Summary and Keywords
Speech acts are acts that can, but need not, be carried out by saying and meaning that one is doing so. Many view speech acts as the central units of communication, with phonological, morphological, syntactic, and semantic properties of an utterance serving as ways of identifying whether the speaker is making a promise, a prediction, a statement, or a threat. Some speech acts are momentous, since an appropriate authority can, for instance, declare war or sentence a defendant to prison, by saying that he or she is doing so. Speech acts are typically analyzed into two distinct components: a content dimension (corresponding to what is being said), and a force dimension (corresponding to how what is being said is being expressed). The grammatical mood of the sentence used in a speech act signals, but does not uniquely determine, the force of the speech act being performed. A special type of speech act is the performative, which makes explicit the force of the utterance. Although it has been famously claimed that performatives such as “I promise to be there on time” are neither true nor false, current scholarly consensus rejects this view. The study of so-called infelicities concerns the ways in which speech acts might either be defective (say by being insincere) or fail completely.
Recent theorizing about speech acts tends to fall either into conventionalist or intentionalist traditions: the former sees speech acts as analogous to moves in a game, with such acts being governed by rules of the form “doing A counts as doing B”; the latter eschews game-like rules and instead sees speech acts as governed by communicative intentions only. Debate also arises over the extent to which speakers can perform one speech act indirectly by performing another. Skeptics about the frequency of such events contend that many alleged indirect speech acts should be seen instead as expressions of attitudes. New developments in speech act theory also situate them in larger conversational frameworks, such as inquiries, debates, or deliberations made in the course of planning. In addition, recent scholarship has identified a type of oppression against under-represented groups as occurring through “silencing”: a speaker attempts to use a speech act to protect her autonomy, but the putative act fails due to her unjust milieu.
1. What Is A Speech Act?
To delineate the subject matter of speech acts, it will help first to exposit a notion it presupposes. This notion is speaker meaning, exemplified in situations that might be described in such terms as:
As case (2) shows, speaker meaning does not require speaking, or even using language, but it does require that an agent behave with the intention of getting across a message, or, barring that, the intention of making his or her state of mind manifest. Some winks, for instance, are done with the intention of expressing interest in another person, or to acknowledge a mutual understanding; others result from a tic or errant dust particle. A person winked at may well wonder which of these descriptions applies, and he might express his question with the words, “What—if anything—did she mean by that?” In so doing, he is using the notion of speaker meaning.
What unites the cases of speaker meaning mentioned thus far is that they are all overt: one not only does something that manifests one’s state of mind, one does something with the intention that one’s intention so to act is clear. Compare the three following cases. In a restaurant, Dakota and Sawyer are vigorously arguing about morphology, and food has just been served for Dakota but not for Sawyer.
(A) Dakota tucks in, quite oblivious to the fact that it’s a bit rude to do so.
(B) Instead of eating, Dakota wafts in the aroma of his dish, making pleased sounds in anticipation of his meal.
(C) Unlike in the previous two cases, Dakota tucks in lustily to the food he has just been served in full knowledge of his rudeness, returning Sawyer’s outraged look with a glare.
In (A), Dakota does not mean anything; he is just satisfying his hunger, while also betraying his lack of manners. In (B), Dakota likely does not mean that he is about to eat, or that the food is going to taste good, but is instead expressing pleasure over the impending meal. Meaning of a kind presupposed in the present account of speech acts only emerges in case (C), where not only does Dakota manifest his desire to eat, he also manifests his intention to manifest that desire. Put differently, his behavior is overt in (C), while it is not overt in either (A) or (B). In what follows, it will be assumed that speakers engage in speaker meaning when, and only when, they overtly manifest their state of mind.1
Part of what generated excitement about speech acts when J. L. Austin (1962) first popularized them in the middle of the 20th century is their ability not just to describe the world but to effect changes in it. Generally, when a speaker describes a situation, her doing so does not appear to create any new states of affairs beyond her having used a few words. However, in some cases, words can do more than this. In saying, under the right conditions, “I declare these proceedings open,” one can make it the case that the proceedings are open; if one is in a position of judicial authority, then by saying, “I sentence you to life imprisonment,” one can make it the case that a defendant is to be imprisoned for life. However, one can perform a speech act without describing oneself as doing so. Just as one can throw a ball without saying that one is doing so, one can assert that it is snowing without describing oneself as doing so: all one need do is say, with the right intentions and further contextual conditions, “It is snowing.”
With these notions of speaker meaning, and of how some utterances can constitute states of affairs, one may explain a speech act as a case of speaker meaning that can, but need not, be carried out by saying that one is doing so. Though, as noted above, one need not be so pedantic, one can assert that it is snowing by saying, “I assert that it is snowing,” and one can promise to be more punctual in the future by saying, “I promise to be more punctual in the future.” This is why asserting and promising are speech acts. By contrast, in light of the present construal of speech acts, convincing is not a speech act. For even if speaker A can convince B of the truth of proposition P by using words, A cannot convince B of proposition P by saying, “I hereby convince you that P.” Likewise for impressing, offending, and intimidating.
Austin distinguished among three elements of a communicative act. The first is the locutionary act, in which a sentence or phrase is formulated in sign language, writing, speech, semaphore, etc. If, as is the typical case, that sentence contains context-sensitive material such as pronouns or demonstratives, then context, and possibly also speaker intentions, will be required to determine what has been said. But even in so saying, a speaker may be performing one of a range of speech acts: predicting, warning, excommunicating, etc. Which act this is will determine which speech act, or in Austin’s terminology, illocutionary act, has been performed. Finally, by performing a particular illocutionary act, a speaker may achieve one or more of its characteristic effects. For instance, by warning a driver of danger ahead, a speaker might dissuade him from proceeding in that direction. Characteristic effects of illocutions are perlocutionary acts. It will not be assumed here that locutionary, illocutionary, and perlocutionary acts occur in any particular temporal sequence.
