Cancellation of Illocutionary Force Potential

1.1 Overview

In real life communication, language (spoken or written) is without doubt the most important media for exchanging thoughts. Despite its pivotal role in human civilization, it was only last century that witnessed the inception of a systemic study of language as an academic field, and even later is that of the study of meaning became an academic pursuit in such fields as semantics, pragmatics, and philosophy of language.

In this quest of the nature of meaning, scholars have come up with different hypotheses and constructed distinctive theories, in which course various technical terms have been coined and used to denote types of meanings claimed to have divergent functions in communication. Among them, the Speech Act Theory and the Gricean Theory of Meaning are the two most fundamental pillars supporting the study of the meaning of natural language.

According to the Speech Act Theory, when a sentence like “I will never cheat again” is uttered, e.g., by a student after caught cheating, what’s communicated consists in two aspects: the encoded linguistic meaning (in Austin’s term, “sense and reference,” and for Searle, “propositional content”) and an illocutionary force of promising. This concept of illocutionary force is central to studies of meaning in terms of the speech act theory. It depicts the function of an utterance intended by a speaker, as opposed to its semantic embodiment. Apparently, this concept seems to be most crucially built upon the speaker’s intention.

In this respect, the Gricean theory of meaning to a great extent inherits and further carries forward the intention-based tradition, and takes the intended meaning as the focus of meaning study. In fact, in his theory, Paul Grice prioritizes the study of what the speaker intends to communicate, i.e., speaker’s meaning (what is said and what is implicated). A successful communication would require cooperative work between all parties of communication, such that the speaker delivers what s/he intends to convey, and the hearer recognizes this intention and parses the meaning. Actually, Grice’s work is considered to have “greatly contributed to the subsequent development of speech act theory.” This being the case, his work upon the speaker’s meaning (what is said and what is implicated), as well as that of his followers, then, mutatis mutandis, could be conducive to refining studies on the illocutionary force of speech acts (Sbisà 2009: 231).

One fascinating fact about language is that people could (and actually it is quite normal for them to) use it to convey (much) more than what is literally encoded in the individual words and sentences that are used. This asymmetry between linguistic expressions and meaning does make communication more economic since people could indeed communicate more with less effort. In the meanwhile, however, the additional information carried might be redundant and troublesome: For one thing, it seems to be more difficult to communicate effectively and convey what is “intended;” For another, it enables people to intentionally manipulate with language what they try to communicate and even mislead the audience’s interpretation.

This set of additional meaning has aroused great interest among philosophers of language and have been examined at great length. Some distinctive types have been recognized and respectively dubbed different names. There is the set of “presuppositions,” i.e., implicit assumptions which, though not explicitly stated nor intended, are taken for granted by both the speaker and the hearer in order for an utterance to be deemed truthful or felicitous. For instance, “John has quit smoking” presupposes John once smoked. “Entailments” depicts the deduction or implication whose truth follows necessarily from the truth of another proposition. There is a set of “entailments,” i.e., the strong semantic relation between propositions, such that in all possible worlds, where proposition A is true (e.g., “Both John and Mary passed the exam.”), proposition B (e.g., “John passed the exam.”) must also be true. These entailments can be logically derived (semantically or pragmatically) regarding what one says, but not necessarily intended by the speaker. Most interestingly, there is a type of meaning which is intended, but is neither literally encoded nor linguistically triggered, associated with some relevant linguistic expressions or propositions the way presuppositions or entailments do. Both the Speech Act Theory and the Gricean theory of meaning have devoted a considerable portion of their literature to its study. In exploring this type of meaning, both theories contributed several classic conceptions and distinctions, such as their distinctions of conversational implicature/ conventional implicature, what-is-said/ what-is-implicated, the locutionary/ illocutionary/ perlocutionary acts, and direct/ indirect speech acts, etc. These distinctions get widely acknowledged and applied in the fields of semantics, pragmatics, and philosophy of language, and have significantly influenced and facilitated the research across a variety of disciplines, such as cognitive science, computer science, education, law, and cross-cultural communication studies.

For the Speech Act Theory, the intended but non-encoded meaning is dubbed the “indirect speech acts” or “indirect illocutionary act” (Searle 1979:31), one sub-type of the “illocutionary act” (Austin 1962), the act performed in saying something. In the Gricean theory of meaning, it is referred to with the cover term “what is implicated.” Specifically, the term “conversational implicature” (CI) is introduced to denote the kind of meaning which is pragmatically derived in light of the Cooperative Principle (CP) and its maxims, while “conventional implicature” refers to the kind of implicature which is neither semantically encoded nor pragmatically inferred. The conventional interpretation of “but” as in “Mary is poor but honest” always implies a sense of contrast, but this sense is conventionally implicated rather than derived.

At the first glance, the difference between the linguistically encoded literal meaning and the set of non-literal non-encoded meaning seems simple and clear, but in effect, such a difference is far from easy to pin down. So, in an effort to spell it out, many criteria have been introduced, disputed and revised. Yet none of them has been unanimously accepted. Nevertheless, there is one of them which have received a lot of attention, i.e., the celebrated classic Gricean “Cancellability Test.”

In “Logic and Conversation” (1975), Grice put forth cancellability as a primary feature of conversational implicatures. According to him,

… a putative conversational implicature that p is explicitly cancellable if, to the form of the words the utterance of which putatively implicates that p, it is admissible to add but not p, or I do not mean to imply that p, and it is contextually cancellable if one can find situations in which the utterance of the form of words would simply not carry the implicature.

(Grice 1989: 44)

As Grice sees it, this is because CIs are not encoded in the truth-conditional content of an utterance; rather, they are pragmatically derived through appealing to the conventional meaning of the utterance and rational principles (CP and its maxims). Therefore, since the presence of CI presupposes the CP being observed, it follows that when one opts not to comply with it, the corresponding CIs can be cancelled (1989: 39, 44). This cancellation can be realized in two ways: explicit cancellation (EC, by adding after the utterance which putatively implicates that p negating expressions such as “but not p,” or “I do not mean to imply that p.”) and contextual cancellation (CC, by putting the utterance which putatively implicates that p in situations in which the utterance of the form of words would simply not carry the implicature). And according to Grice, all CIs are cancellable.¹

This being the case, we are wondering if CT is also applicable to illocutionary force. That is, is illocutionary force cancellable? Specifically, is it possible to cancel it explicitly? Or can it be contextually cancelled? Applying a cancelling clause to the sentence “I will never cheat again,” as with

“I will never cheat again. —But this is not a promise.”

Is such an utterance acceptable in daily conversation? Or in Grice’s words, does illocutionary force constitute part of the truth-conditional content of an utterance? This is an interesting topic to delve into.

Nevertheless, previous studies about cancellability—whether in agreement with Grice or not—have mostly been confined to discussions around types of meaning under Gricean or neo-/post-Gricean classifications, which stimulates our interest about its applicability to a more general conception of pragmatic content. And therefore, strongly motivated to resolve the differences between various (neo-/post-) Gricean taxonomies of meaning, and to restore the effectiveness of classic Gricean CT, the present study endeavors to explore the applicability of CT to a further modified concept of illocutionary force potential, partly to dissolve current debates about division of labor between semantics and pragmatics, and partly to investigate the nature of communication in terms of generation, actualization and cancellation of the illocutionary force.