Main trends in phoneme theory

Now that we know what the phoneme is let us view the main trends of the phoneme theory. Most linguists have looked upon the phoneme as one of the basic language units. But not all of them have described it in the same way. The majority of them agree that the phoneme serves to distinguish morphemes and words thus being a functional unit. However, some of them define it in purely "psychological" terms, others prefer physically grounded definitions. Some scholars take into consideration only the abstract aspect of the phoneme, others stick only to its materiality. This has divided the various "schools" of phonology some of which will be discussed below. Views of the phoneme seem to fall into four main classes.

The "mentalistic" or "psychological" view regards the phoneme as an ideal "mental image" or a target at which the speaker aims. He deviates from this ideal sound partly because an identical repetition of a sound is next to impossible and partly because of the influence exerted by neighbouring sounds. According to this conception allophones of the phoneme are varying materializations of it. This view was originated by the founder of the phoneme theory, the Russian linguist I.A.Baudauin de Courtenay (6) and something like it appears to have been adopted by E.D.Sapir. The same point of view was shared by other linguists, Alf. Sommerfelt (76) for one, who described phonemes as "models which speakers seek to reproduce."

The "psychological", or "mentalistic" view of the phoneme was brought back into favour by generative phonology, and the idea of the phoneme as a "target" has recently been revived, albeit under different terminology by M.Tatham (77).

It is definitely not possible to establish such ideal sounds which do not exist in reality. For this reason the American linguist L.Bloomfield (46) and his followers rejected the view and the English phonetician D.Jones (64), while basically favourable to the view preferred in practice to take a "physical" view. This approach to the phoneme as a clearly idealistic one cannot be taken up by Soviet linguists.

The so-called "functional" view regards the phoneme as the minimal sound unit by which meanings may be differentiated without much regard to actually pronounced speech sounds. Meaning differentiation is taken to be a defining characteristic of phonemes. Thus the absence of palatalization in [I] and palatalization of [1] in English do not differentiate meanings, and therefore [t] and [1] cannot be assigned to different phonemes but both form allophones of the phoneme [1]. The same articulatory features of the Russian [л] and [л' ] do differentiate meanings, and hence [л] and [л¹] must be assigned to different phonemes in Russian, cf. мол — моль, лог — лёг. According to this conception the phoneme is not a family of sounds, since in every sound only a certain number of the articulatory features, that is those which form the invariant of the phoneme, are involved in the differentiation of meanings. It is the so-called distinctive features of the sound which make up the phoneme corresponding to it. For example, every sound of the English word ladder includes the phonetic feature of lenisness but this feature is distinctive only in the third sound [d], its absence here would give rise to a different word latter, whereas if any other sound becomes fortis the result is merely a peculiar version of ladder. The distinctiveness of such a feature thus depends on the contrast between it and other possible features belonging to the same set, that is the state of the vocal cords. Thus when the above-mentioned features are distinctive, lenisness contrasts with fortisness. Some approaches have taken these oppositions as the basic elements of phonological structure rather than the phonemes in the way the phoneme was defined above. The functional approach extracts non-distinctive features from the phonemes thus divorcing the phoneme from actually pronounced speech sounds. This view is shared by many foreign linguists: see in particular the works of N.Trubet-skoy (34), L.Bloomfield (46), R.Jakobson (62), M.Halle (62).

The functional view of the phoneme gave rise to a branch of linguistics called "phonology" or "phonemics" which is concerned with relationships between contrasting sounds in a language. Its special interest lies in establishing the system of distinctive features of the language concerned. Phonetics is limited in this case with the precise description of acoustic and physiological aspects of physical sounds without any concern to their linguistic function. The supporters of this conception even recommend to extract phonetics from linguistic disciplines which certainly cannot be accepted by Soviet phoneticians.

A stronger form of the "functional" approach is advocated in the so-called "abstract" view of the phoneme, which regards phonemes as essentially independent of the acoustic and physiological properties associated with them, that is of speech sounds. This view of the phoneme was pioneered by L.Hjelmslev and his associates in the Copenhagen Linguistic Circle, H.J.Uldall and K.Togby.

