic. In different languages the nucleus of a diphthong may be either the first or the second element. Diphthongs that consist of a nucleus followed by a glide are falling diphthongs, because the total amount of articulatory energy falls towards the second element. Those consisting of a glide followed by a nucleus are rising diphthongs, since the articulatory energy rises towards second element. EnIDish diphthongs are failing. Rising diphthongs are common in Italian. In some phonetic contexts English diphthongs / Ie , V~/ may be pronounced with the second element stronger and more- prominent than the first, and are, consequently, rising. When the diphthong /0 V' I is pronounced as an exclamation with the high r istnq tone, the Iv-/element in It rs as strong and prominent aS/O/. So lov I can be called a level diphthong. D. Jones po ints out that in unst ressed sy Ilables the I I I and / tr I elements in I I a /, / va I may be weaker than the second element / a I. E.g. (5 I a r1 a sl "serious", I'plar1ad / "period", / I nflu e ns/ "influence", I'kp, gr if ant/ "congruent" . 7. Closely connected with the quality of vowels is their quantity, or length. Any speech sound must have certain duration to display its quality, to be perceived as such. According to their length, English vowels are divided into long Ii ~ ,Q: , o : ,u; • 3: I and short / 1 , 1\ , n ,e, V" ,a , ee /. Th is length is h istor ical. It differs from the pos itional length of the same vowels. In connected speech historically tong vowels may be of the same length as historically short ones and even shorter. Cf. Ibi:t/ - /b 1 d/, 15i:t/-/s r t/. (See Chapter 31.


B. Consonants An indispensable constituent of a consonant is noise. The source of noise is an obstruction. There are the following types of obstruction in the production of consonants: 1) complete occlusion (closure), 2) constriction (narrowing) and 3) occlusion-constriction (closure immediately followed by a constriction) . The noise produced by the removal of a closure is that of a plosion, the noise resulting from the movement of the air stream in the narrowing is that of friction. The two effects are combined when closure is followed by a narrowing. 1. According "to the type of obstruction and the manner of the production of noise, English consonants are classified in the follOwing way: Occlusives stops


/ <, nasal
gl Im,n,~1

Constr ict ives fricatives
/. \


Occlus ive-Constrictives


b, t, d, k,



un icentra1 b icentral rned ia1 If,v,e,"O,s,zll_l'J 1 /i. r, wi


oral sonants

(Affri cates)

It/ ' d31


1 II

Obstructions may be formed either by two active speech organs or by one active speech organ (articulator) and a passive organ of speech (point or place of art icu lat ion) _ "2. According to the active speech organ which forms an obstruction, English consonants are classed into: Labial bilabi~ ; Ip,b,m,wl

Lingual foreli~e~iO~klingUal 'ingUal



apical cacuminal It,d,n, Irl s.z.B, (:I I Pharyngeal





3. According to the place of obstruct ion, consonants are classified into den t a I (8 ,ll I, a I v eo I a r It, d, n, I, s, z/, p o s t - a I v e 0 I a r Irl, p a I ata JIll, P a I at 0 - a I v eo I a r If ,j , tf, dJ I, vel 8 r I ")1I 22

4. According to the presence or absence of voice, English consonants are subdivided into v 0 ice d Ib, d, g, v, z; 'tJ ,.) , d3 I and v 0 ice Ie s sIp, t, k, f, s, o , f ' I. 5. According to the force of articulation, English consonants are classif ied as len i sand for t j s. In the articulation of English voiced consonants the muscular tension is weak - lenis articulation. In the articulation of English voiceless consonants the muscular tension is strong - fortis articulation. 6. According to the position of the soft palate, English consonants are subdivided into 0 r a IIp, b, t, d, k, g, f, v, ,J, s, z, S, (j, dJ ' w, I, r, jl and n a s a 11m, n':J I. . In this description of the sounds of the English language we have considered the articulatory characteristics of the sounds, pronounced in isolation. But in connected speech isolated sounds are rather unusual. Sounds are grouped together to form larger un its and in the process of grouping they influence one another (the stages of their articulation merge and interpenetrate), and their articulatory features are modified in various ways. Nevertheless, those characteristics of a sound, which are significant for differentiating meaningful units, are preserved in all positions and combinations.



tJ '



A. Vowels The acoustic classification of speech sounds is based on the analysis of the spectra of the sounds. The spectra of vowels have a sharply defined formant structure and high total energy. The formants in the spectrum of a vowel are determined by certain articulatory positions and movements of the. tongue and the lips. Thus, F, is conditioned by the vertical position of the tongue. When the tongue is high in the mouth, F 1 is low, when the tongue is low, F 1 is high. E.g. I I :1 and lu:1 have F, in the reg ion of 280-300 Hz, whereas I a :1 and I D ! have F, in the region of 600-800 Hz.The second formant (F2, is conditioned by the horizontal position of the tongue and by the position of the lips. F2 is high in the case of a front vowel and it is low in the case of a back vowel. Thus, 1 ( :/ has F2 at about 2500 Hz whereas lu:1 ,has F2 at about 900 Hz. F2 of rounded vowels is lower than that of unrounded vowels, e.g. F2 of I A / is '320 Hz whereas F2 of IV" I is 940 Hz. If the formants F1 and F2 are in the middle of the spectrum, i.e. close to each other as for / a:, 1) ,<e I, the vowels are classified as compact. If the formants are at each of the extremities of the spectrum as for /u:,V", I :,11 the vowels are diffuse (see the spectrogram) . Open vowels are compact, close vowels are diffuse.


OOQ 1.Il10 ilL (a) F. /s/. 7. Spectrograms of / i :/. III 1000 6000 - la .I r / anl'We'li take a dozen".000 1.] jail MOO 1.. al :/. ..000 - '000 2000 - III 1000 - (!J) 18/ 1/1 1000 HI (c) Fig.1 t.J./..

the former are classified as compact noises and the latter as diffuse ones. b. s. labial and dental ones It.J . <e. 11. The characteristic feature of the spectrum of a consonant is the presence of a formant of noise. n. z. Therefore. g'"9. grave. Hence fricat ives are classed as cont inuant no ises. from the point of view of their acoustic characteristics. tj. The spectrum of velar and pa latal consonants tv. vi as grave. Front vowels are acute. d. Th is means that in the spectra of It. a:). The noise peculiar to alveolar and dental consonants It. grave and flat. The vowel/bIis compact. s. 3. 1\. lJ. n. ~ I have the highest formant of noise in the spectrum. zl is diffuse.character ized acoust iea lIy as tense whereas vo iced consonants (len is) as lax. . m. m. Consequently. If it is low. because in their spectrum there is a special nasal formant. f. p. d. dJ I is compact while the spectrum of alveolar. The sonants 1m. compact. for instance.t. s. tJ ' 25 . d. z.Consonants The formant structure in the spectrum of a consonant is not so sharply defined and the total energy is not so high as in the spectrum of a vowel. 1 .31 are characterized by the presence of a noise formant throughout the spectrum. The consonants Ik. n. o :/)a vowel has a dark or grave timbre. There is no no ise in that part of the spectra wh ich corresponds to the articulatory "stop". the vowels are of a clear or acute timbre. s. v. Acoustically.e/. as for l i: . Plosives and affricates (e. and lu:1 is diffuse. Thus.g.0 :. B. vi the formant of noise is lower.! '/ . b. n. e.f. F2 is lower in rounded vowels (as -0. z. vi because it is sharper in character. d. z. so that both F 1 and F2 are in the low section of the spectrum (as for lu .31 have a formant of noise only in that part of the spectra whlcb correspcndsto plosion and friction. since the burst of no ise in vo iceless plosives and the formant of no ise in voiceless fricatives are stronger than those in voiced plosives and fricatives. f. e. I.If the second formant is high. z. s. the vowel lel. d. e. s. S.3. d. is described as acute. u:) than it is in unrounded (as 1:. back vowels are grave. V'. whereas plos ives and affr icates are classed as d iscont inuous. e. til is contrasted with that of lab ial and lab io-dental ones Ip. s. rounded vowels are opposed to unrounded as flat to plain. Vo iceless consonants (fort is) are . b. m. plain. tf' dj I are intermediate in th is contrast.]'. B. b. v. 1 are opposed to a II the other consonants as nasa I to oral. lJ • nl are character ized as acute and Ip. rn. flat. I. v] are low. The frequencies of the noise formant in the spectrum of If.. It. e. tJ I the formant of noise is higher and in the spectra of Ip. The fricatives (alveolar and dental) Is. Fri~tives If.

4) grave . zl. The pr inciple of binary oppos it ions on wh ich the ir classif ieat ion is based is being used not only in phonetics but in other branches of linguistics as well. jl are non-vocalic and non-consonantal.lax. j.Halie is of great theoretical importance td lingu ists. are opposed to Ie./.plain. isquiteoriginal. The scholars established 12 binary distinctive oppositions with the help of which.Fant and M. 5) flat .Halle [45]. because in their spectra the formant structure is not we II defined and the total energy is high. dl as strident to meBow. I. w. 8) discontinuous . lr.non-consonantal. That is why they are classed as vocauc and consonantal. The acoustic information about a definite sound is often found not only in the spectrum of this sound but also in the spectrum of the neighbouring sound. 51 nasal . But in connected speech the acoustic features of the sounds are considerably modified. In the spectrum of strident consonants the intensity of the noise formant is greater than in the spectrum of mellow consonants. The traditional vowel .oral. For the English language. The first attempt to classify speech sounds on the basis of their acoustic distinctions was made by a group of phoneticians A. the following 9 binary oppositions are sufficient: 1) vocalic . which have a flat narrowing. 9) strident . which have a very long history.Fant and M. rl are characterized by a sharply defined formant structure [like vowels).acute. The consonants Is. In comparison with the articulatory classifications. The classification worked out by R.consonant opposition is divided into two oppositions to define the sounds lr. for the construction of speech recognizers as well as machines capable of putting out information in spoken words. Acoustic definitions and classifications of speech sounds are also of great pract ical importance. G.diffuse.Jakobson. They are indispensab Ie in techn ica Iacoust ics fo r the solution of the problem of speech synthesis and sound transmission.. tu. jl are non-vocalic and non-consonantal. as they claim. 7) tense . G. the acoust ic c lassif icat ion of speech sounds. because the air passage is free as in vowels and there is an obstruction as in consonants.The sonants II.continuant. 1/ are vocalic and consonantal. The acoustic characteristics of the sound types of language described above are preserved in the isolated pronunciation of the sounds.mellow. .Jakobson. based on discoveries of modern electro-acoustics. 3) compact . it is possible to classify the phonemes of any language. whereas /w.non-vocalic. but the total energy is low in their spectra (like in consonants). which have a round narrowing. and the affricates It!. The sonants Iw. dJ I are opposed to the pkrs ives It. 2) consonantal .

D /. h. Study the table of Eng lish vowels: 1) to describe the articulatory characteristics of each vowel. 2) to find out the articulatory distinctions between the following pairs of consonants /p-b. k-g. t-d. How are the formants in the spectrum of a vowel related with its articulation? Can the acoustic engineer do without the knowledge of the articulatory features of speech sounds? e. What are the acoustic distinctions between vowels and consonants? How do their acoustic spectra differ? 3. w. f-v. On what factors does the quality of a vowel sound depend? 6. C. /9-h/. . w. What are the articulatory dist ionct ions between vowels and consonants? 2. What are the factors determining the quality of a consonant? 7.EXERCISES A.u :-V'. What articulatory and acoustic features testify to the intermediate position of sonants between vowels and consonants? 4. m. s-z/. Think about the following questions for class discussion: 1. il .D:. 0:-/\ . r. Study the classification of English consonants: 1) to describe the articu latory characteristics of the sounds /?J' n. Does the articulation basis of a language condition the phonetic system of the language? In what way"' 5. 2) to find out the articulatory distinctions between the pairs of long and short vowels / r: -1 .

Those that dist ingu ish words. These various Ipl sounds differ in manner of art icu lation and in acoustic qualities. it is also modified by prosodic features. when opposed to one another in the same phonetic position. In English for example. by the neighbouring sounds). the "dark" I 1 I and the "clear" 11/ are variants. Compare /p/ in "pill" (i.C B APT £ R 3. in Russian there are 6 vowel and 35 consonant phonemes. 28 . In English there are 20 vowel phonemes and 24 consonant phonemes. Their articulatory and acoustic distinctions are cond itioned by their position and their phonetic environment. cannot distingUish words.e.e. in final position). by the position it occupies in a word or an utterance. and tempo of speech. PHONOLOGICAL ANALYSIS OF ENGUSH SPEECH SOUNDS THE PHONEME In connected speech a sound is generally modified by itsphoneticenvironment. though the differences in their production might not be much more notable than the differences in the production of the various Ipl sounds. its allophones (or variants). in "slipper" (i.e. That is why Ip. if one of the various /p/ sounds is subst ituted for another. are realizations of different phonemes./vl and Iwl in English are realizations of two different phonemes because they distinguish such words as "vine" and "wine". the first of which is stressed). A I lop h 0 n e s (or variants) of a certain phoneme are speech sounds wh ich are reali zat ions of one and the same phoneme and wh ich. All the actual speech sounds are allophones (or variants) of the phonemes that exist in the language. after /s/). That is why for the English speaking people it is of no linguistic importance to discriminate the various Ipl sounds. Every language has a limited number of sound types which are shared by all the speakers of the language and are linguistically important because they distinguish words in the language. (i. Those sounds that cannot dlst lnquish words in a definite language and occur only in certain positions or in comb mat ion with certain sounds are realizations of one and the same phoneme. E. therefore. such as stress. in "slip" (I. But it is linguistically important for English speakers to discriminate between Ipl ilnd/bl (as in "pill" and "bi II") or /p/ and Im/ (as in "pi II" and "rn ill") . But they do not differ phonologically. "veal" and "wheel" etc. between vowels. speech melody. or allophones of the same phoneme. in initial position). in "spill" (i.e. ml are different e laments of the English sound system (different English phonemes). Therefore. The subst itut ion of one for another affects communication. b.g. the meaning of the word wi II not change.e. the ph 0 n e m e may be defined as the smallest linguistically re levant un it of the sound structure of a given language wh ich serves to dist inguish one word from another.

3181.Allophones of a phoneme which never occur in identical positions are said to be in complementary distribution. they do not distinguish words in English. Some Hngu ists consider the phoneme to be but an abstraction and deny its material character. For example. None of these aspects of the phoneme can be neglected or disregarded. In spite of their acoustic and articulatory differences they are not perceived by Eng lish speakers as different sounds. lenis consonant. sounds. BCAKoe omensI-IoeeCTb (TaK I·UU~Io1Ha'le) 6LUee.. It I in "Good night" may be either a plosive or a non-plosive sound. On the other hand. BCRKoeoraensace HenOilHO BXOAIo1T B 06w. others consider it to be a class of sounds. for example.. The phoneme can therefore be regarded as a dialectical unity of its two aspects: the mater Ial and the abstracted aspects. whereas he usually pronounces a . This viewpoint is expressed by linguists of the Prague Phonological School. Other linguists overestimate the cYLUecTByeT n."clear" albphone of /II only before vowels and Ij/. Allophones of a phoneme which do occur in the same phonetic position. but can never dist ingu ish words. On the one hand. Though these sounds differ acoustically. is said to 'consist' of its various allophones. D.Jones considers a phoneme to be a family of sounds [84]." [1. they are allophones of the same phoneme and are in free variation.g. The phoneme correlates with its allophones as the un iversa I correlates with the ind iv idual. c. because in speech it is represented by concrete material sounds. It is abstracted from its variants that exist in actual speech and is characterized by features that are common to all its variants (e. That is the materialistic view of the phoneme. for whom a phoneme is but an abstract concept. the phoneme exists in speech in the material form of speecl. the phoneme is an abst ract ion and a general izat Ion. BCRKOO 06w.. For example. real and objective character of the phoneme. are sa id to be in free var lation. real and objective.sepea ornensaoe. which is generally oonsidered 29 .nenbHoM. an RP* speaker pronounces a "dark" allophone of III before consonants in final position. There exist other v rews of the phoneme. as these features are common to aU its allophones) . BCRKoe06LUBe o eelb ('IaCIIo1'1K8 CTopOHaIo1nlo1 Io1nlo1 CYLUHOCTb) nl1Wb np~61l113~TeI1bHO XBaTblBaeT O ace OTAenbHblenpeAMeTbl.. bilabial. to be the orthoepic norm of British English. It means that in English both the "dark" and the "clear" allophones of /11 are in complementary distribution as they are never opposed to each other in identical phonetic positions. "06w. For example.. The Ipl phoneme. In other words. Ibl is an occlusive.WbB oT. *RP stands for "Received Pronunciation". the phoneme is . Therefore.

it is mainly phonobgical analysis. the phoneme performs the d j s tin ct i v e function.The linguistic role of the phoneme is clearly seen from the three linguistic funct ions of the phoneme. because phonemes distinguish one word from another. Among the different types of variation we dlst inqulsh idiolectal. but when we determine the role of those sounds in communication. the con s tit uti ve function of the phoneme. The analysis is primarily phonetic when we describe the articulatory and acoustic characterictics of particular sounds and their combinations. Thus both phonemes and sounds are simply two sides of one and the same phenomenon . On account of the fact that native speakers identify definite combinations of phonemes as meaningful linguistic units (words. The phoneme is. I d j 0 Ie c t a I variation embraces the individual peculiarities of articulating sounds. or lisp (say "thish ish" for "this is").ldiotectai var iat ion may cause a lot of difficulties in communication. diaphonic and allophonic variation. At the same time it enables people to identify the speech of definite individuals. In analysing speech we constantly carry out a phonetic and a phonological analysis. a phonolog leal unit which is represented in speech by phonetic units (the speech sounds). in the form of speech sounds) they constitute morphemes. word combinations. have no mean ing.e.the sound substance of language. all of which are meaningfu I. linguists distinguish a third function of the phoneme . We may now summar ize by say ing that the phorreme is a lingu ist icallv relevant unit that exists in speech in the material form of its allophones. Though the phonemes themse lves. the use of the right allophone is not much less important. It appears that when identifying linguistic units the use of the right phoneme is not the only significant factor. MODIFICATIONS OF PHONEMES IN SPEECH Every phoneme displays a vast range of variation in connected speech. or stammer (say "a f-f-ffine d-d-d-day") . which are caused by the shape and form of the speaker's speech organs and by his articulatory habits. which can be analysed on either the phonemic (functional) levelor the allophonic (variational) level. and the non-asp irated character of Ibl. namely. since. This is why an Enqlishman will often hear "bride" for "pride" when a foreigner uses a non-aspirated Ip/. that make cteer the opposit ion of /p/ and /tll In words like "pie" and "buy". 30 . Besides the constitutive fUnction.the ide n t if i cat 0 r y (or recognitive) function of the phoneme. in the ir materia I form [i. in isolat ion. in English it is the aspiration of /p/ rather than its voicelessness. they are Iingu ist icallv important. the constitutive. Hence. and the identificatory functions l1101. For instance. words. a speaker may mumble. the distinctive. therefore. or phrases). Thus.