In what follows, ‘speech act,’ ‘illocution,’ and ‘illocutionary act’ will be used synonymously. ‘Illocute’ will be used as a verb denoting the performance of a speech act, and ‘illocutionary force’ will refer to a certain aspect of a speech act as described below. Also, this delineation should make clear that speech acts are not to be confused with acts of speech, or even with utterances more generally (where ‘utterance’ is a capacious enough notion to include uses of sign language). For one can produce an utterance, even of an indicative sentence, without performing a speech act. (“Alas, poor Yorick! I knew him, Horatio,” says the actor as she practices for her part in the play, but she is not asserting or otherwise illocuting that she ever knew any person named ‘Yorick.’) Accordingly, just as something that is big, and an elephant, is not necessarily a big elephant, so too, an act that is also a case of speech is not thereby a speech act.
Because speech acts are forms of speaker meaning and thus demand complex intentions on the part of those producing them, one should expect to find many communicatively significant behaviors that are not speech acts. A millenial’s use of upspeak in his vocal register (“Hello, my name is Edison, and I’ll be your server this evening?”), is most likely unintentional, and although his use of that intonation contour may well manifest a desire to affiliate, or at least not to come off as aggressive, it is unlikely that he speaker-means anything in his use of upspeak. Similarly, a smile forming on a person’s lips as he opens a letter from a long-lost friend may well express his pleasure, but it is not likely that in smiling, he speaker-means that he is happy. That is one reason why a smile is not normally thought of as speech act, though in some cases it may be the vehicle of one. The same goes for many gestures, so-called body language, and much of the behavior falling within the scope of animal communication. (This entry does not take a stand on the question whether any non-human animals can perform speech acts.). See Bar-On and Green (2010) for further discussion.
Speech acts are thus only a portion of the broader category of communicative acts and behaviors. That category will also include expressive behavior, which is designed to convey information about an agent’s psychological state. The notion of design at issue here is broad enough to include behaviors resulting from natural selection, from cultural evolution, and from conscious intention. An unintentional scowl will thus express anger because scowls are designed (presumably by natural selection) to telegraph information about their owners’ affective states. A’s saying, “Brrr!” in response to the cold gust of air entering through the window might be designed to manifest her discomfort. If so, A is expressing that discomfort without thereby illocuting. Likewise, although Searle (1969) describes an utterance of ‘Hooray for Tottenham!’ as a speech act, this does not seem mandatory. We could instead take it to be a conventionalized expression of Tottenham support. One consideration in favor of this treatment is that such an utterance would not typically be produced with the reflexive communicative intentions characteristic of speaker meaning.
In speech acts, with their comparatively sophisticated communicative intentions, we may also distinguish what is communicated from how it is communicated. A and B have been arguing about today’s impending weather, and A asserts,
Here one may distinguish between what A says, namely that it is going to rain, and how A says it, namely as an assertion. A could instead have put forth this content (the proposition that it is going to rain) with no force at all, or as a conjecture or even a sheer guess. Each of these would be an importantly different act: in response to an assertion, a speaker is normally entitled to ask, “How do you know that?” By contrast, it would belie a misunderstanding to respond with such a question to either a guess or a conjecture. In his influential Speech Acts (1969), Searle proposes that speech acts may generally be represented along these two dimensions of force and content. He also contends that, for all speech acts having content, all such contents must be propositional, and offers:
as a general form in which speech acts may be represented, where ‘F’ represents the act’s force and ‘p’ its propositional content. However, this is unduly restrictive. As will emerge in Section 3, it is also feasible to allow for illocutionary contents corresponding to imperative and interrogative grammatical forms. Until then, it will be expedient to draw a distinction between illocutionary force and semantic content while leaving open the question of precisely what form that content may take.
2. Felicity Conditions
Speech acts, as discussed above, exhibit a saying-makes-it-so property, yet for a given speech act to achieve this feat appropriate background conditions must obtain. Only an appropriate authority can sentence someone to prison in the U.S. legal system, and only an occupant of an appropriate institutional role can declare proceedings open. Likewise, even if a toddler utters the sentence, “I promise to pay you $100,” she likely will not have made a promise: it is not likely that she understands what promising is, and it may be obvious to her addressee that she has no $100 to give. Austin considered these cases under the rubric of felicity conditions, and he distinguished them into two kinds, misfires and abuses.
Misfires: here a speech act of a certain kind is attempted, but no such act occurs. The toddler may attempt to promise to pay $100 but does not succeed in performing that illocution. So too, A may utter, in front of the Eiffel Tower, “I bequeath this monument to my grandchildren,” but because the monument is not his to give, A has bequeathed nothing. Here again, one observes a locutionary but no illocutionary act. Equivalently, one may say that in a misfire, an act of speech but no speech act occurs. Misfires also can occur in speech acts that require uptake. It is only possible to bet with someone if that person accepts the proffered bet. (Machines act as proxies for the people who own or design them.) As a result, “I bet you my umbrella that it will rain today,” only amounts to a bet if the addressee accepts the wager; otherwise the speaker has tried but failed to bet.
Abuses: here a speech act of the attempted kind occurs, but is still defective in some significant way. If A answers the creditor’s question with, “Yes, the check is in the mail,” when he knows that he has not sent the check, A has still asserted that he has paid the outstanding bill. An utterance of, “I promise to do so and so” may still be a promise, even if the speaker has no intention of keeping it. One central area of potential abuse, then, is failure of sincerity. Many speech acts mandate of their users that they be in a certain psychological state (believing that a certain proposition is true, intending to carry out a course of action, etc.), and they are abused when performed by speakers who are not in such states. Although Austin did not discuss such cases under the rubric of abuses, in may be illuminating to consider speech acts made in inadequate evidential conditions. If A confidently and sincerely assures B that this mushroom is edible when A is no mycological authority, B would have cause to blame A for abusing the practice of assertion if B falls ill after ingesting it.
