Truth Matters, Aesthetically

This paper defends a version of aesthetic cognitivism: the truth of statements expressed, implied, or alluded to by a work of fiction matters aesthetically, and bears upon the work’s aesthetic value. Our aim is to explore a route from truth to aesthetic value that claims, roughly, that, if our engagement with a work of fiction is based on truth, it is more vivid than otherwise, and thereby contributes to the aesthetic value of the work. Whether truth increases the vividness of our engagement with fiction is an empirical question. On the assumption that it does, we spell out some consequences for the aesthetic value, and in particular for the literary value of a work, as well as for critical practice.


I. Introduction
What is the relation between cognitive and aesthetic values? Some hold that cognitive values can bear upon aesthetic ones 1 and others that aesthetic values can bear upon cognitive ones. 2 A third position claims that neither bears upon the other, at least not crucially. 3 In this paper, we set out to defend a version of the first view: that the truth of statements expressed, implied, or alluded to by a work of fiction matters aesthetically, and bears upon the work's aesthetic value. In doing so, the paper defends a particular version of cognitivism with respect to literary works, which is in line with what has been labelled ' aesthetic cognitivism'. In its traditional form, cognitivism is the view that we can gain some sort of knowledge or insight from literary fiction. So-called 'neo-cognitivists' argue that literary fiction can enhance or