"So-so" 15011' so 71'/. dear" (where its features are obv ious}. or reduced to lal. It is the art ificia I departure from th is characterist ic of Eng Iish that is a corruption of the actual pronunciation of the cultivated people. The less noticeable variation of phonemes is a 110 ph 0 n i c variation. Diaphonic var iants do not affect intelligibility of speech. it ranges from a front open {?Ie I in the southern part ot England to lal in Northern England. When viewed Iingu ist iea lIy.o i a ph 0 n i c variation affects the quality and quantity of particular phonemes. no less than the number of phonet ie pos itions and env ironments in wh ieh the ohoneme occurs. The weakening of articulation and shortening of the duration of unstressed vowels results in modification of their quality and quantity. the mistake is merely in spelling. The number of allophones of each phoneme is. le rt I. Or again. the region he comes from} and his social standinp. This phonetic phenomenon is known as red u c t ion. It is caused by concrete historical tendencies active in certain localities. "So late" /so~leltl and in "Not so late" Inn t s a I le 1+t.e. its allophones. J. English vowels are considerably modlf ied: in unstressed svllanles. because in some dialects / ~ I is much longer than the standard sound. Kenyon remarks that when the schoo lbov writes "I wouldn't of gone" he is not making a mistake in grammar.le I involves significant changes in its length. it means that in a speech continuum there appear a variety of realizations of one and the same phoneme. As to its quality. In every language there are positions in which a phoneme can be easily identified because its features are fairly obvious. There are other positions in which the characteristic features of a phoneme are less obvious. 10crl in "Oh!" jAo7f/. The obscuring of vowels owing to lack of stress on certain syllables in an utterance is the result of a perfectly normal linguistic development of English. [87] 31 . Cf. yet they inform the listener about the speaker's origin Ii. For example. therefore. note the differences in /d/ in "Oh. Vowels in unstressed syllables are sometimes partially reduced. The listener easily not ices both idiolectal and diaphonic variants. as in rnnt sa. It has already been mentioned that in connected speech the sounds undergo various modifications under the influence of neighbouring sounds and the intonation patterns they occur in. the diaphonic variation of l. wou Id be a worse blunder than to misspell "have". "Good!" (where it generally loses its plos ive and voiced character) and in "Good-bye" (where it actually loses all its features). He says that to pronounce "have" and "of" different Iy in sentences that requ ire the unstressed forms. but it does not take him much time to 'tune in' to the speaker's manner of speech and understand him. which is conditioned by phonetic position and phonetic environment. The two expressions "I WOUldn't have gone" and "I wouldn't of gone" are exactly alike in standard pronunciation. as in Iso \ Ie HI.

as in "shirt". mainly by the following consonant. When a consonant is a component of a consonent cluster. "cheese". the shortening of li:/ in "cease" (as compared to "seize" or "see"). An illustration of that is the nasal'ization of lei in "men". For example. by the system of phonemes in the Jt 32 . Neg lect of these tendencies results in foreign accents and unintelligible speech. There is always an overlapping of articulatory movements of neighbouring sounds. For example. it is part ially or completely assimifated by the neighbouring sounds. The speaker avoids art icu latory movements wh ich are not absolutely necessary for intelligibility of wou Id be wrong and would simplify the problem. "great"). One should admit that these phenomena 'occur more obv iouslv in rap id and care less speech. reduction depend only on tempo and sty Ie. it depends ml'inly on its position in the intonation group. is known as ace 0 mod at ion. It is generally considered that allophonic modification is caused by "economy of effort". Spectrographic analysis makes it evident that the modifications a consonant undergoes in the final position of an intonation group radically differ from those in the initial position. Thus before li:/ (as in "yield") the central part of the tongue is raised higher than for li:l. but they are also observed in careful speech witli moderate tempo and even when the tempo is slow. "leap" or" June". the height of the central part of the tongue for Ii/ fully depends upon the following vowel. The analysis of the phonetic modificat ions that Occur in the speech continuum reveals the phonet ic tendencies of a language. accomodation. This is one of the main reasons why there is a need for further investigation of this prob lem. The pa lata lizat ion of these sounds is caused by the anticipatory upward movement of the bulk of the tongue for the next sound. English vowels are also modified by the neighbouring consonants. This process is to some extent regu lated by the orthoep ie norm. or a consonant to a vowel. A consonant may be voiced (as in /tra :nz1leI til or devo iced (as in /al 'e1!) k So1f'/) .du: l til. in the final position voiced consonants lose their plosive and voiced character (as in "Goodl "}. Or again. To state that assimilation. that they occur on Iy in rap id care less speech. If'3' tj . whereas in the initial position they retain those features but are modified in another manner by the following vowel. "ten". The process of adapting the articulation of a vowel to a consonant. All English phonemes in various pronouncing conditions undergo ass i mil a t ion.Apart from that. there m1ayeven occur coalescent assim ilat ion wh ich resu Its in a new phoneme (as in Iha'lr d3a au:1 or /wo7. 1/ are slightly palatalized when followed by mid and close vowels or by IjJ. while in "yard" it reaches but the height of le/. it may lose its plosion {as in "that time"). As for the quality of an English consonant.Fntje . in the word "yes" it is es high as when articulating Ij/. or the plosion may become restricted (as in "please.

g.Grammont cia ims that the different phenomena of combinatory phonetics are regulated by what he calls "the law of the stronger" [74]. [68] Whether they really are the most stable sounds in English requires thorough analysis. Ia : I. Whatever are the causes. That can be easily illustrated by an example from the English language: in "of course" 1 a f 'k u :sl the /VI is assimilated by I kl land not v ice versa) . Grammont considers that the analysis of combinatory phenomena in every particular language should aim at revealing the phonemes that resist modifications. and by the system of phonolog ically relevant features of phonemes in the language. c) the interrelationships among the phonemes of a language. there are differences in the opinion on the phonological status of t v:t.6248 33 . because tvt here is at the end of an unstressed syllable and is therefore weaker than the initial Ik/. the stronger phoneme influences the weaker one. English lenis consonants/b. d/. but the "clear" allophone of III does not occur in the positions in which the "dark" allophone of III is used. some phoneticians treat /11 and Iwl as allophones of li:1 and IU:/. but they cannot be replaced by fortis /p. n. or by its pos!tion in the syllable. Consequently. But the inventory of the strong and stable phonemes has not as yet been estab Iished for any part icu lar language. and I :) :1 res pect ive Iy. The stronger phoneme assimilates. /" I and I]) I. "lJ. or accomodates the neighbouring phoneme because of its articulatory strength and stability. Analysis of relative frequency of occurrence of English speech sounds shows that the most frequent consonants are It.. "cab"-"cap".!. J/ b) the identification of the inventory of phonologically relevant stinctive) features of a language. as in English the fortis and lenis consonants distinguish words (e. M. if the latter is used instead of the former. "bag'. it is immed iately detected by the nat ive spea kers as a IocaI accent. d. Some scholars consider that frequency of occurrence of phonemes and phonemic clusters may be a factor of stability in language in the sense that frequent phonemes resist modifications and modify the rare ones. t. For instance.language."back"). g/ in final position can be voiceless. "had"-"hat". It has been noted that / may be "clear" Or "dark" in one and the same phonetic position. (di- 33al<. M. k/. the fact that speech is comprised of a great var iety of allophones compl icates the identificat ion of phonemes in connected speech. Thus. I.According to it. It also accounts for the existence of controversial views on phonological problems. s. some scho lars treat them as a llophones of lu -J. the main problems of phonological analysis are as follows: aJ the identification of the phonemic inventory for each individual language.

THE PHONEMIC INVENTORY OF ENGLISH The first problem of phonological analysis is to establish the phonemes in a definite language. The opposit ion Izl versus I-I is called a z e r 0 (phonological) opposition. it is clearly ev ident that in English [s] and (t] are real izat ions of two different phonemes ("sean-"tea". The pairs of words which differ only in one speech sound are called min i m a I p air s. The distributional method of analysis is a purely formal method of identifying the phonemes of a language. while [t] aspirated and [t] non-aspirated are allophones of one and the same phoneme. The semantic method of identification of the phonemes in a language attaches great sign ificance to mean ing. But it appears to be complicated and the investigators very often cannot do without native speakers to confirm their conclusions concerning the phonemic status of certain speech sounds. The invest igator studies the funct ion of sounds by collecting minimal pairs of words in the language. Such analysis is sometimes referred to as "minimal pair test". But one cannot find [pJ aspirated and [p] non-aspirated in the same phonetic position in English. Therefore in English they are allophones of one and the same phoneme. in its tu rn. Thus. while allophones of one and the same phoneme occur in different positions and. There are two methods to do that: the distributional method andthe semantic method. "rope" . as Ipl and Ibl can free Iy occut in the same phonet ic context (as in "pea" . If two speech sounds distinguish words with different meanings. The semantic method is widely used by scholars all over the world. If not. they form a phonotoqical opposition and are realizations of two different phonemes. because they occur in the same phonet ic pos it ions. The s e man tic method. they are consequently different phonemes. This is why the d istribut iona I method of identification of the phonemes in a language works even when one does not know the language at all. is based on the phonolog ical ru Ie that a phoneme can dist inquish words when opposed to another phoneme or zero in an identical phonetic position."bee". cannot be phonologically opposed to each other. For example. whereas in Chinese the aspirated and non-aspirated stops are regarded as different phonemes. as they cannot distinguish words. they are allophones of one and the same phoneme. The opposition Izl versus It I IS called a ph 0 no Jog i c a J 0 p p 0 s i t ion. "so"-"toe")."robe"}. The dis t rib uti 0 n a I method is based on the phoriolog leal ru Ie that different phonemes can freely occur in one and the same position. This can be carried out only by phonological analysis based on phonological rules. The method is widely used by the American linguists who study the languages of the Red Indians. It was mainly with the help of the semantic method that Soviet scholars identified 34 . therefore.

But does that mean that all of them are monophonemic and should be included into the phonemic inventory? N. because as they say I jl and Iwl can form phonolog leal oppositions with each other and with other phonemes (e. a phoneme is indivisible. Idzl form phonological oppositions and distinguish such words as "eat . "abundant" I a'b" nd a nt/) . We support the second v iewpo int not on Iy for the reason mentioned above.I vs. That made it possible to create written languages for them. "headhedge". "yell" .try"/'die ..the phonemes in the languages of the numerous peoples inhabiting our multinational country. because 1.Jakobson and other American Iingu ists treat them as allophones of /11 and lu! on account of their weakness and unstable articulatory features. One ought to mention the fact that III and Iwl in English occur in phonetic positions that are generally occupied by consonant phonemes. The main rules state that. 3* 35 . The problem is whether there is a schwa vowell a / phoneme in English.illusion I"."well"..foreword It is sometimes oonsidered that la I is an allophone of 1"1. I a I vs. especially when the sound is of a complex nature. whereas la I occurs only in unstressed svllables."meat") .1 is almost exclus iv'ely used in stressed syllables (as in "comfort" / I k /\ mf a t].buds"."buzz . Such difficulties arise when one deals with weakened vowels occurring in unstressed posit ion.solo I a I vs. "tie .: There are controversial views on whether Iii and /wl in English are allophones of I I / and lui or they are separate phonemes. armour . allusion .except. Scholars are not in agreement on this point. "wheat" .offices.tempo solar .. which are partially reduced due to their position in unstressed syllables. Its/."met". Idr/. There are cases when the establishment of phonoloq ical oppositions is not suff icient to determ ine the phonemic status of a sound.dry":'hat . consequently.3 /. Trubetzkoy worked out a number of rules which help to determine whether a sound of a complex nature is monophonernlc. Itr/. E. R.each". Id. as no syllabic division can occur within a phoneme. lovl temper .g.hats". officers . it can form phonological oppositions with a number of other phonemes and can distinguish words.g. Though lal can be opposed only to weakened vowel phonemes. they cannot be considered to be allophones of vowe I phonemes. firstly. In the English language the sounds It! I. [82] Whereas other scholars treat Ijl and Iwl as phonemes. a phoneme is produced by one articulatory effort. "yet" . Time and again there emerge difficulties as to the phonemic status of certain sounds. But to identify all the phonemes of a language is not always a simple thing to do. It primari Iy concerns the schwa vowel /e / in English which occurs only in unstressed position. I I I accept . 13:1 forward .

fa va l. a 1. 36 . There appears to be another analog ical problem.Jones calls them affr icates alongside of Itj I and Id. Itsl. as there is an increase in the force of articulation and intensity not only for the first element. o r. [40] Consequently.) Most phoneticians regard hr! and Idrl as biphonemic clusters. In such a way it has been established that in RP there are 12 vowel phonemes: Ii:. in words like "cheese.Q:. "flower"). Idrl (as in "tree. a:. a 11"/. *Cf. 3 :/.~I . because both acoustic and physiolog lea I analysis provide sufficient evidence that these sounds are produced by one articulatory effort. la £. the duratlon of a phoneme should not exceed that of other phonemes in the language. hand-writing". u:.e. hedge. lsI or Iz/. u. Are they monophonemic or b iphonem ic clusters in English? The syllabic and articulatory indivisibility of English diphthongs* and their duration which does not exceed the duration of English historically long vowels /i:. or If I (as in "share").(el . A . chair. on the one hand. o i. They are not produced by a single articulatory effort.j I. Besides that. dream") their phonemic status will remain undecided until special acoustic and physiological analysis is made. each. D.Thirdly. 3:. e a. on the other.I>-<!J I as in "flier". jail".I"J/ "blowing". /blov. 3 glides to t» I -/ r a . are so close Iy Iin ked in the pronunciat ion of Eng lishmen. ae . jail. their duratton does not exceed the duratlon of either It I (as in "tear"}. combinations of two phonemes).8 diphthongs: 3 glides to 111. 11'".P . I . but for the last element as well. hedge" are monophonemic. clearly determine their monophonemic character in English. and I dl and trt. each. or Idl (as in "dare"). John. On account of that they shou Id be regarded as b iphonem ic clusters of a diphthong with the schwa vowel. / dl.VI/. It! I and Idj I in words like "cheese.a I.e.. As for laI<!J I. As for ltr]. because their duration exceeds the average duration of either Itl. Itzl are obviously biphonemic combinations (i. The syllabic division generally occurs in between the diphthong and the schwa vowel U ar. it has been proved acoustically and phYSiologically that in English they cannot be considered monophonemic. [yc-n 3-MY] "ycnelO"/pleJ-a/ ~ "player" and [n6-MvJ "noio". As It I and lr}. [102] (He distinguished them from sequences It/+ Ir/ and Id/+ trl as in "rest-room. ?f'a/. It concerns the phonemic status of the English diphthongs and the so-called "tr iphthonqs". no syllabic division occurs within the sounds ItJ I and Idj I. 2 g Iides to 1?I'1-/oV'.

.24 consonant



d. s, I,C), v, m, k, w, z, r, b, f, p, h, 'fj,g.j.

i. d3 '


As the diphthong /p a I and the labiovelar fricative /fV\/ are not used by all RP speakers (only some RP speakers differentiate words as "pour - paw", "which _ witch") they are generally called facultative phonemes. But since they are used by many RP speakers as phonemes, V.A. Vassilyev includes them into the phonemic inventory and states that in RP there are 21 vowel phonemes and 25 consonant phonemes. [110) THE SYSTEM OF PHONOLOGICAL OPPOSITIONS IN ENGLISH The second problem of phonological analysis is the ident it icat ron of the inventory of distinctive features on which aU the phonological oppositions in the language are based. Every sound is characterized by a number of featureS,notallofwhichare equally important for commun ieat ion. If one compares some of the allophones of /p/, it appears that all of them have common features and features which characterize only a few of them. The problem is to decide which of the features of a group of common sounds in a certain language are phono log ica Ily relevant and which of them are irrelevant, or incidental. This is important not on Iy for a detai led character lzat ion of the phonemic systems of languages and the identification of their typology. It is also most essential for teaching purposes, since the phonolog icallv relevant features require special attention in pronunciation teaching. We primarily need to explain the following: Each phoneme is characterized by a certain number of p h 0 n 0 log ic a I I y rei e van t f eat u res, which are its constant distinctive features (as they d lst ingu ish the phoneme from a II the other phonemes of the language) . Each allophone of a certain phoneme is characterized by definite phonologically relevant features (which are common to all its allophones) plus a number of i r rei eve nt, or incidental, f eat u res (which distinguish the allophone from all the other allophones of the phoneme) .







bilabial occlusive fortis aspirated

bilabial occlusive fort is aspirated



bilabial occlusive fortis non-aspirated non-plosive unrounded


The phonologically relevant features that characterize the phoneme Ipl are, therefore, bilabial, occlusive and fortis. Aspiration, plosiveness,labialization, etc. are phonolog ically irrelevant features. Phonologically irrelevant does not necessarily mean useless for communi-« cation. It has already been mentioned that the aspiration of Iplhelps the listener to distinguish it from Ibl (as in "pride" - "bride", "pie" - "buy"). The point is that if the speaker substitutes one phonologically relevant feature (say, bilabial) for any other relevant feature (say, forelinguall. the phoneme ceases to be the phoneme it was and becomes a different phoneme (in th is case Ipl is replaced by It/) . Such a subst itut ion is easi Iy perceived by any nat ive speaker whether he had been trained in phonet ics or not. (Cf. "pie" - "tie", "cap" - "cat"). The subst itution of one irrelevant feature for another (say, asp irated for non-aspi rated) resu Its in a different allophone of one and the same phoneme ([p I aspirated and {pI non-aspirated). Such a substitution does not affect commun icat ion. When interpreted phonolog icallv, it means that not all the art ieu latorv featu res of actual sounds are equally important for the ident ificat ion of the phonemes of a language and the words they const itute. Someof the features are' phonologically relevant, others are irrelevant. The phonologically relevant features are normally identified by opposing one phoneme to every other phoneme in the language. But there often occur difficulties, which can be overcome with the aid of physiological and acoustic analyses. Thus, until recently it was considered that the oppositions Ip-b/, It-d/, /k-g/, If-vi. I 1) I. Is-z/, If -31, Itf-djl were based on the presence and absence of voice. But it has been proved that the presence or absence of voice in these oppositions is not a constant distinctive feature, because the so-called "voiced" consonants in many phonetic positions are actually not yoioed throughout. Udl in "Do that" is voiceless in its initial stage, while /d/ in final position is voiceless either in its last phase or throughout, as in "Yes. I did". Ch. Barber states that the so-called "voiced"/b, d. g/ are actually voiceless after diphthongs and historically long vowels, as in "rogue", "feed"). [54) Consequentlv, their voiced character cannot be considered to be a phonologically relevant feature. Yet, the oppositions (p-b/, It-d/, etc. exist in the English language. and "cob" is never pronounced as "cop", "had" can never be subst ituted for "hat", and so on. So there must be at least one phonolog ically relevant feature on which such oppositions are based. Acoust ic and physiological analyses have proved that the so-called "voiced" consonants in English are always len i s (lax, weak) irrespective of their phonet ic environment and position, as compared to the so-called "voiceless" consonants In English wh ich are always for tis (energetic, strong) in all phonetic posit ions. Compare Ip t k/ and Ib t g/. Ipl and Ik/ are pronounced with



the lips quite tense, with aspiration (especially in the initial/p/) and with the vocal cords not vibrating: the Ibl and Igl are pronounced with the lips more relaxed, with no aspiration, and with the vocal cords vibrating only in the initial Ib/. Therefore, the phonolog ical oppos it ions Ip-b/, It-d/, Ik-g/, If-vi, Ie-'di, Is-z/, / Itf -dj / are primarily based on fortis vs.lenis articulation, which are their phonologically relevant features. Besides the energy difference, the fort isllen is corre lat ion also implies that the len is sounds are regu larly shorter than the fort is ones. Thus/z/ is a good· deal shorter and much less enerqet ic than ls] (ct. "lose" - "loose", "as" - "ass"). The presence of voice in /b, d, g, v, z, 3, d31 is an incidental feature that can be neutralized in certain phonetic positions. That is why, when teaching English pronunciation, the importance of the vo iced character of these sounds shou Id not be overestimated, whereas special attention should be drawn to their lenis character. Or again, until recently duration in the English language was regarded as a phonologically relevant feature capable of distinguish ing /i:- 1 /, /u:- 'U" I, 10 :-pl, I ai=s], 13 :-a/. But in Eng lish the quality of the historically lonq and short vowels differs, so length is not the on Iy feature that d ist ingu ishes them. Besides, length differences are conditioned, they therefore cannot be distinctive. Acoustic analysis shows that the length of vowels varies in different phonetic environment and in different positions. /si: - si:d - si :tl It is a well known fact that li:/ in ISi:1 is longer than the same sound in /si:d/, and that li:1 in Isi:tl is the shortest: it is almost as short as IIlin/sId/. It has also been established that a vowel is longer in front of a fricative than in front of a stop. Thus, lee J is longer in liR-sl than in lCEt/. Besides that, vowel length depends on whether the syllable it occurs in is stressed. Stressed vowels are generally longer than the unstressed vowels. Vowel length also depends on the number of svllables in a word. Compare the duration of /ct.1 in "arm" - "disarm" -" disarmament". As length varies and does not characterize all the allophones of a historically long vowel, length cannot be considered its phonologically relevant feature. But there are perceptua I features wh ich constantly dist ingu ish all the English vowels: these are distinctions in their quality, which are based on the slight differences in the tongue positions when producing these vowels, i,e. the ir articulatory differences. Thus, the opposition li:-r/ is based on the following phonologically relevant features: high-narrow vs. high-broad, fully-front vs, front-retracted. The opposition J:) :- P I is based on the following phonologically relevant features: back-advanced vs, fully back, low-narrow vs. low-broad. Duration, though it is an incidental feature and therefore cannot be considered phonologically relevant, is nevertheless a very "important feature that serves as an additional means of identifying English sounds. For example, the

f -" /,


no) flat narrowing vs.she.he) labiodental vs. sonorant (bad .sat. low . lenis (two . 40 . mediolingual (less .wine) fricative vs. bicentral (see . "had-"hat"). see .bag. forelingual (fat. thigh .affncate(tillchill.shy) THE INTERRELATIONSHIPS AMONG THE PHONEMES OF A LANGUAGE Once the phonemes of a language are est anusned and their phonologically relevant featu res are determined.sat. sonorant (vine . do . fricative (pen . backlingual (take . round narrowing (thing .shortening of a vowel generally signals that the following consonant is fortis and vo iceless (ct. presence of a glide. mediolingual (wail.sing) unicentral vs. wet .Jew) plosive vs. there arises another phonolog lea I problem: to describe the interrelat ionsh ips among the phonemes of a language. "seed" -"seat". mat . occlusive sonorant (we .me.when.chair) constrictive sonorant vs. When a monophthong is opposed to a diphthong. there is one extra distinctive feature that differentiates them: it is absence vs. f 1 these oppositions are based on both the above mentioned differences and the absence or presence of a glide. In the system of English consonant phonemes there are oppositions based on the force of art iculat ions fortis vs.) There are oppos lt lons based on the active organ of speech: bilabial vs. backlingual (bay . in the system of English vowel phonemes there are oppositions of monophthongs between themselves and diphthongs between themselves monophthongs diphthongs vs. pharyngal (they .gay) forelingual vs. deck .cake) forelingual vs. etc. There are no phonolog ieal oppositions in the system of English vowel phonemes based on length alone.Yale) bilabial vs.yes) forelingual vs.hay.neck) fricative vs.very) ploslvevs.ten.let) bilabial vs. {these oppositions are based on differences in the movements and positions of the tongue and the lips. van .mad. berry . affricate (share . forelingual (pen . back .than) There are oppositions based on the type of obstruction: plosive vs.

