TY - JOUR
AU1 - GARCÍA‐CARPINTERO, MANUEL
AB - ABSTRACT In recent work, Kendall Walton has abandoned his very influential account of the fictionality of p in a fictional work in terms of prescriptions to imagine emanating from it. He offers examples allegedly showing that a prescription to imagine p in a given work of fiction is not sufficient for the fictionality of p in that work. In this article, both in support and further elaboration of a constitutive‐norms speech‐act variation on Walton's account that I have defended previously, I critically discuss his objections. In addition to answering his concerns and developing the account further, I provide additional abductive support for its explanatory virtues vis‐à‐vis institutional accounts like Walton's and Gricean speech‐act proposals. I. INTRODUCTION: CONTRASTING VIEWS ABOUT FICTION‐MAKING Kendall Walton's work on fiction and the imagination has been deservedly influential on philosophical and empirical work. Walton did not advance his account as conceptual analysis (Friend 2008, 152–154; Woodward 2014, 825). On his view, fictions are artifacts with a socially ascribed function from which specific prescriptions to imagine result, whether or not this fits folk pretheoretical notions; this determines the fictional content of the work, what is fictional according to it: “a proposition is fictional in (the world of) a particular work, W, … just in case full appreciation of W requires imagining it” (Walton 2015, 17). In spite of Walton's skepticism about philosophical analyses, writers influenced by him take their work in that spirit. Gregory Currie (1990) argues that fictions result from a sui generis speech act, fiction‐making,1 satisfactorily addressing in my view Walton's (1990, 85–89) concerns regarding nonverbal fictions or works treated as fictions that were not produced as such. Following Currie, I have also suggested in previous work that we should take fiction‐making to be a specific sort of invitation to imagine, addressed to a specific kind of audience (García‐Carpintero 2007, 2013, 2016, 2019b). Like Currie, I thus reject that literary acts of fiction‐making are just (as Mitchell Green aptly puts it) “acts of speech” (2017b, 54)—say, acts of pretending to do something, bereft of the representational aims of speech acts, as John Searle (1975) has it—as opposed to speech acts proper, with specific force and contents.2 I will briefly revisit these contrasting views about fiction‐making. From the earlier stages of speech‐act theory, there has been a divide between descriptive, psychological theories (favored by H. Paul Grice, Strawson, Stephen Schiffer, and their followers) and prescriptive or normative theories (preferred by Austin, Alston, Searle, and their followers). The former take psychological attitudes to be constitutive of speech acts. On a Gricean view, it is constitutive of an assertion that p that the speaker intends the hearer to believe p, or to believe that the speaker believes p, on the basis of the recognition of that intention. Fiction‐making is analogously characterized by the intention to lead the audience to imagine its content by the recognition of that intention. Kathleen Stock (2017) offers a compelling version of the view, addressing objections raised over the years by Stacie Friend (2008, 2012) and others, which Currie (2014), an earlier proponent, also takes up with hesitant conclusions. Griceans know that we evaluate assertions, criticizing them, for instance, when they are false, and that we similarly criticize fictions when they lack interest: when they fail to thrill, entertain, challenge, or move us, when they cause imaginative resistance, and so on (Stock 2017, 135). Griceans, however, take the norms thereby deployed to be regulative—derivable from non‐illocutionary norms, perhaps moral or prudential ones. Austinian normative accounts take instead some of those norms to be constitutive of the relevant speech acts on the model of games whose natures are thought to be normatively specifiable. Unlike Currie's and Stock's, the account I have defended (García‐Carpintero, 2013, 2016, 2019b) is Austinian: it incases Waltonian prescriptions in a normative speech‐act account. Fictions result from proposals by actual fiction‐makers for their audiences to imagine. They thereby constitutively involve prescriptions for the audience to imagine the propositions that constitute the fictional world, conditional as most prescriptions are on different features: the interests of the intended audience, their belonging to the intended class, their considering such propositions. The propositions whose imagining is prescribed in the relevant sense are specifically those such that the fiction‐maker becomes by his or her action beholden to a norm requiring his or her imagining to be a worthwhile project for the relevant audience.3 More specifically, I took as my model Timothy Williamson's (1996) “simple” formulation for the norms of assertion, deploying a constitutive norm that uniquely characterizes the act by its normative essence. For that act—the one we perform by default when uttering declarative sentences—Williamson advocates a knowledge rule, a norm requiring for correctness that the speaker knows the asserted proposition. An assertion with content p is the act whose product is subject to this rule: it is correct only if the speaker knows p. Other constatives have weaker requirements, bare truth, or justification for guessing or conjecturing, and so on. Norms like this are sui generis: they do not have their sources in moral or prudential rules, but fit in specifically illocutionary codes. They are defeasible and pro tanto: they can be overridden by stronger norms. It is possible to violate them, thereby rendering the acts not nonexistent (Austinian misfires), but wrong (Austinian abuses). There are many situations in which p is asserted but the speaker does not know p, because p is false or proper justification is lacking. The assertion is then wrong, relative to a norm constitutive of such a speech act. In order to apply this to fiction‐making, my proposal is modeled on a normative account of directives derived from Alston's (2000). Roughly, directives are subject to the norm that they are correct only if their audiences are thereby provided with a reason to see to it that their content obtains.4 The reason might be based on different sources, which further specify the nature of the directive: the authority of the speaker in the case of commands, the good will, or the presumed interests of the audience in the case of requests, suggestions, or proposals. Again, the norms are sui generis, defeasible, and pro tanto, and it is possible for them to be broken—as when a speaker invites the audience to do something that has no chance of satisfying their relevant desires. On this basis, my proposal was that a fiction with content p results from an act that is correct only if it gives relevant audiences (audiences of the intended kind, with the disposition to engage with such works) a reason to imagine p.5 The reasons in question have to do with whatever makes engaging with good fictions worthwhile: to experience the emotions generated by well‐drafted, suspenseful thrillers, say, or to put oneself in the shoes of other people in other circumstances, thereby getting knowledge about them. A reason I offered for that view was that normative accounts fare better relative to the intentionalism/conventionalism debate about the interpretation of fictions. There are compelling criticisms of Gricean accounts of core speech acts such as assertion based on cases of speakers performing them, even when lacking Gricean communicative intentions (say, officers in information booths). Analogously, I discussed examples showing that a fiction might have a specific interpretation, even if the author has not intended it to be imagined in its detailed specificity (García‐Carpintero 2013, 2019b). Another reason I have offered (García‐Carpintero 2016, 2019a, 2019b) is that the normative account provides a better understanding of the relation between fiction and truth, answering criticisms by Friend (2008, 2012) and providing a better alternative to her Waltonian institutionalist, anti‐essentialist genre account.6 In recent work, Walton (2015) has abandoned the account of fictionality quoted above; he offers several examples allegedly showing that, while a prescription to imagine p in a given work is necessary for the fictionality of p in that work, it is not sufficient. I cannot accept this. For the relevant prescriptions