Lecture 2-4 (Phon. Schools)
Lecture 2-4 (Phon. Schools)
Lecture 2-4 (Phon. Schools)
LECTURE 2-4
PHONOLOGICAL SCHOOLS
1) Phonological Schools in Russia
(a) Bodouin de Courtenay
(b) L. V. Scerba
(c) The Leningrad Phonological School (LPS)
(d) The Moscow Phonological School (MPS)
(e) Compromise Phonological Theories
(f) 2-step Phonological Theory
2) Phonological Schools Abroad
(a) Ferdinand de Saussur
(b) Leonard Bloomfield
(c) Edward Sapir
(d) Generative Phonology
(e) Structuralist Phonological schools
1) Glossematics
2) the Prague School
3) Descriptive Linguistics
(f) London Phonological School
5. The scholars of LPS distinguish 4 functions that the phoneme performs in the
language: constitutive, identificatory, delimitative and distinctive, the last being of
primary importance and resulting from the fact that the phonemes differ and each of
them is opposed to all other phonemes of the language.
6. Though the phonemes as minimal segmental units cannot be further split linearly,
they can be opposed to each other because they possess certain inner properties –
features. The LPS phonologists distinguish between relevant features (significant for
opposing phonemes) and irrelevant features.
MPS sprang up as a result of the research of the group of Moscow linguists which
includes such names as N.F. Yakovlev, P.S.Kuznetsov, A/M/Sukhotin,
A.A.Reformatsky, R. I. Avanesov, V.N.Sidorov and others. The scholars who entered
the group are talented and creative researchers, eventually the phonological theories of
many of them fanned out and developed independent concepts, still basically they all
can be attributed to MPS on such principal issues as the morphological criterion
towards phoneme understanding, more or less identical interpretation of certain
alternations, envisaging a new type of phonological analysis which includes 3 stages
instead of traditional two.
1. The terms "phonetics" and " phonology" in MPS acquire special implication:
"phonetics" tackles both physiological and some functional peculiarities of sound units,
i.e. it covers the issues which are traditionally distributed between phonetics and
5
phonology, whereas the MPS "phonology" considers exclusively alternating speech
sounds which perform the morphological function.
3. Which features are important for opposing phonemes and which are not is
determined when phonemes are compared in equal conditions, e. i. in the same position.
which is the basic in phonology. According to the character of realization the positions
can be strong and weak. The phoneme is in strong position when it is least conditioned
by the neighbouring phonemes. It is in the weak position when the adjacent phonemes
affect its characteristics. In the strong positions phonemes appear. In the weak –
variants and variations.
4. Among the oppositions, probably, the most interesting are the cases of
neutralization which is "the coincidence of 2 phonemes in one variant" (Reformatsky),
compare: коз – кос /кос/; сома - сама /с'ма/. In English it can be illustrated by
the following: in a phonologically neutral position a speech sound (allophone), as part
of the morpheme, belongs to the phoneme which, in the same morpheme, stands in the
strong position. ACT / k t / - ACTIVITY / k ' tiviti/ - According to MPS the neutral
vowel / / is not the allophone of the phoneme // but the allophone of the
phoneme //, because the allophone // is part of the same morpheme (stem) : act –
activity.
There were several attempts to bridge a gap between LPS and MPS
1. As, for instance, R.I. Avanesov states that the two viewpoints do not contradict
each other and are quite compatible. The shortest sound units are defined by him as "
6
minimal articulatory-auditory elements distinguished while segmenting speech linearly.
Those properties are to some extent independent and at the same time dependent on
their position in the word.
2. M.V. Panov put forward the theory of two phonetics: the Phonetics of
Combinations and the Phonetics of Alternations which have different laws. The laws of
combinations are studied in Syntagmatic Phonetics, while the laws of alternations – in
the Paradigmatic Phonetics. Panov restricts the influence of morpheme upon phoneme
to the paradigmatic phonology.
- Taking into account the above said, Shaumyan assumes the " the correct
description of phonological reality must not include anything other than the systematic
description of abstract operations over symbols… This is the essence of structural
description and " it lies in the basis of the 2-step phonological theory.
7
We must state here that the development of the Phonemic theory or Phonology
abroad has been progressing in a way, different from its development in Russia. The
foreign and Russian scholars quite independently have been working over the notion of
phoneme and its place in the system of language. The most prominent research was
conducted by foreign linguists in methods of distinguishing and describing phonemes.
But it s not an easy matter to devise operational tests for establishing such ideal
sounds, and in his practical analysis Sapir trans'cended (превосходить) the mentalistic view.
