As I pointed out in an earlier section, the competence-performance distinction in current linguistics work differs from the earlier de Saussurian distinction of langue-parole, in that competence is viewed as a generative system while langue is a static one; and this is, of course, a fundamental difference. But there is an aspect of passivity that characterizes linguistics generally, whether generative or structural, that contrasts with the activity-oriented interests in such fields as psycholinguistics and sociolinguistics, Part of this difference is attributable to an interest on the part of linguists in an existing system independently of its particularizations (in performance or parole), but part of it is also due to a near-exclusive emphasis on the system in the abstract or its idealization removed from the particular conditions that necessarily accrue to the linguistic competence of individual speakers (cf. Chomsky's "the ideal-hearer's intrinsic competence"). Granted that performance is variable and affected by non-linguistic considerations which the linguist may not wish to incorporate in his system, it may nevertheless be true that the linguistic competence of actual individuals is not uniform over the set of native speakers of a language, and furthermore, while one would not expect an individual's linguistic competence to vacillate from moment to moment (as does his performance), changes may occur when larger time segments are considered (say years or decades).
It is not unreasonable, it seems to me, to argue that linguistic theory ought to concern itself with such variations whether dialectal (in which case different sub-systems may be involved-probably no argument from the linguists in this case), or idiolectal (including changes over large time segments). Furthermore, one could argue with some merit that linguistic theory ought also to concern itself with the kinds of "varieties" of language that are usually discussed under the rubric of formality, register, and style. After all, such variability is a part of an individual's linguistic competence and has been shown to be regular and lawful (hence, it forms a proper part of the "system").
Similarly, a theory of communicative competence, while not directly concerned with particular instances of communicative acts, must deal with individual differences in the ability to use verbal communication. This will take the form of a description of the relationship between the inferences that are possible in a communicative situation and the resultant communication that is achieved. For instance, the theory will specify that a failure to make a particular implicit or implicative inference in a given utterance, will result in that utterance being assigned a different meaning from that it would have, had the inference been made in just the particular way specified. Furthermore, such a theory would also concern itself with the relationship between experiential factors (e.g., socialization) and the ability to make certain kinds of inferences in verbal communicative situations. In particular, it will have to account for such commonplace facts that people with apparently similar experiential histories (at the social psychological level) fail to communicate effectively in some situations and, as well, the fact that people with quite different backgrounds are able in some situations to communicate quite effectively.
What sense can be attached to the phrase "communicate effectively"? In one well-known approach to communication, the "information theory" paradigm (see (Cherry, 1957), efficiency of a communication system is defined as the fidelity of the message as it goes through the encoding-decoding cycle (from speaker to hearer) and is inversely proportional to the amount of "noise" the system generates Within the approach outlined in this paper, the "fidelity of the message" is given by a comparison between the intention of the speaker and the intention the hearer attributes to the speaker. But this definition is only an approximation and leaves out several important factors. First, some of the intentions attributed to the speaker by the listener are not a proper part of the intentions the speaker intended to reveal (cf. Goffman, 1959, on the distinction between "impressions given" and "impressions given off").
It would appear that the "unintended intentions" form an important part of the meaning of an utterance and, thus, one would want to consider these as part of the message to be recovered (cf. the proverbial "No" that means "Yes" in many delicate situations). Second, because of the characteristically indexical nature of utterances, some of the inferences a person makes about the discourse of a speaker or writer may not have been intended (either deliberately or non-deliberately) by its originator. This is what makes possible the richness and stimulating character of "creative writing" in fiction, drama, and poetry (which leads to the seemingly "odd" situation whereby a creative piece of literary criticism is itself in some cases "richer" than the piece being evaluated). It is problematic, at this stage of our analysis, whether or not one should consider these inferences as a "proper" part of the meaning of the original discourse. For these reasons, it might be that the encoding-decoding paradigm of information theory may not be an adequate model for the analysis of meaning and communicative competence.
