Truth and Speech Acts

Studies in the Philosophy of Language

Dirk Greimann, Geo Siegwart, Dirk Greimann, Geo Siegwart

• Publisher: Routledge
• Year: 2013
• ISBN:
  9781135197599
• Edition: 1
• Topic: Lingue e linguistica
• Subtopic: Semantica linguistica

Whereas the relationship between truth and propositional content has already been intensively investigated, there are only very few studies devoted to the task of illuminating the relationship between truth and illocutionary acts. This book fills that gap.

This innovative collection addresses such themes as:

• the relation between the concept of truth and the success conditions of assertions and kindred speech acts
• the linguistic devices of expressing the truth of a proposition
• the relation between predication and truth.

Part I

The illocutionary significance of the concept of truth

1Illocutionary acts and truth

William P. Alston

This paper will be devoted to considering certain issues about the relation, or lack thereof, of truth-values and illocutionary acts. At the outset I need to stipulate certain parameters of the discussion.

First I am concerned specifically with illocutionary acts (IAs) rather than more generally with the genus of speech acts of which illocutionary acts constitute one species, to be distinguished from, for example, locutionary acts (uttering sentences or surrogates thereof) and perlocutionary acts (producing effects on audiences by utterances). The category of illocutionary acts is not as easy or unproblematical to demarcate. My way of doing this is somewhat different from those of J. L. Austin and of John Searle, though closer to the latter. For a quick indication I could say that to perform an IA is to issue an utterance with a certain “content.” But what do I mean by “content” here? The shortest useful answer involves following Searle in taking the content to consist of one or more propositions and an “illocutionary force.” Thus different IA-types can differ in propositional content, illocutionary force, or both. Telling X that the door is open, requesting X to open the door, and asking X whether the door is open, all express the same proposition but with different illocutionary forces. Whereas the first of the above examples shares illocutionary force with telling X that the mail has come, and telling X that dinner is ready, they differ in propositional content. This way of identifying something as an illocutionary act could be put by saying that it is a speech act a fully explicit sentential vehicle for which would involve one or more propositional clauses and a term for, or other linguistic indication of, illocutionary force.¹

This should give an adequate preliminary indication of how I am thinking of illocutionary acts except for a type-token complexity. The above explanation was in terms of maximally specific illocutionary act types, but what about tokens of such types? In particular, suppose that S tells X the mail has come by uttering the sentence “The mail has come” and thereby gets X to realize that the mail has come. S has, in one breath, performed acts of locutionary, illocutionary, and perlocutionary types. Are we to say, as my last sentence would have it, that S has performed three different act tokens, or should we say that there is only one act token that belongs to three different types? This is a special case of the much discussed problem of act identity, and, more generally, of event identity. For my purposes here I don’t care which position is taken. But since my concern will be primarily with illocutionary act types, it will be most convenient for me to speak of the locutionary. illocutionary. and perlocutionary acts performed simultaneously by the same speaker as different acts. Let it be so ruled.

The other preliminary parameter-setting has to do with truth. I will be employing what I call in Alston (1996) a “realist conception” of truth. Before explaining what that amounts to let me say that I take the primary bearers of truth-value to be propositions. Beliefs and assertions that are true are so by virtue of their propositional content, and if we want to think of (some) sentences as true, as I don’t, it is because of the propositions they express. If any reader is worried about the ontology of propositions, let me say that all it takes to deal with propositions as I construe them is to be able to handle that-clauses. If you can understand, for example, “that it is snowing,” you have all you need to understand talk about propositions. To get back to the realist conception of truth, the basic idea is that the concept of truth is uniquely identified by what I call the “truth schema”: The proposition that p is true (It is true that p) iff p. If you recognize the truth of any substitution instance of this schema, that suffices for you to have grasped the concept of propositional truth. This does not provide for even a contextual definition of “true” unless we allow for substitutional quantification, but that doesn’t prevent the above explanation from providing a unique identification of the concept of truth. There are many other ways of formulating the same basic idea. For example: A belief is true iff what is believed is the case. Again: An assertion is true iff what the assertion is about is as it is being asserted to be.²

