Relational pragmatics: towards a holistic view of pragmatic phenomena.

Science may be described as the art of systematic over-simplification.

Karl Popper 

1. Introduction

The situation in theoretical pragmatics is a peculiar one: first there is the Chomsky -- Kasher cognitivist research program (cf. Sinclair 1995) that concentrates on a narrow view of pragmatic competence as conditions on the use of language; second, Leech (1983) and Sperber and Wilson (1986) propose a performance-oriented pragmatics; and finally, there are the majority of the pragmaticians who are simply "doing pragmatics" with few or no theoretical commitments whatsoever (cf. Levinson 1983).

This paper is intended to provide a theoretical framework for studying both pragmatic competence and performance phenomena. The theoretical suggestions should be seen as a research program for theoretical pragmatics rather than a complete, systematic and justified model. Only the essentials of the inquiry into the problems at the interface of theoretical and practical pragmatics will be presented here (for details cf. Kopytko, in preparation). The research program proposed below will be referred to as Relational Pragmatics (RP).

1.1. Goal, scope and method

Relational Pragmatics puts into focus the analysis of relations between elements of a pragmatic system that consists of (a) interactants or language users, (b) signs or language and (c) context. The scope of Relational Pragmatics is broad.

In contrast to Chomsky's narrow cognitivism supplemented with unspecified conditions on language use, RP takes an integrated system of psychosocial, linguistic and contextual interdependencies as the object of analysis. As can be deduced from the views presented above, the method used in the construction of RP is holistic rather than reductionist (on the problems of methodological reductionism cf. Kopytko 1995). It should be remembered that the reduction of higher to lower level explanations neglects the fact that each level has its own characteristic structure, emergent properties and interrelation qualities.

It should be noticed that some pragmaticians, e.g., Nuyts (1992) or the authors of the articles in Verschueren et al. (1994) postulate an integrative approach to pragmatics or "pragmatic perspective". So far, however, no clear formulation of holism in pragmatics has been offered. Pragmaticians seem to avoid this term in favor of integrative, which is even vaguer, informal and theoretically non-committing.

Among holistic concepts Taylor (1972) includes the following: relationality, interaction, equilibration by means of feedback (i.e., circular causal chains), decision making and goal-seeking behavior. A human being as anthropos telos is seen by Taylor as a self-aware seeker after goals that he/she derives from his view of reality (for a critique of holism cf. Fodor -- Lepore 1991). Holism is usually associated with a multi-disciplinary (or cross disciplinary) approach. Remarkable progress may be noticed in some disciplines that integrate research from many distinct subfields of science. For example, the study of the human mind referred to as cognitive science draws on the insights from cognitive psychology, epistemology, linguistics, computer science, artificial intelligence, mathematics and neuropsychology (cf. Reber 1985). On a smaller scale, a very successful application of the cross-disciplinary approach is the study of induction (cf. Holland et al. 1986) or the integration of motivation and cognition (cf. Higgins -- Sorrentino 1990). In communication research, Krippendorf (1993) sees a remedy for the failures of message-driven explanations in a new convergence of natural language philosophy, ethnography and cognitivism in linguistics, social and radical constructivism, second order cybernetics and reflexive sociology. Scores of other examples of the integrative tendency in scientific research might be adduced. These methodological developments can be viewed as independent evidence for a possible approach for other disciplines. The best conclusion of these remarks is a quotation from Popper's (1940: 413) discussion of the undesirable consequences of atomistic approaches in science:

... if any one of these disciplines is practised in ignorance of the concepts, methods, theories of others ... inadequate sets of descriptive concepts will be used, problems will be poorly formulated ... woolly and obscure metaphors and slogans will pass for explanatory theories, conceptual confusion will go undetected ...

The reduction of complex phenomena in the social sciences and the humanities and the search for a "single cause" of human social behavior does not seem to be a promising line of research. The explanation of language use or human behavior with the "either ... or" dichotomy, e.g. either psychological or social explanation (i.e., cognitivism vs. social constructionism) should rather be replaced by the "both ... and" account in terms of integrated interdependencies. Scheflen (1975) claims that understanding complex phenomena by reducing them to unicausal explanations and upholding them one over the others is illusory.

Finally, it should be remembered that in linguistics it was Noam Chomsky (1965) who first went beyond the narrow atomistic approach of linguistic structuralism by introducing the concept of the speaker-hearer and the cognitive plane of linguistic analysis. This was probably the first step in the right direction. The next steps should be taken by pragmaticians. Many steps have already been made (cf. Mey 1994a). It is to be hoped that a research program referred as Relational Pragmatics will contribute to the integrative cross disciplinary study of social interaction. That language is the basic medium of social interaction cannot be dismissed by any comprehensive account of language use.