The significance of Sapir's theory lies not so much in the psychological interpretation of
the phoneme, but the separation of the two notions: the sound and the phoneme.
The approach is also called Transformational Grammar which has sprung up as the
reaction to / opposition to the preceding traditional or, according to Chomsky,
'taxo'nomic phonemics which is defined as based exclusively on segmentation, on
classification and on observable adequacy ( though when criticizing taxonomic
approaches Chomsky actually means the American descriptive structural theory).
Instead, the scholar seeks to introduce a theory which would belong to a higher
theoretical level and would permit to build more abstract counterparts (аналоги) for the
generalizations that are derived from the actual sound signals or actual articulatory
movements.
Most generative treatments have taken some form of phonetic distinctive features as
their basic elements. They are used not only for the phonetic representation of a
sentence ( a specification of its pronunciation) but also at the phonemic level to give the
sentence what is usually called its phonological representation, which has no
necessary link with actual pronunciation. The link between the two representations –
phonological and phonetic – is established by a set of 'rules' which operate in a fixed
order, adding, deleting or modifying distinctive features. The rules are necessary in
order to generate potential language realities, pertaining to the inner structure, into
actual language realities, pertaining to the surface structure. According to Generative
approach, both Phonological and phonetic level possess linguistic status. Phonology
10
reflects the language reality allegedly existing in the native speaker's mind
(unobservable by any method). The speaker turns this language reality into other
language reality which lies on the surface (directly observable) and is called a Phonetic
level. The Phonetic level further splits into phonetic level proper and physical or
physiological level.
The Chomsky and Halle model is expounded in more detail by such scholars as
Langacker, Fodor, Katz, Harms etc.
Accordingly, the sciences dealing with phonological structure of the language are
divided into Phonetics – the science tackling the sound matter on the basis of material-
substance realities -; and Phonematics (cenematics) – the science about the structural
form of the language where material-substance reality is irrelevant. Phonology is
regarded not as the analysis of the language structure but as a disguised method
'covertly (завуалированный) based on the analysis of the real sound matter like the approach
of Daniel Jones.
The activity of the Prague Phonological Circle is connected with the names of such
linguists as R. O. Jacobson, S. O Kartsevsky, B. Matesius, B. Tranka, J. Vahek and
others. But the most prominent among them, who can be considered the founder and the
head of the school, is Nikolay Sergeevich (Prince Nicolas) TRUBETSKOY. The Prague
approach is based on the interpretation and development of the theory put forward by
Ferdinand de Saussur (though in a different direction than in the Danish glossematics),
besides the influence of Baudouin de Courtenay's ideas on the approach seems quite
profound.
- The linguistic conception of the Prague school can be characterized as having two
specifications. First and foremost, it is a structural approach : the problem of the
structural character of the language and the interdependence of its elements is
introduced into linguistics. And, second, the approach is functional. Language is
declared a functional system in which sounds are distinguished both as objective
12
physical facts and as elements of a functional system. It must be noted here that the
notion of "FUNCTION" is regarded by the Prague phonologists in a different way
than it is understood by glossematicians: if for L. Hjelmslev "function" is certain
dependence and the network of relations, by the Prague school it is defined as a goal,
an objective. In a narrower meaning, i.e. applied for the phonological aspect of the
language, the term "function" is employed in the sense of "differentiating meaning",
"distinction" of meaningful language units. Parallelly, N.S. Trubetskoy distinguishes
delimitative and culminative functions.
-In certain phonetic condition - the position of neutralization one variant can manifest
two phonemes which are perceived as different in other positions. To explain the fact
N.A. Trubetskoy introduces the notion of archi-phoneme /'a:ki'founi:m/ which is a specific
phonological unit and which is understood as a complex of relevant features common
for both "neutralized" phonemes, being, in fact, the relations between the phonemes.
EX. In Russian, consonantal pairs in final positions are neutralized and pronounced
similarly: "суп" –"суб-" /п/- the consonants /п/ and /б/ share the following relevant
features: labiality, occlusiveness, hardness – accordingly, the archi-phoneme,
representing them can be described as a "labial, occlusive, hard" consonant, the
opposition of the features "voiced – voiceless" being neutralized. Thus the variants п
and б will represent not the corresponding phonemes |п/ and /б/, but the archi-
phoneme /п-б/. The notion of the archi-phonemein the Prague School is close to the
notion of hyperphoneme of MPS with the difference that according to Trubetskoy, in the
end of the word /stok/ there will always stand the phoneme /k/ regardless of whether it
ends the word "сток"or "стог".