The views of the phoneme discussed above can be qualified as idealistic since all of them regard the phoneme as an abstract conception existing in the mind but not in the reality, that is in human speech, speech sounds being only phonetic manifestations of these conceptions.

The "physical" view regards the phoneme as a "family" of related sounds satisfying certain conditions, notably:

1. The various members of the "family" must show phonetic similarity to one another, in other words be related in character.

2. No member of the "family" may occur in the same phonetic context as any other.

The extreme form of the "physical" conception as propounded by D.Jones (64) and shared by B.Bloch and G.Trager (45) excludes all reference to non-articulatory criteria in the grouping of sounds into phonemes. And yet it is not easy to see how sounds could be assigned to the same phoneme on any other grounds than that substitution of one sound for the other does not give rise to different words and different meaning. This approach may seem to be vulgarly materialistic since it views the phoneme as a group of articulatorily similar sounds without any regard to its functional and abstract aspects.

Summarizing we may state that the materialistic conception of the phoneme first put forward by L.V.Shcherba may be regarded as the most suitable for the purpose of teaching.

METHODS OF PHONOLOGICAL ANALYSIS

Now that you have a good idea of what a phoneme is we will try to show what the aim of phonological analysis of speech sounds is; what methods can be applied to this sort of analysis and what characteristics of the quality of sounds are of primary importance in grouping them into functionally similar classes, that is phonemes.

As was stated in the introduction to the course, any phonetician should look upon his science primarily as a study of the spoken form of the language. To study the sounds of a language from the functional or phonological point of view means to study the way they function, that is, to find out which sounds a language uses as part of its pronunciation system, how sounds are grouped into functionally similar units, termed phonemes. The final aim of the phonological analysis of language is the identification of the phonemes and finding out the patterns of relationship into which they fall as the sound system of that language.

Here we should note that there are two ways of analyzing speech sounds. For example, if we define [s] from the phonological point of view it would be constrictive, forelingual, fortis. This would be quite enough to remind us of the general class of realization of this segment. For articulatory description we would need much more information, that is: what sort ofc narrowing is formed by the tip of the tongue and the alveolar ridge, what is the shape of the tongue when the obstruction is made (a groove in the centre of the tongue while the sides form a closure with the alveolar ridge) and so on. So if the speech sounds are studied from the articulatory point of view it is the differences and similarities of their production that are in the focus of attention, whereas phonological approach suggests studying the sound system which is actually a set of relationships and oppositions which have functional value.

It is common knowledge that different languages have a different number of phonemes and different allophones representing them. So each language has its own system of phonemes. Each member of the system is determinated by all the other members and does not exist without them. The social value of articulatory and acoustic qualities of sounds for the language as a means of communication is different in different languages. In one language community two physically different units are identified as "the same" sound, because they have similar functions in. the language system. In another language community they may be classified as different because they perform different linguistic functions. This statement can be illustrated by the following example. Despite their articulatory difference the two English [1] and [t] sounds (clear and dark) are identified by English people as one phoneme because the articulatory difference between them does not affect the meaning. We shall most probably find that the native speakers are not normally aware of the difference between these two sounds. This would come as a surprise to na tive speakers that they have difference in English. English speakers do not hear the difference because it is of no importance in the communication process.

In the Russian language a similar, though not identical difference between [л¹] and [л] affects the meaning. If [л] in лук is replaced by [л¹] the result will be a different word люк. So the Russian [л] and [л¹] sounds different on the articulatory level are identified by speakers of Russian as two different phonemes. Analogically, the speakers of Syrian notice a difference between the [t^h] of English ten and the [t] of letter, a difference which is phonemic in Syrian but only allophonic in English.