The methodological assumptions of Relational Pragmatics will allow us (1) the avoidance of a narrow rationalistic approach (cf. Kopytko 1995), (2) the extension of the empirical content of the theory, (3) the offering of independent evidence for the model, (4) to take advantage of the achievements of other disciplines of science, (5) to view phenomena e.g., language use, in their complex entirety i.e. as a system of interrelated elements, (6) to account for both for pragmatic competence and performance phenomena.

2. Relational pragmatics

If the subject-matter of pragmatics is the study of language use in context (which seems to be the received view in pragmatics (cf. Mey 1993, 1994a), then ex definitione pragmatics is a multidisciplinary science rather than an autonomous field of research. As can easily be deduced, the study of language use must assume the existence of at least three basic elements: (a) the language user, (b) the code or language and (c) the context. The language user may be analyzed in terms of different branches of psychology (i.e., cognitive psychology and personality studies) and sociology (e.g., social constructionism cf. Berger -- Luckman 1966). Linguistics will take care of the code or language and context may be analyzed in terms of its sociocultural and environmental features. Thus, an autonomous pragmatics would rather be a contradictio in adjecto. A reductionist view of pragmatics along the lines suggested by Chomsky -- Kasher (cf. Sinclair 1995) concentrates on the cognitive (innate) properties of pragmatic compet ence, which makes only a small fraction of the vast field of pragmatics that needs to be cultivated. Relational Pragmatics aims at showing the problems that all language users/interactants have to solve if she/he wishes to participate successfully in social interaction, and what is more important it also aims at accounting for the communicative-interactive failures of incompetent participants.

The three basic elements of the theory of language use, i.e. the language user/interactant, language and context, do not function in isolation; on the contrary, they form an integrated system of interrelations. For this reason, the study of these interdependencies will be referred to as Relational Pragmatics. The second major feature (or idea) associated with RP is the notion of system. As has been mentioned above, the three elements form a system of relations. The idea of "system" is very productive in different branches of science: viz, linguistic, social or psychological systems (cf. Rokeach 1960). Following Bertalanffy's (1956) general systems theory, it will be assumed here that a system is analyzed in terms of (1) a set of entities, (2) a specific set of relations among the entities, so that (3) two types of deductions are possible, i.e. (a) from some relations to other and (b) from the relations among the entities to the behavior or the history of the system (for the claim that systems, e.g. chemical s ystems such as dissipative structures can be analyzed from the point of view of the history of the system cf. Prigogine -- Stengers 1984).

Relational Pragmatics may be characterized as a multidisciplinary science with a psychosocial, sociocultural, linguistic and environmental interface (for a similar view of pragmatics cf. Mey 1993, 1994a: 3269). Mey defines pragmatics as "the study of language from a user point of view, where the individual components of such a study are joined in a common societal perspective". And further: "Pragmatics thus defines a cluster of related problems, rather than a strictly delimited area of research." (Mey 1994a: 3269). This kind of pragmatic perspective is to a great extent compatible with the claims and assumptions of relational pragmatics.

In summary, the basic postulates of Relational Pragmatics (RP) include the following: (1) RP is a multidisciplinary science, (2) a pragmatic system consists of three pairs of fundamental entities e.g., (a) the language user, (b) the code and language and (c) the context; (see below) (3) the three pairs of elements constitute an integrated system of interrelations and (4) RIP attempts to account both for pragmatic competence viewed as a user's knowledge of language use and performance phenomena, i.e. communicative problems associated with the actual social interaction.

In the next section a fundamental idea of relational pragmatics, i.e. that of the relational pragmatic system or for brevity pragmatic system (PS), will be presented in some detail.

2.1. Pragmatic systems (PS)

As has already been mentioned a pragmatic system (PS) in Relational Pragmatics (RP) will be analyzed with the triad: the language user or interactant (the latter term seems to be more abstract and more convenient when nonverbal communication (NVC) is referred to), language (or code) and context. It follows that three pairs of binary relations (xRy) make a pragmatic system: (1) Interactant (I) [left arrow] [right arrow] Language (L), (2) Language (L) [left arrow] [right arrow] Context (C) and (3) Interactant (I) [left arrow] [right arrow] Context (C). Thus a triangle of interdependencies may be represented as follows:

The most important claim of Relational Pragmatics is the proposition that the three entities mentioned and the interrelations between them that make the pragmatic system in (fig. 1) underlie the pragmatic competence of language users. This means that the knowledge of pragmatic systems is a conditio sine qua non of appropriate language use.

