It is commonplace when discussing putatively anomalous speech acts to claim that engaging in them is like “speaking on-stage,” that is, to compare them to the speech acts made by actors during their theatrical performances. But comparatively little work has been by way of elucidating such speech acts, and without an adequate account of them, such comparisons will ultimately prove to be empty. In this paper, I will defend an illocutionary pretense view, according to which actors pretend to perform various kinds of illocutionary acts rather than genuinely performing them. This is, of course, a fairly intuitive position to take. What I want to argue, however, is that this is the route one must take: there are simply no tenable alternatives.
This paper will consist of four parts. First, I will develop criteria of genuine illocutionary action and explain why the on-stage variant fails to satisfy them. Second, I will sketch the three central approaches to theatrical discourse: Searle’s illocutionary pretense account, Curry’s sui generis theatrical illocution account, and Saltz’ game model. Third, I will present a number of difficulties that arise for the latter two approaches. And finally, I will develop a version of the illocutionary pretense account and defend it from objections.
It is worth noting that Searle’s and Currie’s views are, strictly speaking, offered as accounts of fictional rather than theatrical discourse. But fictional story-telling is best viewed as a species of theatrical performance in which story-tellers portray the narrators of the stories they tell. As a result, the aforementioned theories of fictional discourse can be used without revision as accounts of theatrical discourse.
I: Genuine Illocution and Theatrical Discourse
The question here concerns what kind of speech acts actors make during their theatrical performances. Let me elaborate. The utterances actors make on-stage count as illocutionary actions on the part of the characters they portray. So, for example, an actor’s utterance of “The king is dead” – or “Hand me the dagger” – might count as an assertion that the king is dead – or a request for a dagger – on the part of an assassin she is portraying. But it is important to note that it is the character portrayed, rather than the actor who portrays him, who makes this request or assertion. The actor makes what we might call an “on-stage assertion” or an “on-stage request” by means of her utterance. And our question is what kind of speech act is that.
Now a naïve first suggestion is that on-stage illocutionary acts are genuine illocutionary acts. When an actor makes an utterance during a theatrical performance which counts as an assertion on the part of the character she is portraying that the King is dead, the actor herself asserts that the King is dead. For familiar reasons, however, this suggestion simply will not do. The genuine performance of an illocutionary act of a given kind requires that a number of criteria be satisfied, several of which are not satisfied when actors make their on-stage illocutions.
For present purposes, the important criteria of genuine illocution are the illocutionary-intention criterion and the sincerity-obligation criterion. First, according to the illocutionary-intention criterion, in order to perform an illocutionary act of a certain kind, a speaker’s utterance needs to be prompted by an illocutionary intention appropriate to acts of that kind – in general, the intention to produce some effect upon the listener by means of his recognition of this intention. So, for example, in order to assert that the King is dead, a speaker needs to intend that her listener come to believe that the King is dead by means of his recognition that her intention that he believe it; and in order to request that a listener hand her a dagger, a speaker needs to intend that the listener pass her the dagger by means of his recognition that she intends that he do so.
And second, according to the sincerity-obligation criterion, a speaker who makes an illocutionary act of a certain kind is under an obligation to be in a mental state corresponding to the effect she intends to produce in listeners. So for example, a speaker who asserts that the King is dead – and thereby intends to produce a belief that the King is dead in her listeners – is under an obligation to believe it herself; and a speaker who requests a listener hand her a dagger is under an obligation to desire that the dagger be passed to her. It is important to note, however, that a speaker who fails to meet this obligation may nevertheless make a genuine illocutionary act, albeit one that is insincere. A speaker fails to satisfy the sincerity-obligation criterion – and hence fails to make a genuine illocutionary act – only if she is under no obligation to be in an appropriately corresponding mental state.