There are many other differences which are unimportant on the phonological level of analysis. For example, the realization of the [p] phoneme in the words pie, spy, lamp-post. They are all different because of the phonetic context in which they occur: in the word spy the sound [p] loses its aspiration, in the word lamppost the first sound [p] is replaced by a glottal stop. But phono-logically these sounds are the same. Thus a very important conclusion follows: where languages are concerned everything is relative and statements concerning phonological categories and allophonic variants can usually be made of one variety of a particular language.

So the aim of the phonological analysis is, firstly, to determine which differences of sounds are phonemic and which are non-phonemic and, secondly, to find the inventory of the phonemes of this or that language.

It should be noted that a number of principles have been established for ascertaining the phonemic structure of a language. For an unknown language the procedure of identifying the phonemes of a language as the smallest language units has several stages. The first step is to determine the minimum recurrent segments (segmentation of speech continuum) and to record them graphically by means of allophonic transcription. To do this an analyst gathers a number of sound sequences with different meanings and compares them. For example, the comparison of [stik] and [staek] reveals the segments (sounds) [i] and [as], comparison of [stik] and [spik] reveals the segments [st] and [sp] and the further comparison of these two with [tik] and [task], [sik] and [sаek] splits these segments into smaller segments [s], [t], [p]. If we try to divide them further there is no comparison that allows us to divide [s] or [t] or [p] into two, and we have therefore arrived at the minimal segments. From what we have shown it follows that it is possible to single out the minimal segments opposing them to one another in the same phonetic context or, in other words, in sequences which differ in one element only.

The next step in the procedure is the arranging of sounds into functionally similar groups. We do not know yet what sounds are contrastive in this language and what sounds are merely allophones of one and the same phoneme. There are two most widely used methods of finding it out. They are the distributional method and the semantic method. The distributional method is mainly used by phoneticians of "structuralist" persuasions¹: the fact is that the structuralist model of languages which flourished from the 1930s to the 1950s emphasized "the facts about the language" approach. In fact, these phoneticians underestimated the distinctive function of the phoneme. They consider it possible to discover the phonemes of a language by the rigid application of distributional method, that is to group all the sounds pronounced by native speakers into phonemes according to the two laws of phonemic and 'allophonic distribution. These laws were discovered long ago and are as follows.

1. Allophones of different phonemes occur in the same phonetic context.

2. Allophones of the same phoneme never occur in the same phonetic context.

The fact is that the sounds of a language combine according to a certain pattern characteristic of this language.'Phonemic op-posability depends on the way the phonemes are distributed in their occurrence. That means that in any language certain sounds do not occur in certain positions. For example, [h] in English never occurs word finally, while [rj] never occurs word initially, or [t^h] never occurs word finally or never follows [s] while [t] (unaspirated) never occurs word initially before stressed vowels. Such characteristics permit identification of phonemes on the grounds of their distribution. For example, if a sound occurs in 9 certain phonetic context and another one occurs only in a different context no two words of a language can be distinguished solely by means of the opposition between these two. The two sets of phonetic contexts are complementing one another and the two sounds are classed as allophones of the same pho neme. They are said to be in complementary distribution. For example, if we fully palatalize [1] in the word let it may sound.peculiar to native speakers but the word is still recognized as let but not pet or bet. The allophones lack distinctive power because they never occur in the same phonetic context and the difference in their articulation depends on different phonetic environment, that is why the two phonetically different sounds are perceived as identical ones. To be able to distinguish the meaning the same sounds must be capable of occurring in exactly the same environment as [p] and [b] in pit and bit or [I] and [d] in lay and day. Thus two conclusions follow:

1) If more or less different sounds occur in the same phonetic context they should be allophones of different phonemes. In this case their distribution is contrastive.

2) If more or less similar speech sounds occur in different po sitions and never occur in the same phonetic context they are allophones of one and the same phoneme. In this case their distribution is complementary.

