Modifications of Vowels in Connected Speech

⇐ ПредыдущаяСтр 7 из 29Следующая ⇒

In discussing vowels we should now turn to vowel reduction as it is one of the factors that condition the defining of the phonemic status of vowel sounds in a stretch of speech. The modifications of vowels in a speech chain are traced in the following directions: they are either quantitative or qualitative or both. These changes of vowels in a speech continuum are determined by a number of factors such as the position of the vowel in the word, accentual structure, tempo of speech, rhythm, etc.

The decrease of the vowel quantity or in other words the shortening of the vowel length is known as a quantitative modification of vowels, which may be illustrated as follows:

1. The shortening of the vowel length occurs in unstressed positions, e. g. blackboard [ɔ:], sorrow [зu] (reduction). In these cases reduction affects both the length of the unstressed vowels and their quality.

Form words often demonstrate quantitative reduction in unstressed positions, e.g.

Is -»he or ˎshe to blame? — [hi:] But:

At last he has ˎcome. — [hi]

-»2. The length of a vowel depends on its position in a word. It varies in different phonetic environments. English vowels are said to have positional length, as you probably remember from your practical course of phonetics, e.g. knee — need — neat (accommodation). The vowel [i:] is the longest in the final position, it is obviously shorter before the lenis voiced consonant [d], and it is the shortest before the fortis voiceless consonant [t].

Qualitative modification of most vowels occurs in unstressed positions. Unstressed vowels lose their "colour", their quality, which is illustrated by the examples below:

1. In unstressed syllables vowels of full value are usually subjected to qualitative changes, e.g. man [mæn] — sportsman ['spɔ:tsmən], conduct ['kɒndəkt] —conduct [kən'dʌkt]. In such cases the quality of the vowel is reduced to the neutral sound [ə].

These examples illustrate the neutralized (reduced) allophones of the same phonemes as the same morphemes are opposed.

The neutral sound [ə] is the most frequent sound of English. In continuous text it represents about eleven per cent of all sounds. And if we add the occurrence of [ı] which is closely related to [ə] in unstressed positions we get a figure close to twenty per cent — nearly one sound in five is either [ə] or the unstressed [ı]. This high frequency of [ə] is the result of the rhythmic pattern: if unstressed syllables are given only a short duration, the vowel in them which might be otherwise full is reduced. It is common knowledge that English rhythm prefers a pattern in which stressed syllables alternate with unstressed ones. The effect of this can be seen even in single words, where a shift of stress is often accompanied hy a change of vowel quality; a full vowel becomes [ə], and [ə] becomes a full vowel. Compare: analyse ['ænəlaız] — analysis [ə'nælısıs]; in both words full vowels appear in the stressed positions, alternating with [ə] in unstressed position. It would be impossible to have [ə] in a stressed syllable, and almost as impossible to have a full vowel in every unstressed syllable.

We should point out that in Russian there is a well-developed system of unstressed vowels. All the Russian vowels are regularly subjected to reduction, the vowels [o] and [и] more

often than others. The degree of sound weakening depends on the place of the unstressed vowel in relation to the stressed one. The farther the unstressed syllable is from the stressed one the weaker the vowel in the unstressed syllable is, e.g. молоко [мълʌко́].

2. Slight degree of nasalization marks vowels preceded or followed by the nasal consonants [n], [m], e.g. "never", "no", "then", "men" (accommodation).

The realization of reduction as well as assimilation and accommodation is connected with the style of speech. In rapid colloquial speech reduction may result in vowel elision, the complete omission of the unstressed vowel, which is also known as zero reduction. Zero reduction is likely to occur in a sequence of unstressed syllables, e.g. history, factory, literature, territory. It often occurs in initial unstressed syllables preceding the stressed one, e,g. correct, believe, suppose, perhaps.

The example below illustrates a stage-by-stage reduction (including zero reduction) of a phrase. Has he done it?

[hæz hi-ˏdʌn it]

[həz hi ˏdʌn it]

[əz ı ˏdʌn it]

[z ı ˏdʌn it]

We would like to conclude that certain interrelation which we observe between the full form of a word and its reduced forms is conditioned by the tempo, rhythm and style of speech.

SOUND ALTERNATIONS

In the previous section we saw how the pronunciation of speech sounds can vary according to their position in the word, and that the variation is usually quite regular and can be stated in the form of "rules" which predict the variants or allophones, that will occur in each position. In this section we shall look at variation of a different kind, involving not only interchange between sounds, but also between related phonemes. We shall now touch upon the sound variations in words, their derivatives and grammatical forms of words. These variations are known as sound alternations. It is perfectly obvious that sound alternations are caused by assimilation, accommodation and reduction in speech. Alternations of consonants are mainly due to contextual assimilations: the dark [ł] in spell alternates with the clear [l] in spelling. Vowel alternations are the result of the reduction in unstressed

positions: combine I'kɒmbaın] (n) — combine [kəm'baın] (v) where [ɔ] in the stressed syllable of the noun alternates with the neutral sound in the unstressed syllable of the verb. In Russian alternation may be illustrated by the following pair of words: роса [рʌса] — росы [росы]. The [ʌ] sound in the unstressed syllable of the word роса alternates with the [o] sound in the stressed syllable of the plural form росы. To approach the matter from the phonological viewpoint it is important to differentiate phonemic and allophonic alternations. Some sound alternations are traced to the phonetic changes in earlier periods of the language development and are known as historical.

