Page 445 - Encyclopedia of Philosophy of Language

edge of the language and its rules (as, for example, described in a generative transformational grammar). If this opposition is kept as a valid one, the study of language can simply be divided into two largely independent parts: one a description of its structure (as dealt with in the classical descriptions or gram- mars), the other a description of its use (to be taken
care of by pragmatics).
In Jerry Katz's words: 'Grammars are theories
about the structure of sentence types... Pragmatic theories, in contrast,... explicate the reasoning of speakers and hearers...' (1977:19), when the latter try to establish a relation between what is said and the semantic 'proposition' that is behind it.
A preliminary result of the quest for a definition is that in pragmatics, the language user is at the center of attention. Thus, it is possible to say there is a 'user's point of view' as a common orientating feature for both linguists and philosophers.
However, this is not sufficient to define pragmatics as a science, as witness the varying conceptions in pragmatics of that user's role, as well as of what is implied by the term 'use of language.' For instance, one can either consider 'language use' to be whatever happens when users are 'doing things with words'; or, in a more restrictive procedure, one can demand that pragmatics refer explicitly to a user, whenever lan- guage is discussed.
Furthermore, from a social-scientific point of view, a theory of language as a user's interest should rest on a theory of the user. However, the user being a member of a particular human society, such a theory should encompass everything that characterizes the user as a societal being. This seems to be a logical extension of the notion of pragmatics as a theory of use, even though it entails a 'very broad usage of the term [pragmatics]' (Levinson 1983:2); Levinson also comments, somewhat wistfully, it might seem, that this usage is 'still [!] the one generally used on the Continent' (ibid.).
2.3.2 Communication and Behavior
From yet another point of view, pragmatics can be defined as behavior. This is the approach advocated by some very early proponents of the pragmatic view, Watzlawick, et al., who back in 1968, expressed their discomfort with information science and linguistics, placing the emphasis on one-way transmission of signs without attention to either communication or interaction. As they say, 'from the perspective of prag- matics, all behavior, not only speech, is communi- cation, and all communication—even the communicational clues in an impersonal context— affects behavior' (Watzlawick, et al. 1968:22). These authors' approach to pragmatics is not linguistic, but communicational and behaviorist; in fact, for them, communication (including nonverbal com- munication) and behavior are more or less syn-
onymous: pragmatics is behavior, is communication. Pragmatically speaking, one's communicative behavior is such that one cannot not communicate (1968:72).
No wonder that one looks in vain for traces of influence from these quarters on the linguistic move- ments of the time, most of which were somehow indebted to the 'Chomskyan revolution' with its emphasis on the formal characterization of a lan- guage's syntactic properties. By contrast, the 'hope of abstracting the formal relations between com- munication and behavior' must be said to have been rather remote (1968:13); as a result, the area of com- munication 'has received remarkably little attention,' as Watzlawick, et al. remark (1968:18).
In the tradition of Watzlawick and his colleagues (which to a great extent had its roots in the study of abnormal, psychiatrically treated behavior), the ques- tion is not what a person should say, according to the grammar of a language, but what a person does actually try to communicate, using whatever language he or she has. In this framework, the key words are not 'rules,' but 'information,' 'redundancy,' and 'feed- back' (terms taken from information theory, but not used in the formal abstract sense in which they are defined there); 'imperviousness,' 'paradox in inter- action,' and 'double bind' (actually terms that have their origin in interaction theory and the treatment of the mentally ill). Thus, even the 'craziest' language of the schizophrenic ('schizophrenese': see Sect. 4.4) is said to have its communicative value and import.
Watzlawick, et al. were in many ways pioneers; in other respects, they were the 'voice[s] crying in the wilderness' (Matthew 3:3). Not until the linguists themselves had turned to pragmatics (forced by the paradoxes of their own science) could they begin to understand the double binds that they had been caught in by positing, as the touchstone of their research, a model of a nonexistent, 'ideal speak- er/hearer' (Chomsky 1965).
2.3.3 Grammar and Context
In contrast to these broader uses of the term 'prag- matics,' one finds others demanding a minimum of strictly linguistic involvement before one can begin to talk about pragmatics in the linguistic sense of the term. In Levinson's words: 'Pragmatics is the study of those relations between language and context that are grammaticalized, or encoded in the structure of a language' (1983:9; emphasis in original).
Even though he does not say so explicitly, Levinson seems to detect a conflict between those language- context relations that are, and those that are not, 'grammaticalized' (the process of grammaticalization being understood as the expression of pragmatic relations with the help of strictly linguistic means, such as the rules of a grammar operating on phonological, morphological, and syntactic elements). This, in its
Pragmatics
423