There are three separate reasons for thinking that when actors make on-stage assertions, requests, etc., these criteria fail to be satisfied. First, as matter of fact, actors need not and typically do not find themselves in the requisite mental states. An actor who uttered ‘The King is dead” during a theatrical performance would not ordinarily either believe that the King is dead or intend to instill this belief in either other actors on-stage or the viewing audience. And an actor who uttered “Hand me the dagger” would ordinarily intend that another actor pass her the relevant item – or desire that he do so – only if the script calls for him to do so. There are, of course, exceptions; the members of a guerilla theatre group of a certain kind, for example, might intend to produce such effects in an unsuspecting public. Nevertheless, the fact that actors so often fail to be in the requisite mental states when making on-stage assertions, etc., without their speech acts being, for that reason, defective suggests that the illocutionary-intention criterion (at least) simply does not apply to on-stage illocutions.
Second, there is no reason to think that when they make their on-stage illocutions, actors are under sincerity obligations, that is, that they are obligated to believe what they assert on-stage, to desire what the request on-stage, etc. Not only do audiences refrain from criticizing actors who do not believe their on-stage assertions – or for failing to keep on-stage promises – in most ordinary cases they would be surprised if actors did so. In fact, an actor who believed many of her on-stage assertions in a performance of theatrical fiction – or kept her on-stage promises – would be judged by most of us to simply be deluded.
And third, when a speaker makes a genuine illocutionary act, her utterance is caused by her illocutionary intention. But except when they are engaged in improvisational performances, the on-stage utterances of actors are caused instead by the scripts they are following. An actor who, for example, utters the sentence “The King is dead” at the point in a theatrical performance when the script instructs her to do so typically does so because of this scriptural instruction, and not because of any illocutionary intentions she might have. Moreover, should the demands of the script and the demands of her illocutionary intentions diverge, an actor who followed the latter would rightfully be subject to criticism. 
II: Three Theories of On-Stage Illocution
There are three central alternatives to the naïve account of on-stage illocution critiqued above. The most intuitive view, and the one I favour, is the illocutionary pretense account. According to this view, rather than making genuine illocutionary acts, actors by means of their on-stage utterances merely pretend to do so. The advantage of this approach to on-stage illocutions is that there is no reason to suppose that pretending to make an illocutionary action requires having illocutionary intentions of any kind or that it places a speaker under any kind of sincerity-obligation. In order to pretend to assert that the King is dead, a speaker need not intend that her listeners believe that the King is dead nor is she under any obligation to believe it herself. Moreover, there is no reason to suppose that having one’s utterances caused by a script in any way interferes with one’s capacity to engage in illocutionary pretense. Note: this view will be developed and defended in more detail below.
The second approach takes on-stage illocutions to be a sui generis species of illocutionary action, distinct from and irreducible to familiar illocutions such as assertions and requests. This theatrical illocution is distinguished from other sorts of illocutionary action in terms of the intended effects upon listeners. Whereas the illocutionary intention characteristic of assertion, for example, is listener belief in the asserted proposition, the goal of theatrical illocutionary action is that the listener imagine or make-believe the proposition expressed by her utterance. Although this approach does presuppose that actors’ on-stage illocutions are prompted by illocutionary intentions and that they are governed by sincerity-obligations, the intentions and obligations characteristic of sui generis theatrical illocutions are more plausibly attributed to actors than the intentions and obligations characteristic of more familiar illocutionary actions. In order to make an on-stage assertion to the effect that the King is dead on this view, an actor need only intend that her audience imagine what she says and not believe it; and she is obligated only to desire that her audience imagine what she says and not believe it herself.
The third approach takes on-stage illocutions to be genuine and familiar illocutionary actions prompted by game intentions and subject to game obligations. There are in games are a number of distinct roles with which are associated a number of goals. In chess, for example, associated with each player role is the goal of capturing the other player’s king. As a result, a player who occupies a game-role and who desires to play the game properly will adopt the goals associated with this role. Moreover, during a game a player is under an obligation to adopt the ends associated with the role she occupies: a chess player who does not intend to capture her opponent’s King is violating her obligations by throwing the game. What is important to note, however, is that a player remains motivated to pursue such goals and is subject to such obligations only long as she occupies her game-role. Once the game is over – or she has stopped playing – game intentions and obligations no longer apply.