The three pairs of binary relations in (fig. 1) deserve special attention. The first relation represents the interdependencies between Interactant (I) and Language (L), which may be represented by means of symbols as (I [left arrow][right arrow] L). (At this stage of the discussion, attention will be focused on relations rather than on entities.) As is well known, there are many theories of language but no holistic theory of the language user is available (for a cognitive account see Nuyts 1992). Some suggestions toward achieving such a goal will be presented briefly below. A theory of the language user in Relational Pragmatics will be delineated briefly below. For theoretical and illustrative purposes, a theory of language presented by Noam Chomsky (i.e., a grammar internalized by a speaker-hearer) may be accepted as the model i.e., the government and binding theory (cf. Chomsky 1981) or any new model.) It should be noted, however that many pragmaticians would rather reject Chomsky's view of language of lang uage use (and his distinction between competence and performance) in favor of a functional approach to language proposed by Dik (1989) or Halliday (1985).

It will be assumed that any Interactant acquires linguistic competence, i.e. a grammar and lexicon of his/her speech community. The complex relationship between language and its user (on the cognitive plain) has been the subject of numerous studies in different fields of science e.g., the acquisition of language (cf. Vygotsky 1962), the Sapir -- Whorf hypothesis and linguistic relativity theory (i.e., the claim that linguistic categories structure perceptual and cognitive ones, and that our language structures our perception of the world respectively, cf. Whorf 1956); in cognitive anthropology, or "cognitive sociology" -- a term introduced by Cicourel (1973) to refer to a group of disciplines including ethnomethodology, cognitive anthropology and sociolinguistics.

From the point of view of relational pragmatics, the relation between language and cognition seems to be of utmost importance. It might be analyzed in terms of the following questions: (1) To what extent (if at all) (a) language determines cognition and (b) cognition determines language, (2) How is language acquired? (3) What are the linguistic universals? (4) What part of language is innate rather than acquired in social interaction? Noam Chomsky's theory of language (cf. Chomsky 1965, 1981) attempts to give an answer to some of these questions. However, as has been suggested above, the dimension of "language and cognition" possesses the social and sociocultural interface (especially in the case of language acquisition); for this reason, disciplines such as cognitive anthropology, cognitive sociology or the theory of social constructionism (cf. Berger -- Luckman 1966) might contribute to a holistic explanation of the discussed interrelationship. It should be remembered that the extremist theoretical position s that insist on the "either ... or" option (i.e., either cognition or social construction) are rejected in relational pragmatics both for empirical and methodological reasons. In brief, reality (of any kind whether psychological, social, cultural or physical) is too complex for it to be accounted for it with unicausal explanations.

The second dimension of relational pragmatics, e.g. "language and context", seems to be of special interest to pragmaticians because it refers to the knowledge of language use (also to Chomsky's conditions on appropriateness of language use). A knowledge of language use in context seems to be acquired rather than innate. Again the sociocultural interface seems to be crucial for language acquisition. However, Kasher (1994a: 535) claims that "Use of language is governed by systems of abstract rules that are universally, most probably innately constrained." If Kasher's claim is true, the knowledge of language use would have to be analyzed in terms of innate and acquired cognitive structures (which remains to be proved first).

The basic questions associated with the "language and context" dimension include the following: (1) How language determines context and (2) how context determines language; (3) What is the scope of pragmatic context and (4) how it is represented in the brain.

There are at least three theories that seem to be directly relevant to the questions above: (1) Schank and Abelson's (1977) theory of scripts, plans, and goals, (2) Johnson-Laird's (1983) theory of mental models and (3) Rosch's (1978) theory of prototypes (cf. also Glover 1995).

Chomsky -- Kasher's conditions on the appropriateness of the language use attempt to account for the question in (2) above (e.g., the relation between context and language). Duranti and Goodwin (1992) argue that language can create context. The knowledge involved in such a process may be very useful in strategic language use, e.g. when specific goals of the language user come to the fore. In sum, the problems raised in questions (1) and (2) above can be reduced to the following: What knowledge of language a language user must possess to suit language to context and context to language.

The third dimension of relational pragmatics e.g., the interrelation between Interactant and Context (on the cognitive plane) is perhaps the most neglected aspect of cognitive pragmatics. The questions that should be asked in connection with this dimension of RP include the following: (1) How the internalized contextual knowledge of language use can influence (a) the other cognitive faculties of the language user i.e., perception, reasoning, pragmatic interpretation, the definition of the situation, etc., (b) attitudes and affect and (c) his/her behavior (in general); (2) How can a language user's cognitive skills influence context perception and interpretation? (3) How is pragmatic context acquired and stored in the brain? And finally (4) What are the characteristic features of contextual pragmatic knowledge?

The interdependency between the language user and his/her contextual knowledge of language use is as complex as those of the two dimensions presented earlier. This is so because the internalized contextual (pragmatic) knowledge of an interactant has a cognitive, affective, social and communicative interface (for a summary of research on language in a broad psychosocial perspective cf. Giles -- Robinson 1990).