Avanesov [2QJ. Neutralization of phonological oppositions IS the loss of a distinctive (phonologically relevant) feature by one of the phonemes of an opposition. Those who support this view consider that a phoneme is morphemically bound and. it becomes an allophone of the phoneme the distinctive features of which it acquires. llYK) we deal with the allophones of the phoneme /K/. In the case of [kl.Pike [104J and others) reject the notion of "neutralization of phonological oppositions". Russian voiced consonants lose their voiced character and are pronounced as voiceless in final position (as in "nyr" Ikl "mesrIc]. Scholars term th is phenomenon n e u t r a liz a t ion of phonological oppositions. The supporters of the phonological viewpoint (l. R. Reformatsky [33] and others) claim that a phoneme in a "weak" position may lose one of its distinctive (phonologically relevant) features and. K. Accordingly. Scholars who support the morphonological viewpoint (R. Consequently. Jakobson [82J. They consider that an allophone cannot lose any of its distinctive features. The archiphoneme is an abstraction which combines the distinctive features common to two phonemes. Therefore. in all the derivatives of "nvra" (nvroa. ~This leads to the loss of the distinction between h<1 and lt]. the phoneme is characterized by def inite articu latory and acoust ic characterist ics and can be easi Iy described 41 . nor voiceless. Shcherba and his followers. nvr) we deal with the allophones of one and the same phoneme Ir]. We assume that for teaching purposes the most suitable viewpoint is that of l. [a ] in "addition" is an allophone of the schwa vowel phoneme I~ I (and is not an allophone of [se].Can different phonemes have common allophones? Can allophones of a phoneme lose any of their phonologically relevant features in certain phonetic positions? There are three views on the problem. [K] in "nvr" is an allophone of IK/. therefore. l tl in "walked" is an allophone ~f It/. etc. therefore. Trubetzkoy [40]. S'ut the Russian language is the on Iy languag~ in wh ich the phenomenon of neutralization has been examined more or less in depth. D. lei and 13/. different phonemes may have common allophones and sometimes a sound may be assigned to either of two phonemes. vo iceless" are neutra Iized in Russian. and in all the derivatives of "nvxa" (llYKOM. in word final position the phonological oppositions based On the phonologically relevant features "voiced vs. The third viewpoint is that of N. Kuznetsov (291. it may either be cons idered an allophone of the phoneme /K/ (as in "nYK") or an allophone of the phoneme /rl (as in "nyr"). P. and some other linguists who consider that there are phonological units higher than a phoneme . Shcherba [441. For example.Jones [86).the archiphonernes. as in "add"). lose its distinctive function. A. If it does. Thus. Aocording to this viewpoint both l k l and [r] in "nvx" and "nvr" are assigned to the archiphoneme IK/ which is neither voiced.

TYPES OF TRANSCRIPTION Besides the problems of phonological analysis of speech sounds discussed above. or b r 0 a d. because even its phonologically relevant features appear to be unstable (they can be neutralizedl . transcription. One can class ify the sounds into phonemes disregarding the different degrees of aspiration. The former would increase the number of symbols considerably. there may be different tvpes of transcriptions depending upon the degree of exactness a separate un it of the sound system of language. the transcript ion shou Id prov ide each phoneme with a distinctive symbol to avoid ambiguity. one can differentiate between all those features and classify them as well. This problem is closely connected with the problems of phonological analysis already discussed. On the other hand. and that would "~"$e great difficulties for those who use it. it is always a generalization about them. palatalization and other phonologically irrelevant features of the sounds. or introduce special marks to represent the diHerent features of the allophones. The phonemic data is usually enclosed between virgules (also called d iaponals}: It/. or syllabic writing (with a symbol for each syllable. length. Such an approach hinders the practical application of phonology to teaching pronunciat ion.labialization. because any system of writing is not a simple record of speech utterances. as in Chinese writing). the t ranscr ipt ion shou Id prov ide either different symbols for each allophone. The 42 . If it is exactness in the differentiation of the allophones of each phoneme that is requ ired. writing systematizes and provides a distinctive symbol for each class of sounds it represents. The extent of the gene ral izat ion may vary. Moreover. Be it ideographic writing (with a different symbol for each word. phonology deals with the problem of representing speech visually. Such a transcription is generally called p h 0 n e m i c. It contains as many symbols as there are phonemes in the language. Whereas the other viewpoints treat the phoneme as a phonological unit which is actually devoid of articulatory and acoustic characteristics. as in Englishl. A t ran s c rip t ion. If it is accuracy only in the representation of the phonemes of the language that is requ ired.known fact that language is too compl ieated for a II its features to be described in terms of anyone theory. Consequently. wh ich is a v isual system of notation of the sound structu re of speech. is also a general izat ion of a great variety of sounds that are uttered by speakers of a given language. the phoneme in that sense embraces sounds that can be assigned to other phonemes as well (the so-called "common" allophones). The existence of a number of viewpoints on phonological problems ean be expla ined by the well. or alphabetic writing (with a symbol for each phoneme or combination of phonemes. as in Japanese writing).

au. EI :. lowered variety. The broad type of the International Phonetic Transcription was first used by D. t = American "voiced" etc. 0 .3 . ~ = syllabic n. etc. The modern phonetic transcription that is most widely used now is the International Phonet ic Transcr ipt ion dev ised by the Internat ional Phonetic Association in 1904.or a = sound between a and a raised variety. although it makes the notation rather complicated.a:. <>:. 101 (close lip rounded/e/in French "peu"}. E. One can hardly do without it in foreign language teaching.u. Scholars usua lIy ma ke use of both ways: they prov ide some of the typical allophones with distinctive symbols and introduce special marks (called "diacritic marks") to denote the different features the allophones are characterized by.Jones in his "English Pronouncing Dictionary". The narrow type of the transcription makes use of extra symbols /lJ . + advanced var ietv: u + or ¥ = sound between u and-l*. i. EE . For this reason the transcript ion is often referred to as the "u n i ve r s a I t ran s c rip t ion" of the IPA (International Phonetic Association). under a letter (or over it if the letter has a tail below) means that the sound is syllabic. Most of the symbols it uses are letters of the Latin alphabet. retracted var iatv: a.. 3.g. I (as in "bag") . '£ = nasalized e . " . length mark. 3 = unvo iced n. " . half length. J. ee " " slight aspiration after /. One of the principles of this transcription is to use the fewest possible symbols of the simplest possible shape. s= z . or n a r row transcription. It has always been one of the main concerns of the phonetic science to work out a transcription. 1./y/ (close lip rounded/ i/in German "u"}. A phonetic transcription is essential for scientific and practical use. accents.U: .a . Such a transcript ion is called a p h 0 net i c.L or i = ~ = .Ie . in studying spoken languages. a. dialects. Besides. published in 1917.ou. I <Je.. t.oi.latter is more economical. These are the symbols that he selected for English: Ii:. ']. ia . e T or -r = e. The phonetic data is customarily enclosed in square brackets: [tJ.v .ei. [86J 43 . e. it contains a series of diacritic marks. Th is transcr ipt ion is a phonetic alphabet which may be applied to most of the languages. The first attempts to represent speech sounds visually by means of special symbols were made as far back as the 16th century. such as nasalization. devoicing. I landdiacriticmarks. That is why it contains symbols that stand for phonemes in different languag.

g. compact and diffuse. The "linguistic alphabet" includes new symbols: " !t/ for I" I (e. EXERCISES A. later it was expanded by B. /~I for / J I. "just" Ij'.Bloch. G. rhythm and • /ih/ for 11 ~ I (e. What helps the listener to cope with what he hears? Why is the phoneme concept essential to the phonetic science? 2. " Ijl for /d.g. such as speech melody. L. Why can't allophones of one and the same phoneme be in contrast with each other? Can they occur in exactly the same phonetic environment? Can they d ist ingu ish differences of mean ing? 3. The "linguistic alphabet" is widely used by those who study American English pronunciat ion. hI in law. /~/ for tc] I. etc? Which of the two presen+ tations is preferable for making comparisons between the phonemes of 44 . "near" Inihl). ow. Phonology has developed rapidly and made a profound study of the functions of sounds in most of the living languages.Trager [561 and other American linguists. Scholars are beginning to tackle these and similar problems of intonology. These features have been presented in art ieu latory terms. tempo of speech. Can they also be presented as oppositions between acoustic features. grave and acute.g. Bloomfield [55] was the first to use it. J I suggest that the sounds are really monophonernic consonants. it is doubtful whether the postvocalic glides Iw. /iy/ for /i:/ (e.j/.g. Z • c . hi. But so far it has not as yet fully examined the functions of most of the prosodic features. Every language contains an immense amount of actual speech sounds. e. v vvY If the symbols Is. Think about the following questions for class discussion: 1. Why isn't it possible to make any rigid separation of phonetics and phonology? Does phonetic analysis necessarily involve phonological considerations? Are phonological considerations always based on the analysis of the relevance ·of phonetic features? 4. /~I for 131. ih/ are really allophones of the initial /w.The American linguists use what is often called the "linguistic alphabet". All the similarities and differences between the phonemes of a language can be reduced to a comparatively small number of distinctive features. "mean" Imuynn. uw.

(e) /a r I. The glottal stop is used instead of It/ when the following consonant is a stop. English spelling is by no means a reliable guide to correct pronunciat ion. D. Why isn't it possible to develop an orthography for an unwritten language without an adequate phonemic analysis of the language? 8. for example. Compare. C. the /p/ is exploded only after the glottal stop is released. compare the spelling of / I I in "little". "thought". in "What a hopei" the vowel /0'11" / is cut off by a complete closure of the vocal cords. "enough". an aiioobone of It/? 7. Or again. "women". "'tough".different languages? Which of the two presentations is preferable for teaching purposes? 5. Provide some evidence to prove that in English (a) aspirated and non-aspirated It I are allophones of one and the same phoneme. Find more examples to show how essential a phonetic transcription is for practical purposes. Iev]. Why isn't the glottal stop included into the system of English phonemes in spite of the fact that it is so widely used? Can it be regarded a:. the pronunciat ion of the "ough" cornblnat ion in "cough". The glottal stop is also used in other phonetic positions. Is it important to be well aware of these regularities to identify easily the words and phonemes of the language? What peculiarities of English combinatory phenomena should the learner's attention be directed at? 8. as it uses the same letter or letters for different sounds and it gives the same sound different spellings. "though". For instance. particularly when Ip/ is followed by a pause. "through". [b) / '!) I and Inl are realizat ions of two different phonemes. In every language there are certain regularities in the modifications of sounds in connected speech. /Pl/ are single-unit phonemes. as in "Not good enough". "busy". "thorough". Why isn't there unanimous agreement among the phonologists as to the system of English phonemes? Why can /tr! and /ju/ be treated as sing Ie-un it phonemes? Are there any other problems wh ich have no single solution? 6. It is a characterist ic feature of many RP speakers and non-RP speakers to pronounce the glottal stop [7] before Ipl. fortis It I consonants? Find other examples to illustrate that a phonemic transcription 45 . Compare Isi:dl and Isi:tl. The phonemic transcription suggests that the only perceptual difference between the two words lies in the final consonants. What else contributes a great deal to our recognition of onewora or the other/ Is there any difference in the length of li:1 which is conditioned by th€ len is /d/ vs.

Transcribe the words "spend". . (b) the narrow transcription of the IPA.does not symbolize conditioned differences. "peace". E. ~'jOY" using (a) the broad (phonemic) transcription of the IPA. (c) the American transcript ion (the "I ingu ist ic alphabet"l . "life". "choose".

by the alternation of increases and decreases in articulatory tension. Each language has its own rules of combining its phonemes into syllables. The syllable can be considered as both a phonetic and a phonological unit. "mimsy". areEnglish and "kpo". into wh ich the speech cont inuum is d iv ided. with certainty. "fsple'" cannot be English as far as the combinations of phonemes are concerned [64]. the Greek syllabe. As a phonetic unit the syllable is defined in articulatory. for example note the movements of the tongue and the lips in (su:n( "soon". without any reference to morphology (to the meaning). "slithy". because these units serve to different iate words. while producing a con sonant. the speech organs. "wabe". As a phonological unit the syllable can be defined and described only with reference to the structure of one particular language. And it takes less time to identify a syllable than the isolated sounds contained in it. Kozhevnikov [41]. And because of the specific grouping and d istr ibution of phonemes in d ifferent languages one and the same word may. It is practically impossible to draw articulatory boundaries between them. "something taken together". "together". from syn-. show that the listener can recognize the preceding sound only after he has analysed the whole syllable. On the contrary. Z. So the smallest pronunciation (articulatory) unit is the syllable. are s y I I a b I e s [27]. It has been proved experimentally that the syllable is also the smallest perceptible unit. V. carried out by Soviet linguists L. by "themselves". That is why the boundar ies between the consonant and the vowel are not clearly marked. others are not.Chistovitch. as a result. A number of experiments. "take"). boundar ies between syllables are mar ked by the alternation of openings and closings in sound production and. be interpreted as 47 . But in connected speech sounds are not pronounced separate Iv. and labein. (Iu:z/ "lose". Some combinations are permissible in a language. If we slow down the tempo of utterance and art iculate the sounds d ist inctly we shall see that the sma lIest un its. take all the positions necessary for the following vowel. from Lewis Carol's "Alice in Wonderland".C HAP T E R 4. (Cf. auditory (perceptual) and acoustic terms with universal application for all languages. Dzhaparidze [251. Therefore. "roves" etc. The very term "syllable" denotes particular ways in which phonemes are combined in a language. it is possible to say that such nonsense words as "bulling". When we pronounce a syllable. THE SYLLABLE AS A PHONETIC AND PHONOLOGICAL UNIT GENERAL NOTES ON THE SYLLABLE Sounds (phonemes) are the smallest segments into which the speech continuum is generally divided for purposes of analysis.

vowels are always syllabic.bisyllabic by a speaker of one language. Jespersen. The most sonorous 48 . there are as many syllables in a word as there are chest pulses (expirat ions) made dur ing the utterance of the word. or c h est p u Is e the 0 r y . (Hence the term "consonant".ll. So. I. Consequently. in Czeck . --dp/. audibility or carrying power).vowels and consonants fulfil different functions in speech.def ines the syllable as a sound or a group of sounds that are pronounced in one chest pulse. can also be syllabic because of their strong vocalic features. vowels are always syllabic and consonants are incapable of forming syllables without vowels. such as In. which consists of a vowel alone or of a vowel (or a syllabic sonorant) surrounded by consonants in the numbers and arrangements perm itted by a given language. accompanied by increases in air pressure. considers that sounds tend to group themselves according to their sonority. According to this definition. The ancient Greak scholars noticed that the two main phonological types of sounds . "lighten" jl a 1 -t~/. and therefore. "needn't" itv«: -d~tl. for instance. created by the Danish phonetician O. the syllable is a structural unit. speech sounds differ in sonority (prominence. Each vowel sound is pronounced with increased expiration. But in a number of languages some sonorous consonants. THE PHONETIC ASPECT OF THE SYLLABLE Phoneticians are not always in agreement in their definition of the syllable because in their analysis they proceed from either articulatory or acoustic aspects of the "seeing" lSI.the ex p ira tor y. "castle" Ik a: -s. Pronounced with uniform force. whereas an Englishman would make it trisyllabic . though such words consist of two syllables [71). whereas consonants serve as the margins of the sound combinations.) In other words. length and pitch. One of the phonetic theories .Gimson notes that it is doubtful whether a double chest pulse will be evident in the pronunciation of juxtaposed vowels as.I"} I. because the English language does not permit Iknl as an initial sound combination [99]. Boundaries between syllables are in the place where there occur changes in the air pressure. But it is impossible to explain all cases of syllable formation on the basis of the expiratory theory. which means "soundlllY with something" con+sonant."krk" (neck) and in English "garden" Ig a. The function of a vowel is to occupy the central position in certain combinations of sounds. A. For example.Ik a n o rb e I. m}. and as tr lsvllabsc by a speaker of another language. to determine boundaries between syllables. . for instance. r. The rei a t i ve son 0 r i t y the 0 r y. phonologically. a German pronounces the word "Knabe" as bisyllabic. or the prominence theory.

One peak of sonority is separated from another peak by sounds of lower sonority. e."'"d.1. there are two syllables . Close vowels I I :. and consequently. ":st. I 'Sl<eltl "skate". sfl .1/.e. But there are cases t_ha. r.jl 6. The most widespread among Soviet linguists is the m usc u I art ens jon (or the articulatory effort) the a r y which is known as Shcherba's theory. Here are some more e~lJIples to illustrate the sonority theory: I~I "wooden" 1tiid'1 "filled". Jespers_enclassifies sounds according to the degree of sonority in the following way (beginning with the most sonorous) : 1.a:. less sonorous are sonorants Iw. n. Voiceless fricatives If. I an 'a r sm. w.6. 8.. as it does not say to which syllable the less sonorous sounds belong. Yet. Voiced fricatives Iv. u:1 4. It is evident that the relative sonority theory does not explain the mechanism of syllable formation.Jones and some other phoneticians. ") I and the least sonorous are noise consonants. n. a :. '!) ../~1 "next".contradict Jespersen's theory. isrti':1 "star".d"fl ''f'ddle''.D . e. consonants. A" m a'dres~zl "some addresses" and J\!\m'a 'dres[z/"summer dresses" Nevertheless. e. separated by the least sonorous Itt. O. I. Neither does it explain syllable division. z. Mid-open vowels Ie. I!l. m.8 . The distance between the two points of lower sonority is a syllable.enl "an iceman" and la 'nata'm sen/ r'a nice man". The energy of articulation increases at the beginning of a 43aK. d.<>:1 2. the relative sonority.g.AI 3. I a n 'etrn/ "an aim" and lalneim/"a name". Voiced stops Ib. Accord ing to th is theory a syllable is character ized by var iations in muscular tension. theory has been accepted by D. jl 5. Voiceless stops (p.lei and III. gl 7. Open vowels lae. It only makes an attempt at explaining our perception of a syllable. In the word I metll"metal" there are two peaks of sonority .l. Thus in the word I melt I "melt" ~e is one peak of sonority lei and the word is monosyllabic. I nk_rtol. In these words the sound lsi is more sonorous than It I and IkI and forms the second peak of sonority. m. t..e. Sonants lr. In Czeck words like ""lk". r.. kl Sounds are qrouped around the most sonorous ones. vowels (and sometimes sonants] which form the peak of sonority in a syllable. i. sl are sonorous peaks. i. i.g.g. the words are monosyllabic.~k" and in English the sounds II.f -ba I "october". The number of syll~s is determined by the number of peaks of prominence.6248 49 .sounds are vowels.

m1Idd:el. On the perception level the syllable is defined as an arc of actual loudness. ~ I . L... If in I gnats hav-s/ the sound Inl is initially strong.. ~d. if there is a new onset of muscular tension on the sound In/. therefore. / .!1>he~._ I' E . in the articulation of which both the beginning and the end are energetic whereas the middle is weak. Acoustically they produce the impression of two consonants. in the articulation of which the beginning is stronger while the end is weaker. These consonants occur at the junction of words or m_!l.g. The boundar ies between syllables are determined by the occurrence of the lowest articulatory energy. p s eejc.syllable. : 50 . Cf. ~f. They occur at the end of a closed syllable. ~ E. so there are two arcs of muscular tension and. I "a~it'flalrTI'.. lis used to ISOlate the consonant out of the syllable. . Consonants within a syllable are characterized by different distribution of muscular tension.. the latter belongs to the second syllable. a syllable is an arc of muscular tension. "I. two syllables. showed that the organ immediately responsible for the variations in loudness of a syllable is the pharynx. The experiments carried out by N. In other words. Schherba distinguished the following three types of consonants.. 2) Finally strong consonants. while the loudness of the . The narrowing of the pharyngeal passage and the resulting increase in muscular tension of its walls reinforce the actual loudness of the vowel thus forming the peak of the syllable. and if the new onset of muscular tension is on la1/ .~ I:n. The above theories define the syllable on either the production or perception level. N.:. In accordance with this... If the sound Inl is f ina lIy strong.Zhinkin. the boundary is before it I a 'nais 'ha v sl. the syllabic boundary is after the In// 3 n a is 'ha 1I's/.1 "tar" and Ita:! "tower" (a reduced var iant of Ita?f"fJ j). 1) Initially strong consonants. There are as many syllables in a word as there are maxima of muscular tension in it. So. t1a 1 • a..pO:jt " E .g..g. which takes into account both the levels. The sound I a-] in the second example is pronounced with two articulatory efforts. They occur at the beginning of a syllable.--f. /\:s.• s:ae m: 11-:t """-:d 3) Double-peaked consonants. in the articulation of which the beginning is weak while the end is more energetic.the sound In/ belongs to the first syllable.--:. ltd .t. I I I I The type of consonant is therefore a cue for syllable division. reaches its max imum with the vowe I (or the sonant) and decreases towards the end of the syllable.. 'it: rt. Zhinkin has worked out the so-called loudness theory [26].