Now according to Saltz, the roles occupied by actors on-stage are analogous to the roles occupied game-players: the intentional states adopted by the agents who occupy them derive from rules governing the practice or institution of theatrical performance. In particular, on Saltz’s view,
“… actors can perform real and sincere actions on stage by adopting game intentions that arise if they accept as part of the convention of performance a rule that actors work to achieve the conditions of satisfaction implied by the character’s actions.”
The idea is that the character an actor portrays has a number of intentional states which can be inferred from his actions. And just as rules of chess attach the goal of capturing the opponent’s King to the player-roles, the rules of theatrical practice attach to any given character-role the goal of achieving the character’s ends. So, if an actor desires to fulfill the function associated with a character-role she occupies within the broader theatrical production, she will adopt the goals of the character she is thereby portraying. For example, an actor who portrays a character who intends to instill in the audience the belief that the King is dead – and who desires to properly play the “theatrical game” – will herself intend to instill this belief in the audience. Moreover, an actor is under an obligation to adopt the goals of a character she portrays. After all, if she fails to do so, she is throwing the “theatrical game.” As should be obvious, this view entails that on-stage illocutions are prompted by the illocutionary intentions of actors and are subject to sincerity-obligations. But what is important is that actors have these intentions and are subject to these obligations only while they are on-stage occupying character-roles. Note: Saltz’s view is merely that actors may make genuine illocutionary actions by means of the utterances they make during their theatrical performances and not that they must or inevitably do so. In the critique of this position below, I will argue that actors cannot perform genuine illocutionary actions.
III: Lost Illocutions
There are three central difficulties I wish to present here against the view that on-stage illocutions are sui generis theatrical illocutionary acts. First, Searle has argued that the Functionality Principle – “[in] general, the illocutionary act (or acts) performed in the utterance of the sentence is a function of the meaning of the sentence…” – and the fact that actors use the very same sentences off- and on-stage together undermine the existence of uniquely theatrical illocutions. After all, otherwise the sentences of our language would have to be systematically ambiguous – differing in (literal) meaning as between their on-stage and off-stage uses.
One might, of course, balk at the functionality principle itself. But even if it is ultimately unsatisfactory, a similar argument against theatrical illocutions can be generated using only Hoffman’s weaker (and less contentious) Expressibility Principle: for any illocutionary act, there is a possible linguistic expression whose literal meaning is such that its utterance in the right circumstances is the performance of that act. Assuming, as above, that the sentences of English are not systematically ambiguous, this principle and the fact that actors use the very same sentences off- and on-stage together suggest that the sentences actors utter on-stage do not literally express theatrical illocutions. Rather, there must be distinct (possible) sentences which literally express the theatrical illocutions which the sentences actors utter on-stage figuratively express. Moreover, the most promising candidates for literally expressing theatrical illocutions – sentences such as, “Make-believe that the King is dead!” – actually express commands, suggestions, or the like – distinguished from other acts of those kinds in terms of their intended (perlocutionary) effects upon listeners – rather than sui generis theatrical illocutions.
The second difficulty stems from the fact that actors often generate on-stage pragmatic implications by means of their on-stage assertions and other on-stage illocutions. An actor portraying a character attempting to encourage an assassination conspiracy might, for example, on-stage implicate that the King has no redeeming features by on-stage asserting that the King cheats only infrequently at cards, when asked why he should be allowed to retain the throne. Now ordinarily pragmatic implications are generated by the interplay between speakers’ assertions and various conversational maxims. The trouble is that sui generis theatrical illocutions are not governed by the familiar conversational maxims. Consider, for example, Grice’s maxim of quality:
G-QUALITY: do not say what you believe to be false.