3) Descriptive Linguistics
The Descriptive Linguistics is the leading direction in investigating Linguistics in
general and Phonetics/Phonology in particular in the USA. It arose as the development
of concepts and theories put forward by outstanding American linguists L. Bloomfield
and E. Sapir and employs practically all their basic notions and methods of analysis.
Very often L. Bloomfield is called an 'ante'cessor or pre'cursor (предтеча) of American
Structuralism and the followers of Descriptive Linguistics are correspondingly referred
to as .
- Sometimes the contemporary linguistic analysts find it difficult to attribute all the
followers of the Descriptive Linguistics to one and the same school on the grounds that
the notions and method applications, employed by the scholars, differ in certain details
and treatment. Some Russian linguists (Арутюнова, Климов, Кубрякова, 1964, с.191) class them
into two groups: the Yale Group (Z.Harris, L.Trager, Bloch, Ch. Hockett, G.Glison)
and the Ann-Arbor Group (K.Pike, Nida, Friese, Wonderly). But we would rather share
the opinion of G.V. Voronkova who maintains that "the differences are not as striking
as they first appear. All the variations in description are united by one
method, in which the factor of meaning is less significant than the factor of 'sequent
(последовательный) description and classification of phonemes and their elements." ( "Problems
of Phonology", p.86) All the followers of Descriptive Phonology do not only employ a similar
technique of phonological analysis but they similarly interpret the organization of the
phonological system of the language. Their method is conditioned by approaching the
phonological system as the procedure of establishing the units of the system - from
stating the material-substance facts to linguistic facts on a higher level of the
hierarchic step of the language.
- The distinction between language and speech in this school is not clear-
cut due to differences in analyzing procedures and the ambiguous relations between the
notions of "phoneme" and "sound". More often than not the descriptivists claim that
only speech belongs to linguistics while language is rather referred to semiotic
structures and is treated as one of semiotic codes. Language exists in human minds, not
in reality, and is shared by all the members of the language community; it is social,
dynamic and exists historically (diachronically) – thus, it is a structural system which is
not used in immediate human communication – when people communicate they
employ speech, so speech acts, but not language acts, should be studied as linguistic
phenomena. Adhering to the principles of behaviourism, descriptivists avoided all but
empirical analysis, stating that our only information about the linguistic process of mind
is derived from the behavior of an individual. As speech is always spontaneous and
individual, linguists should study its static characteristics in their syntagmatic relations
on the synchronic level .
15
- It should be noted that American structuralists were somewhat inconsistent in
providing a clear-cut basis for distinguishing between the notions of "phoneme"
and " sound" . Very often these two notions are confounded (смешивать) even
within the works of one author. Such indistinct opposition of phoneme to sound,
absence of any stable position as to how phonemes function in language and speech and
withdrawing these basic concepts from the phonological analysis led to developing
phoneme definitions that are not quite compatible. (G.V.Voronkova,p.22). The differences can
be partly due to the terminological character, partly to the practical aims of analysis.
Generally , the phoneme is understood by descriptivist scholars as the sound which
differentiates the meaning, i.e. a distinctive sound which differentiates between the
meaning of acts of speech (not language!) EX., Hockett writes that, in the phrases
"That's a nice pin" –"That's a nice bin", the sounds (!) 'p' and 'b' differentiate between
acts of speech (Hockett,1955,p.16) "Phoneme is a class of sounds pronounced in a given
dialect" (Bloch,1948,p.5) "Phoneme is a sound (p.67) …but at the same time a meaningful
unit (Pike,1947,p.63) but further, "Phoneme is not a language entity but a result of
phonological analysis" (Pike,1947,p.74) Thus, treating the phoneme as a sound, the
descriptivists consistently remain within the syntagmatic analysis referring both to
syntagmatic level and assigning the distinctive function not to the phoneme but to the
sound (which has , tackled traditionally quite different functions in the language). But
the incoherence of the phonologists in establishing the balance and relations between
material and linguistic sides of the phoneme led some descriptivists to denying the
phoneme as a linguistic reality. See, Phoneme is an exclusively logical symbol and
should be analysed with the help of mathematical logics (Harris,1951,p.18); and Phoneme is
an "abstract fictitious unit (Twaddell,1958,p.67). Summarising, we can say that mixing up
two units (phoneme and sound) belonging correspondingly to different spheres (
language and speech) explains the ambivalence of the descriptivist "phoneme-sound" ,
in which the material reality is not separated from linguistic reality (Voronkova,p.24).
1) Speech continuum is investigated and segmented into the minimal sound units, then
the segmented text is notated with the help of transcription which possess special rules
and should be performed thoroughly as it is given primary importance. While
transcribing the text the analyst should avoid some characteristic mistake which include
excessive detailization, insufficient detailization, incorrect segmentation.