Perceptually. i. R.g.A. In unstressed position a short English vowel is not checked and because of the loose VC transition it may form one syllable and the following consonant may belong to another one.g. and. there is a syllable boundarv after the consonant. The absence of aspiration shows that they belong to different syllab les. In the production of loudness variations of all the speech mechanisms are involved. in "plump eye" /pll\ mp at /. So on the speech productio~ level the correlate of "the arc of loudness" is "the arc of art iculatory effort". toa consonant (VC transition) and rrorn a consonant to a consonant (CC transition) .marginal consonants is weakened./ <l n a is ho. Applied to the syllable. e.Vassilyev {1101. from a vowel. therefore the voiceless plos ives /p. According to the results obtained.e. but none of them is able to explain reliably all the cases of syllab Ie boundar ies. The acoustic aspect of the syllable has been studied by E. it refers to the relationsh ip between the phonemes constituting the syllable. because of the checked character of short vowels under stress. VC transitions are often close in English.Halle. the characteristic tendencies ·in articulatory transitions from a consonant to a vowel (CV transition).g. "The term is suggested by V. In English CV transitions are loose.Zwirner. "plum pie" / pl » m par /. "positive" / ppZ-I-tlV/. and in many cases a higher fundamental frequency. 'I/' 5/) which leads to controversial views on syllbble division.el or a sonant) has a higher intensity than its consonants. k/ before stressed vowels are aspirated. e. e. we cannot but agree with the scholars who point out that each of the existing theories is correct to a certain extent. the peak is louder and higher in pitch. It seems that the phonetic definition of the syllable should also take into account the peculiarties of the articulatory basis of a concrete language. t. Such a VC combination forms one syllable. the peak of the syllable (a vow. 4* 51 . It*The term "structure" denotes the relations between elements of a system.Jakobson and M.g. So the presence of aspiration can indicate that the consonant and the vowel belong to the same syllable. THE STRUCTURAL ASPECT OF THE ENGLISH SYLLABLE To study the syllable as a phonological unit of English is to describe the structure*"'of the syllable by stating the functions of the phonemes in it and the relat ions between the phonemes [99] . These acoustic features easily agree with the physiological definition of the syllable as an arc of articulatory effort (or muscular tension~ In analys ing the above theor ies of the syllable. However the syllabic boundaries are not always well marked in connected speech (e.

g. It is a feature of English that in initial position. because it is due. Ibl'P sm/ I I "blossom'.t s 1.'/ "I ighten". 3) covered syllables (CV (C). as in other languages. that historically short English vowels /i e. sr. 4) uncovered syllables (V (C).:t/"ar:t". they are syllabic. Consonants present particular interest in the study of the syllable. dj/. tense. CVC. /Ialfl "life". e. e. /lalkl "like". It ~l/"t ie". occurrlnq ontv betcre the vowel.g. The sounds Iw. "mixed" I-kstl. e. VC. ordinal number. such as V. CCVCC. lIa!t'. before the vowel. I f:>:J "Shore". It should be noted specially. 7/'.g. and such consonant clusters as /mh. The fundamental syllable type in English is the closed syllable. 2) closed syllables (vCI. when the vowel is preceded by a consonant. The sounds II. "texts"/-ksts/.al can occur as final. there can be any consonant except J "] I: no consonant combinat ions are possible with I ~ . In unstressed position the vowels II.nnot occur initially either. r. the number and arrangement of consonants that the structure of the syllable varies. Depending on the position of consonants (C) in relation to the vowel (V). because they represent grammatical prefixes. . I a.l\rt :-d~ l. in various sound combinations before the vowel. when preceded by a noise consonant. i. If a :1 "far". st1/ca. As to the presence. e. j/. whereas in Russian it is the open syllable. nl normally function as consonants./fskr-. <3 1 never occur in stressed final position without the following consonant. lsi :fsea~. e. This is due to the tact that final clusters are used to express grammatical meanings of plurality. f'win. CCVC. J. And it is largely due to consonants that we understand the utterance. CCCVC. vzr-. lsnl ':« it". sf. "glimpsed" I-mpst/. spw. m. then there is no consonant before the vowel. lpetllI "petal".g.1). when there is no consonant after the vowel. CCCVCC. when the vowel is fo Hawed by a consonant. e. e. there are 4 types of syllables: 1) open svllables ICV) . " . Vowels are always syllabic. e. Iset! "say". kst-/. CV.g.Syllable formation in English. O'Connor notes that final clusters are much more complex in English than initial ones. 52 .g. / j :tl "eat". they occupy a central position in the syllable. despite their strong vocalic features.g. funct ion as consonants. is based on the phonological opposition of vowels and consonants. Ija:d/.e. leItl "eight". The most frequent and fundamental pattern in English is eve. etc. I Aptl "apt". The vowe I may occur aione in a syllab Ie or it may have up to 3 consonants before it and up to 4 consonants after it. fspl-. number and arrangement of consonants there are 23 syllable patterns in English [36]. But in unstressed final position. hr. (99] Phonotactic possibilities of English phonemes predetermine the rules of syllable division. In Russian initial clusters are more complex and more numerous than the final ones. -t.g. Consonants are non-syllabic and marginal.

/k o. If it does occur in initial position in English the syllabic boundary is before it. If ~ mlill "family". fr---. Ik a:1 "car". . ai-ra) "tiny".. ge tn. II-rektl "erect".. In speech the consonant forms a close link between the two syllables. the syllable in which such a vowel occurs. 1'ffu:JI "sh illing". /rl-'gretl "regret". E. because they contain two vowel phonemes. /twen-ttl "twenty". therefore it can occur at the beginnmg of a syllable and the syllabic boundary is before the cluster.-. Idvl do not occur word-initiallv and cannot occur at the beginning of a syllable.g. before the consonant. Igu d=nrs/ "goodness" or within it.g. if it is the only consona~t between the checked vowel and the succeeding vowel. When a free vowel is separated from a succeeding vowel by only one consonant sound.g. . It.. /va1r.3miV "admit". r-v-.g. /~la nsl "science". /ar-drs 1 "idea". E. E. The syllabic boun~a-:t is therefore b~he consonants constituting the dusters... Checked vowels are always followed by a consonant.. I e -'9T1:j "agree". otn. I®l-I-SII "poJicy'-"I. {"hotter". the boundary is most probably. I 1 I E. ""---"'k l. the boundary is between the consonants.. The preceding and following vowels attract this consonant and the consonant is split into two. It lies after the following consonant. When there is a cluster of consonants between two vowels. E. ~ tfet& { IhDta Historically long monophthongs. In the unstressed syl- It 53 . If it doesn't. I . Ig~1 "admission". When a post-stressed short vowel is separated from a succeeding vowel by a single consonant. If a checked vowel is separated by one consonant from a syllabic sonant the boundary between the two syllables is also within the consonant..:t/ "cart". When two vowels are separated by more than two consonants as for example in !ekstra I the boundary may be both before lsi and It I because both Istrl and Itrl occur at the beg inn ing of words and Iks/ can occur in final position.English historically short vowels under stress (checked vowels] occur 'only in a closed syllable. /ka :-tu :n/ "cartoon". So the syllabic boundary never occurs after these yowels. as in I If\v-lr/ "lovely".a 1 "flower".g.. For instance the cluster 19r1 is used word-in itlal1y in English. The clusters Idml. /kWlk-lt! "quickly". E. diphthongs and unstressed short monophthongs (free vowels] can occur both in the open and in the closed syllable. IflaV' . E. I ad-'mala I "admire".g. E.g. /hZ) t-II/ "hotly". I .a 1/ "vowel" The structure of an English syllable depends on whether it is stressed or not. the place of the syllab ic boundary is cond it ioned by whether th is cluster is perm itted at the beginning of words or not. The peak of the 'stressed syllable is always a vowel. beI cause the short vowel is free in unstressed position._. /6I9a'1 "b igger". The so-called triphthongs in English are disyllabic combinations. is always open.b h rr m.PPS-I-'btl-l-tII "possibility".rtl. J ~I "advice".g. "letter"..

words. within a syllable or a sequence of syllables prosodic (or suprasegmental) features of speech are also realized. it is a specific minimal structure of both . For instance. and in duration (tempo. Idnl. It has been mentioned that phonemes exist and function within the syllable. accentual or rhythm ic groups. high. Istll thus making the boundaries between words. When the peak of the stressed lable is a short vowel. Istr/) and others only word-finally. the mOnosyllab ic words Ib i :t/ "beat" and /b I :d/ "bead" differ not on Iy in their consonant phonemes It/ and /d/.table the peak may be a vowel or a sonant. Two aspects of this function can be emphasized. that may be combined into the main three: constitutive. So we can say. but also in the lenqth of Ii :/. Thus syllables may be stressed and unstressed. These are distinctive variations in loudness (stress) in pitch (tone). the syllable is a un it in wh ich segmental phonemes are realized. All these prosodic features are significant for constituting the stress-pattern of a word and the tonal and rhythmic structures of an utterance. Relations of languageunits. the syllable must be "closed" by a consonant. function which is inseparable from the constitutive function. lenght). FUNCTIONS OF THE SYLLABLE syl- As a phono log ieal un it the syllab Ie performs several funct ions. long or short. In forming words and utterances the syllable performs the delimitative . distinctive and identificatory. Itn/. ~. On the one hand. The dis tin c t i v e function of the syllable is to differentiate words and word combinations. that the syllable is not a mere sum of sounds. Bondarko has proved experimentally that the relations between the dist inct ive features of the phonemes and the ir acoust ic correlates ean be revealed only within the syllable [141. The con s tit uti ve function of the syllable manifests itself in the fact that the syllable forms higher-level un its . wh ich is cond itioned by 54 .phonemic and prosodic features. morptteme I I I s 11111/9 I Fig. Therefore words are actually differentiated by the syllable as one articulatory and perceptible unit. On the other hand. utterances. mid or low. B. riSing or falling. L. Some syllables can occur only word-initially (/gr/. In the taxanomical scale of language units the syllable occupies the position between the phoneme and the word [641.

on this account he distinguishes a separate phonological unit . "ice cream".V. their intensity and formant transitions. 10'11'1and In/.the neighbouring fort is and len is consonants.the juncture phoneme. Thus. within one articulatory unit.lilait'rertl n ight'-rate. There are two types of juncture: open and close.-.LleHlile) nd a /(. differenation of the words: ~ I "lightening" (ocBel. i. I ~I tess» Q . Variations in the syllabic structure of one and the same word or a word combination may serve to differentiate styles of pronunciation. I~I I~I "a name" . I b i:tn I "beaten" . Syllable division (syllabification) is very important too in distinguishing words and utterances. The second variants are characteristic of the colloqu ial style. s o~/·1 saw her rise".Iga:dzl "guards". Thus..~I "I scream" . between two articulatory units.e.the syllabeme [11 OJ.l'n/ the open juncture is between III and 101. according to the data obtained by 55 . Therefore.. i. "we loan".. in "we'l! own" /wll o'\.. This juncture may also be called intrasyllabic juncture.Vassilyev notes that the existence of such pairs of words makes it possible to consider syllabicity the only distinctive feature of the words and. For example the word "national" may be pronounced with 3 and 2 syllables -/~I and I ~. and in "we loan" Iwt lovnl it is between If/ and 11/.~I "we'll own" .e. Such wo rds as Iga '''g~rden'' . There are some words in English where syllabicity alone is responsible :dnl are also for the. Open juncture (or open transition) occurs between syllables. Thus in "we loan" Iwt 10'1I'nl the close juncture is between II/ and 10V'I.'t~/ "lighten ing" (Mall Hlil R) . It may also be called intersyUabic juncture. the word combination "little and nice" may have 4 or 3 syllablesI ~I and I~/.G.I b i : ts I "beats" disti ngu ished not only by the phonemes In/versus /zl and Inl versus 151 but by their syllables as bisyllabic and monosyllabic words. The latest acoust ic investigat ions of juncture show that the factors determ in ing an open or a close junctu re are the durat ion of the sounds. T~~inctive role of syllabification is illustrated by examples like Inafheill "n itrate" . V.I "I saw her eyes'!_/at Due to the distinctive importance of syllable division. the syllabic boundary is often regarded by the American descript iv ists as a separate phonological unit .I~I "an aim". the transitions from one sound to another are closer within a syllable than between syllables. Close juncture (or close transition) occurs between sounds within one syllable.

Pike and LLehiste regard the juncture to be a contrastive feature of high-level units but not a phonological unit in its own right [94. What are typically English structural characteristics of the syllable? B. pernit. others consider it a suprasegmental phoneme [78] or a phoneme in its own right [79. fole. 1041. Transcribe the following words and divide them into syllables: 56 . K. That is why learners of Eng Iish shou ld take care not to mispronounce English sounds and not to shift the syllabic boundary as it may cause not only a strong foreign accent.pre-junctural Inl has falling intensity. Why is the syllable considered to be the smallest articulatory unit? Why can't we consider the sound to be the srnallest articulatory unit? 2. Choose one of the nonsense words given below which can be recognized as either English or Russian due to the structural characteristics of the syllables: vzol. What makes us assert that the syllable is also the smallest perceptible unit? Don't we perceive separate sounds in connected speech? 3. kpi. but also misunderstanding on the part of the listener. while the postjunctural Inl has rising intensity. there is less uniformity in the ir phonolog leal interpretat ion of the phenomena. because in the first example Ipl is aspirated and Iml is as long as if it were final.Lehiste [94]. Th. whereas in the second example Ipl is unaspirated and Iml is shorter on account of the following fort is Ip/. Could we describe the structure of the syllable with reference to all the languages? What universal features of the syllabic structure could you name? 6. Some phoneticians consider the open juncture to be a segmental phoneme [771. the iniUal Inl in "a n ice man" is longer than the final/n/ 10 "an iceman". strem. C.e. Think about the following questions for class discussion: 1. Formant transitions of Inl and lail are different in the contrasted pai rs. EXERCISES A. rzhest. 109]. While the phonetic realization of open juncture is described by different phoneticians in approximately the same terms. Is the syllab Ie a lingu istic un it? Is it the sma lIest Iingu ist ic un it? 4.I. mimsy. The listener identifies two syllables in "plum pie" and "plump eye" with the corresponding boundaries before /pl and after /p/. The ide n t if i cat 0 r y function of the syllable is conditioned by the hearer's perception of syllables as entire phonetic units with their concrete allophones and syllabic boundaries. What are the main problems of the phonetic aspect of the syllable? Which of the theories of syllable formation do you think is the most cons istent and he Ipfu 17 5.

require. science. 21 towel. solid. cricket. Collect pairs of words and word comb inations which are different iated by. runner. shower. function. . 31 express. bowl. specific. extraordinary. strengthen. petal. succeed. syllabic. the position of the syllabic boundary as are "plum pie" and "plump eye". E. obstruction. sonority. level.11 river. cluster.

the relative prominence of those syllables differs. Though English words generally retain their stress patterns in connected speech.liab iI ltv" And this correlation of degrees of prominence of the syllables in a word forms the s t res spa t t ern 0 f the w 0 r d. The stress patterns of different words may coincide.xam ihat ion") or more prom inent sy Ilables (as in "unre. which is. it is conditioned only by objective factors: pronunciation tendencies 58 . It is also conditioned by subjective factors: by the spea ker's intention to bring out words wh ich are considered by hi m to be semantically important in the situational context. The stress pattern of these words differs from that of "analyse". The phonetic structure of a word comprises not only the sounds that the word is composed of and not only the syllabic structure that these sounds form.. Monosyllab ic words have no stress pattern. Yet as lexical units monosyllables are regarded as stressed. . Thus.He was 'so unhappy. though their sound structures have nothing in common.He 'remembered those 'unhappy 'days. Cf.isb~have"). which should not be confused with utterance stress. WORD STRESS THE NATURE OF ENGLISH WORD STRESS A word.. And if a word contains more than one syllable. As for the stress pattern of a word. "after" have an ident ica I st ress pattern . On the other hand.L __ . 'unhappy . the stress pattern of a word is only its potential pattern in an utterance. two Unequally prominent sylJab les (as in "e. Thus the words "mother". There may be one prominent syllable in a word as compared to the rest of the syllables of the same word (as in "important"}. "syllable". "happy". People easily distinguish between "subject" and "sub'[ect". "prominent". "tab Ie". Word stress belongs to the word when sa id in isolat ion. The placement of utterance stress is primarily conditioned by the situational and linqu ist ic context. there may be two equally prominent syllables (as in "h. "character". because there can be estab 11shed no correlation of prominence within it.def in ite st ress patte rn. The stress patterns of words are generally perceived without difficulty. there occur numerous instances when the stress pattern of a word is altered. as a meaningful language unit. it a Iso has a. which is often called the a cc e n t u a 1st r u c t u reo f a w 0 r d. The auditory impression of stress is that of prominence. Actual speech does not consist of isolated words. And the stress pattern of a word is deduced from how the word is accented in connected speech.C HAP T E R S.L _ . Whereas utterance stress belongs to the utterance. word stress may be said to be a word-level concept [61]. has a definite phonetic structure.

but when pronounced with a fall in pitch on the first syllable and low pitch on the second syllable means "duck" [g9l. Japanese. duration. Russian word stress is considered to be mainly quantitative though it has been proved that durat ion is not the on Iy parameter that produces the effect of stress in Russian (28]. The mean ing of the words in those languages depends on the pitch leve Is of the ir sy llables. length. The stressed syllables are louder than the unstressed ones. frequency and formant structure are the physical correlates of loudness. In languages with quantitative word stress the effect of stress is mainly based on the quantity of the sound. because such a distortion wi II make speech in many languages the quality of vowels in stressed syllables is unobscur*It is generally accepted that intensity. Swedish word stress is characterized as dynamic and musical. For instance. Acoustic analysis shows that the perception of prominence may be due to definite variations of the following acoustic parameters. formant structure". i. Bes ides those types of word stress. Depending upon which parameter is the principal one in producing the effect of stress. greater length of the stressed syllable. Vietnamese are languages with musical word st ress (or ton ic wo rd stress). the auditory impression of stress is that of prominence. the main acoustic parameter being fundamental frequency. All these parameters generally interact to produce the effect of prominence. because both loudness and pitch variations are relevant factors in producing prominence. lingu ists distingu ish qualitative word stress. In languages with musical word stress prominence is mainly achieved by variations in pitch level. and greater intensity On the acoustic level. Stress in such languages is mainly achieved by a greater force of articulation which results in greater loudness. Chinese. The effect of prominence may be produced by a greater degree of loudness. word stress in languages may be of different types. Ail the other parameters playa less important role in producing the effect of stress in such languages.and the orthoepic norm. 59 . on the auditory level. the Swedish word "Anden" with falls in pitch on both syllables means "soul". In such languages vowels in the stressed sy lIables are a Iways longer than vowels in unstressed sy lIables. its duration. pitch and quality correspond ing Iy. One cannot distort the stress pattern of a word on one's own. some modifications in its pitch and quality. frequency. So a stressed syllable on the auditory level is a syllable that has special prom inence. As stated above. There are languages with dynamic word stress. In different languages stress may be achieved by various combinat ions of these parameters. duration. intensity.e.