In this section we are concerned with typical, well-documented sound changes. A thousand years ago, the English spoke a language which was the direct ancestor of the English we speak today. Yet this language which we call Old English, or Anglo-Saxon, is not intelligible to present-day speakers in the written forms in which it has been handed down to us, nor would it be intelligible, if it were spoken: it has to be learned as if it were a foreign language. There are many reasons for this: there have been changes in the vocabulary, and changes in the grammar, but the primary reason for the differences between Old and Modern English is the sound changes which have taken place over the intervening period. Sound changes occur gradually; the time scale for significant changes is usually measured in hundreds of years, rather than decades. The English spoken in the 1930s sounds old-fashioned, but the changes are in details only. For two stages of "the same" language to resemble different languages require a much longer period. English speakers need assistance in reading Chaucer (fourteenth century), just as Greeks today study the Classical Greek of the fifth century ВС as if it were a foreign language. Historical alternations distinguish grammatical forms of words and lexical units in the process of word-building, e.g. provide — provision (provide was at one time pronounced with [i:]). The [i:] was converted to [aı] by complex rules of diphthongization and vowel shift. These changes, it should he noted, reflect actual changes in the history of English. Here is another example: the OE [k] phoneme split into two separate phonemes, modern [k] and [ʧl. Thus alongside cat, cool, from the OE catt, col, we have choose, chin from OE cēosan, cinn. Minimal pairs such as kin — chin; care — chair establish that Modem English has the two sounds as separate phonemes.

The alternations exemplified above are quite regular. Historical alternations mark both vowels and consonants, though the alternating sounds are not affected by the phonetic position or context, neither are they subjected to stylistic modifications. To sum up, the sound changes which occurred in the process of historical development of the English language are reflected in present-day English as alternations of phonemes differentiating words, their derivatives and grammatical forms. We are going to introduce here phonetic realizations of the most common historical alternations and their functions in word building and word formation. The alternations are often supported by suffixation. The following list of examples presents the most common types of alternations.

Vowel Alternations

1. Distinction of irregular verbal forms:

[i— e — e]: mean — meant — meant

[ı — ʌ — ʌ]: dig — dug — dug

[aı — ɜu — ı]: write — wrote — written

[ı — æ — ʌ]: sing — sang — sung

[ɛə — ɔ: — ɔ:]: wear — wore — worn

[aı — ı — ı]: hide — hid — hidden

[i: — ɜu — ɜu]: speak — spoke — spoken [ɜu — u: — ɜu]: know — knew — known

[ı — eı — ı]: give — gave — given

[e — ɒ — ɒ): get — got — got

[i:- — ɔ: — ɔ:]: teach — taught — taught

[æ — u — u]: understand — understood — understood

[eı— u — eı]: take — took — taken

[eı — ɜu — ɜu]; wake — woke — woken

[u: — ɔ: — ɔ:]: shoot — shot — shot

[e — ɜu — ɜu]: tell — told — told

[ı —æ —æ]: sit —sat —sat

[ı — ɔ: — ɔ:]: think — thought — thought

[ʌ — eı — ʌ]: become — became — become

[aı — ɜu — ı): rise — rose — risen

[ɜu — u: — ɜu]: grow — grew — grown

[u — ɜu — ɜu]: choose — chose — chosen

[aı — u: — ɜu]: fly — flew — flown

[aı — ɔ: — ɔ:]: fight — fought — fought

[aı — au — au]: find — found — found

[i: — ɔ: — i:]: see — saw — seen

[ıə —з: —з:]: hear — heard — heard

and some other less common verbal alternations of this type.

2. Distinction of causal verbal forms: [ı—e]: sit —set

[aı — eı]: rise — raise [ɔ: — e]: fall — fell

3. Distinction of singular and plural forms of nouns:

[æ — e]: man — men foot — feet tooth — teeth mouse — mice woman — women child — children

4. Distinction of parts of speech in etymologically correlated words:

feast — festive class — classify long — length broad — breadth nation — national wise — wisdom hot — heat

This type of alternation is often strengthened not only by suffixation but also by the shifting of stress like in: part — par'ticular, 'climate — cli'matic.

Consonant Alternations

1. Distinction of irregular verbal forms: [d —t]: send — sent, lend — lent

2. Distinction of parts of speech in etymologically correlated words:

advice — advise, house — house, use — use defence — defend intent — intend speak — speech important — importance

[u - i:]: [u: - i:]: [au — aı] [u-ı]: [aı-ı]:

[t-e]: [a: —æ]: [ɒ-e]: [ɔ:-e]: [eı — æ]: [aı-ı]: [d-i:]:

[s-z]: [s-d]: [t-d]: [k-ʧ]: [t-s]:

3. Vowel + Consonant Alternations (often supported by suffixa-

tion and the shifting of stress)

[ı — aı] + [v — f): live — life

[a: — eı] + [Ɵ — ð]: bath — bathe

[e — i:] + [Ɵ — ðj: breath — breathe

[ɒ — u:] + [s — z]: loss — lose

In the Russian language there are numerous types of vowel and consonant alternations illustrated by the examples below:

сидеть — сядь резать — режу плыть — сплав лететь — лечу судить — сужу плакать — плачу