are, in my view, constitutive, definitional of fictionality (see Ohmann 1971, 15). Here, I critically discuss Walton's objections to his own proposal. I set its virtues in relief in contrast with Walton's, Peter Alward's (2009, 2010a, 2010b), and Friend's institutional view and with Currie's and Stock's (2017) orthodox Gricean account. The methodology to address these matters is broadly abductive. With Stock, I assume that studying the specific propositional imaginings that fictions call for or make appropriate—which she calls “F‐imaginings” (2017, 4–9)—is a good strategy to get a better understanding of the imagination itself. She convincingly argues that, when it comes to providing an account of the imagination, intuitions are particularly problematic because systematic agreement is missing, and, hence, that it is a better strategy to investigate the imagination by studying its theoretical role in well‐taken explanatory pursuits (7); Amy Kind (2013) makes a similar point. We are after an account of fictions and the imaginings they trigger with good explanatory payoffs, unburdened by any irrevocable imperative to accept ordinary usage and folk intuitions. I add that this is what I take conceptual analysis to be, in the spirit of Carnapian explications or rational reconstructions. From a Carnapian perspective, intuitions about cases indeed play a distinguished evidential role vis‐à‐vis data from fiction studies, critics, or cognitive psychologists, but they may nonetheless be overridden. In the following sections I first show (Section II) how the normative view of fiction‐making cashes out the intuitively and theoretically significant metaphor of the fictional world. In Section III I argue that it thereby provides us with an intuitively well‐supported distinction between propositional imaginings required by a fiction and those constituting the fictional world. Then I explain in Section IV how this helps us to deal with Walton's cases. II. FICTIONAL WORLDS To address Walton's (2015) objections, I argue that a normative speech‐act account of fictionality allows us to untangle what I claim is an ambiguity in what Stock calls “F‐imagining,” which I articulate in Section III. The key for it, and with it for my account of Walton's cases in Section IV, lies in the role I give to the metaphor of the world of the fiction, which Walton employs when, in summarizing his view in a quotation given above, he says that “a proposition is fictional in (the world of) a particular work, W, … just in case full appreciation of W requires imagining it” (2015, 17). This metaphor is put to several serious theoretical uses. Psychologists use it to explain “transportation” or “immersion” in a narrative—the experience of “suspending disbelief” so as to become concerned with fictional scenarios as if they were part of the actual world, no matter how wild.7 I show in this section how the proprietary notions of a normative view help us to cash out this metaphor. Stock stipulates F‐imagining to be “whatever kind of imagining is appropriate, at a minimum, as a response to fictional content” (2017, 20). This is, I take it, a technical notion well motivated by all accounts of fiction‐making we are considering, which in one way or another understand fictions as representational devices constitutively inviting imaginative responses. The nature of F‐imaginings is to be further specified by the theoretical role they play in the specific theoretical proposals positing them. I also agree with Stock that F‐imaginings have propositional contents (2017, 20–27). I think we should adopt a minimalist attitude toward propositions, such as the view originating with Robert Stalnaker that they are just properties of circumstances of evaluation (see Richard 2013). What are such circumstances? They are complete and consistent possible worlds for Stalnaker, for Lewisians, centered possible worlds. For reasons presented below, I need to take them to be more fine grained, in the tradition of situation semantics updated in the “truthmaker semantics” of Yablo (2014) and Fine (2017). Stock might concur with this, for she prefers to take F‐imaginings to be about “scenarios” rather than complete and consistent fictional worlds (2017, 27), on account, I think, of worries related to my own. I nonetheless keep speaking mostly about the fictional world, because this is more standard than fictional scenario and I need to emphasize the intuitive character of this notion. I fully agree with her in addition that not only can such a “world” not be identified with a complete and consistent possible world, but it cannot be reductively analyzed in such terms either.8 Formal semantics offers tools to better understand the notion of a fictional world. Truthmaker semantics develops the Austinian idea that truth (the standard of correctness for assertoric speech acts) concerns not the full actual world, but more fine‐grained situations made of objects, events, and properties that the representational acts are about (Kratzer 2017). Francesco Berto (2017a) invokes work in that tradition to articulate in a sufficiently precise sense the idea of a fictional world that F‐imaginings are in a similar sense about.9 Berto's account assumes some “explicit input,” as in supposition‐based reasoning, and provides a theoretical account of an acceptable logic of the imagination that leads us beyond that explicit input while we stick “with what the explicit input is about,” offering inferential patterns that he takes to be valid (2017a, 1875). This assumes a version of the standard distinction between direct, or primary fictional truths, and indirect or implied ones, as in Lewis's ([1978] 1983) or Walton's (1990) work.10 In Lewis's view we get the primary content on the assumption that the fictional narrator tells the fiction “as known fact,” and we derive implied truths from either a “Reality Principle” or a “Mutual Belief Principle.”11 This assumes something controversial that I nonetheless grant, to wit, that verbal fictions have fictional narrators, explicit or implicit.12 As Alward nicely puts it, actual storytellers of verbal fictions “portray” them, the way actors play characters: “fictional storytelling is best viewed as a species of theatrical performance in which storytellers portray the narrators of the stories they tell” (2009, 321). Thus, in creating Don Quixote, Cervantes “plays” the first‐personal narrator—whom we may or may not identify with Cervantes (see Wilson 2011, 114–115)—who first presents himself as passing information gleaned from archives in La Mancha, and then his translation of an Arabic historical narrative by a Cide Hamete Benengeli.13 Now, Walton was skeptical that, when it comes to F‐imaginings, there are any such principles that are sufficiently general to be codified (1990, 139). Stock shows that fiction‐making intentions can trump even the most solid inference principles used to derive fictional content (2017, 64, 180). Thus, she would accept the default validity of Berto's adjunction rule (2017a, 1879–1880), for it corresponds to her principle that “fiction is normally conjunctive” (Stock 2017, 27): when both p and q are fictional in a given fiction, they “co‐occur with respect to the same scenario” (27). However, she shows that even this rule has exceptions, in cases of fictions presenting a plurality of worlds (2017, 169–174); this is illustrated below. Nonetheless, this is not a reason to be skeptical about the distinction between direct and indirect fictional content and the notion of fictional world that comes with it, only to conclude that all such inference principles, including the more purely logical ones that Berto studies, are defeasible. The distinction is intuitively plausible, and it can be theoretically grounded. The explicit contents of a given fiction are, roughly, those features of the fictional world fixed by what the author primarily, immediately pretends to do or has other actors pretending to do. This includes non‐illocutionary pretend actions and illocutionary acts. In putting forward a declarative sentence, the author immediately “plays” the fictional narrator asserting what is said by it, similarly for questions, and so on. Implicatures by the actual fiction‐maker that may add to the character of the thereby presented fictional world, like the outputs of inferences based on the aforementioned principles, are not immediate.14 A notion of the fictional world also plays an important role in “semantic” alternatives to “pragmatic” accounts of fictional utterances like mine. I did not mention them