An actor who makes such a speech act is under no obligation to believe what she says and so can generate no implicatures by her failure to do so. One might, of course, invoke communicative maxims specifically governing theatrical discourse such as,
F-QUALITY: do not invite readers to make-believe what you do not intend to be fictionally true,
or, perhaps, very general maxims from which both F-QUALITY and G-QUALITY follow. But in lieu of an independent argument for the existence of such maxims, this strategy has more than a hint of ad hoc-ness to it.
The third difficulty is that the view suffers from a lack of generality in a number of dimensions. Recall: theatrical illocutions are supposed to be distinguished from other illocutionary acts by the intention that the listener imagine the proposition expressed by a speaker’s utterance. Now first, while this may provide an adequate analysis of on-stage assertions, it seems ill-suited to handle on-stage questions, commands, etc. An actor who makes the on-stage request that a listener hand her the dagger, for example, is unlikely to primarily intend that he imagine her in possession of the dagger. Second, although this analysis may give an adequate account of on-stage assertions when they are directed to the viewing audience, it seems ill-suited to handle them when they are directed towards other actors on-stage. An actor who on-stage asserts that the King is dead to another during a theatrical performance is more likely to intend to prompt his next his line rather than intend he imagine anything. And finally, the whole strategy of appealing to uniquely theatrical illocutions to explain theatrical speech acts cannot plausibly be applied to theatrical action more generally. Consider, for example, a play in which one character kisses and then stabs another. The actor portraying this character might reasonably be taken to actually perform the kiss and pretend to perform the stabbing, but there would be little motivation to suppose that in either case she performed a sui generis on-stage kind of action which is neither pretense nor genuine kissing or stabbing.
IV: Game Over
The game model has a number of advantages over the sui generis theatrical illocution account. Since it does not posit a distinct kind of illocutionary act, it sidesteps the Searle-Hoffman worry altogether. And for the same reason, it can accommodate on-stage pragmatic implications without having to appeal to a special set of theatrical conversational maxims. After all, the familiar maxims arguably govern assertions and other non-theatrical illocutionary actions even when they are prompted by game intentions rather than ordinary illocutionary intentions. Moreover, it is not subject to the same failures of generality as the theatrical account. According to the game model, actors can genuinely perform the whole gamut of illocutionary actions in virtue of having the requisite game intentions. And there is no reason to suppose it cannot be incorporated into a broader theory of theatrical action. A wide variety of theatrical actions can be understood as genuine when similarly backed by game intentions.
Nevertheless, the game model does face serious difficulties. First, although it does show how actors could satisfy the illocutionary-intention criterion of genuine illocutionary action without having anomalous mental states off-stage, it fares less well with the sincerity-obligation criterion. This is because even if an actor is under an obligation to adopt the goals of a character she portrays, this does not suffice to place her under sincerity-obligations. An actor under an obligation to adopt her character’s goal simply may be under no obligation to believe what she says, desire what she requests, etc. Consider, for example, an actor portraying a dishonest character whose goals include the deception of various of his conversational interlocutors. In such circumstances, her character-obligations – to adopt the goals of any character she portrays – would require that she disbelieve what she says, rather than believe it as would be required by any sincerity-obligations she might be under. Of course, at various points in a given performance, the demands of an actor’s character-obligations may coincide with the demands of the sincerity-obligations appropriate to the illocutionary actions she putatively makes. If, for example, a character’s goals require that that she desire what she requests on a certain occasion, then an actor who is obligated to adopt the character’s goals is thereby obligated to desire what she requests on that occasion. But this does not suffice to place an actor under a sincerity-obligation. After all, to suggest otherwise is to suggest that one can be under a genuine obligation to engage in a course of conduct if the obligation holds only if the course of conduct furthers one goals.