It should be stressed in connection with the problems formulated in questions (1) and (2) above that contextual pragmatic knowledge stored in knowledge structures is interconnected with other cognitive and linguistic systems that cooperate in the process of the acquisition, retrieval and change of that knowledge. Furthermore, one might suggest that the human knowledge of any kind (including that of contextual language use i.e., about social situations (cf. Argyle et al. 1981), people, physical environment, etc.) seems to be frequently associated with a specific affect or attitude that affects language use. Affect and attitudes, in turn, can bias perception and reasoning, pragmatic interpretation, etc. (cf. Frijda 1988, and Oatley -- Jenkins 1992).

Regarding question (2) above one may briefly comment that context perception and interpretation, and the use of language in general, is guided (or regulated) by a cooperative effort of many nonlinguistic cognitive systems (i.e., perception, reasoning, affect etc.) that exert their influence on linguistic processes both at the level of production and comprehension of language.

In view of what has been said on relational pragmatics so far, it seems easy to predict that in the controversy concerning modularity vs. non-modularity of pragmatics (cf. Fodor 1983; Chomsky 1984, and Sperber -- Wilson 1986), e.g. the question of the existence or nonexistence of an independent pragmatic (module) system, the latter assumption seems to be more promising because the non-modular approach has more explanatory power, for instance, the interrelations between pragmatic knowledge represented in cognitive structures and other cognitive systems show that they are open systems (cf. Rokeach 1960). Furthermore, and what is more important, the non-modularity of the pragmatic system seems to be essential to the possibility of accounting for the processes at the interface of competence and performance.

Concerning questions (3) and (4) e.g., the problem of acquisition and specific features of "pragmatic knowledge" (the concept of pragmatic knowledge will be assumed to refer to the contextual knowledge of language use) the following salient features of the pragmatic system (especially in contrast to the grammatical system) should be noticed: (a) incompleteness, (b) uncertainty and (c) metastability.

The incompleteness of the acquired pragmatic knowledge may be related to two problems: (1) the incompleteness of the acquisition of a specific language use, e.g. in a social situation such as formal reception and (2) the incompleteness of the acquisition in relation to all possible uses of language in a specific society. Usually, a language user masters a limited number of contextual uses of a language to a certain degree. There are conspicuous differences among interactants in the scope of their pragmatic knowledge. The interactants are usually aware of their deficiencies in the contextual mastery of language (which depends on many social skills and social knowledge of the interactant) i.e., giving a speech, chairing a meeting, giving an interview or talking to a person of a high social status, etc.

A question closely related to that of incompleteness of acquisition is that of uncertainty: the outcome of a social encounter is uncertain because (1) the number or scope of the relevant contextual features is unpredictable, and (2) the contextual configuration of features will differ from situation to situation (it should be stressed that the configuration of features depends on its accessibility as well as memory, affect, conation (e.g., the user's goal and motivation) and many other cognitive aspects of a language user's knowledge of language and its processing). One might say that from the cognitive point of view identical (isomorphic) linguistic situations or contexts do not exist (unless one insists on some idealized abstract constructs that represent linguistic situations). Uncertainty will be assumed to be a defining feature of cognitive processes and pragmatic knowledge.

Finally, the third defining feature of the pragmatic knowledge of language is associated with the idea of metastability (cf. Stengers - Prigogine 1984) e.g., a temporary stability of a system. The system of a language user's pragmatic knowledge will be assumed to be metastable e.g., liable to change in time. Metastability of a system is a matter of degree. Three subsystems e.g., (1) Interactant, (2) Context and (3) Language, make up the system of Relational Pragmatics. The linguistic or grammatical system seems to be the most stable, certain and complete, although as all students of linguistics know very well synchronic linguistic variation may lead to gradual or abrupt changes at different levels of linguistic structure. In comparison with grammatical metastability the pragmatic knowledge of language (or knowledge of pragmatics) is incomplete, uncertain and unstable to a great degree. The intertactant's mental system is also metastable and liable to change (cf. Kopytko, in preparation).

2.2. Towards a theory of language user/interactant

Unfortunately, it can hardly be denied that terms such as pragmatics, context and language user belong to the vaguest and imprecise concepts in theoretical pragmatics. The aim of the following sections will be to present a view of language user/interactant, context and pragmatics in the framework of Relational Pragmatics.