in the previous section because, to the extent that they constitute a real alternative, I have serious doubts about them. Thus, Stefano Predelli (1997), François Recanati (2000, 213–226), Marga Reimer (2005), and Alberto Voltolini (2006) have advanced semantic contextualist views of fictional discourse. In a narrative in the historical present, the context in which “[t]he battle unfolded now” is uttered requires us to evaluate the assertion not with respect to the time when the utterance is made, but rather with respect to another, contextually provided time. On these views, the context of fictional utterances similarly leads us to evaluate their truth not at the actual world, but a counterfactual or imaginary one, the world of the fiction.15 These proposals are committed to fictional referents for fictional terms or instead make claims that I think we should not accept,16 so I go along with Walton's that “truth in fiction” is not really a form of truth (1990, 41–42).17 Summing up: there is an intuitive metaphorical notion of the world that a fiction is about, which is deployed for significant theoretical purposes in accounting for fiction. It captures the measure of objectivity that fictional content has, in particular when it comes to nonexplicit content, allowing for error and ignorance in fiction consumers, and even on the side of fiction‐makers themselves. It is also useful to explain fictional immersion, and it has received elucidation in formal semantics. On my view, this helpful metaphor is to be cashed out in terms of the prescriptions to imagine assumed in the normative speech‐act account. Fictions result from the specific norms to which fiction‐makers commit themselves, appraising the particular proposals to imagine they issue in creating them. The fictional world realizes the plot of the story, consisting of the contents on whose imagining by the intended audience a proper appraisal of the fiction‐maker's proposal is to be grounded. In other words, it is the scenario whose imaginative representation the fiction‐maker commits to be worth his or her while for the audience. The fictional world is thus on my view determined by the prescribed fictional truths.18 Fictional worlds are indeterminate, as shown by Beardsley‐like silly questions, such as how many children Lady Macbeth had—which, following Lorand (2001, 428), I distinguish from Walton‐like silly questions, considered in Section III. It is not that they include indeterminate objects, which—putting aside weird postmodernist fictions—they do not (Berto 2017a, 1880–1881); it is just that the grounding facts leave the issue undecided (see Taylor 2018). It is not fictional in the world of Macbeth that Lady Macbeth has exactly one child, and it is not fictional either that she has at least two; Macbeth leaves the question undecided. J. Robert G. Williams and Richard Woodward (forthcoming) point out an ambiguity that normative accounts of fictionality like Walton's or mine have not broached so far. For this indeterminacy can be taken (as they put it) in either a “permissivist” or, rather, a “prohibitionist” way. On the latter reading, one ought not imagine it one way, and one ought not imagine it the other. On the former, one is allowed both. Given that the prescriptions that determine the fictional world are those required for competent appreciation of the work, that is, for proper appraisal of the speech act from which it results, the prohibitionist line is the best take on the indeterminacy that Beardsley silly questions exhibit, I submit: whatever one does is for such purposes irrelevant and therefore fails to characterize the fictional scenario. When it comes to imaginatively construe it, the reader ought not imagine that Lady Macbeth has exactly one child and ought not imagine that she has at least two (granted that it is to be imagined that she has some).19 I disagree with Williams's and Woodward's view that prohibitionism is adequate for cases of “deliberate ambiguity” creating interpretative plurality. A Turn of the Screw is a well‐known example; whether Deckard is a replicant in the world of Blade Runner is another good case. The best account here is permissivist. I take this to follow from the view that the work determines two worlds, only partially overlapping, and that full appreciation requires imagining at least one, and, if one also imagines the other, to avoid conjoining them.20 This is also Stock's view (2017, 105, 170). III. ANCILLARY VERSUS CONSTITUTIVE PRESCRIPTIONS TO IMAGINE The normative view allows us to make the problematic ambiguity that I have announced in Stock's F‐imagining, “whatever kind of imagining is appropriate, at a minimum, as a response to fictional content” (2017, 20). For there are imaginative acts required to understand a fiction—hence appropriate responses to it, and thus, prima facie F‐imaginings by Stock's stipulation—that are merely ancillary or subordinate to the determination of the world that the text invites proper appreciators to imagine, without constituting it. Walton's objections to his previous account, I argue in the next section, also neglect this distinction between two different prescriptions emanating from fictions. I distinguish (merely) a(ncillary)‐imaginings, from the c(onstitutive)‐imaginings that make the fictional world. In this Section I illustrate the distinction with sufficiently uncontroversial examples, and I explain how it fits with the normative account. Mindless stories provide my first example: “once upon a time, when there were no people and things did not have names, there was this little dragon, Urkul ….” On my account of proper names, different homonyms are just different words (García‐Carpintero 2017). The reader is, hence, to imagine that there is a dragon picked out by a specific “Urkul”‐naming practice.21 However, that there is such a practice is clearly not intended to be part of the (main) content of the fiction.22 I thus count it as a mere a‐imagining.23 Alternatively, following Alward (2010b, 357) and Walton (2015, 31–33), we could take a “plurality of fictions” view about these cases, placing the fictional narrator and the assumed naming practice in a scanty embedding fictional world in which the fiction presenting the richer fictional world is told. The worlds of the two fictions are non‐overlapping, unlike the Quixote world—which both the first‐person narrator and his sources (Cide Hamete Benengeli, the La Mancha archives) share with Don Quixote, with whom they causally interact. If we adopt this pluralist alternative, the normative account still helps to explain the relative order of importance of the imaginings: the thin world is the one less relevant for the appraisal of the representational act by its constitutive norms. Authorial intrusions such as Brönte's in Jane Eyre (“A new chapter in a novel is something like a new scene in a play; and when I draw up the curtain this time, reader, you must fancy you see a room …”) provided a second example. Intuitively, they are not meant to constitute the fictional world; the asides are serious assertions by the fiction‐maker, commenting on the story and its characters (see Wilson 2011, 118–119n16). However, readers must entertain (that is, F‐imagine) them in appropriately responding to the fiction. A third illustration is Walton's silly questions; why does Othello speak in such a fine verse, or why do Vietnamese peasants in American films speak English? In contrast with the line defended above on Beardsley's silly questions, I do not think that prohibitionism applies here. Engaging with the fiction requires F‐imagining that Othello speaks in fine verse and the peasants, English, but these are just a‐imaginings. We ought not imagine these as features of the (core) world of the fiction. However, I think we ought to imagine that Othello does not speak in verse and the Vietnamese peasants do not speak English; these are c‐imaginings, determined by reality assumptions. Then again, we might think the issue to be indeterminate (Wilson 2011, 122–125) and these to be cases of multiple stories. If so, as argued above, we would not be adopting prohibitionism anyway, but permissivism. Unreliable narrators, like Nabokov's Kinbote in Pale Fire, offer a final example. He tells us that a Zemblan assassin intending to kill Zembla's deposed king (that is, Kinbote himself) accidentally killed the poet Shade. But we are not supposed to take this to be true in the world of the fiction, i.e., a c‐imagining part of the plot of the story we are presented to imagine. We are to infer that the killer