What oomplicates the matter is that in English a vowel in an unstressed syllable may be non-reduced and longer than in a stressed syllable (as in "pillow". as stress was generally correlated with loudness. variations in the pitch direction will not change the mean ing of 8 word. In other words. the vowels which are the most sonorous l i. His experiment showed that as long as duration and intensity were increased together. and consequently differs greatly from the quality of vowels in unstressed syllables. with no pitch. Bolinger's experiments have shown that pitch. D. But a relatively wide departure from a monotone level is always perceived as a change in the degree of prominence. a 60 . but. duration and formant structure. In this nonsense word tot and loe I are usually judged as the points of greatest prominence. Besides that. A. and varied the relative durations and intensities of the two vowels.stic parameters: intensity. relative prominence in the listener's mind is created by an interaction of four acou. it is the pitch contrast that really matters. [70J D. For example. What type of word stress is English word stress? What is its acoustic nature? Until recently. intensity or length variations but with vowels of different quality. The peculiarity of this interaction st ill remains a controversial problem and a very complicated one. as far as English word stress is concerned.:. Thus. '~bst ract" remain to " be one and the same word. there was agreement on which of the syllables was the most prominent one. for example. [71] This shows what an important role the inherent quality of a vowel plays in producing the effect of prom inence. movement in English is also one of the most important cues for prominence. "compound"}. when increased separately. fundamental frequency. reinforcing each other. Thus. "abstract". Vowels differ in their intensity as welJ.Gimson notes that if a synthesized nonsense'word 1I1'Vlelee/ is presented to English listeners. "abstract". the intens ity of /11 is much lower than that of I 0:/ or I. English word stress was oonsidered to be dynamic.:/. the quantity of English vowels also differs: in identical phonetic environment an open vowel is longer than a close vowel. the most open vowels) will be judged most prominent. duration appeared to be more important than intensity. It is a universal rule for most languages (la: > c : > 3: > u: > i:> ae '> l) > e"> 1f' > II).e.Fry synthesized pairs of words (such as "object=object"] on monotones. and that English word stress is of a complex nature. But numerous invest igations of the acoustic nature of English word stress have made it clear that stress in English does not depend on intensity alone. 8ut it is not the pitch direction that is significant in English.

wave forms of sounds.'port".vowel following a lenis voiced consonant tends to have lower pitch than one wh ich follows a fort is voice less consonant (cf.9. 61 . "bee" "tea"). "dear" . II I ! " I J 1m I I ! Fig."tear". 2 .envelope amplitude of intensity. 3 .curve of fundumantal frequency. Intonograms of the words '~mport" and "in-. Yet an Englishman easily distinguishes a stressed syllable from among the unstressed despite the diversity in the acoustic characteristics of stressed syllables. 1 .

Vassilyev [110]) consider that there are three degrees of word stress in English: primary (or strong stress). "\ mport . "recoqru ze".tion _ cer. Some ·(e. this can also be observed in 62 . "_ ita". As for Russian word stress. There are two v iews of the matter. "austere". All these degrees of stress are linguistically relevant as there are words in English the meanings of which depend upon the occurrence of either of the three degrees in their stress patterns.xam ihat ion". Th is can be clearly seen in verbs end ing in "-ate". or the fundamental frequency may show a distinct rise (or fall). How many contrastive degrees of word stress exist in English? How many degrees of word stress are lingu ist ica Ily relevant in Eng llsh? • Instrumental invest igat ions show that a po IysylIab ic word has as many degrees of prominence as there are syllables in it.cert ifll:<.g.imlport". or duration of the stressed syllable may increase. 24 153 :op a tju mrtt [84] But not all these degrees of prominence are linguistically relevant. V. There may also be a combination of any of these parameters (see Fig. "canteen".by figure 2. the second degree of prominence . '~qualifi~at ion".t if ibat ion". it is qualitative and dynamic [15].Therefore stress in English manifests itself in various ways. "-y" (e. or the spectrum of the stressed vowel may be sharpened.K ingdon [89]. E. ')ha ir-1dresser".g. Besides.9) . either the intensity. E.Jones [84] . Secondary stress is chiefly needed to define the stress pattern of words containing four or more syllables.R. and so on. it is considered to be primarily quantitative (because in Russian a stressed syllable is about 1. D.Jones has indicated the degrees of prominence in the word "opportunity". secondarily. secondary (or partial stress) and weak (the so-called "unstressed" syllables have weak stress). though the syllable it occurs in does not actually bear primary or secondary stress. ". "occupy") and in such words as "portray". But auditory analysis shows that there are certain positions in the stress patterns of English words where the vowel generally remains unobscured and its duration is considerable. D. and compound words. The most prominent syllable is denoted by figure 1. LINGUISTICALLY RELEVANT DEGREES OF WORD STRESS One of the main questions for the linguist is to determine the number of contrastive degrees of word stress in a language. "t.g. The problem is to determine which of these degrees are linguistically relevant.5 times longer than an unstressed syllable) and. "e levate".g.

'. which is the standard of American English. The most common among them are: (words with one primary stress. It is free in the sense that stress is not fixed to any particular syllable in all the words of the language." I primary stress III (as In cup b oar d") .~mination").i.. . secondary stress I". and -L (words with one primary and one secondary stress.Torsuyev [361. .g. The st ress patterns of the bu Ik of English words are regular and stable (e.. I . Word stress in such languages is ·said to be fixed. I * GA stands for General American. tertiary word stress can be taken for a variant of secondary word stress. _ or __ . "ceremony"). "speciali"ze").defined as a correlation of three degrees of stress.Trager [lOBLA-Hili [79)) distinguish four degrees of word stress. They are the most productive types of stresspatterns too. "-any" (e. and one cannot alter it to _. distinguishes more than 100 stress patterns.i. the stress pattern of "happiness" is ..' {as in "discrimination") . On this account some American linguists (G. as in . THE STRESS PATTERNS OF ENGLISH WORDS There are languages in wh ich stress always fal!s on the fi rst svllab Ie (as in Czech and Finnish).yeek-'end").i. while tertiary stress occurs after the primary stress (as in "handbook".).J. or on the last syllable (as in French and Turkish). In yse weak stress I v I (as in "cupbd'ard". as in "after"). "-ory". That is why the stress pattern of English words may be . I' tertiary stress I '/ (as .. "territory". Linguistically. who has made a special analysis of the English stress patterns.. . 63 pronunciation . G.J. Yet English word stress is said to be free. as there are no words in English the meanings of which depend on whether their stress pattern is characterized by either secondary or tertiary stress. (words with two primary stresses.\. which he groups into 11 main types..i.magazine'" .g. . as borrowings and new words that appear in English are generally stressed accordingly.GA* nouns ending in "-ary".. but very often the weakly stressed syllable is left unmarked) . the distinction between secondary and tertiary degrees of stress is too subtle to be noticed by an untra ined ear." ana I' ").. American phoneticians consider that secondary stress generally occurs before the primary stress (as in "ex. __ . as in "hairTTl presser". "dictionary". Though the second view seems to be more exact.

L _.L . The p.indivisible"inefficiency.physiblogy..L (e.L.g.L .L _. ".. but the tendency has a Iso influenced many borrow ings I . "mother".g. In present-day English 74% of words containing two syllables have the stress pattern . __ . 36% of words of four syllables have the pattern _ .g. "mis. '. 2 syllables . the percentages of OCcurrence of woras containing different are to be expected: 1 syllable . "Lnsear'orthy"). . [75] In the English language a considerable part of the vocabulary consists of monosyllabic words. some of which are stressed. . "window") or on the second syllable in words which have a prefix of no special meaning (e.s.possibility") . This created the rhythmic tendency to alternate stressed and unstressed syllables. 3011.interpret").I.g.L • In words of three syllables 55% have the stress pattern . and in 2% the stress falls on the last sy liable [75]. . indeed. "ready". following approximate numbers of syllables 3 syllables . The recessive tendency in stressing words is characteristic of words of Anglo-Saxon or ig in.g. "USA") .g.g.g. __ . "brother" .3%. I .. forgive. (e.'sister". "po'litical.. "unplrcumcision"] .L are less common in English: Though wo rd stress in Engl ish is called free.indirviduali~ation~')._ and 6% have the pattern __ . there are certa in tendencies in English which to a certain extent regulate the accentuation of words. It is the usual way of stressing four-syllabled words (e.. others not*. ".L. garage") .' .L . 64 . remain ing 1% of words have 4 syllables or more ..The remaining types of stress patterns (e. stress is on the 3rd syllable from the end("intensity".L.L . democracy. . I I ~ • • * "In a running text of a conversational kind.. "father".. According to the recessive tendency. T T.29% have the pattern __ . According to the rhythmic tendency..phono'iogical") . The linguists who have made a thorough analysis of English stress patterns have agreed upon the existence of two main accentuation tendencies in English: the r e c e s s i vet end e n c y and the r h y t h m i c ten den c y. "excellent. ~ .. (e.T 'T' (e. In words with more than four syllables we very often find the influence of I both the rhythmic and the recessive tendencies (e.. "become". T....g. -'.I. and 26% have the pattern _ .84%. "USSR") .... combar ison").... T (e..g: "ginge'beer-.J.bottle" .J. T T. whereas only 39% have the pattern _ . (e.L (e. ...L. identify.. bJhind").g. 33% have the pattern .12%.L. stress falls on the first syllable which is generally the root syllable (e." [72.g.

t icu'latory T _ . especially if these have 1 e 1 or 1 I I.perso'nal ity.g. These are the tendencies that to some extent regulate the placement of stress in English words and condition their stress patterns. because there exist different words in English with analogous sound structure which are differentiated in speech only by their stress patterns. Without a definite stress pattern a word ceases to be a word and becomes a sequence of syllables. 'Well-known".L. __ The new variants of pronunciation of these words and many more English 'Words have been accepted and included in Everyman's English Pronouncing Dictionary by D.6248 65 .. 'controversy 'hospitable . Such meaningful prominence is given to negative prefixes "un-". suffix "-teen" (e..g. "misbehave") .'temp(e) ture") . (e. Word stress has a con s tit uti ve function. "in-".'lit (e) rature.I n rapid couoou ial soeech the two tendencies verv nften coincide as one the vowels is elided (e. This tendency is clearly evident in the new pronunciation of the following words: 'exquisite Iprecede nce . Thus. "mis-" (e. "fourteen"). "bad-tempered"). __ 'sonorous 'capitalist . "ter it Io] rv.. as it moulds svlleblesinto a word by forming its stress pattern. "ex-president' "vice-president". "under>"..L.g. This regularity is sometimes called the ret e n t i vet end e n c y in English. There is one more tendency in English: the tendency to stress the most important elements in words. there appears a stress shift with a rhythmic alternation of stressed and unstressed syllables [72]. "sub-editor". 53aK. "thirteen". "inartistic". Word stress has a dis tin c t i v e funct ion in English.Jones as either second or even first variants of pronunciation. "under-mine"). "sub-"..g. icu'latory "T __ -L _ _ or or or or or or or exquisite precedence sonorous capitalist hospitabje _J.. As a result. ~iction (a) ry.. Cf.g.. "v ice-"... semantically important elements in compound words (e.g. "red-hot". such prefixes as "ex-". It has also been not iced that the stress of the parent word is often retained in the derivatives. in some polysyllabic words there ls a tendency nowadays to avoid a succession of weaK syllables. THE FUNCTIONS OF WORD STRESS. "unknown". of ra- The rhythmic tendency remains a strong one and it affects the stress patterns of a large number of words in modern English._ _. 'personal -. __ controversy ar. nat ion -rnat icinality...

On this account he treats the degrees of stress as phonolog ieal units.inventory 4 stress phonemes: primary (or loud). Accentemes differ from phonemes. It is evident that degrees of stress can be perceived only in stress patterns as relatively strong.'import"-"im'port". Word stress has an ide n t i f i cat 0 r y function as well.Torsuyev [37]. Some linguists (G. tertiary (or med Ial) and weak stress phonemes [108. 79]. ')insult" -"in'sult". V. i.accent or accent Is it the different degrees of stress or rather the stress patterns that distinguish one word from another? There exist different viewsof the problem.Gimson [71] and others. It is actually the stress pattern of a word 66 . Hill) regard degrees of word stress as phonological units. This viewpoint appears to be well-grounded. They consider degrees of word stress to be separate phonemes.certifieationcerj if lcat lon". A.Kurath [921. weak stress. it is obvious that word stress performs its linguistic functions only as a structural element of a word. another syllable is less strong but stronger than the weak ones. Another view is expressed by G. A. On this account a stress pattern can be treated as a relevant prosodic unit. it is the stress patterns "pr imary stresstwea k stress" and "weak stress+primary stress" that distinguish words as .e. 1 ·a~cent or accent Verb in'sult abstract or abstract I . Alongside tne qenarallv accepted phonemes they have introduced into the phonemic. Thus. Moreover. secondaryaccenteme.Noun/Adjective 'insult 'abstract I' . weak stress) . A distortion of the stress patterns may hamper understand ing or produce a strange accent. They think that it is the stress patterns of words that contrast with each other rather than degrees of stress. because the stress patterns of words enable people to identify definite combinations of sounds as meaningful linguist. medium or weak stress.ic units. Therefore. He distinguishes three word accentemes in English: primary accenteme. one syllable has stronger stress than any other. is based on the phonologicalopposition of secondary stress vs.vassilyev. according to V. DB!=Irees f stress are o superimposed on syllables just as other prosod ic phenomena. in one stress pattern secondary stress may be stronger than primary stress in another stress pattern. Trager. because they are not segments into which speech may be divided. because accentemes are prosodic phonological units."import" primary stress and weak stress form phonological oppositions (primary stress vs. The distinction in the mean ing of the words '. which can distinguish words. secondary (or reduced loud). wea k accenteme. H.Vassilyev [1101 states that in minimal pairs as '~mport" . which he calls "accentemes". . But it may be argued that degrees of stress can be treated as phonemes.. .

And it is in the stress pattern of a word that the degrees of stress can be differentiated and opposed one to another. "~nsult" and "in~ult". In an utterance only the more important words are accented. There are hundreds of English words with unstable stress patterns. But stress in many lanquages is fixed. unstressed. Compare "an 'outside "b lind" and "he I went outside". Why can't stress in English delimit one word from another as it may in languages with.fixed stress? Can stress in English signal the beginning or end of a word? 7.that performs both the distinctive and the identificatory functions. though the second stress of such groups is usually 67 . 2. Find examples for each of the oppositions. other words are either slurred over or are partially stressed. Is the connection between stress and vowel quality in English very strict? Are there particular sounds which occur only in stressed syllables and other sounds which occur only in unstressed syllables? 5. Is the English stressed syllable always louder. partially stressed.. Is there a single stress per word in English? 3. "a 'Ch inese . 1.Jones.. 3. Since stress is free in English. There are languages in which certain vowel qualities are strictly associated with either stressed or unstressed syllables. fundamental frequency and duration. This produces a certain effect on the perception of stress. it is possible for it to distinguish words like "billow" and" "bililow". EXERCISES A. Are the stress patterns of English words ever altered by utterance stress? 2. Why isn't it possible for stress in those lanquaoes to distinguish between one word and another? B. and fully stressed vs. partially stressed \IS. longer and higher in pitch than the unstressed syllables? 4. Why has a fourth parameter (that of formant structure) recently been included in the list? Does the inherent quality of sounds affect the percept ion of English stress? 6.. Provide evidence to prove the followtnq: 1. Use Everyman's English Pronouncing Diet ionary by D. Stress is generally considered to be the result of an interaction of three acoustic parameters: intensity. Linguists generally distinguish three relevant degrees of stress in English and state that there occur the following oppositions: fully stressed vs. Find more words in which stress may shift from one syllable to another in an utterance. A large number of more or less fixed word combinat ions have what is called even stress. Think about the following questions for class discussion. lantern" and "heknows Chihese". unstressed.

Br. one or more syllables having been lost from the end of the word by reduction and elision.dnrr I ("ordinary"). on the first syllable.. reactionary adminis. .. a locus within a word where the sentence.trative . An examination of the stress patterns in British and American English shows that the British have a predilection for pre+k inet lcstrssses and the Americans for post-kinetic ones. C. .' New 'York. 2) for the lar-ge number of monosyllables in English. own purposes. Collect more examples. 3) for the loss of one or more syllables as in I 'e. mayor may not confer a prominence [57].accounts 1) for the large number of English words stressed. Br. I facade asepttc 'co nsequence 'faeade la-septic [89] for its 3.'Mrs. Cf.. The recessive tendency.'apple'pie. Stress is potential.'Smith. Am. February . February r~ctionary administrative "consequence . 'gold'ring. 'seventy.slightly stronger than the first:'Charles'Dickens. Express your opinion of the following: 1. Am.!five.. . [871 2.

includ ing D. and its decrease lowers the pitch. A great number of phoneticians abroad. There is no single mechanism to which the production of stress can be attributed. level intonation is also a complex phenomenon. R. intensity and duration (see the intonogramms.35. intonation is a complex combination of varying fundamental frequency. It reflects the actual interconnection and interaction of melody. rhythm and timbre are all components of intonation. K. Most Soviet phoneticians [5. This is a narrow approach to the definition of intonat ion.Gimson. These are percept ible qualities of intonat ion. rhythm and timbre in speech. alongside of its phonemic and syllabic structures has a certain prosodic structure. tempo. variations in the pitch of the musical note produced by the variations of the vocal cords"[84. J. In the production of speech mebdy the subglottal. tempo.O'Connor and G. But there is no one-to-one "relation between any of the acoustic parameters and such components of intonation as stress and rhythm. Speech melody is primarily related with fundamental frequency. as lower-level linguist ic units. Another example of the narrow approach to intonation is the definition given by J. or product ion. sentence stress.the utterance.Kingdon. p.Jones. L. stress.Arnold define intonation as the variation of the pitch of the voice. or intonation. 2751.Jones writes: "Intonation may be defined as the variations which take place in the pitch of the voice in connected speech. tempo with durat ion. But phvs loloqlcal correlates of different degrees of stress haven't as yet been established. Acoustically. Speech melody. syllables and words. The definition of intonation given above is a broad definition.C HAP T E R 6. Thus D. i.Armstrong and I. UTTERANCE PROSODY INTONATION AND PROSODY Phonemes. thus reducing it to one component .Pike. 110] define in ton at ion as a complex unity of speech melody. sentence stress.Ward. laryngeal and supraglottal respiratory muscles regulate the subglottal air-pressure.Arnolcl: "When we talk about Enqllsh intona- 69 . which enables the speaker to express his thoughts. O'Connor and G. 70). emotions and attitudes towards the contents of the utterance and the hearer. An increase of subglottal pressure raises the pitch of the voice. rhythm and voice timbre.speech melody. which makes the vocal cords vibrate. A. On the articulatory. tempo. E lectromyographic stud ies of the act ivity of the internal intercostalsshow that bursts of intercostal activity correlate with the nuclear stress of the utterance. p. Every concrete utterance. are grouped by var rous prosodic means rnto a h'lgher unit .e.

or rhythmic unit 70 . loudness. 126].e." 1 . D. loudness.Crystal defines prosodic features as "vocal effects constituted by variations along the parameters of pitch. lo. "Prosody" and "prosodic" denote non-segmental phenomena. Thus L.". Alongside of the term "intonation" the term "prosody" is widely used. loudness and duration. From the definition of prosody and intonation we can clearly see that both the notions include essentially the same phenomena. Some phoneticians apply the term "prosody" and "prosod ic" only to the features pertaining to the syJlable and phonetic word.wave forms of SOWIdI. those which do not enter into the system of segmental phonemes. p.Danes the variations of pitch and intensity. i. tion we mean the pitch patterns of spoken English. Some foreign phoneticians give broader definitions of intonation. 3 . the mus ieal features of English" [100.Fig.curve of fundamental frequency. 2 .the pitch tunes or melodies. of "Mjdsummer day?" and "Midsummer day. But the terms "intonation" and "prosody" are used differently by different linguists. F. duration and silence" [60.I ] . with rhythmicality and tempo closely related.lntonogr. pitch range. p.envelope amplitude of intensity.Crystal .Hultzen includes the variations of pitch.tone.D.

