Foundations of Speech Act Theory Philosophical and Linguistic Perspectives

S L T Tsohatzidis, ed.

Asa Kasher

Modular Speech Act Theory:

Program and Results

Modular Pragmatics

Modular pragmatics is a research program. Every research program is a continuous series of theories that share a conception of objectives, a philosophical "hard core" and an appropriate methodology. Generative linguistics, as it has been portrayed by Chomsky, is a research program that has had syntax in it focus.(1) Truth-conditional theories of meaning, as proposed and defended by Davidson, form another research program, one that has had semantics in its focus(2)Modular pragmatics is meant to be a research program that applies to the field of pragmatics the general modular approach to the study of language and cognition, which in turn is a part of the general cognitive approach to the study of the human mind.(3)

According to the cognitive approach to the study of any competence of the human mind, be it vision, imagery, reasoning or natural language use, the main theoretical objective of such a study is delineation and explanation of that competence within the conceptual framework of cognitive studies.

Within this framework, a human competence is a system of knowledge that is governed by a characteristic set of principles.(4)The system operates as a representation processing device that embodies these principles. This processing device itself is embodied in the human brain.

A delineation and explanation of a competence of the mind, within such a cognitive conceptual framework, takes the form of a theory that answers questions of the following forms:

(C.1) What is the domain of competence C? (C.2) What is the system of abstract principles that govern the system of knowledge that constitutes competence C? (C.3) What is the operating system that embodies these principles of competence C and employs them? (C.4) What is the neural embodiment of the abstract principles of competence C and its processing device? (C.5) What is the adjacent system of growth, development or acquisition that enables competence C to arise in a normal person?(5)

Whether a study of pragmatics is a study of a particular human competence or of a certain system of closely related human competences or of some assortment of human competences, a cognitive study of pragmatics is meant to delineate and explain a system or systems of pragmatic knowledge by a theory that answers the questions of these forms, C.1 to C.5, with respect to a particular competence or some particular competences. If pragmatics is, roughly speaking, the study of language use,(6) then cognitive studies of pragmatics, cognitive pragmatics for short, is a research program whose theoretical objectives are delineation and explanation of some system or systems of knowledge of language use, within the conceptual framework of cognitive studies. Theories that are formulated, investigated and improved within this research program are expected to eventually provide us with adequate answers to questions of the forms C.1 to C.5, about the domain of each system of pragmatic knowledge, its principles, its operation, its embodiments and its arising.

Thus portrayed, cognitive pragmatics shares with other cognitive studies of language as well as with various branches of cognitive science some important features.

First, the conceptual framework pictures every human competence as a four-tier system of knowledge. There is a top stratum of abstract principles. Then, the system of abstract principles is embodied in an underlying stratum that operates as a representation processing device. This device is, in turn, embodied in the brain, the third stratum within this picture. This is a three-tier system that results from the operation of a fourth tier, one that enables the person to reach mature mastery of the competence.

Secondly, these four-tier systems of knowledge show a most important top-down independence: every stratum of the system is independent of the underlying strata. Consider a computer D that runs a program that computes the value of to the n-th digit, for any given n. A full understanding of D competence to compute involves a four-tier system of the above-mentioned form. The abstract, mathematical stratum involves the definition of and a proof that a certain function correctly specifies, given n, the first n digits of . Obviously, this stratum is independent of all the other strata: the flowchart of the program, the underlying electronic system, and the compilation process of the program. Similarly, the flowchart is independent of the structure and operation of the chips of the underlying electronic stratum. In the same vein, the abstract stratum of the human competence of counting and computing is independent of the strata of embodiment in the mind and in the brain. Other human competences, such as language use, reasoning and face recognition, show the same form of top-down independence. Indeed, no bottom-up independence of any stratum of upper strata is expected to appear, since no embodiment of a principle or a device can in any sense be independent of it.(7)

Modular pragmatics is a thread of cognitive pragmatics. It is a research program of the same theoretical objectives, namely, delineation and explanation of some system or systems of knowledge of language use. However, within the conceptual framework of cognitive pragmatics, which is the general conceptual framework of cognitive studies, modular pragmatics is an attempt to portray the whole spectrum of the human knowledge of language use in terms of systems of knowledge of two types only: modular systems and central systems.(8) Since the notions of a modular system and of a central system are playing a major role in modular pragmatics, we move now to a brief clarification of each of them.

