1. The Aim and Structure of the Semiotic Theory of Language
1. (p. 7) All the above estimates are reported in Miller, 1981.
2. (p. 9) A similar example is used in Reformatskij, 1960. The term stratification has been used in other senses by Lamb (1966a, 1966b, 1973a, 1973b, 1982) and Lockwood (1972) (also cf. Makkai, 1972 and Makkai and Lockwood, eds., 1973). Whereas applicative grammar is a self-contained system in its own right, it is ultimately not incompatible with Lambian cognitive stratificationalism. The reader should be most careful in drawing facile comparisons.
3. (p. 11) The Principle of Semiotic Relevance, which constitutes a basis for proper methods of abstraction in linguistics, can be traced back to the Principle of Abstract Relevance (Prinzip der abstraktiven Relevanz) proposed by Karl Bühler (1931: 22-53). Bühler’s principle was meant only as a general point of view; it did not specify the correlation between sound differences and meaning differences as an essential condition for constraining linguistic abstractions.
4. (p. 16) A similar critique of the notion of homonymy is given by Zawadowski (1975: 117-26). Zawadowski correctly points out that different signs have the same form, and therefore homonymy is a theoretically invalid notion; he suggests replacing the notions of homonym and homonymy with the notions ‘polysemous sign’ and ‘polysemy’, which correspond to my notions of ‘polyvalent sign’ and ‘polyvalence’, based on the Principle of Semiotic Relevance and the Principle of Suspension. For lack of an explicit formulation of the Principle of Semiotic Relevance and the Principle of Suspension, Zawadowski does not explain why different signs cannot have the same form and why the same sign can have different meanings. I am at odds with his narrow conception of the sign as merely a sound sequence. Thus, for him, bear is the same sign in (5a), (5b), (5c), and (5d), while I distinguish three signs here: 1) bear; 2) bear+place of a term; 3) bear+place of a predicate.
5. (p. 17) A detailed critique of Saussure’s notion of the bilateral sign along similar lines is given by Zawadowski (1975: 120-26).
6. (p. 30) The latest version of generative grammar is presented in Chomsky’s Lectures on Government and Binding (1981). Although the new version greatly differs from the first version presented in 1957, the basic principles of Chomsky’s linguistic theory have not changed.
7. (p. 31) I must mention a few linguistic theories that view language in the spirit of modern semiotics. These are:
1) L. Hjelmslev’s glossematics (Hjelmslev, 1961);
2) Sydney M. Lamb’s stratificational linguistics (Lamb, 1966a, 1966b, 1973a, 1973b, 1982; Lockwood, 1972; Makkai, 1972);
3) Kenneth L. Pike’s tagmemics (Pike, 1967, 1970, 1982; Pike and Pike, 1982, 1983);
4) M.A.K. Halliday’s systemic grammar (Halliday, 1973, 1974, 1978);
5) I. A. Mel’čuk’s ‘meaning↔text’ model (Mel’čuk, 1974).
Of course, these linguistic theories greatly differ both from each other and from my linguistic theory. Still, the semiotic view of language is their common ground.
2. Phonology
1. (p. 33) Speaking of Eastern Armenian, I mean literary Eastern Armenian as represented by educated Yerevan speech (Yerevan is the capital of Eastern Armenia). It should be noted that in some dialects of Eastern Armenian, pˉ, tˉ, and kˉ are pronounced with glottal closure which is released only after the release of the articulatory oral closure; they belong to a class of consonants called ‘ejectives’. Ejectives occur also in Georgian and thirty-six other Caucasian languages and are found in many languages of the world (Catford, 1982: 70).
2. (p. 65) The reader will find a detailed description of distinctive features from an articulatory, acoustic, and perceptual point of view in Jakobson and Waugh, 1979.
3. (p. 70) As far as Russian language data are concerned, A. Martinet inaccurately points out that in Russian all components of a composite word, with the exception of one, lose their proper stress, because in Russian there also exist composite words every element of which preserves its proper stress, either as a primary or as a secondary stress, as, for instance, kòneférma ‘horse farm’, pàrnokopýtnye ‘artiodactyla’, slàborázvityj ‘underdeveloped’, and others (cf. Russkoe literaturnoe proiznošenie i udarenie, reference dictionary under the edition of R. I. Avanesov and S. I. Ožegov, Moscow, 1960). However, in spite of the inaccuracy, Martinet grasped the phonological essence of the phenomenon: since in Russian the components of composite words preserve their proper stress only in a few cases, the number of stresses in a sentence does not enable us to judge the number of lexemes within that sentence, and, therefore, from the phonological standpoint, in Russian the accentual unit is not the lexeme but the word.
4. (p. 70) A similar approach to phonological syntagmatic units is used in Prieto, 1975.
5. (p. 92) For morphophonology, or morphophonemics, see Jakobson, 1971; Hockett, 1958: chap. 16 and 32; Kuryłowicz, 1977; Lamb, 1978; Martinet, 1965; Matthews, 1974: chaps. 11-12; Stankiewicz, 1979.