This is a minimalist account of the concept of truth, for it makes no attempt to go into details as to what it is by virtue of which the truth-maker for a given proposition does that job. Thus it makes no pretense at being a theory of the property of truth. But it is distinct from a “deflationary” account of truth that attempts to dispose of the claim that there is a property of truth whose exemplification distinguishes true from false propositions. An extreme version of this is the “disappearance” view of truth according to which asserting that it is true that grass is green amounts to no more than asserting that grass is green. My account of the concept is not that minimalist. Though it is not an account of the nature of the property of truth, unlike a correspondence theory with which it is naturally associated, it is open to the possibility of such a theory’s being developed, though it doesn’t imply that this can be successfully done.³

It is not crucial for my purposes in this essay to explain why I call this a “realist conception” of truth. But in case anyone is wondering, the point is that this view makes the truth of a proposition dependent on what the proposition is about, which is usually something that is external to the proposition and its role in thought and discourse, unlike epistemic conceptions of truth according to which for a proposition to be true is for it to have some epistemic status such as being what Putnam calls “ideally justified.” But my account of the concept of truth does not carry with it a commitment to a metaphysical realism, the view that what (many) propositions are about are items that are what they are independently of our thought and discourse about them. That is a further issue.

With this background I can begin to consider various alleged ways in which IAs and truth are related. The most obvious suggestions have to do with assertoric illocutionary acts. That is a large family of acts rather than a single homogeneous type, for it embraces a large variety of distinct illocutionary forces, such as replying that p, admitting that p, insisting that p, remarking that p, objecting that p, and so on. What these all have in common is that they all involve asserting that p. They could all be construed as “merely asserting” that p + some additional feature that distinguishes them from each other. I think of (merely) asserting (stating) that p as a maximally simple assertoric IA, one that exhibits the common feature of assertoric IAs in its purity without any additional feature. Alternatively we could think of assertion as a genus of which the various “special” modes of asserting are species. Dealing with the issues about assertoric IAs I will be concerned with here does not require separate treatment of the different special assertoric IAs. It is their common or generic nature that is relevant to those issues. Hence in discussing them I will restrict myself to speaking in terms of assertions and thereby implying that the points made apply to all the “special” assertoric IAs as well.

Now I can proceed to a fairly popular view of what distinguishes assertoric IA-types as such. It was classically formulated by Frege as follows: To assert a proposition is to present it as true.⁴

But despite its appeal, it is fairly easy to see that this view will not hold up under scrutiny. For one thing, if this position means what it seems to mean, it represents all assertions as having the same main predicate. For if to present a proposition as true is to assert that proposition to be true, then every assertion is a predication of truth to some proposition. But this is a Procrustean bed indeed. We certainly mean to be predicating different properties of different things (usually not propositions), depending on the constitution of the propositional content. And this view would, so to say, prevent our doing what we mean to be doing, and seem to be doing, in making assertions. It takes the propositional content of an assertion and makes that what the assertion is about, thereby foisting a role on it that distorts its proper function in the proceedings. I assert that Mt. Shasta is impressive. Unless I am completely out of touch with what I am doing, then what my assertion is about is Mt. Shasta, and what I am asserting of it is that it is impressive. I could also make an assertion about the propositional content of that first assertion, and assert that it is true. But that would be a different assertion, one of a sort that most speakers seldom make. And so this suggestion is on the wrong track altogether.

But perhaps this is the wrong way to construe Frege and others who take his kind of position.⁵ Perhaps he didn’t mean to imply that all assertions are about propositions, predicating truth of them, but instead to assert some different connection between asserting that p and presenting p as a truth. Here are some possibilities. (1) To make an assertion is to claim that the propositional content of one’s utterance is true. (2) In making an assertion one represents oneself as presenting a true proposition. (3) When one makes an assertion, one would be prepared to claim that one was presenting a true proposition if the question were raised as to what one takes oneself to be doing. (4) If one asserts that p but does not suppose that it is true that p, one is guilty of deceit of some sort – misleading the hearer, misrepresenting one’s beliefs, or the like. None of these connections involve altering propositional contents of assertions to make them predicating truth of a proposition, the defect that was seen to discredit the earlier suggestion. But each of them claims a significant relationship between assertion and truth.