Traditional views of the mind portrayed it as "entirely indivisible": "For, as a matter of fact," said Descartes, "when I consider the mind, that is to say, myself inasmuch as I am only a thinking thing, I cannot distinguish in myself any parts... And the faculties of willing, feeling, conceiving, etc. cannot be properly speaking said to be its parts, for it is one and the same mind which employs itself in willing and in feeling and understanding."(9) A modular approach to the study of the mind portrays the mind as entirely divisible into systems of two types: special faculties, modular systems of knowledge, on the one hand, and general purpose systems, central systems of knowledge, on the other hand.

The notion of modular system was explicated in Fodor (10). Input systems, such as language and vision, are modular on that account, since they possess most or all of certain properties:

(F.1) they are domain specific; (F.2) their operation is mandatory; (F.3) there is only limited central access to the representations that they compute; (F.4) they are fast; (F.5) they are "informationally encapsulated" in the sense of having significantly constrained access to information present in the mind; (F.6) they have outputs that are "shallow" in the sense of encoding significantly constrained information; (F.7) they are associated with fixed neural architecture; (F.8) they exhibit characteristic and specific breakdown patterns; (F.9) their ontogeny exhibits a characteristic pace and sequencing.

At an early stage of modular pragmatics we showed(11)that: (MP.1) If (1) by "pragmatics" we mean a system of knowledge of language use that is reflected in the major facts with respect to (i) forces of speech acts; (12) (ii) performatives; (iii) deixis; (iv) lexical pragmatical presuppositions; (v) conversational implicatures; (vi) politeness principles; (13) and (2) by a "modular system" we mean a system that satisfies conditions (F.1) to (F.9), then pragmatics does not constitute a modular system. At that stage it was natural to replace the issue of the modularity of pragmatics by a family of issues of the modularity of some components of pragmatics, directly related to phenomena of language use drawn from the list (i) to (vi) or other ones.(14) Although each of these components of pragmatics is related to a feature of language use, each of them is related to what seems to be, in some sense, a separate facet of speech activity in context. There is no reason to assume that they are all mutually dependent on each other or even closely related parts of one cohesive system. Pragmatic phenomena such as forces of speech acts, deictic expressions and conversational implicatures have each been the subject matter of an apparently independent branch of pragmatics. The question, therefore, naturally arises, whether the apparent mutual independence of these branches of pragmatics is a result of some methodological decision to discuss phenomena in abstraction from their interrelations or is it rather a manifestation of an underlying substantive independence of each of the different pragmatic phenomena. Naturally, the issue of the substantive independence of some components of pragmatics can take the form of the problem of the modularity of the related systems of pragmatic knowledge.

However, the relation between the intuitive notion of substantive independence and the technical notion of modularity is at least slightly more complicated than it seems. Whereas conditions (F.1) to (F.9) of modularity all specify highly interesting features of cognitive systems, not all of these features indicate substantive independence in any ordinary sense of the term. For instance, there is no reason to assume that substantively independent systems of knowledge have to satisfy Fodor's condition (F.2) of being "constrained to apply whenever they can apply" (15), unlike input systems, whose operation is mandatory.