is the insane Jack Grey who, wanting to kill the judge who put him away, mistook Shade for him (see Wood 1998, 198). To understand the sentences comprising the fiction, we must grasp what Kinbote asserts and, hence, F‐imagine it; but these are not c‐imaginings relevant for its proprietary appraisal. Other contents are c‐imaginings instead, which we only arrive at through inferences deriving them in part from such a‐imagining.24 As granted, these a‐imaginings can be treated as constituting additional thin stories. This is not my view, for in clear examples of fictions with a plurality of interpretations like Blade Runner imagining one of the worlds suffices for appreciation. But if we adopt it, the thin “plots” must still be ranked as lower in relevance for that purpose than, and subordinate to, those constituting the (core) fictional world. They would thus also count as ancillary. The distinction between a‐imaginings and c‐imaginings is thus intuitively well motivated. It also fits naturally the normative framework, for it is of the same kind as others required by it. I have embraced Williamson's hypothesis that constitutive rules are simple. However, as he makes clear, this does not exclude that other norms should also be taken into consideration in a full account of the relevant representational acts. It is just that constitutive rules play the crucial role of individuating their nature or distinctive essence. Among such additional rules there are the general norms that Williamson (2000, 256–257) invokes in order to distinguish primary and secondary propriety.25 A speech act is primarily correct just in case it obeys its constitutive norm and secondarily correct just in case it is reasonable for the agent to think that it does.26 Then there are “sincerity” norms, concerning psychological states expressed when indicating liability to a constitutive norm. These are more specific to particular speech acts, but nonetheless, on the constitutive norms approach (reversing the priorities on psychological views), they are derivative. Finally, there are “conceptual” or “preparatory” norms: those requiring general conditions for being liable to obligations and norms in the first place or grasping the resources used to indicate liability to constitutive norms, such as a natural language, or Gricean cooperative principles. The distinction between a‐imaginings and c‐imaginings belongs in the third group. A subject that makes a promise is prescribed to use a language that his or her audience can understand. He or she is also required to fulfill the promise, on the standard account of the relevant constitutive rule.27 But, assuming that the constitutive norms account is on the right track, both obligations do not have the same status: the latter is constitutive of what promising is, the former merely ancillary to promising. The distinction I have made, assuming a constitutive norm account of fiction‐making, is of the same kind. Summing up, in the previous section I pointed out that fictional worlds play important roles in theoretical pursuits such as the semantics of fictions—aiming to capture the “objectivity” of fictional contents, in particular, implicit ones—and psychological accounts of fiction understanding, and I cashed out the metaphor assuming a constitutive norms account. In this section I have shown how to distinguish in our framework imaginings constituting the (core) fictional world from imaginings merely auxiliary to it. IV. DEALING WITH WALTON'S OBJECTIONS I now use the distinction to address Walton's (2015) examples showing that a prescription to imagine p in a given work is not sufficient for the fictionality of p in that work. In contrast, Waltonian and Gricean accounts overlook the distinction. A first example concerns the exploitation of the sequential nature of our engagements with fictions for “garden path” effects: what seems a feature of the fictional world at a point is eventually revealed not to be so—as when the closing of Alice in Wonderland reveals everything from early on to have been a dream. Intuitively, such cases should be accounted for in that—as Woodward puts it—“the prescriptions generated by a work can change as the story develops, even though what is fictional is what we are prescribed to imagine after we have taken into account all relevant aspects of the work as a whole” (2014, 832). In my terms, imaginings that are prescribed at one stage only to be finally discarded are a‐imaginings, not c‐imaginings determining the world of the fiction. On the plausible assumption that, instead of constituting independent imaginative projects presenting their own isolated fictional worlds, serial fictions should be understood as adding features to the character of a unified world, they raise similar issues. Thus, it is already a feature of the world of the first Star Wars film, Episode IV: A New Hope, that Luke is Vader's son, even though this is only revealed in The Empire Strikes Back, and it would come as a surprise to competent viewers of the first film. Lee Walters offers a compelling formulation of a version of the “prescriptions to imagine” account of fictionality capable of accounting for these cases, which my own account can adopt (2017, 20). Walters's account for such fictions is similar to Stock's own proposal for garden path effects (Stock 2017, 64–66). Given that serial fictions might branch and that there does not need to be a final one supplying a “maximal” fiction, Walters defines fictional truths as those prescribed to be imagined after a given stage, so that the prescription is never revoked at a later stage. Except that, questionably, as I argue, Stock claims that these are fictions presenting a plurality of worlds/plots: they involve at least two different fictional worlds: the one with the misleading features revised at later stages and the “official” final one. Other of Walton's alleged counterexamples to his previous account concern fictions within fictions. He offers an example of a painting by Vermeer, in which a painting of Cupid is depicted; the dream that constitutes most of Alice in Wonderland is another. Appreciators of the Vermeer painting are to imagine that there is a naked, winged boy. But this is intuitively not part of the world of the painting; it is only part of it that there is in it a painting with such a content. The same applies to Alice going down the rabbit hole. As Walton (2015, 26–27) notes, a proposal like the one I am making also works well here: viewers must imagine that there is a winged, naked boy on their way to imagining what the work prescribes, but this is merely an a‐imagining that does not determine what is fictional in the work. Walton objects that the distinction is question‐begging, and it might well be on his account. But, as shown, it is perfectly well justified in the framework of a normative speech‐act account (see Woodward 2014, 831). Incurring an obligation (advancing a c‐imagining, in our case, making an assertion or other specific speech act) may demand incurring others (inviting an a‐imagining; assuming that one's audience has the conceptual resources needed for understanding one, speaking loud enough for them to hear). Given the specific speech act I take fiction‐making to be, it is well motivated to distinguish the fictional‐world‐constituting invitations to imagine from others merely ancillary to it. The same treatment can be given for metaphors, which Walton (2015, 25–26) discusses with a nice example from Katherine Mansfield. On my reading, Clarice Lispector's The Passion According to G. H. provides another good illustration. Readers will hardly forget the central epiphanic experience in the novel, when G. H. puts in her mouth the cockroach she has crushed. On my interpretation, this is just an allegoric metaphor for the character's handling of a more prosaic event alluded to in ways parallel to the cockroach episode—an abortion that leaves the character infertile. The novel requires readers to imagine the cockroach episode, but this does not obtain in the world of the fiction; it is just an a‐imagining, a means to ascertain what is fictional according to the novel. The example also shows that, even if an F‐imagining is only a means to ascertain fictional truth rather than constitutive of it, it can nonetheless contribute to the appraisal of the overall aesthetic value of the work. Of course, the case of Othello's verse makes the same point.28 I have granted that we could treat these cases as fictions presenting a plurality of worlds, that is, fictions with several correct