3. Genotype Grammar
1. (p. 97) At this stage of the discussion, I do not question the conventional notions of subject and object. It will be shown below that these notions are inadequate for the purposes of linguistic theory.
2. (p. 103) The first to formulate the Applicative Principle was the Russian mathematician M. Schönfinkel, who formulated the bases of combinatory logic developed by the American mathematician H. B. Curry (Schösnfinkel, 1924; Curry and Feys, 1958; Fitch, 1974).
3. (p. 106) On the notion of immediate constituents, see Wells, 1947 and Gazdar, 1981.
4. (p. 117) The notion ‘functional transposition’ and notions ‘primary function’ and ‘secondary function’ can be traced back to the works of Slotty (1932), Kuryłowicz (1936), and Bally (1944). The notion ‘functional superposition’ introduced in this book is new.
5. (p. 195) The terms operationality and operationality primitive correspond to Curry’s terms functionality and functionality primitive (Curry and Feys, 1958:262-66). Within the conceptual system based on the notion of the operator, I prefer the terms operationality and operationality primitive, since the operator is a function taken as a non-set-theoretical concept, and the term functionality involves associations with the function conceived of as a set-theoretical concept.
6. (p. 224) Section 3.13.7 of this chapter incorporates some material from the book Theoretical Aspects of Passivization in the Framework of Applicative Grammar by J.-P. Desclés, Z. Guentchéva, and S. Shaumyan (1985), to which I here apply some different theoretical viewpoints.
7. (p. 238) The characterization of reflexivization presented in section 13.8 of this chapter is a continuation of the research started in the paper “Reflexive Constructions: Towards a Universal Definition in the Framework of Applicative Grammar” by J.-P. Desclés, Z. Guentchéva, and S. Shaumyan (1986).
8. (p. 253) Zellig Harris has published books in which he used the concepts of operator and other related concepts (Harris, 1968, 1982). What is presented by Harris is not a linguistic theory proper but descriptive methods applied to a particular language—English. Harris operates with strings in the spirit of his previous works in descriptive linguistics. He is not interested in investigating language on the level of functional units. The level of abstraction is the same as in his previous works on transformations.
The basic notion of Harris’s books is transformation. Harris understands transformations not as operations on trees, as does Chomsky, but as operations on strings. Transformations are systematic relations between sentences conceived of as strings of words. Technically, a transformation is an unordered pair of linguistic structures taken to be structurally equivalent in a certain sense. In order to obtain transformations, a set of constants called operators is applied to sentences. Analysis of transformations in terms of operators, in Harris’s sense, is simply another way of describing the system of transformations presented in his previous works (Harris, 1957, 1965).
9. (p. 254) Among various works on analyses of natural languages from a logical point of view, one can mention, for example:
1) Carnap, R., “Meaning and Synonymy in Natural Languages” (in Carnap, 1956: 233-48);
2) Reichenbach, H., “Analysis of Conversational Language” (in Reichenbach, 1947: 251-354);
3)Quine, W. V., Word and Object (Quine, 1960).
10. (p. 269) It should be noted that a sharp critique of various orthodox versions of transformational grammar has been produced by some transformationalists who have departed far from its tenets in trying to mend its broken fences. Among them, James McCawley (1979, 1980, 1982) deserves mention in the first place.
11. (p. 283) In addition to works on the theory of combinators mentioned above, I would like to mention an excellent book for the general reader by Smullyan, 1985.
4. Phenotype Grammar
1. (p. 284) See the discussion of the notion of the word in Sapir, 1921; Bloomfield, 1933; Lyons, 1968; Matthews, 1974; Mel’čuk, 1976, 1982.
2. (p. 290) Since the main goal of this study is the clarification of the contemporary issues of grammar and phonology, I have been less interested in the otherwise fascinating and rather important area of lexicon and of one of the lexicon subsets, what is called bound or phraseological expressions (such as to kick the bucket, red herring, ham and eggs). In the framework of applicative grammar, rather extensive theoretical studies have been carried out of the organization of the lexicon with a special reference to Russian; see Shaumyan and Soboleva, 1968; Soboleva, 1972, 1973, 1978, 1981, 1983; Al’tman, 1981, 1983; Barčenkova, 1981; Ginzburg, 1973, 1981.
3. (page 297) There is no generally accepted notion of linguistic category. Some linguists define a linguistic category as a set of mutually exclusive meanings shared by a class of linguistic objects. Under this definition, the number in English is a linguistic category, but the singular number or the plural number is not; the grammatical case in Latin is a linguistic category, but the nominative case, the genitive case, the dative case, etc. are not. Under my definition, both the number and the singular number or the plural number are linguistic categories; both the grammatical case and the nominative case, the genitive case, the dative case, etc. are grammatical categories.
5. Linguistic Methodology
1. (p. 307) For a detailed discussion of the concept of ‘psychological reality’, see Black and Chiat (1981). The authors argue correctly that the notion of psychological reality not only has been irrelevant to the practice of linguistics itself but also has had negative effects on the development of psycholinguistic theory. They conclude that the notion of psychological reality must be abandoned.