We, therefore, propose that an alternative notion of modularity will be used within cognitive theories when the issue under consideration is that of substantive independence of a competence or a system of knowledge. Let us dub a system of knowledge that satisfies Fodor's conditions (F.1) to (F.9) an F-module, and refer to a system of the proposed alternative type as a P-module. The alternative notion rests on the above-mentioned questions, of forms (C.1) to (C.5), which a theory has to answer for it to be an adequate delineation and explanation of a competence C of the mind. Intuitively speaking, a competence is shown to be substantively independent when the related questions are answered in a uniquely characterizing way. Accordingly, a P-module will have its independent (P.1) domain (P.2) theoretical principles; (P.3) information processing; (P.4) neural embodiment; (P.5) acquisition process A more accurate specification of conditions (P.1) to (P.5) of P-modularity will clarify for each of them the relevant sense of independence. Some of the features of F-modularity will play a major role. For instance, conditions (F.5), which requires that the system should be informationally encapsulated, and (F.6), which requires that the system should produce shallow outputs, form part of a theoretical clarification of (P.3). An advanced theory, within the framework of the research program of modular pragmatics, will rest on such detailed clarification of each of the conditions of P-modularity. On grounds of such a clarification, one could tackle the problem of whether a certain system of pragmatic knowledge is a P-module or not, as well as more general problems, such as how many pragmatic P-modules are there, and why.(16) A detailed example will be discussed in the next section.

Central systems are interestingly different from modular ones. According to Fodor's suggestions and arguments the major distinction is that modular systems are domain specific (P.1) and informationally encapsulated (P.3), while central systems are domain inspecific and informationally unencapsulated.(17) A system of knowledge for which (P.1) and (P.3) do not hold, can still have its own theoretical principles, neural embodiment and acquisition process. It is, therefore, reasonable to draw the distinction between these two types of system in terms of conditions (P.1) and (P.3).

We assume that a central system of knowledge is governed by abstract, general principles that apply to different domains and that its embodiment takes the form of a representation manipulation device that has access to the central store of beliefs held by the person at the time. Under this notion of a central system of knowledge, it can be shown that:

(MP.2) Conversational implicatures are generated by a central system of pragmatic knowledge.

In an earlier study, we showed that all the Gricean super-maxims and maxims that generate conversational implicatures are derivable from general rationality principles of intentional activity, ones that apply not only to speech but also to painting.(18) The system of knowledge that governs conversational implicature is, therefore, domain inspecific.

Grice pointed out that different kinds of data are used for the generation of conversational implicatures, including "the context, linguistic or otherwise, of the utterance" and "other items of background knowledge".(19) Consideration of certain types of conversational implicature, such as ironical interpretation of utterances in appropriate contexts,(20) clearly shows that any item of background knowledge can be used in working out an implicature. The system of knowledge that governs conversational implicature is, therefore, also informationally unencapsulated. Thus, this system of pragmatic knowledge is a central one.

Modular Speech Act Theory

Modular speech act theory is that part of modular pragmatics that focuses on speech acts. Accordingly, modular speech act theory is a research program. Its theoretical objectives are delineation and explanation of some system or systems of knowledge of speech act use, in both production and understanding. Within the conceptual framework of modular cognitive pragmatics, modular speech act theory tries to depict human knowledge of using speech acts in appropriate contexts in terms of modular systems and central ones. In this section we briefly present and discuss few steps that have been made within this research program and point out some possible future evolvements.

Frege's theory of meaning included in addition to sense and reference two other ingredients, namely, force and tone.(21) The indication of force is an indication of some type of speech act, such as making an assertion, posing a question, issuing a command or expressing a wish. A theory of force, on such an account, is a part of a theory of meaning, but for our present purposes we do not have to commit ourselves to any precisely drawn distinction between semantics and pragmatics. We do, however, adopt the idea that a theory of force has to do with the given variety of types of speech act.

Enumeration of the types of speech act that are performed in English by an appropriate utterance of a first person singular present indicative form is a list "of the order of the third power of 10" by Austin's estimate(22), but such an enumeration won't have any explanatory power. Taxonomies of speech act types provide us with only quite shallow generalizations, such as "every speech act is verdictive, exercitive, commissive, behabitive or expositive", to use Austin's classification(23). Such classifications are also of no explanatory power.

