the aspect of linguistics that deals with the regular patterns of the formation and usage of word forms; often understood more generally as a synonym for linguistics.
The extension of the meaning of the term “grammar” was known as far back as the ancient Greeks and has been preserved to the present day in the combinations “comparative grammar,” “historical grammar,” and “stratificational grammar.” In contemporary linguistic study the term “grammar” is often used in a narrower sense, although the range of phenomena pertaining to grammar is not defined in quite the same way by the various linguistic schools.
The most generally accepted sense divides the science of language into phonology, grammar, and lexicology. In accordance with the traditional approach, the first two divisions deal with general categories (such as vowels, consonants, the verb, and the predicate), and lexicology deals with individual lexical units. It is precisely through this grouping that phonology and grammar proper are sometimes combined under the general category of “grammar” and together are set apart from vocabulary. However, it is more common in contemporary linguistics to exclude phonology from the sphere of grammar. The term “grammar” is then sometimes used in a broad sense, defined as everything in language except phonology and correspondingly as “the science dealing with signs—as opposed to phonology, which deals with the constituent elements of signs.” In a more specific sense, only part of the phenomena on the sign level pertain to grammar. These phenomena are isolated according to various criteria, the application of which, however, sometimes leads to results which generally coincide. Thus, the widely accepted opposition of grammar to vocabulary (within the sign level) is in some linguistic conceptions based upon the feature of the length of the corresponding units; vocabulary deals with words as integral units of the lexicon, and grammar with units either larger or smaller than words. The traditional division of grammar into two sections is accordingly retained: morphology (etymologically, “the science of forms”) and syntax (etymologically, “arrangement together,” or “combination”), the first of which examines the internal structure of words and the second, the rules for combining words into sentences. This division of grammar is linked with recognition of the word as the basic grammatical unit. In traditional grammar it was considered an accepted fact that the “forms” studied by grammar are word forms, and that words then are the units that combine with one another.
Many representatives of contemporary linguistics consider the basic unit of grammar not the word, but rather the smallest meaningful element, usually called a morpheme (or moneme), and they are not inclined to consider as fundamental the difference between combinations of morphemes forming words and combinations forming more complex syntactic units (word groups, sentences). In this sense the necessity of dividing grammar into morphology and syntax is eliminated, and grammar is defined as “the meaningful arrangement of forms” (L. Bloomfield) or as “morphotactics” (that is, the rules of combinability of morphemes), as opposed to “phono-tactics” (that is, the rules of combinability of phonemes, studied by phonology—C. Hockett). The line between vocabulary and grammar in this instance is drawn by some linguists on the basis of whether the corresponding units are part of an unlimited stock or a limited stock.
The opposition of grammar to vocabulary is sometimes argued through the fact that the categories of grammar are general, so that statements about corresponding phenomena pertain to a whole class of grammatically homogeneous units, whereas lexicological statements are of a special nature, bearing upon each separate lexical unit individually. Thus, the nongrammaticality of combinations such as kruglogo stolu pod lezhish’ tolsiyi kniga ([the] round [adjective, genitive masculine] table [substantive, dative masculine] under [taking the instrumental] lie [verb, second person singular] [the] thick [adjective, nominative masculine] book [substantive, nominative feminine]), instead of pod kruglym stolom lezhit tolstaia kniga (under [taking the instrumental] [the] round [instrumental] table [instrumental] lies [third person singular] a thick [nominative feminine] book [nominative feminine]) is determined by their lack of correspondence to the general rules governing the combinability of the units of the Russian grammatical system. In contrast, a statement about the restricted combinability of the adjective peklevan-nyi (fine rye), which can be combined only with the substantive khleb (bread), is a statement of a lexicological order. The use of the terms “grammaticalization” for the further extension of rules with a previously narrower sphere of effect, and “lexicalization” for the opposite process, corresponds to the idea of the general nature of grammatical rules.
The criterion of differentiation between the lexical and grammatical spheres has to do with the characteristic of the meanings expressed by the corresponding units. Thus, a material, concrete character is ascribed to lexical meaning and a formal, abstract character to grammatical meaning. However, in many instances it is difficult to perceive the distinction between these two types of meaning in terms of “concreteness” and “abstractness.” Therefore, the greater abstractness of denotations of quantity as compared with denotations of quality or size is doubtful. However, the distinction between the meanings of the forms dom (house) and domá (houses) is considered to be grammatical, and the distinction between the meanings of the words khoroshii (good) and plokhoi (bad) or domik (little house) and domishche (large house) is considered to be lexical. Many linguists prefer therefore to speak of the distinction between lexical (or nominative) and syntactic (or relational) meanings and about the opposition, independent of this distinction, of grammatical to nongrammatical meaning. The distinction between nominative and syntactic meaning can be reduced to the fact that the first directly reflects (“names”) extralinguistic reality (objects, events, attributes, relationships), whereas the second reflects only a given word form’s capacity to enter into certain types of syntactic relations with certain classes of word forms in the construction of a phrase. From this standpoint the word forms stoly (tables), stol (table), and stolik (little table) possess different nominative meanings, while the word forms stolu (table [dative]) and stolom (table [instrumental]), or begushchii (running [present active participle] and bezhit ([he] runs) possess different syntactic meanings.