I have arranged these claims in an order of decreasing similarity to the discredited thesis. (1) is just like that thesis except for avoiding the claim that the propositional content of every assertion is of the form “The proposition that p is true.” (2) differs from (1) only in taking the same claim to be implicit rather than explicit. (3) puts the claim to the truth of the proposition further away by placing it in how one would answer a question or challenge. Whereas (4) is still further away from an explicit claim to truth by only specifying a result of not supposing the claim to be true. The same order exhibits a decreasing plausibility of construing the thesis as a statement of what it is to make an assertion. (1). I would say, has the same title to that as the original thesis. (2). though less explicit, could with some plausibility claim to be spelling out what it is to make an assertion. It would seem to be less plausible to give (3) that role, since making an assertion seems to be something more overt, more out in the open, than merely a disposition to respond to questions or challenges in a certain way. And (4) would seem to be even further away from an account of the nature of assertion for the same kind of reason. Although those are my intuitions, I don’t want to hang a great deal on their correctness. And since I will go on to reject all four of these suggestions, both as accounts of what making an assertion is, and even as something that is universally true of assertion making, I take the question of which of them is a correct account of what it is to make an assertion as a question that does not arise.

So why do I think that none of the four theses holds universally, much less necessarily, of all assertions. The reason is that they all require the asserter to have command of the concept of propositional truth, and that is an untenable requirement to make of all asserters. The crucial point here is that taking a proposition to have a certain truth-value is a more sophisticated cognitive performance than making the most rudimentary or primitive sort of assertion. In arguing for this I will leave aside the question of what it takes to have a working mastery of a concept of a proposition, since earlier I maintained that being able to deal with that-clauses is sufficient for that. And though it could be contended that even that is something more sophisticated than what is required for the most primitive sorts of assertions, I will not press that point here. Instead I will concentrate on what it is to have and use the concept of truth. In the previous section I maintained that one has the ordinary concept of propositional truth provided one recognizes that for a proposition, p, to be true it is both necessary and sufficient that it is the case that p. But to be in a position to recognize that one must have at one’s disposal the concepts of necessary and sufficient conditions, as well as the concept of something’s being the case. I am not suggesting that the cognitive subject must be capable of spelling all this out explicitly. Again, a working mastery in practice of the relevant similarities and distinctions is all that is required. But if either (1) or (2) is true of a cognitive subject, S, S must be capable of claiming, explicitly or implicitly, that the propositional content of S’s utterance is true. And this requires S to predicate the property of truth to a proposition, and in addition to be able to recognize that proposition as the propositional content of a particular utterance. Thus it requires mastery of a fairly complex conceptual scheme. And think of the most rudimentary example you can of an utterance that would count as an assertion and ask yourself whether one could not be credited with making an assertion unless one is able to wield that complex a conceptual scheme. Take as your rudimentary assertion something like “It hurts” or “I’m tired” or “It’s cold.” Do we really want to say that S could not count as making assertions in these cases if S were incapable of predicating truth of the propositional content of those utterances? Suppose that S could pass the following tests of making a genuine assertion when saying “I’m tired,” an assertion of the type normally and conventionally made by uttering those words, (a) S, on being questioned as to what made him/her so tired, would say that he had been doing a lot of running outside, (b) When asked whether s/he would like to go for a walk, replies “Of course not; I’m tired.” (c) When the mother says “You always say you’re tired when you want to get out of picking up your toys,” S replies “But I often pick up my toys.” This list could be extended indefinitely. Isn’t it clear that a toddler could carry on a conversation like this without being able to claim, as implicitly as you like provided it is a genuine claim, that the propositional content of the original utterance is true, or even that any proposition is true, lacking at that point the conceptual development necessary for making such a claim?

The same decision is to be made with respect to (3). For if one lacks the conceptual equipment needed to claim that a certain proposition is true, one obviously cannot be prepared to claim that one is presenting a true proposition in response to certain questions or challenges. But what about (4)? This differs from the other theses on the list by not representing the asserter as making, or as capable of making, claims of the sort featured in (l)-(3). Instead it rules that the asserter is guilty of deceit if he is not prepared to make such a claim. It doesn’t require the asserter to recognize that or engage in any other cognitive performance that would require the conceptual repertoire I have been suggesting that need not be possessed in order to make the most rudimentary assertions. That is certainly correct. But by the same token (4) as stated is not a good candidate for an account of what it is to make an assertion. If (4) were strengthened into (4A) – If one apparently as...