It has no meaning of its own. . 1'-' v. and to a certain extent..!. u nit (or group) is either one stressed syllable or a stressed syllable with a number of unstressed ones grouped around it.. There are as many rhythmic units in an utterance as there are stressed syllables in it. since the latter are constituted by these units. their functioning in speech and their systematization. is broader than the notion of intonation as it can be applied to the utterance. These prosodic characteristics make it possible to perceive the rhythmic unit as an actual discrete unit of prosody. R. or a c c en t u a I.I. whereas prosody of the utterance and intonation are equivalent notions. the rhythmic group is characterized by a pitch pattern lor tonal contour) and duration pattern (temporal structure)." E. consequently. stress. ILL/u/. Bes ides a definite accentual-and-rhythmic pattern. The notion of prosody. tv. The unstressed svllables are clitics.Ihich are regarded as meaningless prosodic units). and those following it .is to determine the units in which prosodic features are actualized. Prosodic features of the syllable (tone. but it is significant for constituting hierarchically higher prosodic units. The stressed sylJable is the nucieus of the rhythmic unit.g. by the stress patterns of the words.(lII.Jakobson says that prosody is one of the most difficult and controversial problems of modern linguistic studies. the phonic substance of prosody is regarded by all phoneticians as the modifications of fundamental frequency. the word. 1'-" _!_ I. words and rhythmic units. duration) depend on its position and function in the rhythmic unit and in the utterance.enclitics. 12]. A r h y t h m i c. 67]. Whatever the views on the linguistic nature of prosodic phenomena. intensity and duration. 34. The most complicated and unsolved problems of prosody are the interaction between its acoustic properties. the syllable. The s y I I a b Ie is widely recognized to be the smallest prosodic unit. to We adhere to the point of view that prosodic features pertain not only to syllables.. 71 .and oppose prosody intonation (which is a meaningful phenomenon) I5.!. Those preceding the stressed syllable are called proclitics.. PROSODIC UNITS One of the basic problems in the study of prosody . but to the intonation group and the utterance as well I9. etc. The rhythmic unit may be *These patterns are conditioned by the accentual-aoo-rhythmic structure of the whole utterance. Depending on the position of the stressed syllable and the number of proclitics and enclitics in the rhythmic group there exist various accentual-andrhythm ie patterns of it. VI.

"tone-un it" also ernohas ize the role of just one (pitch) component of prosody for the formation of the unit. "sense-group". which is uttered with a single breath. as in The rhythmic unit should. therefore.are thel'"better. be considered a meaningful one. A breath-group usually coincides with a sense-group because "pauses for breath are normally made at points where pauses are necessary or allowable from the point of view of meaning" [84. "tune". de Saussure used this term to mean two or more linguistic elements joined together: two successive morphemes or two elements of a compound word or a noun with an attribute. "intonation contour". The divisible accentual unrt may consist of several indivisible accentual units (rhythmic units).Baudouinde Courtenev applied the term "syntagm" for a word used in a sentence in contradistinction to a word taken as a lexical unit ("a lexeme"). "divisible accentual unit". or the pitch of the stressed syllable is higher than that of the proclitics) the second and the first rhythmic units of the utterance: The'brighter !they . The terms "tone-group". p. 274]. But a pause for breath may be made after two or more sense-groups are uttered. p. "tone-unit" . The term "sense-group" calls attention to the fact that it is a group of words that make sense when put together.singled out of an utterance also due to the meanings expressed by its prosodic features. the meaning of connectedness and incompleteness (when the pitch rises within the stressed syllable. the term "intonation group" [110] better reflects the essence of this unit. But it doesn't indicate its intonational character. which expresses a semantic entity in the process of speaking (and thinking) and which may consist either of one rhythmic group or of a number of such groups is what I call a syntagm" [44.V. In our opinion. "tune". separateness. The in ton at ion g r 0 u P is hierarchically higherthan the rhythmic unit. L.86J. l. so a breath-group may not coincide with a sense-group. It shows that the intonation group is the result of the di72 . newness (when the pitch falls within the stressed syllable or within the enclitics or within both) as in the first rhythmic unit of the following utterance: But'nobodyfknewabout it. It has also been termed "syntagm". The term "divisible accentual unit" [9] emphasizes the role of utterance stress in constituting the unit. though this viewpoint is not unanimously accepted.Bolinger [57] these may be the meanings of assertiveness. "breath-group". The term "breath-group" emphasizes the physiological aspect of the unit. The term "syntagm" has a drawback: it is often used with different meanings which have nothing to do with the prosodic unit under consideration. According to D. F. "tone-group". Shcherba defined the syntagm in the following way: "The phonetic entity.

Different types of head (scale) convey attitudinal meanings. J.e. Auditory observations and the analysis of acoustic data show that pitch characteristics attributed to the first stressed syllable are actua lIy character ist ic of the unstressed syllables following it. But whether the first stressed syllable of an intonat ion group plays a functional role or not is a moot point. form another element . It seems more consistent to treat the first stressed syllable as part of the functional whole . For instance/the effect of the rising tone on the first stressed syllable is frequently conditioned by the higher pitch of the following unstressed syllables. it contains the prehead. finality versus incompleteness. an intonation group consists of one (stressed) syllable . The boundaries between intonation groups are marked by tonal junctures and pauses.Palmer was the first to single out the consecutive structural elements of the intonation group ("tone-group") _. intonation in the broad sense] playa role. Types of prehead differentiate emotional meanings. Theypassed the exam yesterday. A. the nucleus and the tail. The functional role of some of these elements is ind isputable. delimit one intonation group from another and show its relative semantic importance. G. used by Soviet phoneticians. The structure of the intonation group varies depend ing on the number of syllables and rhythmic units in it. the head. The intonation group is a meaningful unit. The most general meanings expressed by the intonation group are. preceding the first stressed one) and the head (the first stressed syllable and the following stressed and unstressed ones) . . A. which he considers to be an indeoenderrt structural element. Maximally.the body. It may be coextensive with a sentence or part of a sentence.Torsuyev. admitting its role as 73 . O'Connor and G. those of completeness. These are the nuclear stress.the nucleus. Thus.Vassilyev. "nucleus" and "tail" [101].vrsron in which not only stresses. E. for instance. Minimally. "head". but pitch and duration (i. The number of structural elements distinguished by different phoneticians is not the same.g.Arnold dist inguish two elements in the pre-nuclear part of their tone-groupthe pre head (unstressed syllables. on the semantically most important word and the terminal tone (i.g.the scale or head (in the broad sense). The stressed and unstressed syllables following the head. The notion of "head" in this sense coincides with the notion of "scale". V. All these features shape the intonation group. indicate the end of the intonation group. H. e. structurally the intonat ion group has some ob ligatory formal characteristics. non-finality. The most conspicuous is the funct ional role of the nucleus :its prosod ic features express communicative and attitud ina I mean ings. pitch variations on the nucleus and the tail if any). H.Kingdon uses the term "head" to mean only the first stressed syllable.e.Antipova and others.Trakhterov.Yesterday [thev'passed their Jxam.

E. E. two or more. The utterance is not the ultimate unit of prosodic analysis. it is necessary to note."But1shall we dis~uss it again'?" 4) a special question .e. that the syllable.. the phonetic paragraphs and texts. "The term "sentence" structure." 21 an imperative sentence ."And 'what is thel matter with 'Jane'?" 5} an exclamatory 'sentence . they are connected and grouped into st ill larger un its . 'listening is an'im'portant'process in 'learning a-!anguage. A higher unit in which prosodic features are actualized is the u t t era n c e. texts. It is characterized by semantic entity which is expressed by all the language means: lexical. Each utterance has a definite prosodic structure which may be coextensive with a sentence".hensibi lity. or ~ith a word combination. thus form ing the prosodic structures of the hyperutterances.the onset that determines the pitch movement within the intonation group. BJsides the/auditory 'process [there are/speakinglreading land'writing of the . Whereas the elements of the intonation group. I ."If'only vou'd'asked me b~fore!t' The utterance may conta in one intonat ion group. the prosod ic structure of the utterance is viewed as a single semantic entity. I One and the same prosodic pattern of an utterance may be used with any syntactical structure: 1) a declarative sentence . The prosod ic features of these higher un its ind icate the re lations between their constituents. In speech single utterances are not very frequent. the intonation group. the rhythmic unit. The study of these units in modern linguistics is in the forefront of scholars' interest. It is also disputable that the tail Is an independent functional element of the intonation group. To summarize.1 . the utterance and the hyperutterance are taxonom ical prosodic units: each hierarchically higher unit consists of one or more units below it.' '\ . On the contrary.g.or with a word.language. The prosodic structure of an utterance is a meaningful unit that contributes to the total meaning gf the utterance.Jane's ability. 'Count on. The utterance is the main communicative unit. "head" and "tail" are non-obligatory elements of an intonation group. Compre. grammatical and prosodic."So tel( them they 'ought to go'back. since its pitch variations are determined by the nuclear tone."The'island is 'far in the 'East.hyperutterances. The "prehead". phonetic paragraphs and 1 . Irrespective of its structural complex ity.g. is used here in the sense of a forma I grammatical 74 . . the degree of their connectedness and interdependence. considered above. i." 3) an interrogative sentence (general question) . whereas the nucleus is an obligatory and the most important funct ional element.

For example. mid and high.Trager. According to circumstances the speaker chances his voice range. it sounds less enthusiastic. G. speech melody is the variations of the fundamental frequency. PROSODIC SUBSYSTEMS units. mid and high levels some phoneticians distinguish the emphatic (higher and lower) and the emotional (higher and lower) pitch levels [88]. pitch ranges.cast e] as. they are not related Pitch The pitch component of intonation.f irst=classt lcket for thls. Pronounced with a low narrow range it sounds but not emotional. If said with a wide range 75 . Besides low. R. direct ions and rate of pitch movement in the terminal zone and pre-terminal part of each of its intonat ion groups. sincere. and also the speaker's attitude and emotions.!ative and are produced on different registers depending on the individual peculiarities of the voice.Pike. nucleus and tail. mid. It may be widened and narrowed to express emphasis or the speaker's attitudes and emotions. in'This . extra-high).. generated by the vibrations of the vocal cords. J. high. Acoustically. for example. or'Wh~re d~d the'char ltable.seas ide resortlfind you? The number of linguistically relevant pitch levels in English has not been definitely established yet: in the works of different phoneticians it varies from three to seven. which they term "pitch phonemes".l was'buikl in the'15th century. These levels are rp. To descr ibe the melody of an utterance it is necessary to determine the relevant pitch levels. It shows the degree of semantic importance the speaker attaches to the utterance (or intonation group) in comparison with any other utterance (or intonation group).qentlernanjwho had a. if "Very good" is pronounced with a narrow (high) range ====- ..-. The pit ch ran g e of an utterance is the interval between its highest-pitched syllable and its lowest-pitched syllable.. American linguists K.far as I r~member.Smith. or speech melody is the variations in the pitch of the voice which take place with voiced sounds.preheaa. Parenthetical phrases and other semantically ess important intonation groups of an utterance are characterized by a lower pitch level than the neighbouring intonation groups. as. The pit chi eve I of the whole utterance (or intonation group) is determined by the pitch of its highest-pitched syllable. nead. In unemphatic speech most phoneticians distinguish 3 pitch levels: low. are autonomous taxonom icallv. The sequence of pitch phonemes in pronouncing an intonation group are called intonation contours.Welis and others distinguish 4 relevant pitch levels (low.

is steeper (the angle of the fali is obtuse) when it is pronounced within a shorter period of time. In rising tones the unstressed syllables of the tail form an ascending sequence (the rise actually occurs on the unstressed syllables. Static tones may have different pitch level of the voice . the mid static tone. and the rising-falling-rising tone. for instance. The rat e 0 f pit c h va ria t ion s may be different depending on the time. Some phoneticians adopt the existence of two significant pitch rangeswide and narrow. Cf. Thus. When the rate of the fall is fast. the fall is steeper. others distinguish three pitch ranges . On the other hand. but the range is wider. is also based on the differentiation of the pitch level of their initial and final points. kinetic tones are subdivided into simple and complex.. The pitch of the "tail" depends on the kind of terminal tone.the high static tone. etc.. when the time of the fall is the same.. in falling tones the unstressed syllables are pronounced on a low level or they form a descending sequence. The falling tone.g. Depending on whether the pitch of the voice varies or remains unvaried tones are subdivided into kinetic and static (or level) [90]. the falling tone sounds more categoric and defin ite than when the rate of the fall is slow. The differentiation of kinetic tones as high falling and low falling. It conveys certain meanings of its own which 76 . high rising and low rising. The basic unit used to describe the pitch component is the ton e. the risingfalling tone.lean you . The most important from the functional point of view is the t e r m in a I ton e of an utterance. Simple tones are unidirectional: the falling and the rising tones. mid and narrow. The peculiarity of the terminal tone in English is that it may occur not only on the "nucleus" but may be extended to the "tail". and on the. during which these variations take place. the carrier of the terminal tone is not only the stressed syllable.wide.range of the variations. It is rather this stressed syllable together with its enclitics. Cf. wait for me? • _ •• In other words. As to the direction of pitch movement. This is an illustration of the interact ion of the pitch and the temporal component of prosody. its range being the same. E. Differences in the rate of pitch variations are semantically important.- • "'\ it sounds both sincere and enthus iast ie. the low static tone. Complex tones are bidirectional: the falling-rising tone. The term inal tone is the "most important functional element of the prosodic structure of the utterance.

e.This tone is used in general questions of the basic type. are definiteness. uncertainty. Fall-Rise. 77 ----- . If you'wait for him.· The High . Due to its linguist ic mean ings and the funct ions that it performs in speech the terminal tone can be treated as a phonological (or rather intonolog ical or prosodemic) unit in the structure of a language. i. while the high falling tone expresses incompleteness. non-finality and is.High The Full . Rising-Falling and Rising-Falling-Rising. E. The number of such to nemes. The meanings of the rising tone are those of indefiniteness. Rise. interested special questions in the first part of alternative quest ions and in the second part of disjunctive questions.Ward d ist ingu ish two tones: Falling and Rising.Jassem distinguished also Full Rise and Fall which differ from the high and low variants in a wider interval between the initial and final pitch levels: High Fall Rise --=. therefore.g. Full Rise 2 \ . contrast etc.Kingdon there are five terminal tones as to the directions of pitch movement: Rising. non-finality. According to R. for example. E.make the whole utterance more concrete and precise. the High Rising. FU.funct iona I var iants of each of these tones: high and low.wait for him.Arnold exclude Rise-Fall-Rise from the system of English tones but include the Mid-Level tone. the Falling-Rising and the Low Rising. W.====-- . if you\.Crystal discriminates seven nuclear pitch movements in the subsystem of tone: Fall. certainty.::===-- . Rise-Fall.Armstrong and I. assertiveness. !What did you say? / Has he? Besides some lingu ists distinguish the Mid Rise (which starts near the bottom of the normal voice range and rises to a level of pitch somewhat above the middle) J . tentativeness. frequently used in non-final intonation groups: E. in English is not unan imously agreed upon. Level. The meanings of the faIling tone. He'II'come. he'll\come.Low Rise ____. which is commonly used to show nonfinality.g.g. O'Connor and G. What did you say? Has he? The Full Rising tone serves to express surpr ise.Palmer has four tones in his system: the Faliing (both high and low] . Depending on the pitch level of the starting and ending points of the tone. aston ishment. Low Fall Falling tone sounds complete and final. Failing-Rising. Concrete tones in which a tonerne is·realized in speech can be defined as a 110tones of the tone me [110]. incompleteness. J. most scho lars d ist ingu ish two. Rising tone is widely used in repeated questions and echo / questions. H.each of which is represented by high and low varieties. Falling.Jones. . The falling-rising tone carr ies the mean ing of reservat ion. FalH-Rise and Rise+Fall.etc. finality. a demand for informat ion. D. D. completeness. relevant term inal tones.a ton e me.IIFall . L. impJicat ion.

and three emphatic heads: Stepping •••••• . H.g. It seems more logical to classify head patterns of English into three major types. which is an emphatic modification of the High head. So. UJmpare the meanings in the following utterances: E. which.s. the criterion in each case being the·general di_o __ " . falling. a Scandent head /'" when the pitch risesor climbs from the mid-level to the highest level of the tonegroup.The tonemic status of Low Fall and High Fall. the Ascending scale .Trakhterov. or aJiotones of two tonemes-falling and rising? Another problem arises in connection with the Fall-Rise Divided.. This highest point is therefore higher in pitch than the pitch 01 any nucleus. The problem of the number of heads as well as the problem of their functions and possible combinations of heads and terminal tones in English hasn't been fu llv solved yet.Vassilyev.• w . It's the:"only sensible1:hing to Clo (the scandent head).an emphatic variant of the Rising head.Arnold classifl four unemphatic heads (scales): Low •••••• .Palmer distinguishes three types of head: an Inferior head . It's the ~nly~sensiblething to'uo (the sl id ing head) . . G. the Scandent scale • _. the Low Level scale the High Levelscate ..which is never higher in pitch than the initial point of the nucleus. another structural element of an utterance. It acts as a unit independent of the nucleus. Low Rise. High •• • .9 head) . Is it a separate tone me or a combination of Fall+-Rise? The functions of these tones as well as those of some others haven't been studied sufficiently.O'Connor and G. Sliding .. one part of the pitch contour of the utterance. It's t~ only Isensible "thing to do (the ascending head).s higher in pitch than the initial point of the nucleus. V.. Significant pitch modifications can also be observed in the head.Palmer distinguishes also a Heterogenious head which is a combinat ion of any of the three foregoing types.. It's the 'on Iy 'sensible 'thing to 'do . The head can a lso predict the communicative type of utterance.rhythmic groups and to convey modal-stylistic meanings. In the works of Soviet phoneticians L. Pitch variations over the head make a subsystem of their own.(the gradually descend i'_!. Mid Rise and High Rise is not clear yet.••• . the Broken Descending scale • . the Sliding scale ----:-~=J. A. 78 .Antipova the following types of scale are distinguished: the Gradually Descending scale' .•• and Rising _. and Climbing •••.Torsuev.~ ... Are they all independent emphatic modification of the Falling head.. we see that the subsystem of terminal tones in English is not fully established. Falling ••. The h e a d is v iewed as one me bd ic shape. a Superior head __ . Phoneticians classify heads in different way. H. The functions of the head are to express relations between its constituent units . rising and level.

acoustic. E. (According to J. two types of the level head: high and low. _But it's in'hedible (surprise). That is why the commun icativedistinctive function of speech melody is widely recognized as a purely linguistic function . melodemes.rection of pitch movement over the head. aooear to contri- 79 .can't have'possiblv'meant it (joy). two types of the rising head: ascending and scandent. which can be reduced to a limited number of relevant melodic contours. A wide scope of the distinctive function of intonation includes also the modal-stylistic (or attitudinal) function. when the fall of the nucleus reaches the lowest level or when the tail continues the mid-level pitch of the nucleus. sliding. As to the pitch movement within each rhythmic unit of the head there are three subtypes of the falling head: stepping. The pre h e a d is normally pronounced on the low or mid pitch level. So the tail is not an independent element of the utterance. It should be treated as a const rtuent element of the terminal tone. there are ten bas ic melod ic contours ("tone-groups"} in Eng lish [100J. the tail is descending when it continues the fall of the nucleus. head (scale) and the terminal tone) . With this broad concept of the dist inct ive funct ion in v iew we can state that the prehead.) To establish the system of melodemes is a complicated task which can be solved with the help of aud itorv. The number of relevant mebdic contours in English. Depending on the context or the comrnunlcat ioo situation some WOl"ric. descending.each of which forms a separate subsystem of utterance prosody. The functional analysis of English prosody in general and speech mebdy in particular shows that the leading role in differentiating communicative types of utterances belongs to the term inal tone. The pitch characteristics of the tail depend on the kind of nuclear tone.g:-The/Rovers:Robert? (aston lshrnent] -He. Utterance Stress Words grouped into an utterance are not all equally important. Arnold.a distinctive funct ion proper. as in other languages. statist ic and linguistic methods. is limited. If it is pronounced on a pitch somewhat higher than the normal pitch' (High Irregular Prehead I or somewhat lower (Low Irregular Prehead) the utterance acquires emphasis and emotional connotations. Various combinations of the elements form invariably complicated and numerous rnelodlcstructures. It may be Ie·vel. the head and the terminal tone are functionally relevant elements of the utterance. Therefore differentiation of melodic contours (tunes) of utterances must undoubtedly take into account the pitch characteristics of their elements (the prehead. The tail is ascending when it is part of the r isinq or falling-rising terminal tones.Q'Connor and G.

e. the acoustic features of intensity and fundamental frequency). The main difference between these three types of stress is the difference in how the syllables that bear them are marked.bute more informat ion than others. unobscured formant structure (d istinct qualltv] . We always know the place of stress in a word.* It has already been stated that stress is part of the phonetic structure of the word. non-nuclear full stress and part ial stress. The subsystem of utterance stress in English includes three basic functional types: nuclear stress. Acoustically. "Accent" combines "stress" and pitch prominence (i. loudness. these types are distinguished by a number of phoneticians as "stress" and "accent". duration and formant structure. utterance stress is determined by variations of fundamental frequency. The special prominence given to one or more words in an utterance is called u t t era nee s t res s. perceived as the most prom inent..Non-nuclear fUlly~stressed syllables are more often marked by static tones. therefore. they do not initiate tones and their pitch characteristics depend on the pitch pattern of the preceding fully stressed syllables. intensity. stress becomes a feature of the utterance. The role of each of these acoust ic parameters in creat ing the effect of utterance stress has been studied experimentally by a number of phoneticians il'lll this country and abroad. When the potential stress pattern is actualized in an utterance. is not qu ite precise. Nuclear and non-nuclear *The term "sentence stress" a syntactical structure only. Those that are semantically more important are made prominent. It appears that fundamental frequency is more efficient in determ in ing stresses in an utterance than intensity. the potential stress pattern of the word. Partially stressed syllables are not pitch prominent. when the word is made prominent. That is why utterance stress is a structura I phenomenon. Because of the difference in the means.e. are variations of pitch. length and quality. both initiate tones. "Stress" is achived by a greater force of articulation. The nuclear syllable is in most cases marked by a kinetic tone and is. The acoust ic structure of stress varies depending upon the type of stress and its position in an utterance. i. As a rule the effect of utterance stress is created not by a single acoustic parameter but by a certain interact ion of different parameters. resulting in greater intens ity on the acoust ic level and in greater loudness on the perceptual level. The means. The sounds of the stressed syllables are character ized by a d ist inct. with the help of which the special prominence is achieved and the effect of stress is produced. as sentence often implies 80 . Durat ion also appears to playa greater role than intens ity. effecting partial and full stresses. Born are pitch prominent.