The line between misfires and abuses would not seem to be a sharp one. A speaker might lose her ability to be taken seriously either because of a pattern of abuses on her part (she is a chronic liar or exaggerator, for instance), or because of bias, oppression, or some other form of marginalization (the men in her company discount her views, or put those views up to an unreasonably demanding standard, say). If either of these first- or second-person types of abuse persists for long enough, the speaker may be unable to have her utterances count as statements; she might instead be taken as just pretending to make statements, or at most, to make guesses or suggestions. However, it is not clear when abuse has crossed over into misfire; instead, there may be cases in which there is no fact of the matter whether one has made a defective statement, or instead, no statement at all but only an act of speech.
Similarly, the vast literature on presupposition leaves open the possibility that some presupposition-failures produce misfires, and others abuses. Russell held that:
is true if and only if there is at least one, and at most one, present Queen of Lesotho, and whoever is a Queen of Lesotho is an actress (Russell, 1919). He also held that these truth conditions are sufficient to capture the meaning of (5). However, Strawson (1950) responds that if a speaker utters (5) in a situation in which Lesotho is no monarchy, the question of whether what she says is true, “does not arise.” (Strawson, 1950, p. 330) This is naturally interpreted to mean that the utterance will misfire. By contrast, a person who rushes in late to a meeting with the words:
may simply be saying something untrue if he was trying to impress, and what did not start was merely a Lancia. If so, his utterance of (6) will be an abuse rather than a misfire.
Ill-considered speech acts (such as assertions made insincerely or on inadequate grounds) can damage speakers’ reputations or, as in the case of a rash bet, incur more tangible costs. A distinctive feature of speech acts, however, is that they can often be retracted once made. By Wednesday, A cannot change the fact that on Tuesday A had made a statement. However, as Sbisà (2007) observes, on Wednesday A might retract Tuesday’s statement, perhaps after realizing it was made on insufficient evidence. So too, for those illocutions requiring uptake, so long as those whose uptake A secured agree to release him from the relevant commitments, A can retract those illocutions as well.
3. Force and Mood
Being cases of speaker meaning, speech acts are manifestations of intentions and other states of mind. But manifesting one’s state of mind is no small achievement, even if the phenomenon is ubiquitous. Unless they share a rich set of background assumptions, one person’s act of wiggling her finger won’t express her desire to go skeet shooting with her friend next Tuesday after lunch, even if she so intends it. Language helps to clarify the content and force of one’s communicative acts. The syntax and semantics of a sentence constitute a powerful clue as to the thoughts of the speaker using it and thereby the content of her utterance, but not an infallible one, since even if the speaker is sincere she may still misspeak. So too, it might be thought that a sentence’s grammatical mood will determine the force with which it will be uttered. But as we have already seen, one might utter a sentence in the indicative mood without its yet being clear whether he is asserting, conjecturing, or supposing that sentence’s content for the sake of argument. Indeed, one may utter that sentence without performing any speech act all, but merely as an act of speech.
That mood is no guarantor of force does not imply that it provides no clue about it. Instead, mood is best seen as an indicator, albeit defeasible, of force. That is why a person’s utterance of:
is evidence that she is demanding or requesting that someone shut the door, and why her utterance of:
is evidence that she is asking how many apples are in the bowl. It would seem that the indicative mood is a tool speakers use to signal their intention to say how things are, that the imperatival mood is a tool for signaling their intention to have something done, and the interrogative a tool for signaling their desire to gather information. Further, some languages encode into grammatical mood what in others is part of locutionary meaning. Hidatsa, for instance, exhibits different forms of indicative mood, one for when the information is common knowledge, one for when it is known first hand, another for when it is passed on from another person’s authority, and yet another for when the speaker refrains from commitment to the truth of the information (Sadock & Zwicky, 1985).
Even when it is not as refined, as in the case of Hidatsa, mood provides evidence not only of the “how,” but also of the “what” of a speech act. As suggested in Bell (1975), an interrogative sentence may reasonably be assigned as its semantic content not a proposition, but a set of propositions, where each member of that set corresponds to one complete answer to the question posed. The semantic content expressed by (8) contains such members as “There is one apple in the bowl”; “There are eleven apples in the bowl,” etc. On one semantic account for imperatives, developed in Portner (2004), such sentences are assigned properties, where the notion of a property is understood in the metaphysician’s sense of any feature that a thing might have. Thus, redness, triangularity, and being larger than St. Louis, are all properties in this capacious sense. Accordingly, on Portner’s view, (7) semantically expresses the property of shutting a certain door. The foregoing are not the only approaches to the semantics of interrogative and imperative sentences, and it need not be decided here which are the most promising among them. Instead, it need only be noted that, rather than assume that speech acts always have propositional content, a more inclusive approach allows a general representation of their structure as F(C), where ‘F’ refers as before to illocutionary force, while ‘C’ refers to content, be it propositional, interrogative, or imperatival. Green (2017) explores a notion of sentential content that includes distinctive contents for imperatival and interrogative sentences as well as the more standard propositional contents associated with indicative sentences.
4. Performatives, Meaning, and Truth
Those illocutions whose force speakers make explicit are performatives. This is normally done with use of a verb phrase that refers to a speech act-type. Thus consider (9)–(11):
These are sentences that can readily be imagined used in the service of a speech act. When (9) is so used, for instance, the speaker is not only promising, but is also making explicit that she is doing so. On the present conception of a performative, then, her utterance counts as one. It is also possible to define a performative sentence as one whose main verb names a speech act, and is in the first person, present, indicative active voice. One may then contemplate how (10) and (11) fall outside this definition while doing so in ways suggesting a widened account.