Sound alternations are also widely spread on the synchroni-cal level in the present-day English and are known as contextual. In connection with contextual sound alternations there arises a problem of phonemic identification of alternated sounds. The functioning of sounds in different grammatical forms and derivatives of words seems very complicated and flexible. The study of the relationship between phonemes and morphemes is called morphophonemics. The interrelation of phonology and morphology in linguistics' investigations is also known as morphophonology or morphonology which is actually the phonology of morphemes. Morphonology studies the way in which sounds can alternate as different realisations of one and the same morpheme. A morpheme is a minimal unit of meaning. We would all agree that such words as windy, dusty, sunny consist of two morphemes. Similarly, demonstration, alternation have two component morphemes. The meanings of wind, dust, sun as well as of demonstrate, situate are obvious. But what function do the morphemes -y and -ion perform? On the basis of the examples, it appears that the function of -y is to convert a noun into an adjective. Similarly -ion converts a verb into a noun. These morphemes have a grammatical meaning, their main purpose is to convert one part of speech into another. Now then what is meant by the identification of alternated sounds? Each set of data below exemplifies a sound alternation in one and the same morpheme of two different parts of speech.

malice ['mælıs] —malicious [mə'Iıʃəs]

active ['æktıv] —activity [ək'tıvıtı]

abstract f'æbstrækt] —abstract [æb'strækt]

conduct ['kɒndəkt] —conduct [kən'dʌkt]

contrast ['kɒntræst] —contrast [kən'træst]

We are interested now in the sound in its weak position. Vowels are said to be in their strong position when they are in stressed syllables and in the weak position when they are in the unstressed ones. Consonants may well be said to be in their strong position before vowels and in the intervocalic position; they are in weak positions when they are word final or precede other consonants.

There may be different solutions to the problem of phoneme identification in weak positions of alternated words. The question arises whether the sound [ə] in the words activity and con'trast is a neutral phoneme or it is an allophone of the [æ] or [ɒ] phonemes (as in activé, 'contrast) which loses some of its distinctive features in the unstressed position. The difference is quite essential as in the first case the neutral sound is identified as an independent neutral phoneme, in the second — it is a neutralized allophone of the [æ] or [ɒ] phonemes of the corresponding alternated words.

It is fair to mention here that the problem is by far more significant for the Russian language because of the widely spread voiced/voiceless assimilation and vowel reduction in the language, e.g.

a) мороз [мʌро́с] - морозы [мʌро́зы]

город [го́рът] - города [гърлда́]

зуб [зуп] зубы - [зу́бы]

b) коса [кʌса́] - косы [ко́сы]

сторона [стърʌна] - стороны [сто́ръны]

слон [сло́н] - слоны [слʌны́]

Scholars of different trends are not unanimous in solving the problem. Though the discussions of the problem are dying down at present the conceptions remain determining for this or that linguistic point of view. The so-called morphological school represented by Soviet philologists R. I. Avanesov, V. P. Sidorov, P. S. Kuznetsov, A. A. Reformatsky supported the theory of neutralization of phonemes.

The concept of neutralization, and the theory related to it derives originally from the Prague School of phonology which flourished in the thirties; it is particularly associated with N. S. Trubetskoy (34) and R. Jakobson (62). A neutralization is said to occur when two or more closely related sounds, which are in contrast with each other in most positions like дом — том, are

found to be non-contrastive in certain other positions, e.g. суд [сут] — судить [суд'йт']. That means that there are environments where the two sounds do not contrast with each other, even though they normally do. When this happens, the opposition between the two sounds is said to be neutralized. The loss of one or more distinctive features of a phoneme in the weak position is called phonemic neutralization. In English, the voicing opposition is neutralized after the initial [s]. We are well aware of the fact that the phonemes [t] and [d], for example, contrast in most environments: initially (tick — Dick), finally (bid — bit); after nasals (bend — bent), after [l] (cold — colt). But after [s], no contrast between [t], [d] is possible, nor, similarly, is there a contrast between [p], [b) and [к], [g] in this environment. The voicing contrast is neutralized after initial [s].

The sound which actually occurs in this environment does not correspond exactly to either the voiced or voiceless sounds, but shares the features of both of them: speak, story, sky. If we represent the [p] of peak, as [p^h], the sound [p] of speak as [p ⁼]¹, and the [b] of beak as [b], the distribution of features is as follows:

[p^h] — peak	[p ⁼] — speak	[b] — beak
Bilabial	bilabial	bilabial
Plosive	plosive	plosive
voiceless fortis	voiceless fortis	voiced lenis
aspirated	unaspirated	unaspirated

This distribution of features shows that [p ⁼ ] is truly intermediate between [ph] and [b]; it shares their common properties, bilabial and plosive, but then it shares one feature (voiceless), with [ph] and the other (unaspirated) with [b]. Does it belong, then, with the phoneme [p] or with [b]? Usually phonologists have assigned it to [p] on the grounds that voicing is more important in English than aspiration, or for other similar reasons. There is a good case, however, for arguing that [p ⁼ ] belongs equally with [p] or [b], as shown by the distribution of features, and therefore that speak could be transcribed equally as [spi:k] or as [sbi:k]. The choice of [p] is usually preferred, because of the spelling, as English has a long-established tradition of spelling these words with sp, st, sk.

The Moscow philologists claim that interchange of sounds

¹ [x̿] — loss of aspiration.

manifests close connection between phonetics as the science of the sound system and morphology of the language which studies grammatical meanings. Alternations are observed in one and the same morphological unit, in a morpheme, and actualize the phonemic structure of the morpheme. Thus the phonemic content of the morpheme is constant. It should be noted here that alternations of morphemes cannot be mistaken for the oppositions of minimal pairs in different stems of words. For instance, [ш] — [ж] in нож [ш] — ножик [ж] is an alternation of sounds within the same phoneme in the same morpheme, where [ш] is an allophonic realization of the [ж] phoneme, while [ш] — [ж] in шар — жар is the opposition of two different phonemes which differentiates the actual meaning of the words. Let us compare the sentences: У него плохой грипп . — У него плохой гриб . The distinction of the sentences is lost as soon as they are pronounced. In the word гриб [n] the final consonant loses one of its distinctive features (voiced/voiceless) but it is associated with the word грибы, it is morphologically bound with them and manifests an allophone of the [б] phoneme. Thus the sound [п] may be an allophone of different phonemes: it is the principal allophone of the phoneme [n] in the word грипп [п] and the subsidiary allophone in the weak position of the phoneme [б] in the word гриб [п].