interpretations, noting that, even if we do so, we need to make a distinction parallel to mine. Stock treats only some of them this way: fictions within fictions, garden path fictions, fictions with unreliable narrators (2017, 169–174). As a challenge to her views, I argue that this is not a stable position. Stock rejects views that rely on applications of Lewis's and Walton's Reality Principle—or, I take it, Friend's Reality Assumption that “everything that is (really) true is also fictionally the case, unless excluded by the work” (2017, 29)—unconstrained by the intentions of the author. She argues that these views have unintuitive consequences, producing fictional truths “which clash with features of the competent reader's experience and practice” (Stock 2017, 52). She illustrates this with silly questions, like why Tintin does not get old or Wodehouse's Wooster does not have cirrhosis: if these views were correct, readers should worry about such issues, but they do not (2017, 54). They populate fictional worlds with irrelevant facts (2017, 55); see also Alward (2010a, 350). Readers should be surprised when exposed to material contradicting well‐known facts, but they are not (Stock 2017, 56). On her account, the alleged fictional truths are blocked because they are not intended by the author (58). This offers a principled explanation, better than claiming that the facts are relatively unimportant parts of the fictional world without indicating the grounds for such ranking (62). However, as far as I can tell, the cases that she does want to treat as involving a plurality of interpretations do not differ in these respects from these. If it is not part of “the” Othello world intended by its author that the character speaks in verse or of the Jeeves and Wooster world that Wooster has cirrhosis or of either of these that Barcelona was called “Barcino” in Roman times, by the same token it is not part either of the world of Pale Fire that Shade was killed by a Zemblan assassin. For we can assume that it was not intended by Nabokov that it should be, and competent readers do not take it to be so. Alternatively, assume that, given that Nabokov must have intended his readers to at least entertain that thought, we must take it that there is a (thin) alternative interpretation presenting a world in which it is true. Why should it not then be the case that, in a similarly thin sense, Hergé did intend us to entertain a Tintin world in which he does get old, but this does not show (he has a Dorian‐Gray‐like picture at home, say), or an alternative one in which he does not—physical laws are suspended, so that he can be in so many different places in a short interval of time? I have argued in earlier work that we do not come to imagine what is true in a fiction along Gricean lines by recognizing an author's specific intention for us to do; rather, both author and audience rely on well‐established fiction‐interpreting practices (García‐Carpintero, 2016, 2019a, 2019b). But if we insist that we are somehow intended by authors to entertain the truths constituting the fictional world, we should also grant that we are similarly intended to entertain those here classified as merely ancillary. It is thus unclear how, on the basis of the materials available to Griceans, Stock can make the distinctions we need. Making them in some cases (unreliable narrators, garden path fictions) is unstably at odds with the reasons she offers to reject doing so in others (silly questions). As part of my global abductive argument, I advance this criticism as a challenge to Stock. As a reviewer notices, she makes a distinction that might be relevant. She says that, just as understanding an assertion does not require believing the content, “for fictional cognition, all the reader needs to do is understand what she is intended to F‐imagine, not F‐imagine it, though most of the time she will immediately F‐imagine it as well” (Stock 2017, 36). For this to work, Stock should explain the difference between entertaining a proposition in understanding, and (propositional) F‐imagining. The reviewer suggests that she could help herself to ideas not far from those I have invoked: “you only really engage with a fiction, or become immersed in it, when you imagine its contents.” Perhaps in those cases that she treats as involving a plurality of interpretations, readers are intended to engage with the thinner fictions she posits along these lines by becoming immersed in them. This suggestion should be elaborated, addressing concerns in the vicinity of those bedeviling Gricean accounts: how could we justify ascribing such discriminating intentions to fiction‐makers and their audience? Can Stock distinguish a sense in which Nabokov intended us to imagine that Shade was killed by a Zemblan assassin while Hergé did not intend us to imagine potential explanations for Tintin's age like those entertained above? We could put back to her the rhetorical question she poses to others, “on what grounds might one insist that no propositional imagining occurs here?” (2017, 125). And we can support the implicated answer the way she does: there are general considerations that explain why reflective readers might well entertain the propositions as appropriate responses to the fictions. As explained at the end of the previous section, on the account developed here what distinguishes c‐imaginings from a‐imaginings are normative facts about the evaluation of fictions by their proprietary norms. This is what on the view intuitions about the world of the fiction track. It is such facts that ground the impression that a‐imaginings do not constitute fictional worlds—or the pluralistic alternative that they constitute thin, subordinate worlds. Stock and Walton hold undiscerning views about F‐imaginings: for Stock they are imaginings intended by fiction‐makers, for Walton imaginings prescribed by fictions given their institutional function. Without further elaboration, these views neglect required distinctions. There may remain concerns about how illuminating the proposal is. I address them in closing, by contrasting it in this respect with Alward's, which also crucially appeals to fictional worlds: “the narrator in a fictional story is … the reader's informant regarding the fictional world described or generated by the text” (2010a, 357). Such a narrator is an explicit or implicit construct like Lewis's ([1978] 1983) fictional teller: an “informant” who “reveals” the fictional world (Alward 2010a, 358). We do not need to go into the specifics of Alward's view here.29 Now, Alward acknowledges a commitment to cash out the world metaphor: “Although I am committed to the utility of the ‘fictional worlds’ idiom, I do not take it to have ontological import” (358). This is what he says to discharge it: “The fictional world embodied by a text is just the collection of fictional truths embodied by it” (357). But the fictional truths were on his account just those that the narrator “reveals” in “informing” us about the fictional world! This is too tight a circle for it to be illuminating: fictional worlds consist of fictional truths … which are the contents of the fictional narrator's assertions revealing them. How does this help us to address the previous worries? How does it help, say, to decide whether it is part of the world of Pale Fire that a Zemblan assassin killed the poet Shade, as Kinbote tells us, or the insane Jack Grey instead, or somehow both? I have not provided a reductive account either. I have assumed that we already have a sense of the contours of fictional worlds, which informs different theoretical undertakings in cognitive psychology and formal semantics. However, by grounding fictional worlds on constitutive norms for fiction‐making, I have given an account establishing wider connections with other familiar notions. On my proposal, truth in fiction is grounded on norms deployed for the evaluation of fictions as such. It is true what we must entertain the thought that a Zemblan assassin killed the poet Shade to properly appreciate Pale Fire, but the “must” here is like the one in stating preparatory conditions for speech acts. In the sense expressing constitutive norms, we must instead imagine that the killer was Jack Grey. Of course, one might reject constitutive norms accounts and primitive normativity, but that is another concern. A reviewer says, “compare Lewis's analysis, which does not seem prey to the same kind of circularity or primitivity—as he analyzes truth in fiction in terms that don't assume it. … [D]oes it not retain