In their turn non-nuclear full stresses (secondary accents) signal greater semantic value of the words than partial stresses. The specific character of word stress and utterance stress is conditioned by the domain of their functioning: word stress is an essential part of wordshape. Each of the above three types of stress has functionally significant degrees depending on the modal-stylistic factors of speech. Their acoustic structure is different. "Have vou'brouqht the ·form/with you?" or 'llwhen d'you·intend. I I.full stresses are referred to as pr imary accent and secondary accent. "He'went'out. same accentual patterns. 63aK. The semantic centre of the utterance is singled out by the nuclear stress (or primary accent).. Form words are likely to be unstressed. the meaning which the utterance is intended to convey. "At the. Well done". The crucial factor in determining the location. whereas utterance stress is a feat"llre of the utterance. respect you get"on.">" 'John went'out". Due to the rhythmical organization of the utterance notional words may be unstressed. may be stressed. The accentual structure of an utterance is conditioned to a certain extent by the stress patterns of its words. "Re-write". "It is 'not at lall'interest ing".leaving?" Notional words. For instance. the inverted word order for expressing interrogation requ ires stress on the auxiliary verb. But in special conditions.g. Each type of stress also has different positional variants: e.g. The distribution of stresses in an utterance is also affected by the rhythmical laws of the English language. "Are you very busy just now?" or "Let me hear. Torsuyev points to the following factors: semantic. if the meaning requires. are predisposed to be stressed in an utterance. .e. The semantic. E. type and degree of stress in an utterance is the semantic factor. This type of stress is opposed to the nonnuclear stresses by its greatest semantic importance. Thus. Word stress and utterance stress are in close relation."I t'is interest ing". stresses in emphatic speech are stronger than those in unemphatic speech. the semantic factor being the main one.g. due to their function in the language. form words may become stressed. grammatical and rhythmical factors are closely connected with one another. i.entrancelthere were'many'people.6248 81 . The distribution of stresses in an utterance depends on several factors. '\ e. prenuclear and post-nuclear partial stresses." The grammatical structure of the utterance also determines its accentual structure. e. The word and the utterance may have the 1'1' t. grammatical and rhythmical." (but "At the 'entrance to the/theatre!there were'many 'people") .g. when they are semantically important.and form words. may become unstressed. Cf. On the other hand notional words. Whenever utterance stress occurs it will normally fall on a syllable which also has word stress. on the contrary. CT. G.

Stresses in an utterance fulfill the same three functions as other components of prosody: constitutive, distinctive and identificatory. In their constitutive function stresses form the utterance by integrating words. They form the accentual structure of the utterance, wh ich is the basis of its rhythm and part of its prpsod ic structure. Whi Ie integrating words into utterances" stresses of different hierarchy segment the speech continuum into rhythmic (accentual) units, intonation groups and utterances, and delimit them one from another thus carrying out the segmentative and delimitative functions. The distinctive function of stresses manifests itself in differentiating utterances as to their meaning, which is conditioned by the position and type of stress. E.g. ')Don't you 'find it/difficult" and"'bon't,.you find itdifficult?" The opposition of degrees of utterance stress carries out a modal-stylistic function. In its identificatory function utterance stress provides a basis for the hearer's identificat ion of the important parts of the utterance and for his understand ing of the content.

Rhythm Rhythm has been defined as regularity or periodicity in the occurrence of a particular phenomenon in an utterance. Languages differ in their rhythm mainly because of this phenomenon. In some languages the recurring phenomena are stresses, in others - syllables. So languages may be characterized either by stress-timed or syllable-timed rhythm [103]. English is considered to be mostly a language with stress-timed rhythm. Though occasionally it may display syllable-timed character as well f103, 60]. Stress-timed rhythm presupposes that utterance stress serves as a basis for the rhythm leal organ izat ion of speech and that stresses segment the speech continuum into units of more or less equal length. These are accentual, or rhythmic uni ts. The units tend to follow one another in such a way that the lapse of time between the stressed syllables is somewhat uniform. Since the rhythmic units differ in the number of syllables they are comprised of, the syllables of the longer groups are compressed by very rapid pronunciation and, those of the shorter ones are lengthened to conform to the same interval of time. This produces perceptible isochrony of rhythmic units within the limits of a given intonat ion group [35, 110, 90J . But there is no direct relation between perceptible (subjective) and acoust ic (objective) isochrony. Regarding isochrony as a characteristic feature of English rhythm G. Torsuyev points out that this rhythmic tendency of the English language does not mean mechanical equality of intervals between peaks of prominence even within one and the same intonation-group [35].


A number of special invest igations show that isochrony of rhythmic groups is rather approximate. The lapses of time between stressed syllables (peaks of prominence) are not absolutely equal. Perfect isochronism can be realized very rare Iy, only when definite cond itions are fu tfllied. So English rhythm can't be said to have objective isochronv of its units. One can only speak about a tendency to isochrony which results in the modifications of the length of syllables and vowels and in modifications of the stress patterns of words. Since the approximate isochrony of intervals between stressed syllables is regarded as a measure of Engl ish rhythm. a great number of phoneticians (A.Classe, D.Abercrombie, H.Halliday. J.Pring) define the unit of rhythm as a sequence of syllables from one stressed syllable to another. But this formal rhythmic division does not reflect the relations between prosodic units and the units of the other subsystems of the language, as the syllables of one and the same word may be parts of different rhythmic units. E.g. Sel1mantic iml portance, G.Torsuyev. V.Vassilyev, R.Kingdon. J. O'Connor, W.Jassem and other scholars represent another approach to rhythmic division. According to this approach the boundaries between rhythmic units are determined by the semantic and grammatical relations between the words of an utterance. With such rhythmic division the syllables of a word always belong to the same rhythmic unit. form words [o in the stressed syllable as proclitics and enclitics, depending on their semantic links. When analysing periodicity of rhythm in particular, both the formal and the semantic approaches may be accepted. Percept ible isochrony is characteristic of formal rhythmic units as well as those based on the semantic pr.inciple [90J. But the second definition seems more correct and consistent for its reference to meaning. Correct rhythmic division is of great importance since division of utterances into rhythmic units can playa distinctive role. E.g. ·Shalll11put itlbnlmysetf? and IShall I"put itlon myself? To single out rhythmic units one should be also guided by their perceptual prosodic structures. The rhythmic unit is a perceptible unit which can be isolated due to its prosodic features and meanings. (See pp.71-72). We have been considering English rhythm in its auditory (perceptual) aspect. What is rhythm from the point of view of its acoust ic structure? A great number of phoneticians believe, that certain temporal modifications, though they are very important. cannot be treated as the only characteristic feature of Eng lish rhythm. Acoust ica lIy. rhythm is a complex of var iat ions in frequency, intensity and duration. Since the basis of rhythm is stress, which. is a structural acoust ic phenomenon. rhythm is a structural acoust ic phenomenon too and it is achieved by the same acoustic parameters that produce the effect of stress. 6*


Rhythmic units form a certain hierarchy, since stresses, on which they are based, are not equal in their prominence, position and function. The most prominent, as a rule, and functionally more important is the nuclear stress. Therefore the nuclear rhythmic unit is the most important in an utterance. A rhythmic unit formed by full stress together with partial stress can be defined as a complex rhythmic unit [3]. A sequence of full and partial stress indicates closer semantic relations between the words than a sequence of two fultstresses. (Cf. "It was a'twenty minutes-walk"alJd "It was a'twenty '-minutes 'walk"). Rhythm ic units formed by post-nuclear (partial) stress may be regarded as wea.k, for example the second groups in 'i'yes"lhelsaid. "You are right, I,Mary" [103]. Prosodic characteristics of rhythmic units are determined by their position and function in an intonation group 'or in an utterance. The prosodic structure 'of the nuclear rhythm ic un it differs from that of the onset rhythmic unit, the rhythmic unit within the head (scale) is different in its prosodic features from that of the tail and so on. D.Bolinger has described three pitch-patterns of rhythmic units, or pitch accents, as he calls them, with the relevant falling, rising and level direction of pitch movement. These pitch accents are differentiated by their meanings and possible position in higherlevel un its [57]. D.Abercrombie has analysed three durat ion patterns of the formal bisyllabic rhythmic unit (foot) [49]. However, the full inventory of prosodic characteristics as well as the number and types of rhythmic units in English has not been established so far. Rhythm ic un its are noth ing but elements of rhythm. Rhvthm as regu tar ity of occurrence of stressed syllables manifests itself in hierarchically higher prosodic units - intonation groups and utterances. 1't has been shown by a large number of wor ks that rhythm performs important lingu istic functions. It is the most important organizing factor which makes for the exposure of exact sense of speech and its maximum percept ibilltv, It unifies smaller language units into utterance, hyperutterance and text, and indicates relations between them. It also performs a distinctive role due to the prosed ic character ist lcs of its un its. There may be different rhythm ic patterns in a language depend ing on the number and types of stresses in an utterance as well as on the degree of prosodic contrasts between its stressed and unstressed syllables. In the English language prosodic contrasts between stressed and unstressed syllables are sharper than, for example, in Russian and Byelorussian. But in the English rhythmicality system these contrasts vary as well. On account of the prosodic characteristics of English rhythm D.Crystal distinquishes 3 pairs of contrastive rhythmic structures. These are: rhythmicl arvthrnic, spiky!glissando, staccato/legato [60]. Rhythmic utterances are very common in everyday speech. They are used when the speaker rephrases a


.: Who would come on a night like this spiky glissando . are rather rare. 85 . the relations betvveen stressed and unstressed sy~ labtes within rhythmic units. Tempo of speech may be determined by different factors. the acoustic qualities of the room.. as a result the actual semantic importance of what we say... . E. thus underlining the semantic . . and the relations between stressed syllables of different rhythmic units on the other. It is common knowledge that by slowing down the tempo of speech we can make an utterance or part of it more prominent. "'l In staccato and legato the transition between stressed and unstressed syllables differs as to their loudness and duration. \ . by increasing the speed of utterance we diminish prominence and. It may vary depending on the size of audience. Arhythmical utterances. E. J J _"t .. ~ . Legato is analogous to glissando in the smooth transition between the loudness and duration characteristics of stressed and unstressed syllables. on the contrarv. They result from certain variations in tempo and hesitation pauses. the individuality of the speaker and other extralinguistic factors. on the one hand. loudness and pitch.don'tworrv. told you to1leave \ that jam alone. I Noth ingata ~ser ious. importance.g. On the contrary. Staccato is analogous to spiky as it is characterized by sharp contrasts between heavily prominent stressed syllables and very light unstressed syllables. whereas glissando is characterized by smooth and usually slow glides between them. Spiky rhythm is characterized by sharp and rapid pitch transitions between stressed and unstressed syllables. But most significant for the linguistic study is how variations in tempo correlate with changes in meaning. Te m po The tempo of speech is the rate at which utterances and their smaller units are pronounced.. On the acoust ic level tempo is generally measured by the number of sylJables per second. J "I "I • . It is quite evident from the analysis of the rhythmic contrasts that English rhythm is determined by all the prosodic features: duration. The types of rhythm mentioned above (except arhythmicality) are typical manifestations of English rhythm." . v: . E. Further investigation is needed to clarify the structure of English rhythm and its concrete functions.. ..• particularly important point of information. 'Hundreds of 'times have I.g. of it.g.

drawled and held syllables are articulated less rapidly than normal. as in "There are your slippers. A silent pause is a stop in the phonation (a stop of the work of the vocal cords. joy. allegrissimo. He distinguishes between simple and complex tempo systems. pauses of percept ion and vo iced (or filled) pauses.Tempo can also be used to express the speaker's attitude or emotion. The complex tempo system is realized in polysyllabic stretches. For example. "Clipped syllables are articulated at a more rapid speed than normal. In polysyllabic stretches of utterance D. and there. deviations from which affect mean ing. fast tempo may express excitement. 154]. in stop consonants as in "Perfectly" [ppa. drawled and held syllables which are generally used for emphasis. Pauses are closely re lated with tempo: the number and length of pauses affect the general tempo of speech. etc. It is the main function of a pause to segment connected speech into utterances and intonation groups to delimit one utterance or intonat ion group from another.Shaw's "Pygmalion".lento. In monosyllables the speeding up and slowing down of the duration of the syllable is perceived as clipped. The simple tempo system is manifested both in monosyllables and polysyllabic stretches of utterance. as for instance in "Well. p. They may also serve as a style-forming and style-differentiating device. Slow tempo shows relaxation or calmness. Phoneticians distinguish 3 main types of pauses: silent pauses. and two degrees slower than the norm . Phoneticians generally dist inguish normal tempo and two departures from the norm: fast and slow. quite blue" [kkwar t] . which results in the cessat ion of sound) . anger. The distinction between drawled and held syllables is that in the former a sound is lengthened as in Iff a ml and in the latter a sound is articulated with the onset of articulation delayed. for example. lent iss imo. we can both breathe a sigh of relief" when pronounced after having done something hard or unpleasant. Everybody's speech has some norm of tempo. In the complex tempo system there are acee lerando a gradual increase in tempo. These contrasts of tempo correlate with changes in mean ing.a gradual decrease. And may you never have a day's luck with them" as pronounced with anger by Eliza in B. reserved and phlegmatic attitude on the part of the speaker. 86 . and rallentando .Crystal d ist inguishes two degrees faster than the norm .allegro. and very lax" [60. in a very tense kt l i ] or "Quite. Pauses The speech continuum is divided into units of different length and hierarchy by means of pauses.Crystal gives a more detailed classification of variations of tempo [60]. D. so that the auditory impression of length is produced through unexpected silence.

What is the prosodic system of a language? 3) Give your interpretation of the term "structure". Pauses are very important constituents of intonation. as there is no penod of silence. The extra-long pause is used as a rule to separate two paragraphs. or by variations in duration.IMiss.P"'uses of percept ion are not a stop in phonation. rauses of perception are generally marked by a wavy line which is used at the junction of intonation groups. stress and tempo each makes a system of its own (a prosod ic subsystem) ? B. It may vary depending on the general tempo of speech. The short pause is mainly used to separate two intonation groups. Short pauses ind icate closer relations than long ones. What is the proso-d ic structure of the utterance? of the rhythmic un it? What is the accentual and rhythmic structure of the utterance? 4) What is the smallest meaningful prosodic unit? What meanings does it convey? 5) Is the prosodic (intonation) structure of the utterance related with a definite syntactical structure? What factors determine the prosodic structure of an utterance? 6) On what grounds do phoneticians assert that tone. Select pairs of utterances with identical syntactical structure and with different prosodic features to illustrate the inventory of 87 . Voiced pauses have usually the quality of the central vowe I [ 3: (~) ] with or without nasalization [ a (rn) ]. pitch level. E.Theeffectofa pause is produced by a sharp change of pitch direction. pitch range. The long pause which is approximately twice as long is generally used to delimit two utterances. It should be noted that the duration of pauses is relative. Silent pauses are subdivided into several types according to their length: short. > EXERCISES A. But the main factors that determine the occurrence of the type of pause are the semantic relations between the prosodic units. The teacher says John is very bright.g. or both. f Cf. Besides their segmentative and delimitative functions they also perform a unifying function showing the relations between utterances or intonation groups. not absolute. Think about the following questions for class discussion: 1) What is the difference between prosody and intonation? 2) Interpret the term "system". IThis is my'niece. long and extra-long. They are used to signal hesitation or doubt and are therefore called hesitation pauses.Smith and 'This is mynieceIiMiss$mith.

IMary would have 'left for'London.(Cf.~ 2) the tendency to avoid stressing several syllables in succession. 4) compression of syllable length in rhythmic units as a manifestation the tendency to isochrony. ~and'ln the/middle of it . Mod e 1: In the'middle of the/oom . words (monosyllabic or polysy lIabic) in success ion. 2) the subsystem of stress. 4) the subsystem of rhythm. 'Mary has'left for\London. of .'un'natural) 3) the tendency to avoid stressing several.. H the tone subsystem.C. Mod e I: 'Mary 'left for'London. Collect examples analogous to the model to illustrate 11the dependence of stress on the rhythm of the utterance. He is a'very punctual'person.. Mod e I :'Read text'ten. 3) the subsystem of tempo. Mod e I : tt's'qutte unnatural.. 5) the subsystem of pauses..

"Fire?" . Prosody is.C HAP T E R 7.utterances. the ever present constitutive factor of the utterance [65]. i. depending on the situation in which it occurs. I. Prosody is the only language device that fransform words as appelative units tvocabulary items> into communicative units . J. assertiveness . e. THE FUNCTIONAL ASPECT OF PROSODY MEANINGS AND FUNCTIONS OF PROSODY The functions and meanings of prosody should be described with reference to the utterance as the bas ic cornmwn ieat ive un it. The inherent meanings of prosody which are of a general character (such as definitenessuncertainty. The prosody of the utterance performs a number of functions. A succession of words arranged syntactically is not a communicative unit until a certain prosodic pattern is attached to it. distinctive and identificatory.32]. etc. the most elementary. the most common. statement. the meanings of the prosod ic structures in the utterances "11'jike'that" and '~C!ever /iiren't you?" with the challenging or antagonistic Rise-Fall are opposite to the meaning of the words. p.a. The can s tit uti ve function is to form utterances as communicative units.statements. nonfinality.e. "l t'rnav be/so" (But I'm not qu ite sure) The falling-rising tone is in harmony with the modal verb. Whereas in "It'may be so" (I'm absolutely sure about it) the falling tone makes the statement sound categoric. separatness . "Fire!" is a command or an exclamation. Or again. "Intonation". questions.reservations.e. regardless of the words and the grammatical structure of the utterance. uncertainty.t provides important information which is not contained in any of the other features of utterance" [100. "gives greater precision and point to the meaning.O'Connor and G.connectedness. It forms all communicative types of utterances . For example. non-categoric attitude. There may be cases of correlation and harmony between the inherent mean ings of prosody and the mean ings of words and grammatical structures as well as disbalance and disharmony.g. "Fire". . E.g. The prosody of an utterance (intonation) carries independent meanings of its own. etc.a question. 1. categoric statements.Arnold mark. Prosody unifies words into utterances. thus giving the latter the final form without which they cannot exist. non-categoric. surprise. until it acquires a certain pitch-andstress pattern. exclamations and modal (attitudinal) types: . per- 89 . therefore.G. Hence the role of utterance prosody in commun ieat ion. The prosody of the utterance is polysemantic. Due to its structural complexity it can express a number of different meanings of interrogation. imperatives. In written speech prosodic features are to some extent indicated by punctuation marks. the basic of which are constitutive. "Pete has left for Leningrad" is not a communicative unit until it is pronounced.) are specified and concretized when interacting with the grammatical and lexical meanings of the utterance.