In the alternation вода — воды the sounds [ʌ] — [о] are allophones of one and the same phoneme [о]. So [ʌ] is the allophone of the phoneme [o] in the weak position of the word вода . It might be an allophone of some other phoneme, [a] for instance, in the word трава. In other words one and the same sound may belong to different phonemes.

In the English word activity the neutral sound [ə] is the allophonic realization of the phoneme [æ] in its weak unstressed position. It alternates with the principal allophone of the phoneme [æ] in the words of the same root-morpheme: act, active. And [p⁼] in speak is actually an allophone of the [p] phoneme.

The supporters of the morphological trend define the phoneme as follows: «Это функциональная фонетическая единица, представленная рядом позиционно чередующихся звуков» (25, р. 107).

The notion of «фонетический ряд», suggested by R. I. Avanesov (1), demonstrates positionally determined realizations of the phoneme. Positionally alternating sounds are grouped into

4—3483

one phoneme whether they are similar or have common features (that is common allophones) with other phonemes.

The Russian preposition с + noun may have the following realizations:

с Колей — [с] с Шурой — [ш]

с Тимошей — [с'] с Женей — [ж]

с Галей — [з1 с Чуком—[ш']

с Димой — [з' ]

In the morphological conception the alternations of the phonemes are not analysed apart from the morpheme, as form and content make dialectical unity. The phonetic system is not isolated from the grammatical and lexical structure of the language, and the unity between the form and content cannot be destroyed. Yet as an answer to the problem it is not entirely satisfactory since ordinary speakers are in no doubt that the sound which occurs in a word like гриб is [б] not [п], and in the English word speak [p ⁼ ], mentioned above, is nothing but [p]. The perception of the listener makes us find the morphological conception too discrepant and conflicting.

The so-called Leningrad phonological school (L. V. Shcherba and his followers L. R. Zinder, M. I. Matusevitch) assert that the phoneme is independent of the morpheme. So [ʌ] in вода belongs to the [a] phoneme while [o] in воды to the [o] phoneme; [c] in the word мороз belongs to the [c] phoneme and [з] in мо розы — to the [з] phoneme respectively. The supporters of this conception claim that the phoneme cannot lose any of its distinctive features.

In the line of words of the same root-morpheme гриб — гри бы — грибов the sound [п] in гриб is an allophone of the [п] phoneme and the sound [б] in the derivatives manifests the phoneme [б].

Consequently, the consonants [б] and [п] do not lose any of their distinctive features and represent different phonemes: [п] in гриб and [б] in грибы. They are: a plosive bilabial voiceless fortis stop/a plosive bilabial voiced lenis stop. They are capable of creating phonological oppositions like пыл — был, столб — столп .

As far as the English language is concerned, the neutral sound [ə] in the word activity and the sound [æ] in the words act, active of the same morpheme belong to different phonemes. As a phoneme does not lose any of its distinctive features the sound [ə] in the

word activity is an allophone of the [ə] phoneme and the sound [æ] in the words act, active is the [æ] phoneme. It seems that according to this point of view the unity between the form and content is destroyed, thus phonology is isolated from morphology.

N. S. Trubetskoy (the Prague phonological school) (34) arrived at an original solution of the phonemic status of a sound in alternations. To overcome the difficulty he introduced a broader phonological unit than a phoneme and named it an archiphoneme. An archiphoneme combines the distinctive features of two different phonemes into one in variations where one of the phonemes is neutralized in the weak position. An archiphoneme is defined as a combination of distinctive features common to two phonemes. According to Trubetskoy it consists of the shared features of two or more closely related phonemes but excludes the feature which distinguishes them. The archiphoneme of [p], [b] consists of the features: bilabial, plosive, but excludes voicing which separates them. In our examples вода — воды [ʌ] and [о] manifest one archiphoneme as well as [п] and [б] in гриб — гри бы — another archiphoneme. For transcription purposes the symbols of capital [O] is used to represent the archiphoneme [ʌ — о] and capital [П] — of [п — б] respectively. One of the disadvantages in extending the notion of an archiphoneme is that the Prague school phonologists limited neutralization to closely related phonemes. A neutralization can be said to occur only if there is uncertainty about the identity of the sound in the position of neutralization. Before two phonemes can be neutralized, they must have common qualities which do not occur in other phonemes. Thus [p], [b] can neutralize because they are the only labial plosives in the language, they share these two features, but no other sounds share them. However, [n], [ŋ] cannot neutralize because their common property (nasality) is also shared by [m], so any neutralization of nasals must involve all the three of them [n], [ŋ], [m], and there cannot be a neutralization of [f], [h] before [1] since the features common to these, voiceless, fricative, are also shared by other phonemes, namely [ʃ], [s], [Ɵ].

In English initial [s] can precede a plosive, but initial [ʃ] cannot. Since (s), [ʃ] contrast in most other environments in English, they share the features voiceless, fortis, constrictive, fricative, and are the only sounds to have these features, it may seem somehow that the contrast is neutralized, though no one will doubt the sound that occurs in the word speak is [s] and not [ʃ].

In conclusion we have to admit that the described conceptions are arbitrary, none is ideal. The morphological conception; as well as the theory of the archiphoneme seems complicated, though the former appears to be most optimum for theory and practice.