We would point out here that the identification of phonemes on the ground of complementary distribution is not without its problems. There are cases when two sounds are in complementary distribution but are not referred to the same phoneme. This is the case with the English [h] and [rj]. [hj occurs only initially or before a vowel while [rj] occurs only medially or finally after a vowel and never occurs initially. In such case the method of distribution is modified by addition of the criterion of phonetic similarity/dissimilarity. The decisions are not made purely on distributional grounds. Articulatory features are taken into account as well.

As we know allophones are supposed to share distinctive features. But the problem is that there is no obvious criterion for deciding whether the two sounds are sufficiently similar or not; how similar they must be in order to qualify them as a single phoneme. Alternatively, how different must they be before we say the sound X and the sound Y cannot, be possibly allophones of the same phoneme. Let us consider sounds which are similar on the articulatory level in the words.

[k^h æ p^h] — aspirated [p] — [k^h æ p^h]

[k^h æ p^h] — unreleased voiceless (loss of plosion) — [k^h æ?]

[k^h æ b°] — unreleased voiced (loss of plosion) — [k^h æ? ]

Here the articulatory difference between [p^h] and [p] is greater than between [p] and [b]. Yet the native speaker associates the former pair as allophones of [p] but distinguishes the latter pair as two phonemes [p] and [b] (cop — cob). To understand what is happening here we must briefly examine speech as a communicative process and we shall see that phonetic similarity can be viewed from two perspectives: one concerning the speaker, the other the listener. For example, [p^h] and [p] are treated by the native speaker as allophones of the same phoneme though both on the auditory and the articulatory levels they are different.

Phonetic similarity can be determined either on the auditory or on articulatory basis. Looking again at the case of [h] and [n] m English we can see now that these sounds are dissimilar both in articulation and in perception, and therefore have no basis at all for being treated as phonetically similar. That makes it very unlikely that they could be members of the same phoneme, though they never occur in the same phonetic context.

So far we have considered cases when the distribution of sounds was either contrastive or complementary. There is, however, a third possibility, namely, that the sounds both occur in a language but the speakers are inconsistent in the way they use them, as for example in the case of the Russian шкаф — шкап, калоши — галоши. In such cases we must take them as free variants of a single phoneme. But since the situation seems somewhat unusual we would take some trouble to find the reason for the variation in the realization of the same phoneme. We could explain it on the basis of "dialect" or on the basis of sociolinguis-tics. It could be that one variant is a "prestige" form which the speaker uses when he is constantly "monitoring" what he says while the other variant of pronunciation is found in casual or less formal speech. If all explanations fail then we have truly free variants, but this statement is in a sense of admission that the critical factors at work have not yet been explained.

It would be unfair to imply that phoneticians of "structuralist" persuasions take no account of native speakers' opinion. When purely distributional approach failed they considered native speakers' opinion to know if the analyzed words differed in meaning (the so-called differential meaning). So in practice they usually admitted consulting a native speaker. But in their anxiety to emphasize the need for proper (scientific) method and procedures some of them come to regard the procedures as ends in themselves, rather than a means towards understanding language.

There is another method of phonological analysis widely used in Soviet linguistics. It is called the semantic method. It is applied for phonological analysis of both unknown languages and languages already described. In case of the latter it is used to determine the phonemic status of sounds which are not easily identified from phonological point of view. The method is based on a phonemic rule that phonemes can distinguish words and morphemes when opposed to one another. The semantic method of identifying the phonemes of a language attaches great significance to meaning. It consists in systematic substitution of the sound for another in order to ascertain in which cases where the phonetic context remains the same such substitution leads to a change of meaning. It is with the help of an informant that the change of meaning is stated. This procedure is called the commutation test. It consists in finding minimal pairs¹ of words and their grammatical forms. For example, an analyst arrives at the sequence [pm]. He substitutes the sound [p] for the sound [b]. The substitution leads to the change of meaning. This would be a strong evidence that [p] and [b] can be regarded as allophones of different phonemes. Minimal pairs are useful for establishing quickly and simply the phonemes of the language. If we continue to substitute [p] for [s], [d], [w] we get minimal pairs of words with different meaning sin, din, win. So, [s], [d], [w] are allophones of different phonemes. But suppose we substitute [p^h] for [p] the pronunciation of the word would be wrong from the point of view of English pronunciation norm, but the word would still be recognized as pin but not anything else. So we may conclude that the unaspirated [p] is an allophone of the same [p]-phoneme.