this advantage over the author's?” To be sure, it does. Lewis's modal realism about possible worlds holds a similar advantage over the Stalnakerian account of the modal realm I endorse, which helps itself to primitive modality. But to many of us, in both cases this partial advantage does not add to a global one (García‐Carpintero 2007; Alward 2010a, 349–351; Stock 2017, 49–61). Lucid as always, Lewis himself comes close to admiting it, anticipating Walton's concern about embedded fictions ([1978] 1983, 280). To wrap up, in some cases (mindless fictions), truth in fiction/the fictional world is fixed by what the fictional narrator tells us as known fact, but the telling itself is not part of the fictional world. In some other cases (embedded fictions, unreliable narrators), only the telling or other intentional acts constitute the fictional world, not their contents. In still other cases, both what the narrator tells us as known fact and the telling constitute the fictional world, in spite of the fact that, if we were told such things in actuality, we would utterly refuse to accept them. Matravers nicely illustrates this with supernatural tales like The Master and Margarita (1997, 79). Or we candidly accept implied contents that we would not even contemplate if the fiction was really told as known fact, as I (García‐Carpintero 2007) illustrated with Cortázar's “Continuity of Parks.” I have shown that these facts, left unexplained by Lewis's account as much as by others I have discussed, are accounted for on the assumption that they are grounded on the normativity of fiction‐making: on the fictional world being fixed by the contents to which fiction‐makers commit themselves for the proper appraisal of their acts.30 This offers abductive support for my normative speech‐act variation on Walton's account of fictionality. It allows replies to Walton's objections to his views and provides explanatory benefits over Alward's and Friend's institutional proposals and Stock's Gricean account.31 REFERENCES Alston , William P. 2000 . Illocutionary Acts and Sentence Meaning . Cornell University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Alward , Peter . 2009 . “ Onstage Illocution .” The Journal of Aesthetics and Art Criticism 67 : 321 – 331 . Google Scholar Crossref Search ADS WorldCat Alward , Peter . 2010a . “ That's the Fictional Truth, Ruth .” Acta Analytica 25 : 347 – 363 . Google Scholar Crossref Search ADS WorldCat Alward , Peter . 2010b . “ Word‐Sculpture, Speech Acts, and Fictionality .” The Journal of Aesthetics and Art Criticism 68 : 389 – 399 . Google Scholar Crossref Search ADS WorldCat Berto , Francesco . 2017a . “ Aboutness in Imagination .” Philosophical Studies 175 : 1871 – 1886 . Google Scholar Crossref Search ADS WorldCat Berto , Francesco . 2017b . “ Impossible Worlds and the Logic of Imagination .” Erkenntnis 82 : 1277 – 1297 . Google Scholar Crossref Search ADS WorldCat Byrne , Alex . 1993 . “ Truth in Fiction: The Story Continued .” Australasian Journal of Philosophy 71 : 24 – 35 . Google Scholar Crossref Search ADS WorldCat Camp , Elizabeth . 2017 . “ Perspectives in Imaginative Engagement with Fiction .” Philosophical Perspectives 31 : 73 – 102 . Google Scholar Crossref Search ADS WorldCat Cumming , Samuel , Gabriel Greenberg, and Rory Kelly. 2017 . “ Conventions of Viewpoint Coherence in Film .” Philosopher's Imprint , vol. 17 , www.philosophersimprint.org/017001/. Google Scholar OpenURL Placeholder Text WorldCat Currie , Gregory . 1990 . The Nature of Fiction . Cambridge University Press . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Currie , Gregory . 2014 . “ Standing in the Last Ditch: On the Communicative Intentions of Fiction Makers .” The Journal of Aesthetics and Art Criticism 72 : 351 – 363 . Google Scholar Crossref Search ADS WorldCat DeRose , Keith . 2002 . “ Assertion, Knowledge, and Context .” Philosophical Review 111 : 167 – 203 . Google Scholar Crossref Search ADS WorldCat Everett , Anthony . 2013 . The Nonexistent . Oxford University Press . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Fine , Kit . 2017 . “ Truthmaker Semantics .” In A Companion to the Philosophy of Language , 2nd edition, edited by Bob Hale, Crispin Wright, and Alexander Miller, 556 – 577 . Oxford : Wiley Blackwell . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Friend , Stacie . 2008 . “ Imagining Fact and Fiction .” In New Waves in Aesthetics , edited by Kathleen Stock and Katherine Thomson‐Jones, 150 – 169 . Basingstoke : Palgrave Macmillan . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Friend , Stacie . 2012 . “ Fiction as a Genre .” Proceedings of the Aristotelian Society 112 : 179 – 209 . Google Scholar Crossref Search ADS WorldCat Friend , Stacie . 2017 . “ The Real Foundations of Fictional Worlds .” Australasian Journal of Philosophy 95 : 29 – 42 . Google Scholar Crossref Search ADS WorldCat Friend , Stacie . Forthcoming . Matters of Fact and Fiction . Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gale , Richard M. 1971 . “ The Fictive Use of Language .” Philosophy 46 : 324 – 340 . Google Scholar Crossref Search ADS WorldCat García‐Carpintero , Manuel . 2007 . “ Fiction‐Making as Gricean Illocutionary Type .” The Journal of Aesthetics and Art Criticism 65 : 203 – 216 . Google Scholar Crossref Search ADS WorldCat García‐Carpintero , Manuel . 2013 . “ Norms of Fiction‐Making .” British Journal of Aesthetics 53 : 339 – 357 . Google Scholar Crossref Search ADS WorldCat García‐Carpintero , Manuel . 2016 . “ To Tell What Happened as Invention: Literature and Philosophy on Learning from Fiction .” In Literary Studies and the Philosophy of Literature: New Interdisciplinary Directions , edited by Andrea Selleri and Philip Gaydon, 123 – 147 . London : Palgrave Macmillan . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC García‐Carpintero , Manuel . 2017 . “ The Mill‐Frege Theory of Proper Names .” Mind 127 : 1107 – 1168 , C10.1093/mind/fzx010. Google Scholar OpenURL Placeholder Text WorldCat García‐Carpintero , Manuel . 2018 . “ John Woods, Truth in Fiction: Rethinking Its Logic .” Notre Dame Philosophical Reviews , https://ndpr.nd.edu/news/truth-in-fiction-rethinking-its-logic/. Google Scholar OpenURL Placeholder Text WorldCat García‐Carpintero , Manuel . 2019a . “ Assertions in Fictions: An Indirect Speech Act Account .” Grazer Philosophische Studien 96 (forthcoming). Google Scholar OpenURL Placeholder Text WorldCat García‐Carpintero , Manuel . 2019b . “ On the Nature of Fiction‐Making: Austin or Grice? ” British Journal of Aesthetics 59 : 203 – 210 . Google Scholar Crossref Search ADS WorldCat García‐Carpintero , Manuel . Forthcoming. “ Do the Imaginings that Fictions Invite Have a Direction of Fit? ” In The Language of Fiction , edited by Emar Maier and Andreas Stokke. Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Grant , Robert . 2001 . “ Fiction, Meaning, and Utterance .” Inquiry 44 : 389 – 403 . Google Scholar Crossref Search ADS WorldCat Green , Mitchell . 2017a . “ Assertion .” Oxford Handbooks Online, https://doi.org.10.1093/oxfordhb/9780199935314.013.8. Green , Mitchell . 2017b . “ Narrative Fiction as a Source of Knowledge .” In Narration as Argument , edited by Paula Olmos, 47 – 61 . Cham, Switzerland : Springer . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Kania , Andrew . 2005 . “ Against the Ubiquity of Fictional Narrators .” The Journal of Aesthetics and Art Criticism 63 : 47 – 54 . Google Scholar Crossref Search ADS WorldCat Kania , Andrew . 2007 . “ Against Them, Too: A Reply to Alward .” The Journal of Aesthetics and Art Criticism 65 : 404 – 408 . Google Scholar Crossref Search ADS WorldCat Kind , Amy . 2013 . “ The Heterogeneity of the Imagination .” Erkenntnis 78 : 141 – 159 . Google Scholar Crossref Search ADS WorldCat Kratzer , Angelika . 2017 . “ Situations in Natural Language Semantics .” Stanford Encyclopedia of Philosophy , Winter 2017 Edition, edited by Edward N. Zalta. https://plato.stanford.edu/archives/win2017/entries/situations-semantics/. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lackey , Jennifer . 2007 . “ Norms of Assertion .” Noûs 41 : 594 – 626 . Google Scholar Crossref Search ADS WorldCat Lewis , David . ( 1978 ) 1983. “ Truth in Fiction .” In Philosophical Papers , vol. 1 , 261 – 280 . Oxford University Press . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Lorand , Ruth . 2001 . “ Telling a Story or Telling a World? ” British Journal of Aesthetics 41 : 425 – 443 . Google Scholar Crossref Search ADS WorldCat Matravers , Derek . 