Within the framework of cognitive pragmatics, the theoretical objective of the research program is not to specify a family of verbs present in a dictionary that reflects, say, the current usage of certain words in expressions by members of the group of speakers of the cultural entity called, say, Modern British English. Cognitive studies of language do not address cultural entities of the nature of Modern British English, but rather cognitive entities, idiolects that persons have in their minds and brains.(24) Accordingly, cognitive pragmatics does not aim at any enumeration or classification of verbs and correlative speech act types of languages such as Old English or Modern Hebrew, which are cultural entities, but rather at an adequate delineation and explanation of the class of cognitively possible speech act types, i.e., types of speech act whose existence is compatible with the cognitive constraints that are imposed on human linguistic activity by the human cognitive system or systems of pragmatic knowledge.

A theory of speech act force, within the conceptual framework of cognitive pragmatics, is descriptively adequate and explanatorily powerful, if it shows how the class of the cognitively possible speech act types is delineated in terms of:

(SAT.i) A general conception of speech act type as a rule governed practice whose system of rules satisfies a restricted class of conditions; (SAT.ii) a restricted class of cognitively possible basic speech act types and (SAT.iii) a restricted class of basic amplifications of speech act types, that is to say, operations on speech act types functions) that generate cognitively possible speech act types systems of rules) when applied to cognitively possible speech act types (systems of rules).

Searle's essay on speech acts(25) made several steps towards a general conception of speech act type, as required by (i). The most important contribution of that essay was the deep insight into the nature of the systems of rules that govern speech acts: (MP.3) Human speech acts are governed by systems of rules that are constitutive. (26)

Of a lesser significance has been the classification offered by that essay of the rules that govern speech acts of a particular type. We argued(27) that such systems of rules have a structure that reflects their being systems that govern intentional acts: (MP.4) Human speech acts are governed by systems of rules that specifyfor each speech act type T:

(i) The literal ends of acts of type T;(28) (ii) the verbal means that may be used for obtaining those ends; (iv) (iii) the conditions that a person has to satisfy in order to play the role of the agent of an act of type T; the literal product of an act of type T.

Preliminary work has also been done in an attempt to delineate the class of basic speech acts. A natural distinction has been drawn between basic speech acts and things done with words.(29) Roughly speaking, the former are speech acts of types whose knowledge is part of one's knowledge of language itself, whereas the latter are speech acts of types whose knowledge requires more than the knowledge of language. Examples of the latter kind abound: "I name this ship the Queen Elizabeth," uttered while smashing an appropriate bottle against the stern of the ship, is a famous example. A speaker of any idiolect of English is not required to know qua speaker that this is how ships are named. One's knowledge of the practice of naming ships transcends one's knowledge of language, though the practice of ship naming involves an utterance of an appropriate sentence. We have used an argument that had been put forward by Dummett for other purposes to show that assertion is a basic speech act type. There are reasons to assume that question and command are also basic in the same sense and perhaps request too.(30) Our working hypothesis is that (MP.5) Basic speech act types are identifiable by syntactic or intonational features of the related sentences.

Work towards delineation of the class of basic amplifications of speech act types has been scarce, but some parts of what has been done in Searle's and Vanderveken's more recent works seems useful in this respect.(31) For instance, one component of the suggested analysis of the notion of illocutionary force is degrees of strength by which a distinction is drawn between similar speech act types such as request and supplication.(32)

Thus far speech act force theory has been sketched within the conceptual framework of cognitive pragmatics. Within the more specific conceptual framework of modular pragmatics, a theory of speech act force will address issues of P-modularity and centrality of related systems of knowledge. There are three different levels on which such issues could arise. On the most general level we have all speech act types under consideration, whether they are basic ones or things done with words. On the least general level we have certain types of speech act under consideration, one at a time.

