Experimental phonetics

/p - tl - pool /pu:l/ - tool /tu:l/,

/1 — s/ — let /let/ - set /set/

/г — 1/ — right /rait/ - light /lait/, etc.

The words used to illustrate the phonological oppositions are known as minimal pairs of words or guasyhomonyms (the term suggested by L. V. Shcherba)

There is a classification of phonological oppositions according to the relationship between the oppositions, between the members of oppositions and the force of oppositions'. This type of classification of phonological oppositions is based on logic and linguistic categorization of phonetic data. Besides there is a principle of preliminary phonological analysis suggested by V.A. Vassilyev2. According to this principle all the phonological oppositions are classified into two-member oppositions but the main difference between the members of opposition is based on the number of the distinctive features: single - when there is one distinctive feature (e. g. /p - 1/), double - when there are two distinctive features (e. g. /p - d/) and complex when there are more than two distinctive features (e. g. /p - if). There are also non-phonological oppositions which cannot serve to distinguish words. For example, the difference between aspirated /pA, t , кА/ and non-aspirated /p, t, к/ sounds is non-phonological. The feature aspirated - non aspirated is non-distinctive or phonologically irrelevant in modem English. A feature which is distinctive in one language may be non-distinctive in another. The concept of a distinctive feature is important to analize the character of phonological oppositions. The articulatory, acoustic and perceptual feature which can distinguish two phonemes is called a distinctive feature. The feature which cannot serve this purpose is known as a non-distinctive feature.Every phoneme in relation to the other phoneme may be characterized by this distinctive and non-distinctive feature. Thus, a phoneme is a bundle of distinctive features. In relation to the phoneme the same phoneme's allophones have non-distinctive feature such as the relation between /pV aspirated and /р/ non-aspirated may be characterized by a non-distinctive feature. But the common features of /р/ and /рл/ generalize the phoneme /р/ which is bilabial, plosive-occlusive, voiceless3. As a linguistic unit the phoneme functions to distinguish lexical and grammatical forms and in this way performs its communicative function in a language. Every phoneme with its allophones is a member of a phonological opposition. The exact number of the phonemes also called the inventory of the phoneme which exists in a certain language is established by using the method of commutation. This method is defined as substitution or replacing one speech sound by another in the same position of minimal pairs o f words. For example: pet- bet - set - let - met - net - jet - get - vet.

Sometimes it is very difficult to discover minimal pairs. As in the case of /tj/, /d3/, /0/, /3/ etc. it is possible to find a few minimal pairs but no minimal pair exists for the opposition / J / —

/3/. But if there is no minimal pair in the language we must not omit the phoneme from the inventory. Such as in Russian and borrowed from it in Uzbek the phoneme /ts/ cannot be used in minimal pairs but we describe it as a special phoneme which has limited distribution. By the term distribution we mean all the positions and combinations in which a certain speech sound - a representative of the phoneme is used. There are four types of distribution.

1. If two elements cannot be used in the same position and replace each other in one position they are considered to be in a complementary distribution. For example, aspirated /p \ t \ khl sounds can be used only before stressed vowels if they are not preceded by /s/ and in the intervocalic position. But in all other positions non-aspirated /p, t, к/ sounds are used. Thus /p \ t \ khl sounds cannot replace /p, t, к/ sounds in the same position. They represent the allophones of the /p, t, к/ phonemes. It is possible to establish the allophones of the phoneme using complementary distribution.

2. Two elements (sounds) may be used in one and the same position and serve to distinguish the words. For example, bill /bil/ - till /til/, sight /sait/ - bight /bait/ - night /nait/ - right /rait/ -light /lait/ - might /mait/ etc. Using contrast distribution it is possible to establish the number of phonemes in a given language.

3. The elements (sounds) used in one and the same position and which cannot distinguish the meanings of words are considered to be in free variation. In such cases every sound manifests the free allo-phone of the phoneme. This type o f distribution is known also as an equivalent distribution. For example, some speakers pronounce /e/ sound either half-close /e/ or half-open in one and the same position but it cannot distinguish the words.

4. Two various sounds may be used in one and the same position. In such cases one of the sounds represents the free allophone of the other. For example, the word phonetics may be pronounced as /fonetiks/, /founetiks/ and /fanetiks/ where the sounds h i, h i represent the free allophones of the phoneme /ou/. In reality each of them is an allophone of the separate phoneme. The distributional method is very important in phonological analysis of the sound structure. It is necessary to show also what clusters of sounds the pattern of a language admits. The branch of phonology which studies the possible clusters of sounds in words and morphemes is known as «phonotactics»1. Using the statistical method it is possible to establish the exact number of phonological oppositions in a language and the number of sound clusters in initial, medial and final positions of
the words. For example, in English, out of a theoretically possible 11,000 initial three member consonantal clusters at the beginning of a syllable, only about 40 occur. Of 576 possible combinations of two consonants, only 137 are utilized by the language2. There are no initial three member consonantal clusters in Uzbek. Thus it is difficult to teach the Uzbek students the pronunciation of the initial three member consonantal clusters of English.