Stratificational phonology is part of a wider theory of language. Developed in the USA in the 1960's,it falls within the broad tradition of Saussurean structuralism and shows particular influence from glossematics, notably the emphasis on language as a network of relationships rather than a set of elements. The stratificational view is that language is organized on distinct levels or "strata", the one of most relevance to phonology being the " phonemic statum".The units of this stratum, phonemes, are presented as points in a network which links each phonemes.
In the version of stratificational theory presented by Lockwood (1972), there
are six strata: those of gnostemics, sememics, lexemics, morphemics, phonemics
and phonetics. Of those the first two are chiefly relevant to semantics. The
lexemic tactics (lexotactics) defines the syntactic possibilities of the
language, while the morphotactics defines the possibilities of word structure
and to some extent also the structure of phrases. Morphophonemic phenomena are
handled by realizational statements between the morphemic and phonemic strata,
with the help of the phonotactics. The phonetic stratum is said by Lockwood to
“provide for non-distinctive phonetic facts”; its details have apparently
not yet been worked out, but presumably the tactics of this stratum
will define a set of possible phonetic representations for the language in
the one hand there is mixing of levels in analysis, where the linguist uses higher-level (e.g. grammatical)
information to justify a particular lower-level (e.g. phonological) description.
On the other hand there is mixing of levels in presentation,
which is, to a greater or lesser extent, a characteristic of many theories.
Generative phonology is particular liberal with this kind of level-mixing; it
allows both grammatical conditions on phonological rules and phonological
conditions on syntactic transformations. Stratificational grammar developed in
the 1960s, and one of its main claims is that mixing of levels in presentation
is an offence against the way language is organized.
Stages of Encoding and Decoding Process
can be viewed in various ways – a set of habits, a set of sentences, a set of
rules or organizing principles, are among the possibilities. To Sydney Lamb, the
founder of Stratificational Grammar, language is essentially a code.
Like any other code, it links a set of messages
with a set of signals. In the case of
language, the messages are what can be labelled conceptual structures, and the
signals consist of phonic material, either as articulated by the speaker or as
received by the hearer. The speaker encodes
the message into a signal; the hearer decodes
the signal into what, errors and ambiguities apart, should be the speaker’s
original messages. It is an axiom of SG that the encoding and decoding
mechanisms are describable by one and the same set of statements, merely read in
SG holds that both the encoding and the decoding processes take place in a small
number of discrete stages, and that each stopping-place, as it were, corresponds
to a distinct level of linguistic organization, or straum.
It follows that a linguistic description must consist essentially of two
encoding-decoding device consists of a set of statements defining the
relationship of the units of each stratum to the strata “above” (nearer the
conceptual level) and “below” (nearer the phonic level). These statements
between them describe the realization
of a message as a signal, or vice versa, and collectively they are termed the realizational
portion of the grammar. In the domain of phonology they correspond roughly
to morphophonemic and allophonic rules.
the right angles, so to speak, to this “vertical” organization, each stratum
has its own “horizontal” organization independent of the other strata. This
defines the elements of the stratum and their possible combinations, up to the
largest unit relevant at that stratum; it is termed the tactics
of that stratum (cf.“phonotactics”).
Stratificational phonology is an impressive outline of a structuralist perspective which incorporates many important concepts such as levels of organization, phonemes and phonetic features. Its bold attempt to formalize the entire work of relationships is an improvement upon the simpler varieties of North America phonemics, and it deserves better than to be submerged in the shifting of the phonemic concept itself towards a more morphophonemic notion.
David G. 1991. Autosegmental and Stratificational
Models of Phonology.
In: Bredn-Ruth-M. (ed.).
The Eighteenth LACUS Forum 1991. Lake Bluff, IL:
Ling. Assn. Of Can. & U.S., 1992, 301-13.
Alan H. 1977. Modern Phonology. London: Edward Arnold.
Clark, John and Yallop, Colin.2nd edition.Blackwell,1995 ,396- 398.