What seems to us to be the most interesting level on which issues of P-modularity and centrality arise is an intermediate level, where under consideration we have (i) some system of pragmatic knowledge of language use that involves only basic speech act types, and (ii) some system of pragmatic knowledge of language use that involves the whole variety of possible speech act types of things done with words, which are not basic speech act types. The existence of such systems is not self-evident. It is, actually, our hypothesis that:

(MP.6) There is a P-modular system of pragmatic knowledge of language use of basic speech act types. and (MP.7) There is a central system of pragmatic knowledge of language use of things done in words.

We sketch some evidence for parts of these hypotheses. Evidence for (MP.6), the P-modularity hypothesis of the basic speech act system, will take the form of evidence for the independence of this system on one of the above-mentioned levels (P.1) to (P.5). Evidence for independence on level (P.2) of theoretical principles can be found in the philosophical characterization of the class of basic speech act types. Such speech acts are governed, according to (MP.3), by constitutive systems of rules. By its very nature, each constitutive system of rules is, in a sense, independent. If the constitutive systems of rules that govern basic speech acts employ only sentences of certain syntactic and intonational properties, as we assume [(MP.5)], then these constitutive systems of rules are also independent of all other constitutive systems of rules, what, in a sense, lends them independence on the level (P.2) of theoretical principles.(33)

Some evidence for the independence of the basic speech act system on level (P.4) of neural embodiment has been recently found in a study made within the framework of our Neuropragmatics Project, with Eran Zaidel et al.(34) Separate detailed tests, each related to one of the speech act types of assertion, question, command and request, have been administered to appropriate groups of right-handed right brain-damaged and left brain-damaged subjects. Preliminary analysis shows a trend for the latter to perform worse than the former.

Evidence for (MP.7), the centrality hypothesis of the system of things done with words, is evidence for the system having general principles, for its being domain inspecific or for its being informationally unencapsulated.

We assume that the above-mentioned recursive amplifications of speech act types are the principles of this system. Clearly, amplification can take the form of an adding operation, i.e. adding to a system of rules that governs speech acts of a certain type yet another rule. Since there do not seem to be any constraints imposed on the content of such added rules, recursive amplification results in domain inspecificity and information unencapsulation.

Evidence for the existence of a system of principles of amplification has been found in another study made within the same framework of the Neuropragmatics Project, with Eran Zaidel et al. In a series of cases we have been able to teach a new type of speech act, another one of the things done with words, to subjects of various types of brain damage.

We expect that future studies will shed light on the hypotheses we have put forward, within modular speech act theory, namely (MP.5) to (MP.7). The combined philosophical and linguistic studies of the related theoretical principles, psychological studies of processing and acquisition, and neuro-psychological studies of brain embodiment seem to provide us with an appropriate interdisciplinary setting for a fruitful research program in pragmatics.

References

Austin,J. L. (1976), How To Do Things With Words, 2nd edition, Oxford, Oxford University Press.

Brandl, Johannes and Gombocz, Wolfgang L. (eds.) (1989), The Philosophy of Donald Davidson, Amsterdam and Atlanta, Rodopi.

Chomsky, Noam (1980), Rules and Representations, Oxford, Blackwell.

Chomsky, Noam (1982)The Generative Enterprise, a discussion with R. Huybregts and H. van Riemsdijk, Dordrecht, Foris.

Chomsky, Noam (1984), Modular Approaches to the Study of the Mind, San Diego, San Diego State University Press.

Chomsky, Noam (1986), Knowledge of Language, New York, Praeger.

Chomsky, Noam (1988), Language and Problems of Knowledge, Cambridge, MA and London, MIT Press.

Descartes, R. (1641), Meditations on First Philosophy, Meditation VI, In: E. S. Haldane and G. R. T. Ross (trans.), The Philosophical Works of Descartes, New York, Dover, 1931.

Dummett, Micahel (1973), Frege, Philosophy of Language London, Duckworth.