Another reason for widespread interest in performatives is their seeming ability to bridge a gap between semantics and pragmatics. Searle, for instance, argues that any utterance of (9) made under what he terms “conditions of successful utterance,” is a promise (1968, p. 407). He rightly concludes from this that some locutionary acts are also illocutionary acts, but infers in turn that, for some sentences, their locutionary meaning determines their illocutionary force. This last inference is, however, a non sequitur. Rather, when that sentence is uttered in such a way as to constitute a promise, what determines that force is the meaning of the sentence together with such factors as the speaker’s being sincere and other contextual conditions being met.
In one of his most striking claims about performatives, Austin held that performative utterances are neither true or false. His reason was that when a sentence such as (9) is used to perform a speech act, the speaker is engaging in a promise and not describing himself as doing so. After giving some examples of performative sentences, Austin wrote:
In these examples it seems clear that to utter the sentences (in, of course, the appropriate circumstances) is not to describe my doing of what I should be said in so uttering to be doing or to state that I am doing it: it is to do it. None of the utterances cited is either true or false … When I say, before the registrar or altar, &c., ‘I do,’ I am not reporting on a marriage: I am indulging in it. (Austin, 1962, p. 6)
Many authors have observed in response that engaging in an act and describing oneself as so doing need not be incompatible. Lemmon (1962) argues that performative utterances are true on the ground that they are instances of a wider class of sentences whose utterance guarantees their truth. If sound, this argument would show that performative sentences may have truth value when used in performative utterances, but not that those utterances are assertions. Sinnott-Armstrong (1994) also argues that performatives can have truth value without addressing the question of whether they are also used to make assertions. Reimer (1995) contends that while performatives have truth value, they are not also assertions. Adopting a similar strategy, Jary (2007) aims to explain how utterances of sentences such as (9) can succeed in being promises. In so doing, he draws on the account of Green (2007), according to whom such utterances show (rather than describe) the force of the speaker’s act. Because ‘show’ is factive (if A shows that P, then if A occurs, then P must true), if such an utterance shows its force, then it must have that force.
Challenges to Austin more typically construe performatives as assertions and explain their properties in that light. According to Ginet 1979, performative verbs (‘promote,’ ‘rescind,’ etc.) name the kinds of acts that one can perform by asserting that one is doing so, and he elaborates on why this is so. Ginet thereby provides an account of how performatives work that depends on the assumption that performative utterances are assertions. Bach (1975) likewise contends that “I order you to clean the latrine” is an assertion, and explains on this basis how the speaker is indirectly also issuing an order. This account depends on the speaker’s being able to rely on the addressee’s competence to discern the speaker’s communicative intention. It is later refined in Bach and Harnish (1979, and 1995), with a notion of standardization, on which a sufficiently common practice of issuing assertions with performative effect enables speakers and hearers to bypass complex inferential reasoning and jump, by default, to a conclusion about the illocution being performed. Reimer (1995) challenges this approach on the ground that hearers do not seem to impute assertoric force to the indicative sentences speakers utter with performative effect; her criticism would seem to carry over to Ginet’s proposal as well. Reimer contends by contrast that performative utterances rest on systems of what she terms illocutionary conventions to achieve their performative effects.
Searle (1969) had argued that a performative formula, such as “I promise to … ” is an “illocutionary force indicator,” meaning that it is a device whose role is to make explicit the force of the speaker’s utterance. Making something explicit, however, would seem to involve characterizing a state of affairs that obtains independently of one’s utterance. As a result, Searle’s account presupposes that speakers can imbue their utterances with the force of, say, promotions or excommunications; yet this is what was to be explained. Acknowledging this, Searle (1989) offers an account that depends on the view that, in uttering a sentence with a performative formula, a speaker manifests an intention to perform an act of a certain kind: in uttering the words, ‘I order you to close the door,’ A manifests an intention to order his addressee to close the door, etc. Searle also takes it that manifesting an intention to perform a speech act is sufficient for the performance of that act. On this basis, Searle goes on to attempt to derive the assertoric nature of performatives, holding that when uttered in such a way as to say something true, they are also assertions.
5. Convention versus Intention
It is natural to wonder how a speaker’s describing himself as doing something can be sufficient for his doing that thing. One’s spoken utterance of ‘I am now speaking’ suffices to make that sentence true, because that sentence is spoken and refers to the time and location at which it is being spoken. But one’s utterance of (9) seems to require more, not least because promising is a fairly momentous act: if a person has promised to do something, then she is in danger of committing a moral infraction if she doesn’t keep that promise, and may risk a consequent loss of status in her community (Green, 2009). To bridge the gap between locutionary and illocutionary acts, Austin (1962) invoked extra-semantic conventions, that is, conventions going beyond those imbuing a linguistic community’s words with meaning. On this force-conventionalist view, one who utters (9) only makes a promise if she invokes a convention to the effect that appropriate utterances of certain forms of words are sufficient for the making of promises, and likewise even for (3): utterance of that sentence must rely upon a convention relating forms of words to assertion for it to count as one.
Strawson (1964) challenges force-conventionalism by noting that while some speech acts, such as excommunicating and declaring offsides, seem to depend on extra-semantic conventions, matters are less clear with others. Accepting (a bet, or offer, for instance) is a speech act on our definition, but it would seem that B can accept A’s proffered morsel of food by manifesting her willingness to do so: wide open eyes, hand outstretched, and the like would seem to suffice, and it is not clear that any convention must be invoked for this transaction to succeed.