1997 . “ The Paradox of Fiction: The Report versus the Perceptual Model .” In Emotion and the Arts , edited by Mette Hjort and Sue Laver, 78 – 92 . Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Ohmann , Richard . 1971 . “ Speech Acts and the Definition of Literature .” Philosophy and Rhetoric 4 : 1 – 19 . Google Scholar OpenURL Placeholder Text WorldCat Predelli , Stefano . 1997 . “ Talk about Fiction .” Erkenntnis 46 : 69 – 77 . Google Scholar Crossref Search ADS WorldCat Predelli , Stefano . 2017 . Proper Names: A Millian Account . Oxford University Press . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Recanati , François . 2000 . Oratio Obliqua, Oratio Recta: An Essay on Metarepresentation . Massachusetts Institute of Technology . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Reimer , Marga . 2005 . “ The Ellipsis Account of Fiction‐Talk .” In Ellipsis and Nonsentential Speech , edited by Reinaldo Elugardo and Robert J. Stainton, 203 – 215 . Dordrecht : Springer . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Richard , Mark . 2013 . “ What Are Propositions? ” Canadian Journal of Philosophy 43 : 702 – 719 . Google Scholar Crossref Search ADS WorldCat Sainsbury , Mark . 2014 . “ Fictional Worlds and Fiction Operators .” In Empty Representations: Reference and Non‐Existence , edited by Manuel García‐Carpintero and Genoveva Martí, 277 – 289 . Oxford University Press . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Searle , John . 1975 . “ The Logical Status of Fictional Discourse .” New Literary History 6 : 319 – 332 . Google Scholar Crossref Search ADS WorldCat Stock , Kathleen . 2017 . Only Imagine: Fiction, Interpretation, and Imagination . Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Sutrop , Margit . 2002 . “ Imagination and the Act of Fiction‐Making .” Australasian Journal of Philosophy 80 : 332 – 344 . Google Scholar OpenURL Placeholder Text WorldCat Taylor , David E. 2018 . “ A Minimal Characterization of Indeterminacy .” Philosopher's Imprint 18 ( 5 ): 1 – 25 . http://hdl.handle.net/2027/spo.3521354.0018.005. Google Scholar OpenURL Placeholder Text WorldCat Voltolini , Alberto . 2006 . “ Fiction as a Base of Interpretation Contexts .” Synthese 153 : 23 – 47 . Google Scholar Crossref Search ADS WorldCat Voltolini , Alberto . 2016 . “ The Nature of Fiction/al Utterances .” Humana‐Mente: Journal of Philosophical Studies 25 : 28 – 55 . Google Scholar OpenURL Placeholder Text WorldCat Walters , Lee . 2017 . “ Fictionality and Imagination, Revisited .” The Journal of Aesthetics and Art Criticism 75 : 15 – 21 . Google Scholar Crossref Search ADS WorldCat Walton , Kendall L. 1990 . Mimesis as Make‐Believe: On the Foundations of the Representational Arts . Harvard University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Walton , Kendall L . 2015 . “ Fictionality and Imagination—Mind the Gap .” In In Other Shoes: Music, Metaphor, Empathy, Existence , 17 – 35 . Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Williams , J. Robert G. , and Richard Woodward. Forthcoming. “ The Cognitive Role of Fictionality .” Philosophy and Phenomenological Research . OpenURL Placeholder Text WorldCat Williamson , Timothy . 1996 . “ Knowing and Asserting .” The Philosophical Review 105 : 489 – 523 . Google Scholar Crossref Search ADS WorldCat Williamson , Timothy . 2000 . Knowledge and Its Limits . Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Wilson , George . 2011 . Seeing Fictions in Film: The Epistemology of Movies . Oxford University Press . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Wood , Michael . 1998 . The Magician's Doubts: Nabokov and the Risks of Fiction . Princeton University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Woods , John . 2018 . Truth in Fiction. Rethinking Its Logic . Cham, Switzerland : Synthese International Publishers . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Woodward , Richard , 2014 . “ Walton on Fictionality .” Philosophy Compass 9 : 825 – 836 . Google Scholar Crossref Search ADS WorldCat Woodward , Richard . 2016 . “ Fictionality and Photography .” The Journal of Aesthetics and Art Criticism 74 : 279 – 289 . Google Scholar Crossref Search ADS WorldCat Yablo , Stephen . 2014 . Aboutness . Princeton University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Zucchi , Sandro . 2017 . “ Games of Make‐Believe and Factual Information .” Theoretical Linguistics 43 : 95 – 101 . Google Scholar Crossref Search ADS WorldCat It is sui generis in that it is a specific speech act, with its own individuating definition, along with others such as promises, guesses, conjectures, and so on. But it belongs in one of the highest genera for such acts in a proper taxonomy, that of directives (cf. Alward 2010a, 356, 2010b, 390). ‘Speech’ is to be broadly understood; “speech” acts can be made with pictures: the account of fiction summarized below is meant to apply to films. Ohmann (1971), Grant (2001), Sutrop (2002), and Stock (2017) also support the speech‐act view; Gale (1971), Searle (1975), Alward (2009, 2010a) and Friend (2012) object to it. In García‐Carpintero (2019a) I discuss Green's less clear‐cut view. He does not posit a specific speech act of fiction‐making, but he explains fictions in terms of the speech act of supposing. I oscillate between deontic and axiological vocabulary throughout. For our purposes, it does not matter which one takes precedence, if any. That the audience fulfills an order does not make it correct (it might have been utterly wrong in the first place), and that it remains forever unfulfilled does not make it incorrect. Alward (2010b) provides a nice metaphor for what on my account are fiction‐making speech acts from which fictions result: the deployment of “word‐sculptures.” (Better, “representation‐sculptures,” not to exclude films, pictures, and, indeed, sculptures.) He provides an institutionalist account of fictionality that, like Walton's, is consistent with my normative speech‐act account—the normative speech‐act kinds themselves are on my view the relevant institutions. Like Walton, however, Alward opposes his account to speech‐act views. This is because he assumes that “what is characteristic of illocutionary action is the intention to produce an effect in an audience by means of the latter's recognition of this intention” (395). The objections he makes on this assumption echo Walton's (1990, 85–89). They only apply to Gricean views like Currie's or Stock's; normative accounts like mine reject that assumption. The only intended effect in the audience is in my view their recognition that by her act the agent becomes beholden to a norm; this typically depends on the institutional features that Alward, Friend, and Walton rightly emphasize. See García‐Carpintero (2013, 2016) and Stock (2017, 163–167). The constrained role that the intentions of fiction‐makers play in a normative account (García‐Carpintero, 2019a) deals well with one of Currie's (2014, 359) reasons for pessimism about prospects for speech‐act accounts of the fictionality of works. The account supports a more holistic approach than Currie's on what fictional utterances are, which is also helpful. See Friend (2017, 31; forthcoming, ch. 4). I take Lorand's (2001) and Sainsbury's (2014) skepticism to be addressed to views like Lewis's ([1978] 1983) that attempt to reductively analyze the world metaphor in possible‐worlds ideology. I also talk of the plot of the story. I am unconcerned with the circularity which Lewis (265) sees in it, grounded on his aiming for a possible‐worlds‐based reduction of fictional content (Sainsbury 2014, 283). Below (n21) I address another, Kripke‐based argument that Lewis invokes to reject a plot‐of‐the‐story account of fictional content. The proposal is made precise by resorting to traditional possible worlds semantics; Berto (2017b) develops it in a way that deals with what I, like he and Stock (2017, 141–144), take to be perfectly possible inconsistent F‐imaginings. In any case, illuminating a view by giving a precise ideal model for it does not amount to providing a reduction. Indeed, as Lorand suggests, “the main function of the ‘world’ metaphor in theories of fiction is to supply a framework for justifying the tendency to go beyond the given text and include in the story more than that which is explicitly stated” (2001, 427). This is the core of what Matravers (1997, 79) calls “the report model”: “in reading a novel, a reader makes‐believe he is being given a report of actual events. In other words, he makes‐believe the content of the novel is being reported to him as known fact by