The basic question associated with the notion of the language user seems to be that of knowledge relevant to social interaction. Thus, Chomsky's (1965) speaker-herarer was endowed with the tacit knowledge of language. A pragmatician following Chomsky's tradition might claim that the language user is endowed with a tacit (?) knowledge of appropriate language use. However, in connection with that postulate a number of problems arise, especially about a view of pragmatic knowledge delineated above, i.e. characterized in terms of its incompleteness, uncertainty and metastability. Moreover, the social aspects of language use that are conditioned by social structure (and the dynamic processes within it, e.g. social struggle) usually make the notion of appropriateness in the sociopragmatic context inappropriate and questionable, especially in view of the social postulates of equality or justice. On the other hand, the forces of social control exert pressure on the nonconformists to comply with the "linguistic order" of society. The notion of "appropriate language use" implies, among other things, a body of social and psychological knowledge that underlies a language user's verbal behavior. This knowledge could be analyzed not only with psychosocial variables such as power -- (P), social distance -- (D), intimacy -- (I), goal -- (G) or affect -- (A), but also as cognitive-affective states, i.e. understanding, empathy, compassion, etc. (cf. Kopytko 1993a and in preparation). Furthermore, the idea of "appropriate language use" implies not only the user's knowledge of the "other" and context, but also self-knowledge, e.g. the user's self-concept (cf. Marcus -- Wurf 1987). It seems rather obvious that the interactant usually knows who she/he is and what her/his relation (e.g., status, position, social role, etc.,) to the "other" in a social situation is. If this is not the case, the interactants usually negotiate (or fight) their relative status in an interaction.

Strangely enough, the concept of self-knowledge has not so far been employed explicitly in any approach to language use. In Relational Pragmatics the concept of user/interactant will be analyzed in terms of the sociocognitive notion of the self and its properties (for a discussion of the self-concept cf. Banaji -- Prentice 1994).

The self-system may be represented as an integrated cognitive-affective-conative system, the latter associated with motivations and goals, (for the tripartite division of the self see Hilgard 1980 and Kopytko, in preparation). Nota bene, the three integrated parts of the self seem to be correlated on the linguistic plane with the three basic functions of language, e.g. cognitive (representative), emotive and conative (cf. Buhler 1934; Jakobson 1960, and Lyons 1977). In this connection, one might suggest that language is acquired in its cognitive, affective and conative context. If this is so, there are some practical consequences of such a state of affairs for language use and interpretation: for instance, the user will be assumed to possess the knowledge of the cognitive, affective and conative language use (the knowledge of conative language use is usually closely associated with the strategic use of language, i.e., social influence, persuasion, argumentation, etc.). It should be emphasized that the three u ses of language frequently overlap and can hardly be separated one from another.

The claim that language is used in its cognitive, affective and conative context deserves some more attention and explanation. In Relational Pragmatics the cognitve context of language use will make possible an analysis of language in relation to (1) other nonlinguistic cognitive structures i.e., knowledge structures (and their accessibility to a language user), (2) reasoning (or rationality) associated with language use, (3) dynamic processes of information processing and (4) processes related to other cognitive faculties, i.e. thinking, perception or attention.

Similarly, the affective context regulates language use to a considerable degree. The affective context is associated with the user's personality (i.e., extrovert vs. introvert), attitudes, needs, etc. On the other hand, the conative context can be viewed as the language user's motivations and goals that underlie the strategic (and probably the communicative) use of language.

In brief, one might conclude that the self-system viewed as an integrated cognitive-affective-conative system makes the internal (mental) context of language use. Rationalistic approaches to pragmatics (cf. Kopytko 1995), i.e. Brown and Levinson (1987), Grice (1975) or Kasher (1991), postulate a very narrow form of what has been referred above as the internal context of language use, namely, the rationality principle. Brown and Levinson (1987: 58) suggest that a Model Person (MP), e.g. a fluent speaker of a natural language is "endowed with two special properties - rationality and face." Rationality is seen as "a mode of reasoning from ends to the means that will achieve those ends", and face is defined as "the want to be unimpeded and the want to be approved of in certain respects."

The self-system in Relational Pragmatics also contains the concepts of rationality and face (cf. Kopytko 1993b), the former in the cognitve and the latter in the affective system of the self (e.g., the language user's needs). However, the meaning and function of these constructs in RP are different. In RP they are seen as pragmatic notions whose basic properties are the following: (1) incompleteness, (2) uncertainty and (3) metastability. As a consequence, their function (role) in language use is less categorical and has to be, in principle, contextually evaluated. That is, the rationality of action (cf. Kopytko 1995), or of human behavior in general, including verbal behavior, should be contextually bounded rather than viewed as an instance of abstract decontextualized reasoning. From the point of view of RP decontextualized reasoning is theoretically impossible because all mental processes are embedded in their contexts, i.e. the integrated cognitve-affective-conative system.

In conclusion, pragmatic phenomena that involve abstract constructs such as rationality, face or the self-concept seem to require a lower level of analysis than that associated with the general or universal aspects of human cognition and behavior (i.e., thinking, reasoning, linguistic universals, etc.). For easy reference it will be useful to label the latter as the categories of Level 1, and the former as members of Level 2 (to which language as competence belongs). From a theoretical point of view, it is important to notice that Level 2 makes an interface for performance phenomena, i.e. linguistic performance, human interaction, etc., e.g. the phenomena of Level 3 (or social level). The investigation of phenomena and processes at Level 3 will make a fascinating study of language use, interpersonal and intergroup communication (interaction) and a host of social phenomena related to language use. A theory of the language user as an integrated cognitive-affective-conative self-system presented above, may prove to be a useful tool for the analysis of performance phenomena at Level 3, (or to be more specific the interrelations between the categories and processes of Level 2 and 3). On theoretical grounds, Level 3 may be seen as capable of providing empirical evidence, e.g. corroboration or falsification (cf. Popper 1963, 1972) for the phenomena and processes on Level 2. In other words, the mental vs. social interface offers an escape from rationalistic theorizing to an empirical endeavor. It should be kept in mind, however, that studies of mental phenomena, i.e. their organization and structure (especially of the entities of Level 1 and 2) belong to the most speculative aspects of the study of the human mind, and are most likely to remain so in the years to come.