As was mentioned earlier, the phonemes of a language form a system of oppositions in which any phoneme is usually opposed to other phonemes in at least one position, in at least one minimal pair. So to establish the phonemic structure of a language it is necessary to establish the whole system of oppositions. All the sounds should be opposed in word-initial, word-medial and word-final positions. There are three kinds of oppositions. If members of the opposition differ in one feature the opposition is said to be single, e.g. pen — ben. Common features: occlusive — occlusive, labial — labial. Differentiating feature: fortis — lenis.

If two distinctive features are marked, the opposition is said to be double, e.g. pen — den. Common features: occlusive — occlu-sive. Differentiating features: labial — lingual, fortis voiceless — lenis voiced.

If three distinctive features are marked the opposition is said to be triple, e.g. pen — then. Differentiating features: occlusive — constrictive, labial — dental, fortis voiceless — lenis voiced.

We should remind you here that the features of a phoneme that are capable of differentiating the meaning are termed as relevant or distinctive. The features that do not take part in differentiating the meaning are termed as irrelevant or non-distinctive. The latter may be of two kinds: a) incidental or redundant features, for example, aspiration of voiceless plosives, presence of voice in voiced consonants, length of vowels: b) indispensable or concomitant features, for example, tenseness of English long monophthongs, the checked character of stressed short vowels, lip rounding of back vowels. It is well to remember that a single opposition remains single if its members differ from each other not only in a distinctive feature alone, but also in distinctively irrelevant both incidental and concomitant features.

It would be a mistake to assume that the use of commutation test is without its problems. The difficulty is that the theory is based on the assumption that sounds in any sequence are discrete replaceable units. But a phonemic analysis needs to take account of the overall pattern of sounds and their structural relationship in the language concerned: phonemes are not isolated, unrelated phenomena in the same environments. In reality speech sounds are modified under, the influence of context in which they are used. For example, in such sequences as [aekt] and [aept] the realization of the allophones of [k] and [p] are alike: acoustically on the perception level it is silence of the same duration. Although the two phonemes are in mutual opposition they are not capable of distinguishing the meaning of these two words in a straightforward way. It is the character of transition from [ae] to [k] in the first case, and from [as] to [p] in the second, that differentiates the meaning. The two opposed phonemes fulfil their distinctive function in an indirect way, that is the differentiation of meaning is provided not by the phonemes themselves but by the adjacent sound. Another example: [bed] — [bet]. In this pair of words it is mainly the length of the proceeding vowel [e] that provides the differentiating characteristic of the word because normally distinctive opposition between voiced and voiceless consonants is neutralized.

Semantic method of phonological analysis is now widely used in Soviet linguistics as well as by overwhelming majority of foreign analysts.

We have indicated so far that the phonological analysis of the sounds of a language is based on such notions as contrastive distribution, complementary distribution, minimal pairs, free variation, phonetic similarity. To these we must add one more concept, that is of native speaker's knowledge. The fact is that all the rules referred to above should account for the intuitions of the native speaker and that is the real reason why we adopt them, for the aim of linguistic analysis is to explain and to take account of native speaker's feelings about his language as far as this is possible. For example, that [n] and [h] are in contrastive distribution is a fact about the language but it would be of little interest in itself and that is not the real reason why we treat these two sounds as separate phonemes. The real reason is that the native speaker feels they are different phonemes.

In the final summing up we might say that the phonemic system of a language is patterned. It is the aim of phonological analysis to attempt to systematize the sounds of a language, that is to group them into functionally similar classes. It is of primary importance for learning and teaching a foreiqn language because the more consistent, logical and concise the description of the phonetic system of a language is the more effective its acquisition will be.

Дата добавления: 2015-05-31