a narrator.” My reasons are essentially Wilson's (2011, 116‐7), addressing Kania's (2005) skepticism. Kania (52) says that a fictional world “is just as stipulated … by the artist through the work,” but he does not tell us how this stipulating goes. In fact, the best account for how verbal fiction‐makers convey the explicit features of the fictional world (to my mind, Walton's 1990 book) has them simulating the assertions that would be made in the relevant context by means of the declaratives they put forward—these are crucial “props” in verbal games of make‐believe. These assertions have assertors, explicit as in Don Quixote or implicit instead. Zucchi provides a related reason, based on the behavior of indexicals in discourse constituting fictions, in contrast with discourse reporting on it (2017, 96–99). Wilson doubts that the view generalizes to all verbal cases, mentioning works consisting mostly of dialogue like Compton Burnett's The Present and the Past (2011, 112). But its very first lines show such dialogues to be explicitly reported by an implicit teller: “‘Oh, dear, oh, dear!’ said Henry Clare. His sister glanced in his direction ….” Novels consisting only of interrogatives or imperatives would provide content through presuppositions, also acts in the “assertive family” (Green 2017a). Regarding films, Wilson (2011, 126–139) himself or Cumming, Greenberg, and Kelly (2017) offer adequate justification. The distinction below between a‐imaginings and c‐imaginings ward off the difficulties that fictions positing prima facie mindless worlds create for Lewis's “idealism” (Byrne 1993); see the discussion of “Urkul” in Section III. See Ohmann (1971, 18) and Gale (1971, 337). I thus agree with Searle (1975) that verbal fiction‐making is illocutionary pretense. Alward, however, argues on this basis against the speech‐act view, on the questionable assumption that “theories of fictional discourse can be used without revision as accounts of theatrical discourse” (2009, 321): given that the sui generis speech‐act view is not a good account of the latter, it follows that it fails for the former too. But the assumption question‐beggingly overlooks a distinction related to the one I make, between ancillary and constitutive imaginings. In performing illocutionary acts, actors (including storytellers of verbal fictions) play two roles, an actor role, and a storytelling role. I agree with Searle and Alward that the former is just pretense, but this does not apply to the latter. In the first role, an actor playing Don José has him saying that Carmen loves him, while in the second the storyteller conveys though his acting that Don José wrongly believes so in his infatuation. In “playing” an unreliable narrator like Kinbote, the storyteller (Nabokov) in the actor role portrays a fictional narrator (Kinbote) asserting p while, in his second role, he invites his audience to imagine something else (see Section III). See García‐Carpintero (2019b). I count manifest implicatures that the fictional teller is portrayed as making as contributing to immediate content, but this raises issues I cannot address here. Unreliable narrators do not require any special qualification (see n24). Predelli and Reimer formulate their views in a possible‐worlds semantics; this can be understood once more not as a reductive proposal, but done merely for the sake of offering a precise model. Voltolini takes the notion as primitive in his earlier work, but he provides a cognitive elaboration in a more recent presentation (Voltolini 2016). Woods (2018) also takes a primitivist view; see García‐Carpintero (2018). Reimer (2005) disclaims the ontological consequences of the view by ascribing to fictional utterances truth conditions but not propositional contents. My deflationary take on propositions leaves me no room for that distinction. Everett (2013, 48) suggests that the difference between these “semantic” accounts and “pragmatic” ones is not big. I have been scare‐quoting “semantic” and “pragmatic” because the indication that a declarative is used for fiction‐making and not for asserting is by my lights as “semantic” as the indication that it is used for guessing or supposing. “In inviting the reader to constitute speech acts to go with its sentences, the literary work is asking him to participate in the imaginative construction of a world—or at least as much of one as is necessary to give the speech acts an adequate setting” (Ohmann 1971, 17). See also Camp (2017, 77–78). In a possible‐worlds analysis of fictional worlds, this corresponds to Lewis's ([1978] 1983, 277) “method of intersection” for dealing with inconsistent fictions. This is my official view, but, as it transpires below, I am not sanguine about it. I am open to the “plurality of interpretations” line below and, hence, the permissivist view and the “method of union” that goes with it. This corresponds to Lewis's ([1978] 1983, 277) “method of union.” This account of names answers Lewis's ([1978] 1983, 265) Kripkean objection to explaining fictional truth relative to the plot of the story. The Holmes stories prescribe imagining that he is called with the specific homonym for which Conan Doyle invites us to imagine a reference‐fixing naming practice (in this case, one also part of the fictional world). The fact that in the actual world someone called by another homonym (“improbably, incredibly”) carried out all Holmes's exploits does not thus make it one of those “where the plot of the stories is enacted.” See Predelli (2017, 135 ff.) on “peripheral” dubbings for a similar take on this, and Wilson's (2011, 125, 115n10) distinction for these and related cases of “heterodiegetic” narrators between what is fictional “in the work” and what is fictional “in the story,” which I take to be parallel to mine between ancillary and constitutive F‐imaginings. The distinction disposes of Byrne's (1993) objection of “idealism” to accounts, like mine, granting Alward's (2009) view that all fictions have fictional narrators: the “Urkul” world itself does not have naming practices or rational agents (dinosaurs aside). They are thus not fictional‐world‐constituting fictional truths, and hence they are not explicit fictional truths either (n14); they are only prima facie explicit fictional truths. The terminology is DeRose's (2002, 180). Lackey (2007) questions the distinction. I take it that what is controversial is only the use of it to respond to alleged counterexamples to the knowledge rule for assertion (say, cases of untrue assertions for which the speaker has very strong justification). This is not my own view. In parallel with the case of directives, I take it that one must promise that p only if one gets thereby a reason to see to it that p: that one fulfills a promise does not make it correct, if it was inadequate in the first place, and it may have been correct even if it remains unfulfilled. But for present purposes we can stick to the traditional view. Woodward (2016) provides compelling responses to some other specifically visual cases that Walton invokes, which I take to be fully compatible with the line I am pushing. I critically discuss it elsewhere (García‐Carpintero forthcoming). Earlier (n12) I supported Wilson's (2011, 111) “minimal” take on fictional narrators in his polemic with Kania. But when it comes to Kania's (2007) polemic with Alward, I am on his side. I fully agree with him (406) that only actual fiction‐makers truly “reveal” the fictional world to us; “playing” fictional narrators is just one of the resources they use for it. Financial support was provided by the DGI, Spanish Government, research project FFI2016‐80588‐R; by the award ICREA Academia for excellence in research, 2013, funded by the Generalitat de Catalunya; and from the European Union's Horizon 2020 Research and Innovation programme under Grant Agreement no. 675415, Diaphora. This work received helpful comments from Filippo Contesi, Alberto Voltolini, Lee Walters, Richard Woodward, and the editors and referees of this journal. Thanks to Michael Maudsley for his grammatical revision. © The American Society for Aesthetics This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) © The American Society for Aesthetics
TI - Normative Fiction‐Making and the World of the FictionGarcía‐Carpintero
JF - The Journal of Aesthetics and Art Criticism
DO - 10.1111/jaac.12660
DA - 2019-08-04
UR - https://www.deepdyve.com/lp/oxford-university-press/normative-fiction-making-and-the-world-of-the-fictiongarc-a-carpintero-6FdLIOEEP0
SP - 267
EP - 279
VL - 77
IS - 3
DP - DeepDyve
ER -