Stylistic Modifications of Sounds

So far we analysed sound modifications typical of connected speech mostly within the norm of the English sound system but without any reference and special attention to their stylistic; distinctions. It is generally known that variability in the sound., realizations is caused by the environment, distribution of sounds in a speech chain, the accentual and the rhythmical structures of the utterance and other linguistic factors and extralinguistic factors. Phonetic distinction marks different forms of communication: monologue, dialogue and polylogue; reading and spontaneous speaking and also forms of speech activity.

Stylistic oppositions have long been observed in linguistic literature in the two marginal types of pronunciation: formal and informal. Formal speech suggests dispassionate information on the part of the speaker. It is characterized by careful articulation and relatively slow speed. A. C. Gimson defines it as careful colloquial style (57), G. Brown describes it as formal slow colloquial style of speech (48). V. A. Vassilyev labels it normal-speed colloquial style of speech (79). Other Soviet field researchers call it full style (9). Informal speech implies everyday conversation.. The following definitions are also used: rapid colloquial speech conversational style.

As was mentioned earlier, stylistic modifications of intonation do not coincide with those of sounds.

It is not always easy to draw a borderline between different stylistic realizations as within each style speech may differ in accordance with extralinguistic factors. For instance, it is generally known that spontaneous form of speech differs greatly from reading on the phonetic level. The character of the outcome of speech in reading and speaking results in different phonation. The difference seems to be basically in the prosodic parameters. Spontaneous speech is usually characterized by the variations of pitch, tempo, loudness and delimitation of phrase, numerous

100

hesitation devices. Segmental modifications are less recognizable. The same sound modifications often mark both speech activities, reading and speaking, though the use of the simplified sound form is often more typical for speaking than for reading; that means that in speaking it is more frequently used than in reading.

Now let us turn to different forms of communication. As has already been mentioned, a monologue often presupposes public speaking with a considerable distance of the addresser (the speaker) from the addressee (the listener) or a piece of calm narrative. Dialogues are more often private, personal and intimate. Monologuing is characterized by more phonetic precision. On the other hand speech may vary in numerous ways. The interaction of the extralinguistic factors may arrange the opposite situation: the speaker's highly excited narration of some critical situation will become full of slurring while a dialogic discussion of problems between colleagues will be phonetically most precise.

Stylistic sound variations seem to have the tendency towards the increase of the sound modifications in speech with the quickening of its tempo and the weakening of the carefulness, e.g. government ['gʌvənmənt -> 'gʌvəmnt -> 'gʌvmnt -> 'gʌbmnt].

Phonetic means which are stylistically relevant depend on the extralinguistic situation of the discourse.

The first thing that counts in the stylistic modifications of sounds is the character of relationship between the speaker and the listener and the degree of formality in their discourse. Speech continuum reflects the amount of attention that the speakers give to their speech. It is assumed that in formal situations the participants will monitor their linguistic behaviour. If the speaker wants to be clearly understood (like while producing a lecture with an educational aim), he should sound explicit and his pronunciation may be characterized as supercorrect. In informal situations, where speakers are more relaxed, less attention will be given to speech and more natural and simplified it will sound. Consequently, the degree of simplification of speech (assimilation, reduction, elision) may be looked upon as a style forming means.

We should admit that the character of sound modifications in relation to situational factors of speech communication is only beginning to be the object of thorough instrumental analysis today. The investigations are usually based on the contrastive

101

principle, and we shall follow the same principle in our description of sound modifications. We would like to illustrate sound simplifications in informal conversational speech, which are not registered in the listener's mind but actually occur in the speaker's discourse. The listener is not usually aware of the changes since they do not affect the meaning. If you look at the transcribed variant of the text you will observe the actual phonation in informal speech. Mind not all the stages of assimilation, reduction and especially accommodation can be reflected in the symbols of the transcription. Here is an extract from a conversation of two friends in a restaurant:

Brenda: Ah, right, here we are!

[ˎa: | ˎraıt ⌇ ^→hı(ə) wi: ˎa:]

Bob: This is the place I was telling you about.

[^→ðıs iz ðə ˎpleıs a wz telıŋ jə əba(u)t]

Br.: Yeah, could you ask the waiter if we can sit near the

window?

[ˎja: | ˋkudʒu a:sk ðə 'weıtə if wi kn ˌsı(t) nı(ə) ðə ˌwındɜ•]

Waiter: Good evening, sir. [gud ˎi:vnıŋ sə]

В .: We'd like to sit near the window if that's possible.

[wid ^→laı(k) t(ə) sı(t) nı(ə) ðə ˎwindɜ(u) ⌇ if _→ðæts ˌpɒsıbl]

W.: Er... Ah! ... er... I'm afraid all the tables there are taken. Would you mind sitting near ... nearer the bar? [> з: ⌇ ˎa > ɜ: m →fre(ı)d əl ðə ˌteıblz ðɛə ˎteıkn || ˋwudʒu → main sıtın >nı(ə) ⌇ nıəıə ðə ˌba:]

В .: Oh, yes, all right. That suit you?

[ˎɜu ⌇ jes ɔ: ˎraıt || ðət 's(j)u:ʧu]

Br.: Mm, fine. [> m: ˎfaın]

В.: Good. [ˎgud]

Now let us pass over to a brief observation of the possible sound modifications which occur in oral speech.

We shall try to illustrate the existing marginal varieties of the sound modifications which appear regularly in informal speech, the slow formal English being taken as the starting point in the

102

comparison. We would like to start the description with vowel sounds.