2.3. The knowledge of context

It has been assumed above that the internalization of conditions on appropriate language use, i.e. the acquisition of pragmatic competence implies the existence of some sort of contextual knowledge that is represented in knowledge structures, e.g. frames, scripts, plans, etc. It has also been suggested that contextual knowledge is usually incomplete, uncertain and metastable. One should notice, however, that there are some areas of pragmatics, e.g. indexical pragmatics, that allow for a certain degree of formalization (which is an exception rather than the rule in pragmatics). One of the consequences of the approach to pragmatics in the framework of RP is a qualitative difference between the conceptualization of linguistic competence (cf. Chomsky 1965) and the pragmatic competence in RP. The former may be viewed as relatively complete, certain and stable (concerning linguistic structure), while the latter may be characterized as incomplete, uncertain and metastable. These features of pragmatic phenomena and p rocesses can be represented (or reflected) on the cognitive plane as mental models (cf. Johnson-Laird 1983; Holland et al. 1986).

A view of pragmatic competence as a qualitatively equivalent: to linguistic competence that may occasionally be distorted by performance phenomena has been rejected here in favor of the claim that the imperfect and incomplete acquisition of contextual knowledge and the three defining features of pragmatic context (e.g., incompleteness, uncertainty and metastability) are phenomena sui generis associated with pragmatics. This is so because (1) the scope of the relevant context is usually unpredictable, (2) context is a dynamic phenomenon that may be constructed (regulated) by interactant, and (3) no two contexts or situations can be claimed to be identical (this is also true of the mental context discussed above).

One might claim that the pragmatic competence of a language user can be analyzed in terms of his/her knowledge of contextual language use. In other words, the knowledge of the context of use is the sine qua non of pragmatic competence. In Relational Pragmatics which aspires to account for both competence and performance phenomena, the scope of context is broad. Basically, pragmatic context can be categorized as (1) internal vs. external, (2) linguistic vs. nonlinguistic, (3) the context of speech production vs. context of speech interpretation and (4) the context of interaction (responsible for the emergent contextual phenomena in verbal interaction). It is the internal context, e.g. a language user's knowledge of pragmatic context, that is directly relevant to pragmatic competence. In RP the internal context will be viewed as a global internalized knowledge of language use interrelated with the self-knowledge (self-concept) analyzed as an integrated cognitive-affective-conative system. Such an interrelation makes possible the internal (mental) interface between competence vs. performance phenomena and between linguistic vs. nonlinguistic context. The external context (physical or social), the context of (1) language use and (2) language interpretation, and the context of interaction belong to the study of performance (communication). (These will be discussed in detail in a separate publication on performance phenomena in pragmatics.)

3. Conclusion

In view of what has been argued for so far, pragmatic competence may be represented as a three-dimensional, interactive system analyzable in terms of relations between Interactant (I), Language (L) and Context (C), i.e. (1) I [left arrow] [right arrow] L, (2) L [left arrow] [right arrow] C, and (3) I [left arrow] [right arrow] C (see fig. 1). Morris's (1938) vague definition of pragmatics as the relation between signs and their users could be redefined more accurately as the relation between signs (language) and the language user's knowledge of contextual language use associated with the internal context e.g. the knowledge of language interrelated with the self-knowledge represented as an integrated cognitive-affective-conative system.

So far, a narrow view of pragmatics analyzed as pragmatic competence has been discussed. Relational pragmatics aspires to go beyond the theory of linguistic (pragmatic) competence and investigate the phenomena at the competence-performance interface (including an analysis of the actual use of language in its natural context). Such a study may also prove to be very important to the investigation of pragmatic competence (especially for its scope), because it seems that many of the phenomena that are associated with the appropriate use of language (e.g., pragmatic competence) reside somewhere in the pragmatic context (that is created in the process of human interaction) rather than in the linguistic cognitive structures to which it orthodoxically belongs. In addition, the study of performance phenomena (e.g., human communication) is crucial to "humanistic pragmatics", dedicated to defining and solving numerous social problems associated with language use, i.e. social inequality, interpersonal or intergroup confl ict (cf. Grimshaw 1990), political correctness, etc.