. 2. certainty and uncertainty questions.Shakhmatov.. etc. In the first case the melodic contour agrees with the word content and the grammatical struc- *Some linguists. "may be".first instance questions ('Where did he \find it?). "Wevcanlif wevwant·to" and "Wei can if we~want to". definitely"prom. exclamations. questions.. intonat ion groups. A. A. showing relations between them. imperatives. and simultaneously delimits them one from another.ised" and "He. the emotional function. interested attitude.*. It segments can nected d iSCQU rse into utterances and.) and so on. These are communicative-distinctive. repeated quest ions. requests (Don't be. 90 . e. within imperatives . statements.def initelv. "perhaps". But utterance prosody may disagree with word content and is. such modal words as "sure". culminative ("themerheme") distinctive.e. The dis tin c t j ve function of prosody manifests itself in several particular functions. etc. "definitely".late.. e. Cf. In constituting an utterance. answers (It was a Ivery hot "after'ttoonl . definiteness versus indefiniteness) and the speaker's att itudes (for instance. "He. quizzical statements. or antagon istic versus fr iendly attitude and so on) . The mod a 1.g. and communicative subtypes: within statements statesments proper (It was a'verv'hot'af'tefnoon] .announcements. then.g.promlsed".: within questions .V. It also signals the semantic nucleus and other semantically important wonts of an utterance {or an intonation group). Various modal meanings can also be expressed and differentiated by lexical and grammatical means. a reserved. insistent questions. Prosody also constitutes phonetic styles of speech [23].Smirnitzky include intonation in the definition of the sentence and treat it under syntax as grammatical phenomenon. Usually. "undoubtful". Into this function some phoneticians include differentiation of the speaker's emotions.distinctive and stylistic-distinctive funct ions . V.functory statements. i.dis tin c t i v e funct ion is to d ifferentiate the communicative types of utterances. The com m u n i cat i v e .dis tin c t i ve (attitudinal-distinctive) function of prosody manifests itself in differentiating modal meanings of utterances (such as certainty versus uncertainty.d ispassionate versus involved. Cf.. the crucial factor in determining the modal meaning of the utterance. syntactical . prosody at the same time performs the s e g men tat i v e and del j mit at j ve funct ion. depending on the meaning which is differentiated..commands ( 100n't be \Iatel)·. "probably" and modal verbs "may". modal-distinctive. the speaker's attitude corresponds to the contents of the words he chooses.. (Where did he\find it?) echo questions (Where did he"find it?). Vinogradov. "might" and so on.

. The adherents to the theory of "sentence perspective" claim that in this way prosody indicates the "theme-rheme" organization of an utterance. ~Stnilin9l'Tomlentered the'hall ("smiling"is an adverbial modifier). The second utterance sounds indefinite and non-categoric. that intonation gives us a clue to the grammatical relations in utterances. 91 . The same intonation as in '.Bolinger says. E. Her IS ister.the "theme-rheme" structure of the utterances is distinguished purely by prosody. That is why in actual speech the listener is more interested in the speaker's "tone" than in his words.. This function is often called idl'Mary was a 'well-known 'actress (a complex sentence with an object subordinate clause). Disambiguation of the syntactical structure is a der ivat ive effect which this function may have within grammar.dis tin c t i v e funet ion of prosody man ifests itself in differentiating the location of the semantic nuclei of utterances and other semantically important words.e. Grammar uses intonation on those frequent encounters... But it is an accidental effect of two possible semantic (themerheme) organizations of the utterances..SmilingrTom'entered the'hall" might be used to emphasize the separate importance of "Smiling" if the listener hasn't heard it. p. "The encounters between intonation and grammar are casual. D. not causal. E.dis tin c t i v e function of prosody is to differentiate syntactical types of sentences and syntactical relations in sentences. So the first utterance sounds definite and categoric. Toml~ntered the'hall ("smiling" is an attribute). t. there is no direct relation between prosody and grammar. it d ist ingu ishes between what is already known and what is new in the utterance.ture. But it is disputable whether prosody performs in such cases a primarily grammatical function. 'Smiling . pred tcat ive and accentual. In "Thank you"the _rising-falling tone adds an antagon ist ic note to the utterance. So. but intonation is not grammatical" [58. Her. The s y n t act i c a 1. Utterance prosody cannot be defined and described in terms of the syntactical structures with which it occurs. the mer hem e The'teacher has 'come rheme theme Theteacher has r come The semantic nuclei in these utterances are different or.37]. In '1-hank you" the high falling to~e is in harmony with the word content and expresses genu ine gratitude. 'said'Mary.G. The cui min a t i v e .g. according to the theory of sentence perspect ive. whereas in the second case it does not. The primary function of prosody is to stress the internal coherence of the items within it.lwas a'well-known'actress (a compound sentence I .

[65. the problem is to establish the system of prosodic units or patterns on the abstractional level analogous to phonemes.Kuznetsov.20] or prosodemes by P. capable of distinguishing meanings are defined as intonemes by V. S. the linguist ic character of prosody ean be summarized in the following way: prosody of speech is significant and meaningful.Gaiduchik (29. its semantic and syntactical structure with the situation of the discourse. Thus. i. Thus the patterns of one language are not the same in form as those of other languages. Sty lis tic . The ide n t if i cat 0 r y function of prosody is to provide a basis for the hearer's identification of the communicative and modal type of an utterance. It onlysignals contrastive emphasis.Vassilyev [5.e.Artvornov.The same can be said about the relation between prosody and the meaning of a word. I have 'certain proofs (undoubtedly true).34.32. K.Kingdon says: "Intonation is the soul of a language while the pronunciation of its sounds is its body . Each language has a certain limited number of such units. so that the foreign learner has a certain latitude in this fiek:!. they should produce the prosodic patterns correctly and use them in appropr iate situat ions..10].Nork. The sounds of English as it is pronounced by different speakers and in different dialects vary within wide limits. E. There is a practical reason why it is advisable to pay more attention to intonation than to pronunciation. Nor do they necessarily express the same meanings. p.dis tin c t i v e function of prosody manifests itself in that prosody differentiates pronunciation (phonetic) styles. 3. V.Reformatzky. 511 But prosody does not determine the meaning directly. They show that utterance prosody is lil1gu ist iea lIy sign if ieant and mean ingfu l.O'Connor and G. xiv]. prosody is 92 . Learners of English should bear in mind both peculiarities of forms and meanings of English prosody. 'Give me some'apples (a few). R. 10 tone-groups of J. These units. All the functions of prosody are fulfilled simultaneously and cannot be separated one from another. though there may be resemblances in some cases.g. Emphasizing the role of intonation in speech. determined by extral ingu ist ic factors (see Chapter 10) . 'Give me'some'apples (any).Arnold in English (1001).Baryshnikova. A. I have certain'proofs (Some proofs). Prosody can differentiate between two possible meanings of a word. Now that the funct ion of utterance prosody (intonat ion) are analysed and recurrent prosodic (intonation) structures are described (e. p.. O. but in most dialects stressing and intonation conform fairly closely to the same pattern" [88.g.

since in one individual speaker there exist two distinct language systems. As for prosodic interference. it should be given a more detailed description. Features 01' phonemic interference in Russian-English "class-room" bilingualism have been well described by G. very often features of prosodic interference occur in a bilingual's speech even if he has a good command of the second language. Learners of English. The major manifestation of bilingualism is interference. should bear in mind that the prosodic patterns of their native language may.Weinreich. grammatical and lexical). In speech prosodic interference manifests itself in deviations from the prosodic norm of a language which result from the influence of the other language. involving whole language communities (as. Prosodic interference is more essential with regard to intelligibility than phonemic interference.Torsyev and V. who are trained to be artificial bilinguals. as "those instances of deviation from the norms of either language which occur in the speech of bilinguals as a result of their familiarity with more than one language. in Russian-Byelorussian bilingualism). Bilingualism may be acquired "naturally" and "artificially". On the speech level interference is defined.and prosodic.Vassilyev as typical phonemic and allophonic mistakes of Russian speakers of English. BILINGUALISM AND PROSODIC INTERFERENCE More than one language can be used in communication by the same persons. and very of- 93 . languages is ca lied bi lingualism [112J. as a result of foreign language learriinq [the so-called "classroom" bilingualism). Of all kinds of interference (phonemic. Languages used by bilingual persons are said to be in contact. for instance. word-building and morphological-syntactical) prosodic interference is the most stable and widespread.e. On the phonetic level there are two types of interference: pnonemic . lexical-semantic. in Russian-English bilingualism in a study-groupl. Prosody is a characteristic feature of each concrete language and cannot be used in speaking another language [1 00]. phraseological. as a result of language contact" [112] . and the speech of bilingual persons is thus the focus of the contacts.g. according to U. Language contacts may be of a "mass" character. The pract ice of alternate use of two. Interference takes place on all the levels of language (phonetic.systematic: it IS not invented in speaking but produced according to the system of prosodic structures of a given language. Language interference is a process and a result of the interanction and mutual influence of the language systems being in contact. i. or they can take place in "individual" or "group" bilingualism (e.

In addition.". 't"mrne.. "Shut the door I ~ ba. O'Connor and G. For instance. the second utterance was sometimes pronounced with the high falling tone instead of the low rising tone. because their meanings may be quite different from those in the native language. sound wrong when applied to English [100}. d ear. J.Whose is it then? . dear. This is more serious than a foreign accent. .rrnne.. 21. "English speakers are able to make a good deal of allowance for imperfect sound-making. nation in their own language.Arnold point out. rsn t -'Whose is it then? -'Whose'is it then? We can conclude that the linguistic skills of the bilingual speaker should be tested according to his "communicative competence".' ·t1. i. the knowledge and expedience in applying the rules which are governed by definite situations. dear. . rsn /. 94 .hindyou" instead of "Shut the/door behind you" .. N0. It probably belongs to one of the boys. but this wtH be done at the cost of getting only part of the meaning of the utterance.t. they are much less able to make the same allowance for mistakenly used tunes. but being for the most part unaware of the far-reaching effects of into . it isn't mine. .. and not being too sure about that. Our research into phenomena of prosodic interference in the English speech of Russian [31] and Byelorussian speakers of standard English has shown that over 10% of English prosodic intonation patterns that were pronounced by those speakers did not conform to the situation of discourse. The result is that they may hold the foreigner responsible for what his intonation seems to say .. and in the third utterance nuclear prominence was given to the word "whose" instead of the word "is". Wrong realizations of prosodic patterns give a foreign accent as do bad sound articulations. p. in the dialogue .I don't know.. A native listener may learn in time to consciously ignore the inappropriate prosodic patterns.even though the tune does not faithfully reflect his intention" [100. The English listener may get a bad they would rightly hold an Englishman responsible in a Similar case.No. Wrong Correct . The use of wrong prosodic patterns can cause vexation and misunderstanding on the part of the listener.ten do. Imagine one saymg "Good bye" Instead of "Good bye". I. . "This is very important".Is this book yours? .e. No. L 1 .' \. the prosodic patterns used by non-native speakers may be inappro-' priate for the given situation of discourse. since he will probably assume that the effect made by the tune was given by the speaker deliberately.

3) lower final pitch level of the falling tone.g. much higher. there are no sharp pitch contrasts (transition) between the structural elements at the utterance. This is due to the fact that in Russian and. 5) wider pitch range of the utterance. because the pitch level of the utterance in the Byelorussian language is lower than in Russian~ 7) narrower pitch intervals between the prehead and head. Bes ides that.g.e. The obvious similar ities of the prosod ic systems of Russian and Byelorussian. (Normally the beginning of Russian and Bye loruss ian utterances is lower than that of English.g. Byelorussian speakers of English pronounce the 1st stressed syllable and the whole utterance on a slightly lower pitch level than Russian speakers. in Byelorussian the pitch movement is more monotonous.FEATURES OF RUSSIAN-ENGLISH AND BYELORUSSIANENGLISH PROSODIC INTERFERENCE Interference from the native language in English utterances spoken by Russian and Byelorussian learners of English is observed in all the subsystems of prosody on the auditory and acoustic levels.* a) Melody The influence of the pitch patterns of the mother tongue in English utterances produced by Russians and Byelorussians manifests itself in the following deviations from the English prosodic norm: 1) higher initial and final pitch levels of the rising tones and the fallingrising tone. 8) the use of the pitch patterns which are not characterist ic of English.). the degree of the deviations may slightly vary (e. 95 . 2) wider pitch interval (the distance between the starting and the ending point) of the rising tone. i. very much higher) . result in typologically common features of Russian-English and Bye loruss ian-English prosodic interference. e. the rising tone /' instead of _/ All the above features result in Russian or Byelorussian accent. head and nucleus. which are closely related languages. to a greater extent. This occurs also when the *The direction of the deviations being mainly the same (e. 4) lower initial pitch level of the utterance. higher than in the norm). learners of English very often exaggerate the pitch contrasts characteristic of English which results in affected speech. 6) lower pitch level of the 1st stressed syllable and the lower pitch level of the utterance. nucleus and tail.

ive contrast with the unst ressed sy lIables. Though Byelorussian rhythm is stress-timed as well as English the tendency to pronounce stressed syllables at approximately equal intervals of time is weaker in Byelorussian.tst stressed syllable of the utterance is pronounced higher than is normally used by Englishmen. This results from the immediate influence of the tempo of speech which is character ist ic of the nat ive language. 96 .the variability of the duration of the rhythmic unit in English utterances spoken by Byelorussians is greater than in the same utterances pronounced by Englishmen. achieved by a wider interval of the terminal tone and a greater quantitat. Exaggerated prominence of the stressed syllables in the prenucfear part of the utterance as a result of their greater duration and also the affected pitch contrasts. stress and rhythm are the areas of the greatest number of deviations from the prosed ic norm of the non-native language of the bilingual speaker. In Engl ish utterances pronounced by Russian and Byelorussian speakers the following features of interference can be observed: 1. Distortions in rhythm are also due to the inadequate (slowed down) pronunciation of some difficult sounds and sound combinations. As a result. the absence of loss of plosion. the lengthening of fricative consonants. Distortions in perceptible isochrony of rhythmic units under the influence of the syllable-timed tendency of Byelorussian rhythm. 5. Exaggerated prominence of the nuclear syllable. 3. and what is more important. somewhat slower than with Russian speakers) . 4. c) Tempo The tempo of English utterances pronounced by Russian and Byelorussian learners of English is slower than in the speech of Englishmen (with Byelo-russian speakers it is. 2. This shifting is known as a characteristic feature of the Russian and Byelorussian languages. The shifting of the nuclear stress to the left. as a rule. towards the beginning of the utterance. The Occurrence of a greater number of stresses and the resutt ing d istortions of the rhythmic patterns of English. bl Stress and Rhythm According to the experimental studies of prosodic interference. distortians in the accentual patterns display a very high degree of stability and communicative relevance. the absence of reduction. These results apparently emerge from the functional and structural differences between the accentual systems of the languages being in contact.

exaggerated or too small prom inence of stressed sy lIables. The character of deviations is essential for intelligibility. Stressed syllables are also longer than they should be. characteristic of form words. The importance of further pedagogically-oriented experimentation in the field of phonetic interference is quite obvious. "Is'this book yours?" Distortions in the temporal structure of the utterance are also observed at the juncture of intonat ion groups. the inadequate realization of the rising tone in a general question or in a nonfinal intonat ion group may appear to be sufficient for the nat ive listener'S evaluat ion of the utterance as unnatural.Unstressed syllables are not reduced to the same degree as in the English norm. On the other hand. 73aK.z l instead of [wa zl in the utterance "So yesterday was the 31st". can gain a lot from what has been found about prosodic interference. The slowing down of the nucleus and the tail may occur in utterances pronounced with the rising tone (the interval of the rise. The degree of familiar ity with the prosodic system of the second language is primarily evaluated. The number and the character of deviations are dependent on the type of utterance.g. Research in the field of language interference in general and prosodic interference in particular proves valuable to the theory of language contacts as it reveals regularities in the functioning of language systems being in contact. lwo. containing one intonation group. In statements containing more than ·one intonation group. by native speakers by the number of perceptible deviations. in part icular. It is also important from the point of view of practical application of the data obtained. Thus. The absence of necessary reduct ion is. Research along the lines of phonemic and prosodic interference may be also helpful to the acoustic engineer who sets up to construct a bilingual synthesizer .g. The foreign language teacher. there is a tendency with Russian and Byelorussian speakers to over lengthen the final syllables of the preceding intonation group and to shorten the initial syllable of the following intonation group which is perceived as an inadeqoate (very longl pause. being wider than in the norrnl .6248 97 . Changes in the number and in the character of prosodic deviations from speaker to spea ker ind icate the degree of the speaker's fam iliar ity with the second language. faster or s lower tempo may not affect a rather high evaluation of the non-native speaker's apparatus that would "speak" alternately two languages. e. in disjunctive and general quest ions there are more features of interference than in special questions and statements. e. whose ultimate goal is to train his students to be art ificial bilinguals.

C. an utterance. D. How can you illustrate the polysemantic and polyfunctional character of the prosodic structure? B.EXERCISES A. with word markers and grammatical sequence markers as the solid cake" [99. Express your opinion of the following: 1. 3) the character of rhythm. 4) the stylistic function. Think about the following questions for class discussion: 1. "The attitude markers are sometimes thought of as luxuries. Are the functions and meanings of the prosodic structure of the utterance equivalent notions? 2. 2681. p. 2. Compare semantically identical Russian (Byelorussian) and English utterances as to 1) the number of stresses in them. 2) the form of the terminal tone. 2) the modal (att itud ina I) funct ion. The learner of English will always be understood (irrespect ive of the situation) if he pronounces the model English prosodic patterns. What meanings can the prosodic structure have? What do you think are the denotative and connotative meanings of the prosodic structure? 3. 3~ the culminative (accentual) function. the icinq on the top. 5) the role of prosody in modifying the basic meanings of the words in. Select examples from tape-recordings to illustrate 1) the communicative function of prosody. .

That is why notation systems of prosodic phenomena are equally important both for research work and language teaching.Ward [531. Even the narrow notation is to provide the scholar with means of indicating what is linguistically more relevant and what is less relevant. interlinear staves with dots. The aim of this chapter. but to acquaint the reader with the most widely used ones.Smith [108]. These notations reflect the differences in the theoretical assumptions of scholars and the ir interpretat ion of prosod ic data. and these are the features that are to be adequately represented. articles. PROSODIC NOTATION SYSTEMS linguistic analysis and the teaching of spoken language cannot disregard its prosody. One should look for distinctive features in prosody. There are var lous degrees of significance that it is necessary to discriminate. adequate representation of distinctive features is the most essential requirement a system of prosodic notation is to satisfy.Armstong and I. A narrow notat ion is intended to be more detailed and precise. For research work it is essential to discriminate all the linguistically relevant prosodic contrasts. whereas for teaching purposes it is generally sufficient to discriminate only the features that are more significant. both major and minor. These symbols should represent intonation visually as clearly as possible and act as a pictorial stimulus for immediate reproduction. if it does not reflect the differences in the degrees of sign if icance of the prosod ic data wh ich it denotes. It has already been mentioned (see Ch. The extent of the generalization may vary. Thus. so that he may read and reproduce them accurately. the head and nucleus system (H. dashes and arrows (L.Jones [84ll. frequent and eas ily perceived.Kingdon [88] I. is not to discuss all types of systems. 3) • that any system of notation is iii generalization of a great variety of linguistically relevant sound phenomena. D.Fonagy and K.Magdics [6911.C HAP T E R 8. the tonetic stress-mark system (R. it may be broad or narrow. There is a fairly wide var iety of notat ions that are be ing used in printed matter (papers.HaJliday [76]) and others.Trager and H. Depending on what the notation is intended for. however. etc.l . M. A broad notat ion is intended to reflect only the most relevant prosodic features by using the fewest possible symbols.Pike [103] I. Phonetic precision is not the main point.Palmer [102J I. the intonation contour system (K. That is perhaps the main reason why one of the first attempts 7* 99 . the numerical or number system (G. Unt iI recently intonat ion was norma lIy defined as pitch movement (or melody) alone. A highly detai led notat ion may turn out to be too campi icated and inadequate even for the scholar. There is a number of means to denote prosodic features: the musical notation (J. textbooks. One should not be misled into thinking that the more detailed the notation is • the more valuable it is.

"nucleus" and ta iI" used special symbols to ind ieate the pitch of the syllables that precede the nucleus. those of the head. who used arrows inserted in the text to mark the pitch change in the nucleus (the most prominent word in a tone-group). The next important development in the system of denoting prosodic phenomena was a not a t ion wit h in the lin e of the text. !): J Oh I . the degree and d irection of pitch change that matters. The tonetic symbols he used were as follows: '\ . ~"V if on-Iy I could ~ If ~J '1 ~ I J i. not only for the practical reason that it is difficult to read.. tired after your walk? '111 didn't notice any difference. Small dots correspond to unstressed syllables and thick dots mark the stressed syllables. Scholars as far back as the 19th century used musical symbols to ind icate the pitch changes in the voice.e.but also for the theoretical reason that the voice does not pass from one pitch to another at def in ite interva Is as in singing.b see her before go [69] Musical symbols are sometimes used even now. It is the level.. to transcr ibe what they considered to be intonation. The syllables following the nucleus (called the tail) are not marked.. Do you feel . the falling represent intonation visuallv was the introduction of a m u s j c a J not a t ion. i..the scandent head. 100 . He marked the head thus: the super lor head. It was devised by H..e. -. Besides those tonet ic ind ieat ions H..PaJr). But such a notation is unsatisfactory. Palmer also used a fuller notation in which the tune of the whole tonegroup is shown by dots ind ieat 1ng the relat ive pitches of each syllable in the "head".\ '). 1J the falling-rising nucleus. J the high rising nucleus. H. the infer ior head. the falling nucleus with intensification. i. the low rising nucleus. not the absolute tones of definite frequency which exist in mus ic.

Sign up to vote on this title
UsefulNot useful