And second, although the game model may provide an adequate account of speech acts which occur during improvised theatrical performances, it runs afoul of the causal role of the script in non-improvised performances. Insofar as an actor performs genuine illocutionary actions, as the game model would have it, her utterances are caused by her illocutionary intentions rather than the script. Moreover, insofar as she genuinely adopts the goals of the character she portrays, her illocutionary intentions themselves are the product of her beliefs about how to best achieve these ends rather than the script. But in non-improvised performances, it is the script – and not any goals they might adopt – which cause actors’ utterances.
Saltz attempts to alleviate such worries by noting that by manipulating “…[the] definition of the fictional context and the way actions are performed…” during rehearsals, an actor can ensure that she adopts a set of goals which will yield the verbal (and non-verbal) behaviour required by the script. The idea is that in rehearsal an actor can try out various hypotheses about the mental states of the characters she is to portray and various ways of uttering her scripted dialogue. The hope is that this will enable her to discover a set of ends which, if adopted, would cause her to utter her scripted dialogue. And once, after sufficient rehearsal, she is able to reliably adopt such ends, she can engage in performances wherein her speech acts are caused by her illocutionary intentions but at the same time conform to the script.
The trouble with this suggestion is that it mischaracterizes the obligations of actors in at least normal scripted performances. On Saltz’s view, an actor’s obligation is to act in accordance with her adopted ends. And an appropriate process of rehearsal can render it highly likely that doing so will yield behaviour that conforms to the script. But if unforeseen circumstances arise, acting in accordance with her adopted ends may require a significant departure from the script. And according to the game model, under such circumstances an actor is obliged to make the requisite departure. But as a matter of fact, the obligations of an actor in a normal scripted performance are to conform to the script under such circumstances. An audience which attends a performance of a play in which the actors make significant departures from the script for the sake of “Saltzian authenticity” will rightly feel cheated. Of course, small departures from the script can be tolerated, and a performance that does not feel forced is preferable to one that does. Nevertheless the primary obligation of the actors is to perform the play that was advertised.
More generally, the obligations of an actor are determined by the purpose or function of the broader performance in which she participates. It is, of course, not always an easy matter to discern the purpose of a performance, and in some instances there may be no determinate purpose. Nevertheless, there are many cases in which the purpose is clearly not served by actors with an overriding interest in authenticity. Even in improvised performances actors may have obligations which override their putative obligations to pursue their adopted ends; consider, for example, improvised comedy. There may, of course, be some performances whose purposes place actors under an overriding obligation to adopt the goals of the characters they portray, but this certainly does not hold across the board. Moreover, as we have seen, actors in such performances are not subject to sincerity-obligations and hence do not perform genuine illocutionary acts in any event.
V: Illocutionary Pretense
In light of the difficulties raised here, the prospects for treating on-stage speech acts as a species of genuine illocution – either as a kind of sui generis theatrical illocutionary action or as familiar illocutionary acts backed by game intentions – appear grim. The natural alternative is to take actors to be engaging in a kind of illocutionary pretense – pretending to make assertions, requests, etc. – rather genuinely performing such acts. The sense of pretense at issue here is that of pretending to do, as opposed to pretending to be or pretending that.  Now in order to pretend to do something an agent needs to actually do something, but the type of action in which she actually engages is typically distinct from the action-type she pretends to perform. And a general account of pretending-to-do requires an account of the conditions that need to hold in order to for an agent to pretend to engage in an act of one type by means of actually engaging in an act of another type. Our somewhat more delineated task is to determine the conditions that need to hold in order for actors to pretend to make familiar speech acts of various kinds by means of the utterances they actually make on-stage.
Now Searle has suggested that a necessary condition of pretending to do something is that the act in which the agent actually engages be a constituent part of the act in which she pretends to engage:
“It is a general feature of the concept of pretending that one can pretend to perform a higher order or complex action by actually performing lower order or less complex actions which are constitutive parts of the higher order or complex action.”