Finally, it is important to highlight the differences between the present approach to pragmatic competence and what will be referred to as the "standard view" associated with the Chomsky -- Kasher view of pragmatics (cf. Chomsky 1984; Kasher 1994a) and their followers.

Firstly, in contrast to the "standard view", Relational Pragmatics is viewed as a non-modular pragmatic system (cf. Leech 1983; Sperber -- Wilson 1986). The system is endowed with an interface with other nonlinguistic cognitive structures, i.e. knowledge structures (indispensable for the production and interpretation of meaning) and the self-knowledge that constrains and regulates the use of language. These functions of self-knowledge are associated with the concepts of Level 2, e.g. cognition (perception), conation (goals and motivations) and affect (positive or negative) that is usually involved in the cognitive and conative processes. Similarly, the acquisition and use of language seem to be frequently (or perhaps always) associated with a specific affect which, in turn, depends on the internal (mental) and external (social or physical) context of language acquisition or use. In brief, Interactant (I) (or rather the self on the mental plane) is analyzed as an interrelated cognitive-affective-conative syste m that makes the internal context for language use. The internal context seems to be essential to language use and interpretation because it adds one more layer of contextual meaning that will be referred to as the metapragmatic meaning that is abducted (e.g., inferred as a hypothetical abduction) from intertactant's goals, motivations, attitudes, features of language use (i.e., educated vs. the non-educated register of language, etc.).

Secondly, RP is presented as a system of interdependencies analyzed in terms of the relations between the Interactant (I), Language (L) and Context (C).

Thirdly, pragmatic competence (in RP) in contrast to grammatical competence is characterized as incomplete, uncertain and metastable.

Fourthly, RP postulates an interface between the language user's mental and social context. This interface accounts, among other things, for the phenomena and processes associated with language acquisition and use.

Fifthly, RP attempts to study pragmatic phenomena holistically e.g., within a pragmatic and metapragmatic context (which is analyzed for the theory of pragmatic competence in RP as the interrelated (nonlinguistic) cognitive-affective-conative system).

In conclusion, a holistic view of pragmatic analysis suggested above requires further elaboration, refinement and critique. A theory of language use (or linguistic interaction) remains to be formulated. Such a theory would have to incorporate a theory of the language user and extend its scope to account for complex pragmatic and psychosocial phenomena taking place between interactants in real speech situations.

REFERENCES

Argyle, Michael - Adrian Furnham - Jean Ann Graham

1981 Social situations. Cambridge: CUP.

Asher, R. E. (ed.)

1994 The encyclopedia of language and linguistics. 10 vols. Oxford: Pergamon Press.

Banaji Mahazarin R. - Deborah A. Prentice

1994 "The self in social context", Annual Review of Psychology 45: 297-332.

Berger, Peter - Thomas Luckmann

1966 The social construction of reality. Garden City, NY: Doubleday.

Bertalanffy, Ludwig von

1956 "General systems theory", General Systems 1: 1-10.

Brown, Penelope - Stephen C. Levinson 

1987 Politeness: Some universals in language usage. Cambridge: CUP.

Buhler, Karl

1934 Sprachtheorie. Jena: Fischer.

Chomsky, Noam 

1965 Aspects of the theory of syntax. Cambridge, MA: The MIT Press.

1981 Lectures on government and binding. Dordrecht: Foris Publications.

1984 Modular approaches to the study of the mind. San Diego, CA: San Diego State University Press.

Cole, Peter - Jerry L. Morgan (eds.)

1975 Speech acts. New York: Academic Press.

Cicourel, Aaron

1973 Cognitive sociology. Harmondsworth: Penguin.

Dik, Simon

1989 The theory of functional grammar. Dordrecht: Foris.

Duranti, Alessandro - Charles Goodwin

1992 Rethinking context: Language as an interactive phenomenon. Cambridge: CUP.

Fodor, Jerry

1983 The modularity of mind. Cambridge, MA: The MIT Press.

Fodor, Jerry - Ernest Lepore 

1991 Holism. Oxford: Blackwell.

Frijda, Nico H.

1988 The laws of emotion. American Psychologist 5: 349-358.

Giles, Howard - Peter W. Robinson (eds.)

1990 Handbook of language and social psychology. New York: Wiley.

Glover, Kelly D.

1995 "A prototype view of context and linguistic behavior: Context prototypes and talk", Journal of Pragmatics 23: 137-156.

Grice, H. Paul

1975 "Logic and conversation", in: Peter Cole - Jerry L. Morgan (eds.), 41-58.

Grimshaw, Allen D. (ed.)

1990 Conflict talk. Cambridge: CUP.

Halliday, Michael A. K.

1985 An introduction to functional grammar. London: Arnold.

Higgins, E. Tory - Richard M. Sorrentino (eds.)