Typical character of sound simplifications in relation to the degree of formality is the great qualitative stability of vowels in slow formal speech and more frequent sound variability in informal spoken English. Both front and back vowels in less explicit articulation tend to be changing towards neutralized sounds, especially in grammatical w6rds.

Spelling Formal Informal

it's not its 'nɒt əts 'not

because bı'kɒz bıkəz

according to ə'kɔ:dıŋ tə əkədıŋ tə

I think he was aı'Ɵıŋk hi· wəz ʌ 'Ɵıŋk ı wz

The historically long vowel [i:] tends to lose its diphthongization; as the next stage it undergoes quantitative reduction and finally changes its quality as well.

Spelling Formal Informal

I don't believe it at 'dɜunt bı'li:v it ʌ dɜun(t) bə'lıv it

it seems to be it 'si:mz tə bi· it 'sımz tə bı

The similar process of reduction is likewise observed in [u:] simplified to [u].

Spelling Formal Informal

a few more words ə 'fjiı: 'mɔ: 'wɜ:dz ə fju mɔ• 'wɜ:dz

a new aspect ə 'nju: 'æspekt ə 'n(j)u 'æspekt

As to labialization of vowels the amount of rounding varies greatly between the individual speakers. The vowel [o:] seems to retain lip rounding as a rule. The vowels [ɒ] and [ɔı] have very little, if any, rounding at all in informal speaking. The vowels [u:], [u] seem to lose the rounding altogether.

Diphthongs are very often monophthongized in informal speech.

The diphthong [ɛə] tends to be simplified to [ɛ(:)], e.g.

Spelling Formal Informal

Where wɛə wɛ

here and there 'hıər ənd 'ðɛə 'hı(ə)r ən 'ðɛ

In an unstressed position it is further modified to [e], e.g. there is an opinion [ðer ız ən ə'pınjən].

103

The diphthong [ıə] often gets a sort of central vowel realization [з].

Spelling Formal Informal

really strange 'riəlı 'streınʤ 'rзlı 'streınʤ

serious action 'sıərıəs 'ækʃn 'sзrı(ə)s 'ækʃn

experienced worker ıks'pıərıənst 'wз:kə ıks'pзrıənst 'wз:kə

The [u] ending diphthongs [au] and [зu] are simplified into [a] and [з] accordingly. The various stages of their realizations are found both in stressed and unstressed positions. The quality of the initial element is retained and the second element, the glide is obscured or lost.

Spelling now they South of Italy going ahead yes or no

Formal nau ðeı 'sauƟ əv 'ıtəlı 'gзuiŋ ə'hed 'jes ɔ• 'nзu

Informal 'na ðe(ı) 'saƟ əv 'ıtəlı 'gзŋ ə'hed 'jes ə'nз

Unstressed positions are sometimes marked by the next stage of qualitative reduction. The diphthong [au] is realized as some kind of [ʌ].

Spelling and now we've come to mark how different it is

Formal ənd 'nau wi•v 'kʌm tə 'ma:k hau 'dıfərənt it ız

Informal ən nʌ wi•v 'kʌm tə 'ma:k hʌ 'dıfrənt it ız

The diphthong [зu] is sometimes completely neutralized in the unstressed position.

Spelling Formal Informal

so we've discussed sзu wi•v dıs'kʌst sə wiv dıs'kʌst

hope to settle it hə tə 'setl it hзup tə 'setl it

Vowel elision is very frequent in informal conversational style. It often goes with other processes involving assimilation and elision of consonants. Elided neutral sound [ə] is very common in the unstressed syllables of polysyllabic words, like:

104

Spelling Formal Informal

collective kə'lektıv 'klektıv

different 'dıfərənt 'dıfrənt

prisoner 'prızənə 'prıznə

political pə'lıtıkl 'plıtıkl

phonetically fə'netıkəlı 'fnetıkəlı

In the last three examples the loss of [ə] in the initial unstressed syllable of a word causes the initial consonant form a duster with the consonant of the stressed syllable. Vowel reduction mostly occurs in extended utterances in sequences of words. The loss of the neutral sound [ə] in the preposition to or the particle to preceded by a consonant is a very common pattern.

Spelling Formal Informal

next to Liverpool 'nekst tə 'lıvəpu:l 'nekst 'tlıvəpu:l

back to London 'bæk tə 'lʌndən 'bæk 'tlʌnd(ə)n

to see them tə 'si: ðəm 'tsi: ðəm

future situation 'fju:ʧə ˌsıtju'eıʃn 'fju:tʃə 'sıtjueıʃn

this afternoon ðıs 'a:ftə'nu:n ðıs 'a:ftnu:n

after all 'a:ftər 'ɔ:I 'a:ft'rɔ:l

In the majority of spoken utterances beginning with its the initial [ı] is elided when the phrase runs on without a marked pause after the previous saying.

Spelling Formal Informal

it's paid well its'peıd wel ts'peıd wel

it's necessary its 'nesəsərı ts 'nesəsərı

it's counted as its'kauntıd əz ts 'kauntıd əz

Likewise in polysyllabic words beginning with the unstressed ex- it is often simplified to [ks].

Spelling Formal Informal

Extremely iks'tri:mlı 'kstri:mlı

Extraordinary ıks'trɔ:dnrı 'kstradnn

Excluded ıks'klu:dıd 'ksklu:dıd

As we have already mentioned vowel reduction often results in regular consonant clusters like [tr], [fr], [pl], [kl] typical for the English sound system. Cf. tram, try, tree and interesting, aft(e)r all; please, play and p(o)litical; clay, cloud, circle and c(o)llective; friend, from and diff(e)rence.