For example, one can pretend to eat an apple by holding one’s empty hand by one’s mouth while moving one’s jaw as if masticating. Saltz, however, rightly balks at the generality of this suggestion. Children, for example, often pretend to shoot one another “by pointing their index finger at their victims with their thumbs upward, and shouting the word ‘bang’” but these behaviours are not constitutive parts of the higher order behaviour of genuinely shooting someone. Alternatively, one might suppose that an agent needs to imitate an action in order to pretend to engage in it – or, equivalently, that a necessary condition of pretending to do something is that the agent perform an act that resembles it. But again, as Saltz points out, it is possible to pretend to engage in an action by actually doing something with bears no non-trivial resemblance to it. Consider, for example, a play or children’s game of make-believe in which saying “Cheep, cheep!” counts as flying.
Saltz does, however, make a positive suggestion about “theatrical pretense” which may ultimately prove fruitful. On his view, an actor who pretends to murder “… is (really) committing an action which counts as murder within the conventions of the play.” The core general idea is that pretense requires the existence of conventions according to which actually performing an action of one type counts as pretending to engage in an act of another type. These conventions can be local – as in the case of the conventions of a particular production of a given play – but they can also hold quite generally. In my view the appeal of Searle’s “constitutive action” analysis of pretense, as well as the imitation account, is due to the existence of very general conventions according to which one can pretend to engage in an action by engaging in a constituent part of it or by imitating it. It should be noted, however, that an agent’s pretense can be facilitated by such conventions only if they are operative in the context in which she acts. This is why one can always pretend to fly by imitating or engaging in a constituent part of flight behaviour but one can do so by means of uttering “Cheep, cheep!” only if one is performing in the right kind of play.
Now illocutionary pretense typically involves general conventions that are operative in a very wide range of contexts. One can almost always pretend to make an assertion or request, etc. by means of producing an utterance (or inscription) which is highly, or even exactly, similar to the kinds of utterances speakers’ use to make genuine assertions, etc. Of course, by means of the very kind of same verbal behaviour in the very same contexts speakers can also make genuine illocutions. The difference between the two cases is that in the former speakers invoke or deploy the requisite pretense-conventions whereas in the latter they do not.
What needs to be addressed, however, is how one goes about invoking such conventions. Searle’s view seems to be that what is required is the intention to do so on the part of a speaker: “[one] cannot truly be said to have pretended to do something unless one intended to pretend to do it.” And while I agree with Searle that this is all that is required, I think, contra Searle, that this opens the door for deceptive illocutionary pretense which is to be distinguish from other sorts of deception like lying and other failures to satisfy one’s sincerity obligations. Let me elaborate. Normally when engaging in illocutionary pretense, a speaker indicates to her audience that she is doing so. This can involve explicit statements to that effect, various kinds of gestures, shifts in tone of voice, or even speaking from the vantage of a literal stage. The important thing to note is that one can intentionally invoke pretense-conventions without giving one’s audience any indication to that effect; consider, for example, a guerilla theatre company of the kind suggested above. And in so doing a speaker deceives her audience. But rather than lying or, more generally, attempting to get them to falsely believe that she has satisfied her sincerity-obligations, such a speaker deceives her audience by attempting to get them to believe she is genuinely making an assertion or a request, etc., and, hence, is even under sincerity-obligations. Orson Welles’ famous radio broadcast of The War of the Worlds arguably counts as a deception of this kind as opposed to an outright lie.
As should be obvious, the illocutionary pretense account of on-stage illocution avoids the difficulties beset its competitors. First, since, on this view, actors do not make genuine illocutionary actions at all, it simply sidesteps any requirement that they have appropriate illocutionary intentions or be under sincerity-obligations. In order for an actor to assert on-stage that the King is dead, she need not intend to produce this believe in her audience nor be under any obligation to believe it herself, she need only pretend to so intend and to be so obligated. Second, again because actors do not make genuine illocutionary actions, there is no reason to suppose that their utterances are caused by anything other than the script in non-improvised performances. A theatrical performance can be viewed as a kind of choreographed pretense in which the participating actors use the pretended actions of their dramatic cohorts as cues to engage in the pretense required of them by the script. And third, since the illocutionary actions they pretend to perform are governed by the familiar conversational maxims, actors can, on this view, generate on-stage implications by means of their on-stage assertions. An actor who pretends to assert a proposition which implicates, in the fictional situation which the performance in which she is participating generates, that a certain King has no redeeming qualities thereby pretends to make this latter implication.