2) To verify the correctness of segmentation results, the scholar should check the
distributional models in three steps: picking out probable pairs (with sounds that can
be allophones of one phoneme), putting forward a hypothesis (whether these allophones
belong to one or different phonemes) and verifying the results. For the second step the
distribution of each probable pair, which is conditioned by the adjoining or
neighbouring phonemes, is analysed. The verification consists in charting distribution
tables or schemes for each sound considering all available facts.
The phones, representing one phoneme, are called allophones – they are in
complementary distribution, i.e. do not occur in the same position. The phones,, which
are in free variation and which occur in the same position, are called phonemes.
Applying the substitution method the descriptivists seek not only to segment a phoneme
(which in practice equals a segmented sound) but to determine material-substance
characteristics of phonemes in oppositions (however, in practical analysis they
distinguish only material-substance characteristics of allophone).
One thing should be taken into account – these methods of phonological analysis are
employed with the aim of establishing the phonemic inventory of a foreign language
under research, whereas the identification of the phonemes of the native tongue does
not possess any such procedure because "the sounds of the (native) tongue are
automatically and subconsciously organized by a native speaker into the structural units
which we call phoneme (Pike, 1947, p.57) i.e. the native speaker intuitively establishes for
himself the phoneme of the native tongue.
The London Phonological school is in the first place connected with the name of the
prominent British linguist Daniel Jones. As the scholar concentrated his effort
primarily on producing transcriptions for the languages without writing traditions on the
one side, and on developing approaches to English orthographic reforms on the other,
he seeks to create a phoneme theory which could be applied to practical research of
languages. "… phonetics can be neither studied nor applied without using phonetic
transcription,… adequate transcription systems require, for their construction, the
phoneme theory" ("Phoneme: Its nature and usage" ,1950 ,p. VII).
a)The various members of the 'family' must show phonetic similarity to one another,
i.e. must be related in character (Jones,1950,p.10)
b) No member of the 'family' may occur in the same phonetic context as any other. This
condition is often referred to as the requirement of complementary distribution.
The 'Physical view' excludes all reference to non-phonetic criteria in the grouping of
sounds into phonemes. Furthermore, the fact that members of different phonemes are
capable of differentiating meanings whereas members of the same phoneme are not, is
said to be a result (consequence) of the definition of the phoneme, and not its basis.
- Though D.Jones treats semantic differentiation as secondary (see the above), he does
not reject the functional role of the phonemes altogether and admits that the phoneme
can perform a distinctive function in oppositions. "The sounds of a single phoneme do
not obligingly differentiate words, but they are capable of doing it" ( 1950,p.13-14).
However, in some cases of the distributional analysis, which determines the distinctive
function of the phoneme exclusively on the basis of binary oppositions employing the
method of substitution, D. Jones singles out such pairs of sounds that cannot be opposed
in their distribution (EX., /h/ and//), accordingly he concludes that the distinctive
function of these phonemes cannot be possibly determined. It can be stated that D.Jones
is certainly right registering this weak point of the distributional analysis though he does
not attempt to explore it, which can be presumably explained by the fact that D.Jones
does not generally differentiate between the semantic function of the phoneme and the
functional relevance of the sound and assigns the role of the semantic differentiator to
18
the sound but not to the phoneme, maintaining that the sound can directly perform a
linguistic ( ex., distinctive) function thus becoming semantically functional. All the
above said testifies to the fact that the notions of the sound and the phoneme in the
'physical phoneme theory' are not consistently differentiated.
- The view upon the phoneme, advocated by D.Jones, may be characterized as the
extreme form of the 'physical' concept of the phoneme as his thesis that "the phoneme
is a family of sounds" is the keynote of his phoneme theory. Yet in the world
phonological research there exist phoneme interpretations which, in one way or another,
take into account the physical/material characteristics of the phoneme.
a) In Russia the materialistic approach towards the phoneme analysis was taken up by
L.V.Scerba who treated it as a sound type. But, unlike D.Jones, Scerba emphasized that
the integrity of this material unit is ensured by its functional relevance.
b) The phoneme theory put forward by L. Bloomfield, on the one side, is close to
Jones's concept, since the former defined the phoneme as a distinctive sound, but
Bloomfield was the first to assign the structural validity to it.
c) The practical analyses of both British scholars and American Structuralists are of
pragmatic character involving the description of different languages and employing the
physical interpretation of the phoneme. But they differ in the goals of their research. If
for the London school pragmatics determines the theoretical direction of analysis, for
American descriptivists the theory conditions the practical investigation and the
structural analysis of the phonetic phenomena becomes, to a large extent, the goal in
itself самоцель).