105

Alongside with regular clusters in informal careless speech we find phonetic facts which seem impossible for the English pronunciations namely consonant sequences [tsn], [tsk], [tsp] and others.

Spelling Formal Informal

it's not exact its 'nɒt ıg'zækt ts 'nɒt ıg'zækt

it's close to its 'ktɜus tə ts 'klɜus tə

it's perhaps you its pəhæps 'ju: ts pəhæps 'ju:

These sequences never occur in speech where the words are uttered clearly and explicitly but in the stream of informal speech in the least prominent parts of the utterance. These facts represent the natural process of compression, or simplification which are known in other languages. In the Russian language, for example, the number of consonant clusters as a result of vowel reduction increases in informal spontaneous speech.

Spelling Formal Informal

Пожалуйста пʌжа́лустъ пʌжа́лстъ

Сейчас с'и^эч'а́с ш̌'а́с

Хорошо хърʌшо́ хрʌшо

In the normal course of conversation the reductions themselves go quite unnoticed by the listener as the distinctive features of phonemes are not lost. The listener is mainly interested in the meanings the speaker aims to convey and not in the precise phonetic detail with which he conveys them. So long as the meaning is recoverable, the listener is satisfied. We have to regard the omissions and reductions then as a kind of economy on the part of the speaker who aims not to give more information than is necessary. The speaker assumes usually correctly that the listener will not notice the omissions. Paradoxically, this makes the omissions difficult to observe, being so used to ignoring it.

We shall now turn to the most common "tendencies in the stylistic modifications of consonants. The process of different sorts of assimilations typical for the English language is usually not so simple as the replacement of one member of phoneme by another. The mechanism of assimilation is a complex of alternations of segmental realizations within the cluster, which is difficult to exemplify in the symbols of the accepted form of transcription, especially when the described sound is only partially "there".

106

The assimilations of consonants according to voiced (lenis) — voiceless (fortis) principle are not so common in English as they are in Russian. Still the degree of voicing or devoicing of consonants increases passing gradually through several stages from slow careful reading before a large audience to informal careless conversation and ends with the elision of the sound, e.g. must be [mʌst bi•-» mʌst b^pi• -» must pi• -» mʌs pi•]; don't get [dɜunt get -» dɜunt ^kget -» dɜunt ket].

In the intermediate stages the cluster is represented by a series of sound alternations which reflect the adaptation to the neighbouring sound. The elision of "t" is often met in the position between two consonants.

The consonants are also markedly different in informal conversational style according to their place of articulation. Word final consonants [t], [d], [n], sometimes [m], [s], [z] immediately followed by a velar or labial consonant undergo a sort of adaptation.

Spelling Formal Informal

great burden 'greit 'bɜ:dn 'greip 'bɜ:dn

that man 'ðæt 'mæn 'ðæp 'mæn

American a'merıkən ə'merıkəŋ

Government 'gʌvnmənt 'gʌv(ə)mənt

hundred places 'hʌndrıd 'pleısız 'hʌndrəb 'pleısız

taken gladly 'teikn 'glædlı 'teıkŋ 'glædlı

Instead of the closure for the [t] a marked glottal stop [ʔ] is also observed before the modified plosive consonant.

Spelling Formal Informal

Great Britain 'greıt 'brıtn 'greıʔ 'p^brıbn

didn't go 'dıdnt 'gɜu 'dıdŋʔ'k^gɜu

couldn't come 'kudnt 'kʌm 'kudŋʔ 'kʌm

The illustrated modifications could be summarized in the following way.

[t] => [p] before [p], [m] that place ['ðæp 'pleıs]

that might ['ðæp 'maıt]

[t] => [k] before [k] don't question ['dɜuŋk 'kwestʃ(ə)n]

107

[d] => [b] before [p], [b], [m] good morning ['gub 'mɔ:nıŋ]

would be ['wub bi:]

[d] => [g] before [k], [g] Good God ['gug 'gɒd]

good cook ['gug 'kuk]

[n] => [m] before [p], [b], [m] on me [ɒm 'mi:]

in business [ım 'bıznıs]

[n] => [k] before [k], [g] in quite [ıŋ 'kwaıt]

can get [kəŋ 'get]

We should strongly emphasize the idea that the students are not recommended to imitate these extreme forms of the existing ways of adaptation in very rapid careless speech.

A definite and very frequent process of assimilation is observed when [s], [z] sounds are followed by the palatal [j] in the unstressed part of the phrase. The alveolars tend to become palatoalveolar in informal conversational style.

Spelling Formal Informal

this year 'ðıs 'jıə 'ðıʃ 'jıə

as you əz ju: əʒ ju:

as yet əz jet əʒ jet

The palatal [j] is strong enough to affect the manner of articulation of the preceding [t], [d] sounds. In accordance with the tempo and style of speech, individual fluency, number of recipients and other situational factors the assimilated segment preceding [j] may consist of several sections with gradually changing features. The process most often leads to an affricate:

would you [wudju: -» wud'ju -»wuʤu]

could you [kudjır -»kud'ju < kuʤu]

mind you [maındju: -» maınd'ju -» maınʤu]

can't you [kantju: -» ka:nt^ʃju -» ka:nʧu]

about you [əbautju: -» abaut^ʃju -» əbauʧu]

The elision of consonants is no less frequent process in informal speech than a vowel elision. The most common consonants to find involved in elision are [t] and [d]. Elision usually occurs in a syllable final sequence when the sound stands between two consonants. It is said to be more common for [t] and [d] to be elided between the other two consonants than it is for them to be pronounced.