Strawson proposes instead that some speech acts rely for their occurrence not on (extra-semantic) conventions, but rather on the speaker’s manifesting appropriate intentions. One who utters (3) in such a way as to manifest an intention that it be taken as a prediction rather than, say, a supposition for argument’s sake or a conjecture, thereby asserts that it will rain. This view may acknowledge that certain other speech acts such as excommunicating and promoting depend on extra-semantic conventions. For the remainder, Strawson advocates explaining them as types of speaker meaning. Thus, asserting that P can be achieved by uttering a sentence whose literal meaning is P, with the further intention of producing a belief that P in an audience. However, merely doing something while intending that it produce a belief, or manifest one’s own state of mind, is not enough for speaker meaning. Instead one must make that intention available for detection by others. McDowell (1998) speaks to this need in describing speech acts as publications of intentions. (To publicize one’s intention one must make it there for the grasping even if no one does grasp it.) One might publicize an assertoric intention with a firm tone of voice and a steady gaze, or instead publicize one’s intention to put forward a content as a conjecture with a rising intonation contour and an upward glance. Green (2013) explores this notion of publicizing one’s intentions for illocutionary purposes in further detail.
6. Indirection and Implicature
In speech acts, speakers (speaker-) mean more than they say. That feature is shared with cases of conversational implicature, so the attentive reader may wonder how these two phenomena relate to one another. The brief answer is that conversational implicata are generated only from the performance of full speech acts, so they are a next step after illocutions in a cascade of speaker meanings. It is widely held, however, that one can indirectly illocute in either of the following senses: (a) by making as if to perform one speech act, one performs another (perhaps by pretending to greet an unpunctual guest, a host requests that he leave; as in “So nice of you to come!” said in a situation in which the would-be guest’s presence would only have been appreciated some hours ago); and (b) by performing one speech act, one also performs a distinct speech act (perhaps by remarking that B is standing on A’s foot, A is also requesting that B move). In type-(b) cases, two speech acts are performed, while in type-(a) cases only one speech act is performed while another is feigned. Further, on some accounts of type-(b) cases, the speech act performed indirectly is conversationally implicated.
This suggests that speech acts and implicature are mutually dependent, though not in a way that raises concerns about circularity. The best-known treatment of indirect speech acts is Searle (1979), whose theory in effect invokes implicature, though he does not use this term. Searle’s core insight is that by calling an addressee’s attention to the conditions prerequisite to the felicitous performance of an illocutionary act, a speaker can sometimes succeed in performing that act even when mentioning those conditions seems to violate a norm of conversational relevance. As mentioned in Section 2, a felicity condition for one’s asserting that P is that one believe that P. This, explains Searle, is why one can assert that P by saying, “I believe that P,” even though such an avowal is ostensibly about oneself instead of the subject matter of proposition P. So too, assuming that a felicity condition of a request is that the speaker has some interest in seeing the request fulfilled, “I wouldn’t mind if you got off of my foot,” may be used to request that the addressee get off the speaker’s foot and not just to remark on what the speaker would not mind her doing so.
Since they are defined in terms of speaker meaning, speech acts are creatures of intention. As a result, to be plausible, Searlean accounts of type-(b) indirect speech acts will need to populate speakers’ minds with intentions sufficient to make a single utterance both a statement and a request (say). Nothing prevents a speaker from having such intentions. Yet for many cases of indirection, a more parsimonious approach to the phenomena seems viable. Recall the “Brrr!” example of Section 1. Though not illocuting, the speaker is using a conventional means of expressing discomfort. Suppose she had instead said:
Suppose further that she is making an assertion, and thus illocuting. Then, as with “Brrr!”, she is also providing a designed indication of her discomfort. In our account of that notion, she is thus expressing discomfort. That may well be enough to suggest to her host to turn up the thermostat, without requiring a further speech act beyond her assertion of (12).
In this abstemious spirit, Bertolet (1994) observes that, just because x serves as a y, it does not follow that x is a y. (A might contort her body so as to play the role of an umbrella to keep a baby out of the sun; this does not imply that A is an umbrella, even for a moment.) So too, even if A’s remark that Selim is standing on his foot serves as a request, this does not imply that it is a request. In fact, it would seem incorrect to report what A has done with the following words:
As with (12), in remarking that Selim is standing on my foot, A might also express a desire that he move, without also performing a second illocutionary act beyond that of remarking. Parsimony, then, suggests that, in many cases, what much pragmatics literature terms indirect speech acts, are better described as expressions of attitudes that tend to elicit appropriate responses in cooperative addressees.
A similar strategy for type-a cases of indirection is also appealing. In pretending to compliment B’s sartorial choice, A might, with the appropriate complex of intentions, also be asserting that B’s clothing is unattractive; yet a more modest approach suggests that in so doing, A may merely be expressing a derogatory attitude toward it. In light of the gloss offered above of expression as a designed manifestation of a psychological state, we need only conjecture that A’s choice of words is designed to make light of B’s fashion sense. Making an evidently implausible comment on B’s clothing choice would be a good way to do that.
7. Speech Acts and Conversations
Speech act theory has developed in relative isolation from other lines of research in pragmatics such as the influential account of implicature associated with Grice (1989), who posits a set of norms implicitly governing much conversational practice. One such over-arching norm is the Cooperative Principle—“Make your contribution, such as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk-exchange in which you are engaged,”—(Grice, 1989, p. 26), which in turn subsumes individual Conversational Maxims, such as those enjoining speakers to be relevant, truthful, sufficiently informative, and clear (pp. 26–27.) Grice employs this apparatus to account for the ability of speakers to speaker-mean more than they say, but implicature theory encourages students of communication to consider the conversational milieu in which speech acts occur. The Cooperative Principle makes reference to the purpose or purposes of the conversation in which interlocutors are engaged, and as Green (1995) argues, such purposes seem to fall into distinct kinds. One characteristic aim of a conversation is to answer a question about what is the case; another is to formulate a plan of action. In yet other cases, one speaker aims to convince another with either a how-things-are answer or a what-to-do answer to a question. A dimension central to the development of any of these conversations will be common ground (CG), defined as that set of propositions that are accepted, and mutually known to be accepted, among all parties to a conversation. From this kinematic perspective, one making an assertion characteristically proffers its content for entry into CG.2 As CG evolves, questions are answered, plans are formulated, and doubters are convinced.