1990 Handbook of motivation and cognition: Foundations of social behavior. Vol. 2. New York: The Guilford Press.

Hilgard, Ernest

1980 "The trilogy of mind", Journal of the History of Behavioral Sciences 16: 107-117.

Holland, John H. - Keith J. Holyoak - Richard E. Nisbett - Paul A. Thagart

1986 Induction. Cambridge, MA: The MIT Press.

Jakobson, Roman 

1960 "Linguistics and poetics", in: Thomas A. Sebeok (ed.), 350-377.

Johnson-Laird, Philip N.

1983 Mental models. Cambridge, MA: Harvard University Press.

Kasher, Asa

1991 "On the pragmatic models: A lecture", Journal of Pragmatics 16: 381-397.

1994a "Chomsky and pragmatics", in: R. E. Asher (ed.), 535-536.

1994b "Pragmatics and rationality", in: R. E. Asher (ed.), 3280-3282.

Kendon, Adam - Richard M. Harris - Marry Richie Key (eds.)

1975 Organization of behavior in face-to-face interaction. The Hague: Mouton.

Kopytko, Roman

1993a Polite discourse in Shakespeare's English. Poznan: Adam Mickiewicz University Press.

1993b "Linguistic Pragmatics and the concept of face", Vienna English Working Papers 2: 91-103.

1995 in preparation "Against rationalistic pragmatics", Journal of pragmatics 23: 475-491.- Conflict discourse.

tion

Krippendorff, Klaus

1993 "The past of communications hoped-for future", Journal of communication 43, 3: 34-44.

Leech, Geoffrey

1983 Principles of pragmatics. New York: Longman.

Levinson, Stephen

1983 Pragmatics. Cambridge: CUP.

Lyons, John

1977 Semantics. Cambridge: CUP.

Margenau, Henry (ed.)

1972 Integrative principles of modern thought. New York: Gordon and Breach.

Marcus, Hazel - Elissa Wurf

1987 "The dynamic self-concept: A social psychological perspective", Annual Review of Psychology 38: 299-337.

Mey, Jacob L.

1993 Pragmatics: An introduction. Oxford: Blackwell.

1994a "Pragmatics", in: R. E. Asher (ed.), 3260-3277.

1994b "How to do good things with words: A social pragmatics for survival", Pragmatics 4, 2: 239-263.

Morris, Charles H.

1938 Foundations of the theory of signs. Chicago: Chicago University Press.

Nuyts, Jan

1992 Aspects of a cognitive-pragmatic theory of language. Amsterdam - Philadelphia: John Benjamins.

Oatley, Keith - Jennifer M. Jenkins

1992 "Human emotions: Function and dysfunction", Annual Review of Psychology 43: 55-85.

Popper, Karl

1940 "What is dialectic?", Mind 49.

1963 Conjectures and refutations: The growth of scientific knowledge. London: Routledge and Kegan Paul.

1972 Objective knowledge. Oxford: Clarendon.

Reber, Arthur

1985 The Penguin dictionary of psychology. London: Penguin.

Rokeach, Milton

1960 The open and closed mind: Investigations into belief systems and personality systems. New York: Basic Books.

Rosch, Eleanor 

1978 "Principles of categorization", in: Eleanor Rosch -- Barbara B. Lloyd (eds.), 27-48.

Rosch, Eleanor - Barbara B. Lloyd (eds.)

1978 Cognition and categorization. Hillsdale, NJ: Erlbaum.

Schank, Roger - Robert Abelson

1977 Scripts, plans, goals and understanding: An inquiry into human knowledge structures. Hillsdale, NJ: Erlbaum.

Scheflen, Albert E.

1975 "Models and epistemologies in the study of interaction", in: Adam Kendon - Richard M. Harris - Marry Richie Key (eds.), 63-91.

Sebeok, Thomas A.

1960 Style in language. Cambridge, MA: The MIT Press.

Sinclair, Melinda

1995 "Fitting pragmatics into the mind: Some issues in mental pragmatics", Journal of Pragmatics 23: 509-539.

Sperber, Dan - Deirdre Wilson

1986 Relevance: Communication and cognition. Oxford: Blackwell.

Stengers, Isabelle - Ilya Prigogine 

1984 Order out of chaos. New York: Bantham.

Taylor, Alastair M.

1972 "Integrative principles in human societies", in: Henry Margenau (ed.), 211-289.

Verschueren, Jef - Jan-Ola Ostman - Jan Blommaert (eds.)

1994 Handbook of pragmatics: Manual. Amsterdam - Philadelphia: John Benjamins.

Vygotsky, Lev S.

1962 Thought and language. (Edited and translated by E. Hanfmann and G. Vakar.) Cambridge, MA: The MIT Press.

Whorf, Benjamin L.

1956 Language, thought and reality: Selected writings of Benjamin Lee Whorf. (Edited by J. B. Carroll.) New York: Wiley.