category (n.)
A general term used in LINGUISTICS at varying levels of abstraction. At its most general level, categorization refers to the whole process of organizing human experience into general concepts with their associated linguistic labels; the linguistic study of this process (in SEMANTICS) overlaps with that of philosophers and psychologists. In the field of GRAMMAR, categorization refers to the establishment of a set of classificatory UNITS or properties used in the description of language, which have the same basic DISTRIBUTION, and which occur as a structural unit throughout the language. In the course of language change, there may be alterations in the category status of a unit (recategorization). The term category in some approaches refers to the CLASSES themselves, e.g. NOUN, VERB, SUBJECT, PREDICATE, noun PHRASE, verb phrase (any associated abbreviations being referred to as category symbols). More specifically, it refers to the defining properties of these general units: the categories of the noun, for example, include NUMBER, GENDER, CASE and COUNTABILITY; of the verb, TENSE, ASPECT, VOICE, etc. A distinction is often made between grammatical categories, in this second sense, and grammatical FUNCTIONS (or functional categories), such as SUBJECT, OBJECT, COMPLEMENT.
While both of these senses of ‘category’ are widespread, several specific applications of the term have developed within individual theories. For example, in SCALE-AND-CATEGORY GRAMMAR, ‘category’ is used primarily to refer to the notions of CLASS, SYSTEM, UNIT and STRUCTURE, which the theory recognized as basic. It is also distinguished from SEGMENT in Chomsky-ADJUNCTION. Most distinctive of all, perhaps, is the special status given to the term in theories of categorial grammar, a type of FORMAL GRAMMAR devised by logicians in the 1920s and 1930s, and developed by several linguists in the 1950s (in particular by Yehoshua Bar Hillel (1915–75)). Its distinctive mode of operation involves the deriving of categories from more basic categories: for any two categories, P and Q, there is a complex category of the type P/Q, which represents the operations which may be performed on a given word. For example, given the basic categories N (noun) and S (sentence), an item such as go would be assigned N/S, thereby capturing its INTRANSITIVE status (i.e. go can combine with a preceding N to produce S). More complex structures can be reduced to simpler ones using a set of syntactic operations, in which the notion of ‘cancellation’ is especially important (e.g. P followed by P/Q reduces to Q).
In GENERATIVE grammar, the set of PHRASE-STRUCTURE RULES in a grammar may be referred to as the categorial component, i.e. that part of the BASE component of the grammar which specifies such syntactic categories as S, NP, VP. A categorial rule is a RULE which EXPANDS a category into other categories. Also, in some MODELS of generative grammar, the term category feature is used to refer to a type of CONTEXTUAL feature, i.e. a syntactic feature which specifies the conditions relating to where in a DEEP STRUCTURE a LEXICAL ITEM can occur. Category features specify which NODE will be the one to DOMINATE directly the lexical item, once it is introduced into the PHRASE-MARKER (replacing the corresponding EMPTY (DELTA) symbol, e.g. [+N], [+Det], [+V]). A category variable is a symbol which stands for any lexical category. A related term in this model is strict sub-categorization, referring to features which specify further restrictions on the choice of lexical items in deep structure.