Finally, one might worry whether the illocutionary pretense account is compatible with the kind of authentic performance wherein actors “act in unison” with the characters they portray, which so exercises Saltz. Saltz seems to presuppose that theatrical authenticity requires that actors literally adopt and be moved to act by their characters’ mental states. According to the illocutionary pretense account, in contrast, the only mental state required of an actor is that she intend to invoke certain pretense-conventions; and she can be moved to act by her own desire to follow a script, her desire to be funny, or any number of other things. But as we have seen, Saltz’s requirements are too stringent: they render authentic performance impossible. Moreover, there is what I take to be a fruitful way of understanding authentic performance which is compatible with the illocutionary pretense account of on-stage illocution: an actor can imagine being/ sharing the mental states of the character she is portraying; and from this imaginative perspective, she can imagine she is making genuine illocutionary acts while she is in fact merely pretending to do so.
Department of Philosophy
Alward, Peter (manuscript), “Truth in Fiction: The Story Concluded.”
---- (forthcoming), “For the Ubiquity of Non-Actual Fact-Telling Narrators: A Reply to
Kania,” Journal of Aesthetics and Art Criticism. Journal of Aesthetics and Art Criticism
Austin, J. L. (1979), “Pretending,”
in Philosophical Papers, 3rd
Edition, J.O. Urmson and G.J. Warnock, eds.,
Currie, Gregory (1990), The Nature of Fiction,
Grice, Paul (1957), “Meaning,” Philosophical Review 66: 377-88.
---- (1975), “Logic and
Conversation,” in Syntax and Semantics,
vol. 3: Speech Acts, P. Cole and J.L. Morgan (eds),
Hoffman, Sarah (2004), “Fiction as Action: Currie and Searle on Speech Act Theory and the Nature of Fiction,” Philosophia 31: 513-29.
Lewis, David (1978), “Truth in Fiction,” American Philosophical Quarterly -46.
Saltz, David (1991), “How to Do Things on Stage,” Journal of Aesthetics and Art Criticism 49 (1): 31-45.
Searle, John (1969), Speech Acts: An Essay in the Philosophy of
---- (1975), “The Logical Status of Fictional Discourse,” New literary History 6: 319-332.
---- (1983), Intentionality,
 Saltz (1991) is a notable exception.
 Searle, 1975.
 Currie, 1990.
 Saltz, 1991.
 See Alward, forthcoming, manuscript. Even when telling fictional stories which lack explicit narrators, story-tellers are best viewed as portraying non-actual tellers who tell the story as fact rather than as fiction.
 Searle (1975) and Currie (1990) attempt to utilize their accounts of story-teller speech acts to mark the boundary between fiction and non-fiction. In my view, this reflects a failure to properly distinguish between fictional composition and story-telling proper.
 See, e.g., Grice (1957) and Searle (1969, 1983).
 Searle (1983) invokes a more nuanced account of the intended effects upon listeners. Such subtleties, however, do not concern us here.
 See Saltz (1991), p. 36 ff.
 Searle, 1975.
 Currie, 1990. Strictly speaking, Currie is concerned with fictional rather than theatrical discourse. But if, as I have suggested above, fictional discourse is just a species of theatrical discourse, Currie’s account should straightforwardly carry over to the latter.
 Saltz, 1991.
 Saltz (1991), p. 39.
 Saltz (1991), p. 39.
 Searle (1975) p. 324. Again, Searle’s focus is on fictional rather than theatrical discourse.