Speech act theory motivates a refinement of this kinematic approach to conversation. CG is defined in terms of the notion of acceptance, but different forces correspond to different ways of accepting a proposition. Scientists might accept a conjecture into CG, but wish to keep its content separate from that part of CG representing what is definitely known. (The latter might be used to justify calculations on which human life will depend, whereas conjectures would only be so used as a last resort.) Accordingly it may be most useful to posit multiple conversational common grounds, each corresponding to a different force with which contents have been proffered and accepted. So too, a group of interlocutors’ common ground might contain a question they are all committed to answering. With questions represented as sets of propositions, it may be possible to represent the way in which many conversations aim at resolution by ruling out all but one element of a question-set.
Attention to illocutionary force also makes it possible to discern different proprieties for conversational challenges. This is due to the fact that members of the assertive family (assertion, conjecture, sheer guess, educated guess, and the like) differ from one another in respect of the norms governing their conversational role. In response to A’s assertion that P … , it is appropriate for B to offer the challenge, “How do you know that?”, normally obliging A to provide strong reason in support of P even if that involves deferring to a distinct authority (“I read it in today’s paper,” “Susan told me,” etc.). Had A put forth P as an educated guess, however, such a challenge would be inappropriate. A challenge to an educated guess would be evidence against the claim, while an appropriate challenge to a sheer guess would instead demand strong counter-evidence.
8. Illocutionary Oppression
Research on speech acts has traditionally assumed that all interlocutors are equally able to contribute their illocutions to an ongoing conversation. However, recent scholarship has challenged this assumption by noting that, in a variety of situations, one speaker’s ability so to contribute may be compromised. This may be the result of a pattern of illocutionary abuses: a speaker’s long history of making claims not justified by the evidence, or that he does not believe to be true, will deter others from taking his utterances at face value. An extreme case is one in which his stating that P provides others with no evidence at all either for that proposition’s truth or for his belief in its truth. Likewise, a pattern of not paying off one’s lost bets may make it impossible to bet with others, simply because of their refusal of uptake.
Another possible source of illocutionary disablement lies in social conditions that are no fault of the speaker’s. Widespread bias concerning race, gender, disability, or sexual orientation may result in a speaker being treated as if she has a record of reneging on her bets. Langton (1993) argues that, in a milieu in which women’s refusals of sexual advances are not taken seriously by men (who might tell themselves that such coy behavior is all part of courtship), an attempt to reject such an advance with the words, “No, thanks,” or even, “Get lost!” might misfire in the sense delineated above: in so speaking, the woman performs a locutionary but no illocutionary act. Bird (2002) challenges this claim by observing that we need, and thus far lack, an argument that refusing and rejecting are like betting (which requires uptake) rather than warning (which does not: a homeowner can warn potential intruders by posting a Beware of Dog sign in her yard, and has still warned all such intruders even if no one takes heed). So too, it is not clear that refusing and rejecting require uptake. Nevertheless, even if bias does not cause putative speech acts to misfire, it may still weaken their perlocutionary effect. The source of the romantic overtures may continue his unwanted advances in the face of one’s ignored (if genuine) refusals.
In light of the discussion in Section 5 of the overlap of speech act theory and conversational kinematics, it emerges as well that, since assertions and their kin are typically proffered for entry into CG, bias may result in a speaker’s being put to a higher standard than are others before her contributions are accepted for use in CG. Her male counterparts may demand that a female employee provide proof, rather than solid evidence, that her policies will help their corporation gain market share. A jury may find a non-white defendant guilty-until-proven-otherwise even if they are not conscious of this bias. As such, that defendant’s testimony, even under oath, may be discounted even if it contains no misfires.
9. Critical Analysis of the Scholarship
In emphasizing the many uses of language other than stating how things are, the Ordinary Language movement that dominated English-speaking philosophy in the middle of the 20th century laid the groundwork for the growth of speech act theory. Wittgenstein (1968) and Ryle (1949) were central figures in this movement. Rorty (1968) anthologizes other influential work in this vein. However, Austin’s (1962) charismatic presentation brought unprecedented prominence to the discussion of speech acts, and offered hope, not only that the topic would enable philosophers to confront ancient philosophical problems with a new set of tools, but also that they might unearth a subject matter with a rich internal structure such as was concurrently being revealed by linguists in their investigation of syntax. After Austin’s untimely death, his student, Searle, defended a reformulation of Austin’s treatment of speech acts in his influential work of 1969. Searle objected to Austin’s distinction among five categories of speech act (verdictives, exercitives, commissives, expositives, and behabitives) on the grounds that (a) it investigates its subject matter through an examination of speech act verbs available in particular languages rather than the concepts that those verbs pick out, (b) it is based on no clear principle(s) of distinction, and (c) many illocutionary acts occur in more than one category. Searle proposed instead to investigate the range of all possible illocutionary acts, including those unnamed in any languages, and did so via the notions of (a) the point of the speech act, (b) the normative structure underpinning it, and (c) its sincerity condition, among others. He extended and refined his approach with widely discussed treatments of metaphor and indirect speech acts (reprinted in Searle, 1979). Searle and Vanderveken (1985) provides a formal approach to speech acts purporting to give a complete characterization of each such act in terms of how it fulfills seven parameters, which the authors call illocutionary point, degree of strength of illocutionary point, mode of achievement, content conditions, preparatory conditions, sincerity conditions, and degree of strength of sincerity conditions. Many of the essays in Tsohatzidis (2007) critically examine Searle’s mature theory of speech acts.