Dummett, Micahel (1981), The Interpretation of Frege's Philosophy, London, Duckworth.

Dummett, Micahel (1991), The Logical Basis of Metaphysics, London, Duckworth.

Fodor, Jerry A. (1983), The Modularity of Mind, Cambridge, MA and London, MIT Press.

Frazier, Lyn (1988), 'Grammar and Language Processing', in: Frederick J. Newmeyer, ed., Linguistics: The Cambridge Survey, Volume II: Linguistic Theory: Extensions and Implication, Cambridge, Cambridge University Press, 15-34.

Gardner, Howard (1985), The Mind's New Science, 2nd ed., New York, Basic Books.

Garfield, Jay L. (ed.) 1987, Modularity in Knowledge Representation and Natural-Language Understanding, Cambridge, MA and London, MIT Press.

Grice, Paul (1989), Studies in the Way of Words, Cambridge, MA and London, Harvard University Press.

Kasher, Asa (1976), 'Conversational maxims and rationality,' in: Asa Kasher, ed., Language in Focus: Foundations, Methods and Systems, Dordrecht, Reidel, 197-216.

Kasher, Asa (1977), 'What is a theory of use?' Journal of Pragmatics Vol. 1, 105-120; republished in: A. Margalit, ed., Meaning and Use, Dordrecht, Reidel, 37-55.

Kasher, Asa (1980), 'The institutional man; Beyond religion and language,' Diotima, Vol. 8, 54-59.

Kasher, Asa (1981), 'Minimal speakers and necessary speech acts,' in: Florian Coulmas, ed., Festschrift for Native Speaker, The Hague, Mouton,93-101.

Kasher, Asa (1982), 'Gricean inference reconsidered,' Philosophica(Gent) Vol. 29, 25-44.

Kasher, Asa (1984), 'On the Psychological Reality of Pragmatics,' Journal of Pragmatics Vol. 8, 539-557. A revised version, under the title 'Pragmatics and the Modularity of Mind,' appears in: Steven Davis, ed., Pragmatics, New York and Oxford, Oxford University Press, 1991.

Kasher, Asa (1989), 'On art circularity: Logical notes on the institutional theory of art, in: Du Vrai, du Beau, du Bien (Festschrift for Evanghelos Moutsopoulos), Paris, Librairie Philosophique J. Vrin, 184-191.

Kasher, Asa (1991a), 'Pragmatics and Chomsky's research program,' in: Asa Kasher, ed., The Chomskyan Turn, Oxford, Blackwell, 122-149.

Kasher, Asa (1991b), 'On the pragmatic modules: A lecture,' Journal of Pragmatics Vol. 16, 381-397.

Kasher, Naomi (1978), 'Deontology and Kant', Revue Internationale de Philosophie, Vol. 126, 551-558.

Levinson, S. (1983) Pragmatics, Cambridge, Cambridge University Press.

Salkie, Raphael (1990), The Chomsky Update, London, Unwin Hyman.

Searle, John R. (1969), Speech Acts, Cambridge, Cambridge University Press.

Searle, John R. and Vanderveken, Daniel (1985), Foundations of Illocutionary Logic, Cambridge, Cambridge University Press.

Vanderveken, Daniel (1990), Meaning and Speech Acts, Volume I: Principles of Language Use, Cambridge, Cambridge University Press.

Vanderveken, Daniel (1991), Meaning and Speech Acts, Volume II: Formal Semantics of Success and Satisfaction, Cambridge, Cambridge University Press.