 Currie (1990), pp. 14-5.
 Hoffman, 2004.
 Strictly speaking, the conclusion that can be drawn here is that if actors express theatrical illocutions by means of their on-stage utterances then either their on-stage utterances do not literally express theatrical illocutions or their off-stage utterances do not literally express non-theatrical illocutions. The latter disjunct is, I take it, implausible on its face.
 One might, of course, simply adopt the view that on-stage illocutions are commands or suggestions or invitations to make-believe as an alternative to the theories presented here. This idea, however, runs into two distinct difficulties. While it may yield an adequate analysis of on-stage declarative utterances, it seems ill-suited to handle on-stage interrogatives, etc. And although it may give an adequate account of such utterances when they are addressed to the viewing audience, it seems ill-suited to handle them when directed towards other actors on-stage.
 Grice, 1975.
 Alward, forthcoming, manuscript.
 Of course, certain theatrical actions – stabbings, e.g. – are usually not genuine but merely pretended.
 Strictly speaking, in Saltz’ view, actors have an obligation only to act as if they have the beliefs of the characters they portray. (Saltz 1991, p. 40).
 The analogy here is a little off. The case at hand is one in which an actor is putatively under an obligation which holds only if the conduct furthers the goals of the character she is portraying. But is should be emphasized that, on the picture at issue, the actor adopts the character’s goals as her own.
 Saltz (1991), p. 40.
 Note: Saltz’s emphasis is on how actors’ utterances can be caused by illocutionary intentions in scripted performances; my emphasis is on how they can be caused by the script if they are genuine illocutionary acts.
 Suppose, for example, a prop bridge collapses foiling the opportunity of an actor portraying an assassin to kill a character who is just briefly on the other side. In order to satisfy his adopted goal, the assassin-actor will have to find extra-scriptural means to terminate his intended victim.
 There might, of course, be “non-theatrical” obligations – moral or otherwise – which override this obligation in certain circumstances.
 I understand the performance of actions here broadly enough to include omissions.
 In fact, my view is that the action-type actually performed in pretense is necessarily distinct from the action-type the agent thereby pretends to perform. Someone who is actually washing windows, for example, cannot be pretending to wash windows (although she can pretend to be a window-washer). This issue is orthogonal to our present concern.
 Searle (1975), p. 327. Searle does not take the satisfaction of this condition to be sufficient for engaging in pretense, requiring additionally at least that the agent intend to do so. The psychological component of pretense will be discussed in more detail below.
 Saltz (1991), p. 32.
 Saltz (1991), p. 42.
 Saltz (1991), p. 42. Strictly speaking, Saltz offers this suggestion as an account of what actors “really do” on stage rather than as a necessary condition of what I am calling theatrical pretense.
 Arguably, at least, they are inoperative in certain very solemn contexts like funerals or criminal trials. Of course, one could rejoin that they are operative even in these contexts but to invoke them in such contexts would be inappropriate.
 Note: following Searle, we might instead take such utterances to be constituents of genuine illocutionary actions. Nothing in what follows will hang on how the verbal behaviour by means of which actors engage in illocutionary pretense is characterized.
 Contra Searle (1975, p. 326) I do not see any point in taking pretense-conventions to be “horizontal conventions” which suspend the normal requirements of “vertical” linguistic conventions which establish connections between language and the world. Simply understanding them to be pretense-convention suffices to establish that the normal requirements of genuine illocutionary action are inoperative.
 Searle (1975), p. 325.
 Searle (1975, p. 326) says, “ [what] distinguishes fiction from lies is the existence of a separate set of conventions which enables the author to go through the motions of making statements which he knows to be not true even though he has no intention to deceive.”
 I am assuming here that pretense is transparent in the following sense: if an agent pretends to do A, and doing A entails doing B, then the agent pretends to do B as well, whether she is aware of so doing or not.
 See note #41.
 Saltz (1991), p. 42.