Academic journal article Studia Anglica Posnaniensia: international review of English Studies

Prototypes and Invariants in Linguistic Categorisation

Academic journal article Studia Anglica Posnaniensia: international review of English Studies

Prototypes and Invariants in Linguistic Categorisation

Article excerpt

1. Introduction

Since Ross (1972) and Rosch (1973), prototypes and concepts with fuzzy borders have been discussed in linguistic semantics. Some linguistic theories, especially generative linguistics and logical semantics have taken no notice of prototype effects, which are claimed by cognitive linguistics to permeate linguistic categorisation at all levels, i.e. from a phoneme (Jaeger 1980) up to a clause (Ross 1973, Kalisz 1981). Cognitive linguistics, especially in Lakoff's (1977, 1982 and 1987) version, treats prototype as a fundamental concept in categorisation of linguistic and psychological phenomena. Lakoff (1987) formulated the notion of a radial category where we can distinguish central members, exhibiting all or the majority of properties devised for a category and other members of the category which depart from the central cases family resembling (the term comes from Wittgenstein 1953) those category members which are regarded as being prototypical. Such category members can be shown to depart from prototypical c ases in a radial manner. The examples which are discussed most often are bachelor and mother

Wierzbicka (1996) is not against the concept of prototypes in linguistic theory; however, she claims that the notion of a prototype is abused in linguistic analyses, especially by cognitive linguistics. She tries to show that often for the cases which are given a prototypical account, it is possible to give an account in terms of invariants, i.e. non-fuzzy categories with clear borders. She also tries to make a distinction between genuine cases of prototypicality and cases which do not need a prototypical account and can be analysed in terms of semantic definition containing exclusively invariants.

In this paper I try to reconsider Wierzbicka's (1996) claims and I will try to demonstrate that although not all linguistic (here semantic) phenomena have to be given a prototypical account, the array of cases where such account is necessary seems to be larger than she assumes. It is also indispensable to state that a prototypical account itself is insufficient for an analysis of semantic or other linguistic phenomena. A prototypical account has to be accompanied by the concepts of ICM (idealised cognitive model), folk models and other grounding devices employed by cognitive linguistics (e.g., Lakoff 1987; Langacker 1991).

My general claim is that an account in terms of invariants is possible; however, such a possibility constitutes a special marked case whereas a prototypical structure of a concept reflects a general, unmarked case.

2. Prototypes and invariants in an analysis of the concept bachelor

Wierzbicka (1996) and Posner (1986) claim that a prototype may be and is overused as a linguistic tool in situations where it is possible to provide a precise definition. They claim that it leads to neglect of an intellectual effort necessary for a precise semantic description of a given concept. Wierzbicka (1996: 167) writes that if a prototype is treated as a magical key to open all doors without effort, the chances are that it will do more harm than good. Posner (1986: 58) claims that "these new ideas have been treated as an excuse for intellectual laziness and sloppiness". These ideas could be right provided that it is really the case that "prototype saves" which is a travestation of McCawley's (1981: 215) slogan "Grice saves" quoted in Wierzbicka (1996). Whether "prototype saves" is a legitimate, ironic slogan and whether the application cf the concept of a prototype really spares intellectual effort and leads to laziness and sloppiness will be considered later. In the next sections I will try to conside r some criticism and some concrete analyses provided by Wierzbicka (1996). She treats some cases as representing indeed prototypical effects required in her definitions. However, the first part of cases which she takes up are to be understood as misuses and misapplications of the notion of a prototype where a definition in terms of invariants is possible in her view. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.