NOTES

1. See: Chomsky (1982) and (1984) and Salkie (1990).(back)

2. See: Brandl and Gombocz (eds.) (1989). The editors of this volume portray Davidson's work as a program in philosophy of mind.(back)

3. See: Kasher (1991a) and (1991b).(back)

4. A human competence is held to be a system of knowledge rather than an ability to act. On this use of "knowledge", see: Chomsky (1980).(back)

5. See: Chomsky (1988).(back)

6. See: Levinson (1983) and Kasher (1977).(back)

7. Thus explained is the interdisciplinary nature of cognitive science, mentioned by Gardner as a characteristic feature of it. See Gardner (1987:6-7).(back)

8. A third possible type of system, namely, interface ones, will not be presently discussed. See: Kasher (1984) and Frazier (1988).(back)

9. Descartes (1641: Meditation VI, 196).(back)

10. Fodor (1983). See also Garfield (ed.) (1987).(back)

11. Kasher (1984) and the (1991) revised version.(back)

12. By "force" it will always be meant what is usually called "illocutionary force".(back)

13. To be sure, these are at least somewhat imprecise labels of linguistic families of phenomena. In most cases it takes at least a little theory to introduce the very family of facts to be under consideration.(back)

14. As a prime example of the latter, we take "talk in interaction" as studied by Schegloff and others.(back)

15. Fodor (1983: 53).(back)

16. See: Kasher (1984), (1991a) and (1991b).(back)

17. See: Fodor (1983: 101-119), where the notions of isotropic and Quineian systems are also introduced and applied.(back)

18. See: Kasher (1976) and (1982).(back)

19. Grice (1989: 31). In this context we take "knowledge" to mean belief.(back)

20. Grice (1989: 34, 53-4).(back)

21. See: Dummett (1973), (1981) and (1991).(back)

22. Austin (1976: 150); see also the related footnote.(back)

23. Austin (1976: 151).(back)

24. For Chomsky's distinction between cognitive i-languages and cultural e-languages, see: Chomsky (1986).(back)

25. Searle (1969).(back)

26. The nature of constitutive systems of rules has been analyzed in Searle (1969) and N. Kasher (1978).(back)

27. Kasher (1977). Vanderveken has also used the idea of a recursive analysis, e.g. in his (1990), but his theory is couched in terms of those kinds of rule that Searle used in his essay on speech acts (1969), such as "sincerity" conditions and the like. The cluster of rules or conditions that play a role in a system that governs speech acts of a particular type seems arbitrary, at least in comparison with the following analysis, which is couched in terms of the natural ingredients of intentional activity in general. Many of the details of Searle's and Vanderveken's works are, nevertheless, very useful.(back)

28. Since the same act can serve a series of related ends, a distinction has to be drawn between the ends that a speech act serves in a direct way by the linguistic rules that govern it and other ends. For the former type of ends we use the term "literal". See: Kasher (1977).(back)

29. Strictly speaking, basic speech acts are also things done with words, but here we mean by "things done with words" those that are not basic.(back)

30. See: Kasher (1981). We are interested in analyzing the general conception of speech act as well as the basic speech acts without resorting to communication as an essential ingredient of language, but this part of the philosophical "hard core" of the present research program is beyond the scope of this paper. See: Kasher (1991a) and (1991b). We are aware of some reports that have been made to the effect that some languages do not include one basic speech act or another, but we believe that the alleged linguistic facts can be explained in a way that is compatible with our tentative identification of some basic speech acts.(back)

31. See: Searle and Vanderveken (1985), Vanderveken (1990) and (1991).(back)

32. See: Vanderveken (1990: 119ff).(back)

33. These systems of rules are as far as we know unique among cognitive systems that have a strong innate ingredient. Constitutive systems abound within various cultural spheres, such as games, morality [see: N. Kasher (1978)], art [see: Kasher (1989)], and religion [see: Kasher (1980)].(back)

34. This project has been supported by grants from the US-Israel Binational Science Foundation (88-00116), The Basic Research Fund of the Israel National Academy of Science and the Humanities (773-92) and Tel-Aviv University. Research has been done by Eran Zaidel (UCLA) and the present author, in collaboration with Dr. Rachel Giora, Dr. Nahum Soroker and David Graves, as well as Sharon Agam, Tamar Etkes, Gila Batori and Mali Gil. (back)