Austin and Searle shared the assumption that speech acts are essentially conventional acts whose conventionality goes beyond those imbuing words with literal meaning. As such they were vulnerable to a challenge made possible by Grice’s (1989) elucidation of meaning in terms of communicative intentions. Strawson (1964) expressed this challenge forcefully, and it is extended by Bach and Harnish’s (1979) systematic treatment of speech acts in intentionalistic terms.
Tsohatzidis (1994) contains influential contributions from linguists, computer scientists, and philosophers, while Clark (1996) is a classic work situating speech acts within a larger theory of cognition and communication. Carston (2003) systematically investigates the ways in which pragmatic factors help determine locutionary meaning rather than illocutionary force and perlocutionary effects. In the last decade, some scholars have looked critically at the longstanding distinction between illocutionary force and semantic content; others have sought to explain aspects of illocutionary force in terms of how it modifies conversation: this approach is thus analogous to dynamic semantics. Sbisà and Turner (2013) contains detailed investigations of particular speech acts such as requesting, asserting, and complimenting, as well as articles on the role of speech acts in legal theorizing and in ritual.
Scholars of ethics, social and political philosophy have turned in the last decade to speech acts as an area for investigating the oppression of marginalized groups; Hornsby (2000) and Anderson, Haslanger, and Langton (2012) provide overviews, while Maitra and McGowan (2012) features work from a number of prominent contributors to the topic.
Alston, W. (2000). Illocutionary acts and sentence meaning. Ithaca, NY: Cornell University Press.Find this resource:
Anderson, L., Haslanger, S., & Langton R. (2012). Language and race. In D. Fara & G. Russell (Eds.), Routledge companion to the philosophy of language (pp. 753–767). London: Routledge.Find this resource:
Austin, J. L. (1962). How to do things with words (2d ed.). Cambridge, MA: Harvard University Press.Find this resource:
Bach, K., & Harnish, R. (1979). Linguistic communication and speech acts. Cambridge, MA: MIT Press.Find this resource:
Carston, R. (2003). Thoughts and utterances. Oxford: Basil Blackwell.Find this resource:
Clark, H. (1996). Using language. Cambridge, U.K.: Cambridge University Press.Find this resource:
Fogal, D., Harris, D., & Moss, M. (Eds.) (2017). New work on speech acts. Oxford: Oxford University Press.Find this resource:
Green, M. (2015). Speech acts. Stanford Encyclopedia of Philosophy. Stanford, CA.Find this resource:
Grice, H. (1989). Studies in the way of words. Cambridge, MA: Harvard University Press.Find this resource:
Holdcroft, D. (1978). Words and deeds: Problems in the theory of speech acts. Oxford: Oxford University Press.Find this resource:
Hornsby, J. (2000). Feminism in the philosophy of language: Communicative speech acts. In M. Fricker & J. Hornsby (Eds.), The Cambridge companion to feminism in philosophy (pp. 87–106). Cambridge, U.K.: Cambridge University Press.Find this resource:
König, E., & Seimund, K. (2007). Speech act distinctions in grammar. In T. Shopen (Ed.), Language typology and semantic description (pp. 276–324). Cambridge, U.K.: Cambridge University Press.Find this resource:
Maitra, I., & McGowan, M. (Eds.) (2012). Speech and harm: Controversies over free speech. Oxford: Oxford University Press.Find this resource:
Recanati, F. (1987). Meaning and force: The pragmatics of performative utterances. Cambridge, U.K.: Cambridge University Press.Find this resource:
Roberts, C. (2017). Speech acts in discourse context. In D. Fogal, D. Harris, & M. Moss, (Eds.), (forthcoming). New work on speech acts. Oxford: Oxford University Press.Find this resource:
Rorty, R. (1968). The linguistic turn: Essays in philosophical method. Chicago: University of Chicago Press.Find this resource:
Ryle, G. (1949). The concept of mind. Chicago: University of Chicago Press.Find this resource:
Sbisà, M., & Turner, K. (Eds.) (2013). Handbook of pragmatics, Vol. 2: Pragmatics of speech actions. Berlin: de Gruyter-Mouton.Find this resource:
Searle, J. (1969). Speech acts: An essay in the philosophy of language. Cambridge, U.K.: Cambridge University Press.Find this resource:
Searle, J. (1979). Expression and meaning: Studies in the theory of speech acts. Cambridge, U.K.: Cambridge University Press.Find this resource:
Searle, J., & Vanderveken, D. (1985). Foundations of illocutionary logic. Cambridge, U.K.: Cambridge University Press.Find this resource:
Smith, B. (1990). Toward a history of speech act theory. In A. Burkhardt (Ed.), Speech acts, meaning and intentions: Critical approaches to the philosophy of John Searle (pp. 29–61). Berlin: Walter de Gruyter.Find this resource:
Stalnaker, R. (2017). Dynamic pragmatics. In D. Fogal, D. Harris, & M. Moss, (Eds.), (forthcoming). New work on speech acts. Oxford: Oxford University Press.Find this resource:
Tsohatzidis, S. L. (Ed.) (1994). Foundations of speech act theory: Philosophical and linguistic perspectives. London: Routledge.Find this resource:
Tsohatzidis, S. L. (Ed.) (2007). John Searle’s philosophy of language: Force, meaning, and mind. Cambridge, U.K.: Cambridge University Press.Find this resource:
Wittgenstein, L. (1968). Philosophical investigations (3d ed. (G. E. M. Anscombe, Trans.)). London: Blackwell Publishers.Find this resource:
Austin, J. (1962). How to do things with words (2d ed.). Cambridge, MA: Harvard University Press.Find this resource:
Bach, K. (1975). Performatives are statements too. Philosophical Studies, 28(4), 229–236.Find this resource:
Bach, K. & Harnish, R. (1995). Standardization versus conventionalization. Linguistics and Philosophy, 18(6), 677–686.Find this resource:
Bach, K., & Harnish, R. (1979). Linguistic communication and speech acts. Cambridge, MA: MIT Press.Find this resource:
Bar-On, D., & Green, M. (2010). Lionspeak: expression, meaning, and communication,’ with Dorit Bar-On. In E. Rubenstein (Ed.) (2010). Self, language and world: problems from Kant, Sellars and Rosenberg (pp. 89–106). Atascadero, CA: Ridgeview, 2010.Find this resource:
Bell, M. (1975). Questioning. The Philosophical Quarterly, 25(100), 193–212.Find this resource:
Bertolet, R. (1994). Are there indirect speech acts? In S. Tsohatzidis (Ed.), Foundations of speech act theory: Philosophical and linguistic perspectives (pp. 335–349). London: Routledge.Find this resource:
Bird, A. (2002). Illocutionary silencing. Pacific Philosophical Quarterly, 83(1), 1–15.Find this resource:
Carston, R. (2003). Thoughts and utterances. Oxford: Basil Blackwell.Find this resource:
Clark, H. (1996). Using language. Cambridge, U.K.: Cambridge University Press.Find this resource:
Ginet, C. (1979). Performativity. Linguistics and Philosophy, 3(2), 245–265.Find this resource:
Green, M. (1995). Quantity, volubility and some varieties of discourse. Linguistics and Philosophy, 18, 83–112.Find this resource:
Green, M. (2007). Self-Expression. Oxford: Oxford University Press.Find this resource:
Green, M. (2009). Speech acts, the handicap principle, and the expression of psychological states. Mind and Language, 24, 139–163.Find this resource:
Green, M. (2013). Assertions. In M. Sbisà & K. Turner (Eds.), Handbook of pragmatics, Vol. 2: Pragmatics of speech actions (pp. 387–410). Berlin: de Gruyter-Mouton.Find this resource:
Green, M. (2017). How much mentality is needed for meaning? In K. Andrews & J. Beck (Eds.). The Routledge handbook of the philosophy of animal minds. New York: Routledge.Find this resource:
Green, M. (-2017). A refinement and defense of the force/content distinction. In D. Fogal, D. Harris, & M. Moss (Eds.) New work on speech acts. Oxford: Oxford University Press.Find this resource:
Grice, H. (1989). Meaning. In Studies in the way of words (pp. 213–223). Cambridge, MA: Harvard.Find this resource:
Hornsby, J., & Langton, R. (1998). Free speech and illocution. Legal Theory, 4, 21–37.Find this resource:
Jary, M. (2007). Are explicit performatives assertions? Linguistics and Philosophy, 30(2), 207–234.Find this resource:
Langton, R. (1993). Speech acts and unspeakable acts. Philosophy and Public Affairs, 22, 293–330.Find this resource:
Lemmon, E. (1962). On sentences verifiable by their use. Analysis, 22, 86–89.Find this resource:
McDowell, J. (1998). Meaning, communication, and knowledge. Reprinted in Meaning, Knowledge and Reality. Cambridge, MA: Harvard University Press.Find this resource:
Portner, P. (2004). The semantics of imperatives within a theory of clause types. SALT XIV Proceedings: Semantics and Linguistic Theory.Find this resource:
Reimer, M. (1995). Performative utterances: A reply to Bach and Harnish. Linguistics and Philosophy, 18(6), 655–675.Find this resource:
Roberts, C. (2004). Context in dynamic interpretation, In L. R. Horn & G. L. Ward (Eds.), The Handbook of Pragmatics (pp.197–220). London: Routledge.Find this resource:
Roberts, C. (2012). Information structure in discourse: Towards an integrated formal theory of pragmatics. Semantics and Pragmatics, 5(art. 6), 1–69.Find this resource:
Russell, B. (1919). Descriptions. In B. Russell (Ed.), Introduction to mathematical philosophy London: Chatto & Windus.Find this resource:
Sadock, J. M., & Zwicky, A. M. (1985). Speech act distinctions in syntax. Language typology and syntactic description, 1, 155–196.Find this resource:
Sbisà, M. (1995). Speech act theory. In J. Verschueren, J. Östman, & J. Blommaert (Eds.), Handbook of pragmatics (pp. 495–506). New York: John Benjamins.Find this resource:
Sbisà, M. (2007). How to Read Austin. Pragmatics, 17, 461–473.Find this resource:
Searle, J. (1968). Austin on locutionary and illocutionary acts. Philosophical review, 77(4), 405–424.Find this resource:
Searle, J. (1969). Speech acts: An essay in the philosophy of language. Cambridge, U.K.: Cambridge University Press.Find this resource:
Searle, J. (1979). Indirect speech acts. In Expression and meaning: Studies in the theory of speech acts (pp. 30–57). Cambridge, U.K.: Cambridge University Press.Find this resource:
Searle, J. (1989). How performatives work. Linguistics and Philosophy, 12(5), 535–558.Find this resource:
Sinnott-Armstrong, W. (1994). The truth of performatives. International Journal of Philosophical Studies, 2(1), 99–107.Find this resource:
Stalnaker, R. (2014). Context. Oxford: Oxford University Press.Find this resource:
Strawson, P. (1950). On referring. Mind, 59(235), 320–344.Find this resource:
Strawson, P. (1964). Intention and convention in speech acts. Philosophical Review, 73(4), 439–460.Find this resource:
(1.) Grice’s original term ‘non-natural meaning’ has been replaced in recent years by ‘speaker meaning.’ Also, certain kinds of anonymous communication may be cases of speaker meaning as well. If A finds a stick of antiperspirant prominently displayed on his desk at work, he will likely infer that a colleague means he should attend better to his hygiene, without knowing who is the source of the message.
(2.) This generic claim is compatible with the existence of exceptions, such as when one asserts what is already part of CG, to make sure that all interlocutors are paying attention to it, or when one makes an assertion in full knowledge that others will not accept what one says. For a fuller account of conversational common ground see Stalnaker, 2014.