The teaching of "effective" speaking and writing would seem to involve two distinct aspects. One is related to characteristics of conceptualizations, the other, to linguistic characteristics of utterances (vocabulary choice and syntactic structure). While these two aspects do not operate independently (as discussed above), they are nevertheless distinguishable. Remedial instructional activities may require different techniques in the two cases. With some students, both aspects may be deficient (by criteria the teacher has), only one with some others. Certain remedial programs for the "culturally disadvantaged" assume a deficiency in both areas (cf. Bereiter and Engleman, 1966), but this is denied by some writers (cf. Kochman, 1969). A theory of communicative competence which will specify the nature of the inferences involved in the analysis of meaning would help resolve this issue by making possible studies which investigate the relationship between experiential factors and the ability to engage in certain specific types of inferences.
The possibility has been expressed by some writers that certain languages are more suited for certain purposes than others (e.g., that English is more "logical" and analytical, therefore better for expressing scientific thinking). Today, not much credit is generally attached to these views, especially since linguists emphasize the underlying similarity of all natural languages and their mutual intertranslatability.17I think it is important to distinguish between the claim of unsuitability of a language as a system or code for the expression of certain conceptualizations (a strong version of the Whorfian Hypothesis), on the one hand, and on the other, the weaker claim that speakers of a particular language or dialect thereof show evidence of the absence of certain types of syntactic relations and constructions that are believed to underlie certain types of conceptualizations. The weaker claim may be true even if the stronger one is false. There are, it seems to me, cogent reasons for considering the stronger claim implausible and the weaker one plausible. The reason for the implausibility of the stronger claim has already been suggested: the biological foundation of human languages and their specific innate components that must he considered universal-unless one rejects the assumption that any human infant can learn any human language.
Thus, it would be strange to assume that any particular natural language is incapable of encoding certain human cognitive processes that are codable in another particular language. the reason for the plausibility of the weaker claim is that the syntactic relations and constructions habitually used by a speaker and which constitute part of his communicative competence are related in some way to the nature of the conceptual inferences he habitually makes and in which he develops facility. Just as there are individual differences in communicative competence within a sub-culture, so also are there differences between subcultures in the types of communicative and conceptual inferences habitually made. For instance, the system of formal grammatical markings for indicating status relations between; two speakers is more extensive in languages such as Japanese, Thai, and Korean than in English. It is plausible, therefore, to assume that speakers of these languages are more skilled than English speakers in making inferences about the status distinctions between individuals, and thus, an American learning Japanese must improve his communicative skills in this social area. Speakers of Standard American English, if they are to develop competence in Black English, must learn certain new communicative skills that are habitual for Black Americans (e.g.,."loudtalking,18 see Mitchell-Kernan, 1969).
Speakers of Black English, if they are to learn Standard American English, must learn certain new communicative skills that are not habitually required in their sub-culture. There is no evidence to suggest that the new communicative skills one is required to develop when learning another language or variety thereof cannot be developed within the individual's own language. Thus, Americans can become more attuned to status differences (some are!) without learning Japanese, many of the communicative devices used in "rapping" can be used by speakers of Standard American English (many do! cf. talking ironically, by innuendo, and "one-upman-ship") and the kinds of inferences required of a pupil in a white, middle-class American school can be taught through Black English. In each of these cases, the issue of competence is not linguistic, but communicative and conceptual.
17. It is interesting that Descartes, who was also impressed by the universal underlying features of language, nevertheless subscribed to such a view. Similarly, Diderot. See the discussion in Chomsky, 1965, pp. 6-8 and Chomsky, 1968.
18. The term loud-talking is applied to a speaker's utterance which by virtue of its volume permits hearers other than the addressee, and is objectionable because of this" (Mitchell-Kernan, 1969, p. 129). Loud-talking is used as a "put-down" device, as a measure of retaliation to insult or pestering. or as a means of exerting "public" pressure on an individual. Other forms of speech acts which appear more highly developed in Black English than in Standard English include "signifying." "shucking,'' "playing the dozens" and the whole gamut of "rapping" which constitutes the highly valued and prestigeful verbal art in the street culture of the inner-city Black world. See Kochman, 1969.
Back to the Titlepage