The opposition of grammatical to nongrammatical meaning is based upon the property of necessity, inherent in the former and absent in the latter; thus, the general-categorial meaning of “parts of speech” in those languages in which the speaker is obliged to represent the corresponding extralinguistic matter as an object, attribute, or action (that is, in which the speaker is forced to make some selection from a limited number of possibilities for representing grammatically the given signified, even when the distinction between one or another means of representation is not essential for the speaker himself). Thus, the same situation (“it’s freezing outside”) may be expressed in Russian by the phrases na ulitse moroz (nominal construction), na ulitse morozno (impersonal adverbial construction), and na ulitse morozit (verbal construction), but the designation of the phenomenon of freezing cold must necessarily be represented as an object, attribute, or action, in view of the impossibility of designating this phenomenon without corresponding specification. The selection of some grammatical representation implies in turn the presence of certain necessary (that is, grammatical) meanings. For example, the meaning of number in connection with substantives is grammatical in Russian (since any Russian substantive is either in singular or plural form) and nongrammatical in Chinese and Japanese, since in these languages a noun may be used to designate both one and several objects, as long as the corresponding specification does not enter into the speaker’s intention. In accordance with the opposition of nominative to syntactic meanings, division has been proposed for the study of the plane of content into syntax and the theory of nomination (vocabulary, onomatol-ogy). Since both nominative (for example, number, for Russian substantives) and syntactic (for example, gender, number, and case, for Russian adjectives) meanings can be grammatical, grammar must occupy an intermediate position between vocabulary and syntax: it studies both lexical and syntactic meanings, but only those whose expression is necessary in the given language. Given this division, it is expedient to leave the study of the means of expression of meanings to morphology.
A division that seems successful is a suggestion which is comparatively recent but which corresponds, in general, to linguistic tradition; specifically, it is suggested that those means of expressing any linguistic meanings that are realized within the word boundaries (affixation, alternation, reduplication, incorporation) be considered as morphological, and those realized outside the word (such as by means of syntactic words and word order) be considered as nonmorphologi-cal. Thus, the long-standing controversy over whether all morphology or only a part of it (excluding, for example, word-formation) should be included as grammar is being resolved according to whether a given morphological method serves a grammatical or nongrammatical function. The property of necessity of expression of grammatical meaning, however, seems a more universal criterion (that is, one independent of language type) for defining the sphere relevant to grammar in the proper sense. It may be noted that the fact that grammatical units belong to a limited stock and the fact that grammatical “rules” have a general and regular nature (permitting the reduction of all the diversity of linguistic utterances to limited complexes of systems and structures) are in essence a consequence of this property of necessity.
The methods of modern grammar have their origin in ancient Indian philological science, the most well known representative of which was Panini (fourth to third century B.C.). The system of notions and categories of modern “school” grammar, including even its terminology (for example, the names of the parts of speech and the cases), goes back to the grammatical theory of the ancient Greeks (Aristotle, the Stoics, the Alexandrian school). Of the Roman grammarians, Varro (116–27 B.C.) is the most notable. Greco-Roman grammatical theory was adopted by European philologists of the Renaissance and Enlightenment by way of the Late Latin grammars (M. V. Lomonosov produced the first Russian grammar in 1755; the first Church Slavonic grammars appeared in 1591 and 1596), so that both the concepts and the categories of Latin grammar were transferred to the grammars of new languages. In the 17th and 18th centuries interest in the logical and philosophical foundations of grammatical theory (for example, the problem of “universal” grammar) grew significantly. The development of typological research and the creation of the first morphological classifications of the languages of the world in the early 19th century stimulated the creation of differentiated notional systems for describing languages of different grammatical systems; however, the first systematic work in this direction was not begun until H. Steinthal and continued by the neogrammarians. The idea of “emancipating” the grammars of new languages from the Latin-Greek grammatical system in essence began to penetrate the descriptive grammars of real languages only by the early 20th century; in particular, the grammatical system elaborated by F. F. Fortunatov has been used in Russian grammar.
The basic lines of development of grammar in the 20th century have been concerned not so much with the methods of describing specific languages (although even this aspect has been given some consideration, for example, in descriptive linguistics) as with the problem of grammatical theory itself.
T. V. BULYGINA