Lecture 1 Phonetics as a Linguistic Science Plan

Bloomfieldian descriptive phonology is also called the relative-acoustic theory, as it is based on the analyses of structural functions and acoustic features of phonemes. According to L.Bloomfield, a phoneme is a minimal distinctive unit of a language, which has no meaning itself but may be determined as a special unit, owing to its physical and structural contrasts in relation to all other sounds types of a particular language. His other definition of the phoneme as a minimal unit of the phonetic feature is purely a phonetic one. He sometimes mixed up the notions of a "speech sound" and a "phoneme". His idea on the primary and secondary phonemes was very important in the further classification of segmental and suprasegmental phonemes. He also gave descriptions of the phoneme combinations in initial, medial and final positions of the words.
L.Bloomfield’s theory was developed and improved by a number of linguists and is called the post-Bloomfieldian theory of descriptive phonology. The representatives of this are Z.Harris, Ch.F.Hockett, H.A.Gleason. According to this theory a phoneme is a class of sound or a class of allophones (phones) which have both phonetic similarity and functional identity, in the sense that the substitution of one for another in the same context does not change its syntactic or semantic function, i.e. makes no change in its meaning. This theory defines a phoneme on the basis of the distributional method. Usually the phoneme is defined as the repsentative of phones in free variation or complementary distribution, which are phonetically similar. The allophones of phonemes may also be determined on the basis of the distributional method. Some representatives of this trend define a phoneme as a sum of distinctive features. They state the physical and functional aspects of the phoneme from the centralistic point of view, as their theory is based on the stimulus-response segments that are the same or different.
American tagmemic school of linguistics advanced its own phonological theory which differs from the theories of descriptive phonology. According to the tagmemic trend a language is the result of verbal behavior and mind and it consists of three levels: grammatical, lexical and phonological. Each of these levels has its own units: morpheme, tagmeme and phoneme, the latter is a minimum unit of the phonological level.
The phoneme is characterized as composing disjunctive, emic portions of the verbal behavior phonetically represented. A phoneme is not a class of sounds, but a phonetic unit with particular features, which is connected with the units of the other levels. A new unit, which is called a tagmeme, is defined as the implication of a slot or position for a functional meaning and a morpheme. The head of the tagmemic school of American linguistics Kenneth Pike uses the term "archiphoneme" in a different sense that N.S.Trubetzkoy did. Two phonemes, which cannot be identified with the phoneme, is called an archiphoneme. For example, in the English words night-rate and nitrate we may predict the medial unaspirated long /t/ as opposed to the aspirated short/t/. Without using junctures they form an archiphoneme in such morphemes. K.L.Pike states: "Phonemes cannot be analyzed without some knowledge - though it may be very slight of grammatical facts". In his work “Coexistent Phonemic Systems” (1949) K.L.Pike attempts to demonstrate the possibility of two or more phonemic systems in monolingual speech. The tagmemic theory is also based on behavior. The phonological theory, which was suggested by K.L.Pike, is called phonotagmemics according to which all languages have a phoneme level, most have a syllable level, a pause group level and the level between the syllable and pause group. Intonation takes its characteristics on the phonological phrase level. Thus, the relation between the levels of a language, strictly speaking, the sublevels of speech, is very important in phonotagmemics.
The theory which is being developed in modern American linguistics is a generative-transformational phonology (often called a "generative phonology"). Generative phonology is one of components of generative grammar as a syntactic component and component of lexicon. Generative phonology serves to provide phonetic representations of utterances in a language. It studies the phonological form of morphemes and morph listed in the lexicon and determines the rules of how the phonetic units (sounds, syllables, stress and partly intonation) are pronounced in various environments in which they are found. The resulting phonetic representation level provides a transcription of a sound segment used in actual utterances. According to generative phonology distinction between phonemes and allophones requires levels of phonological representation to be recognized: the level of pronunciation (the phonetic level) and the level of contrast or opposition (the phonemic level). As to articulatory and acoustic feature they fulfill three functions: 1) they are capable of describing the systematic phonetics - a phonetic function; 2) at a more abstract level they can differentiate lexical items - a phonemic level; 3) they define natural classes, that is, those segments, which, as a group undergo similar phonological processes. The main aim of generative phonology is to find the rules and answer the following questions: 1) What segments change? How do they change? Under what conditions do they change? This theory cannot be universal as each language requires its specific rules for phonological analysis.
Almost all phonological theories in USA regard variations in phonological form at or across morphological boundaries as the morphophonemics of a language. N.Chomsky and M.Halle suggest the principle of cycle to predict accent elements in their work "The Sound Pattern of English" (N.Y., 1968). Discussions on the problems of adequacy and predicative power in recent phonological theories are still going on among American linguists.