108

Spelling Formal Informal

second group 'sekənd 'gru:p 'sekəŋ 'gru:p

first five 'fɜ:st 'faıv 'fɜ:s' faıv

next point 'nekst 'pɔınt 'neks 'pɔınt

best judge 'best 'ʤʌʤ 'bes 'ʤʌʤ

the fact that ðə 'fækt ðət ðə 'fæk ðət

second term 'sekənd tɜ:m 'sekən 'tɜ:m

[d] elides even more readily than [t]. We find the loss of [d] in a syllable final sequence preceding another consonant but immediately following a vowel.

Spelling Formal Informal

that it would be ðət it wud 'bi: ðət it wu 'bi

he said some words hi- 'sed sʌm 'wɜ:dz (h)i 'se səm 'wɜ:dz

about abaut əbaut

Other consonants tend to be elided in some definite environments. For instance, the consonant [v] is often elided when it is final in an unstressed form word have or of and immediately precedes another consonant.

Spelling Formal Informal

lists of the students 'lists əv ðə 'stju:dənts 'lists ə ðə 'st(j)u:d(ə)nts

we've been studying wiv bin 'stʌdııŋ wı bin 'stʌdııŋ

of course əv'kɔ:s, əf'kɔ:s ə'kɒs

The definite article [ðə] is often realized as the neutral sound alone. It occurs in cases when the definiteness of the noun is clearly established and [ə] can only be interpreted as the realization of the definite article [ðə].

Spelling Formal Informal

and the way he did it ənd ðə 'weı hi•'did it ən(d) ə "weı (h)ı'did it

and the reason for it ənd ðə 'ri:zn fər it ən(d) ə 'ri:zn frt

and the Scotchman ənd ðə 'skɒʧmən ən(d) ə 'skɒʧmən

The elision of [1] is restricted to the position after the vowel [ɔ:]. This process was established in the earlier periods of the English language which is reflected in the pronunciation of the words talk, walk; sometimes in the word certainly.

109

Spelling Formal Informal

all right ɔ:I 'rait ɔ:'raıt

already ɔ:l'redı ɔ:'redı

always 'ɔ:lwız 'ɔ:wız

also 'ɔ:lsɜu 'ɔ:sɜu

The elision of [l] in words beginning with all is typical even for slow full speech style.

We have tried to exemplify the extreme variants of sound modifications in the informal conversational style to help Russian learners comprehend spoken English. It is hardly a teachable model. The junior trainees should not be required to produce the extreme forms given in this section but to recognize them and understand sequences of words in which they occur Refinements of pronunciation of this kind should be left for advanced students to adopt by themselves. We should like to make it quite clear that the beginners are not supposed to be taught the extreme forms of careless speech like [mʌs pi•] for must be but that they should properly use the established forms of sound modifications (reductions, assimilations) of connected speech and articulate must be with the loss of plosion.

The described stylistic variability conditioned by the extralinguistic factors has a certain theoretical aspect. It concerns the problem of a subtle structural relationship of phonetic units: a phoneme, an allophone, a phone. The phoneme is viewed as the smallest indivisible phonetic unit, it is the smallest contrastive unit, a unity of co-existing distinctive features, e.g. [ted] — [ded]. As you know, an allophone preserves all the distinctive features of a phoneme but acquires additional non-distinctive contextual features, which are predictable as they are received in various positions of sounds in connected speech in different phonetic context, e.g [ded] — [dred], [ten] — ['i:tn], [keık] — [skeıt]. They are universal since they are obligatory for all people of English-speaking community in any form of presentation and style. An allophone differs from a phoneme in the degree of abstraction.

Still allophonic variations do not cover all segmental modifications observed in a speech discourse. There exist sound modifications due to extralinguistic factors of communication. For instance, in spontaneous speech of relatives or intimate friends the meaning of the phrase "I think so" is absolutely clear to the listener. He perceives it as [aı 'Ɵıŋk sɜu] though the speaker actual

110

ly pronounces [ʌ 'Ɵıŋk sɜ]. The [ʌ] realizition here is a stylistic variant of the phoneme [aı].

The analysis of stylistic modifications of sounds allows us to speak of an intermediate stage between an allophone and a phone (a sound realization) that is a variant which implies allophonic and stylistic variations of sounds. The term "variant" suggests the realization of stylistic features which are universal for every speaker in any type of discourse.

We cannot deny that every actual sound realization is a unique and individual ideophone. Apart from the distinctive, contextual and stylistic features it differs in the timbre and personal voice qualities of every speaker which make his speech recognizable though we may not see the speaker but only hear him over the radio or in a telephone talk. Thus the sound realizations of phonemes are marked by personal features in addition to distinctive, contextual and stylistic. In the most general way the relationship between these phonetic units may be illustrated in this scheme.

Phoneme →	Allophone →	Variant →	Phone
Distinctive	Distinctive	Distinctive	Distinctive
features	features	features	features
	Contextual	Contextual	Contextual
	features	features	features
		Stylistic	Stylistic
		features	features
			Personal
			features

So, a phoneme, an allophone, a variant and a phone form a kind of hierarchy of phonetic units in discourse.

As was mentioned at the start of this section, the degree of formality or in other words the character of relationship between participants of the discourse proves to be most significant in the stylistic modifications of sounds. The role of extralinguistic factors as style forming ones seems to be different on segmental and suprasegmental levels and wants further investigation.

111

⇐ Предыдущая 2 3 4 5 678 9 10 11 Следующая ⇒

Последнее изменение этой страницы: 2019-03-22; Просмотров: 1301; Нарушение авторского права страницы