II. Truth and Vividness
In his Mimesis as Make-Believe, Kendall Walton comments on the relation between the truth of what is imagined in our engagement with fiction, imaginative vividness, and ' artistic' purposes. He notes that, first, knowing that what one imagines is true may enhance the vividness (or 'vivacity', as Walton has it) of one's imaginings and, second, that truth may thus serve ' artistic' purposes: The vivacity of the reader's imaginings may be enhanced by the knowledge that what he imagines is true, by his realization of the reality of the setting of a story and its characters and events. Even if The Executioner's Song [by Norman Mailer] -or Tolstoy's War and Peace, for that matter -is read mainly as a tragic fable or an adventure story and not with an eye to improving (in any direct way) one's knowledge and understanding of the actual historic events, it is likely to be more exciting, more gripping, if one takes it to be largely accurate than it would otherwise be. […] To the extent that ' artistic' purposes are distinct from cognitive ones, they may be served by getting the facts right. The enormous efforts novelists sometimes expend in researching the settings of their fictions and the historical figures they use as characters are not to be explained solely, sometimes probably not at all, by an interest in informing the reader about them. The purpose may be -to use an exceptionally obscure word -to enhance 'realism '. 7 Let us start by noting two things, which can then be set aside. First, Walton notes that 'getting the facts right' may be among the author's artistic purposes, but he does not go into much detail as to how knowledge of these facts gets across to the reader. The sentences of the work may be straightforwardly true, but truths may also be otherwise expressed, implied, or alluded to by the work, or they may be a presupposition for understanding what the work is about, and so on. These differences do not matter for our purposes. Further, as Walton notes in a cross-reference, the readers' knowledge that what they imagine is true may also stem from the fact that they imagine something ' of some actual object' that is in their presence. 8 Since we are concerned with literary fiction, the case of immediate perception is not relevant for us. However, as we shall see, the vividness of the reader's engagement with the work may very well benefit from having perceived some object in the past, and hence from taking to be true that this object has certain perceptual properties. Second, Walton endorses a particular idea of the connection between fiction and imagination as 'make-believe'. 9 But we should note from the outset that, for the purpose of exploring a route from truth through vividness to aesthetic value, nothing much hangs on the precise contours of either the notion of imagining or the relation between our imagination and works of fiction. While we take it that an account that assigns the imagination a central place in our engagement with fiction is basically right, 10 we think that our view is independent of any theory about the attitude 7 Ibid., 93-94. 8 In this cross-reference, Walton links his observation to his discussion of 'props', where he reports the 'impression which I have, and which I suspect others share, that an experience of imagining a bear is likely to be more "vivid" if one imagines of some actual object that it is a bear than if one does not' (ibid., 26). 9 See ibid., chap. we typically take (or are prescribed to take). What we are interested in is a qualitative feature of our engagement with fiction that is assessable independently. Any theory concerning the attitude that characterizes our engagement with fiction has to account for the fact that our engagement with fiction can be more or less vivid. This should not be surprising, since it is commonly held that, besides imagination, memories, beliefs, or dreams can be vivid, and sometimes even real-life experiences. Because the authors we discuss hold that imagination is the central mental attitude in our engagement with fiction, and for reasons of simplicity, we refer to the relevant attitude as 'imagining'. We can now turn to a more precise characterization of our central thesis concerning the relation between truth, imagining, and vividness in order to pave the way for an explanation of the bearing of this relation on the aesthetic value of a literary work of fiction. Concerning the connection between truth, imagining, and vividness, what we are dealing with is a psychological dependency thesis of the form 'X tends to be larger if it is based on Y than if it is not based on Y'. 'X' is a placeholder for the vividness of our imagining some proposition p as part of our engagement with works of fiction. 'Y' is a placeholder for 'taking p to be true' or, for short, 'believing p'. As can be seen from its constituents, the psychological dependency thesis is an empirical thesis. We can therefore begin its clarification by noting that it should be expected to exhibit four regular features of such theses.
First, there are certain background conditions that need to obtain for the dependency between X and Y to hold. Research indicates that people show individual differences in their responses to fiction more generally, as well as in their capacity to imagine something vividly. 11 Also, whether or not someone imagines something vividly is expected to depend on situational factors, as well as on the very proposition in question. Thus, imaginings may be blocked from being vivid by some situational distractions, and some truths that are regarded as dull or abstract will simply not be candidates for vivid imagining.
Second, the positive dependency between 'taking p to be true' and 'imagining p vividly' should be thought of as probabilistic. It is an empirical prediction to the effect that, if we look at sufficiently many instances, we will see the predicted dependency.
Third, the comparative claim ('larger with Y than without Y') of the dependency thesis should be thought of as governed by a ceteris paribus proviso: the predicted effect will occur, other things being equal, that is, only if no other factors are interfering with it. Also, the thesis does not imply that other factors may not override the predicted effect.
Fourth, as with any empirical claim, the psychological dependency thesis is in need of empirical confirmation. While empirical testing is beyond the scope of this paper, in what follows we provide some reasons to think that the psychological dependency thesis is in fact true. Doing so comes as part and parcel with explaining the notion of vividness involved. This is the task to which we turn now.
Walton gives but one hint as to how an increase in vividness may be understood in the present context, namely as imaginings that are 'more exciting, more gripping'. An experience that is both exciting and gripping can be understood as one that not only catches but also sustains our attention and keeps us focused on what is imagined. Presumably, it also does so in an affectively charged way, as is the case when we develop an emotional interest in the theme of the literary work, or in some of the subject-level particulars such as a certain character. For instance, we may hope that things turn out well for an adored character, or fear that they might not. And, indeed, there are good reasons to think that the relationship between truth and vividness holds in this regard. Our imaginings may be more vivid in the sense of being more exciting and more gripping when they are based on what we take to be true because we might believe that someone, possibly we ourselves, could be affected by what we imagine. When imagining something about, say, Gandalf (a character in J. R. R. Tolkien's fantasy genre fiction), a reader would usually not think that they could be affected in any way that goes beyond their engagement with the novel. But imagining, say, that a notorious mass murderer is on the run will keep the reader focused and be more emotional if they also believe that the murderer exists and may even be in the vicinity. The reader's excitement may consist in fear for themselves or others, in hopes to evade the murderer, and so on. So, the reader's imagining may be more vivid in the sense of being both attentively focused and affectively charged when it is based on what they believe is true. 12 In addition to this, there are at least four further ways in which the vividness of our engagement with fiction may be affected by the truth of what we imagine. First, our imaginings may be more vivid in the sense of being motivationally charged. An experience that is vivid in this sense is one in which you have a tendency to act upon what is on your mind.
The truth of what we imagine may play an important part here, too. This is so because our belief set is not decoupled from our motivational system. If we believe some proposition p, we will be disposed to act upon it (if appropriate, given the content of p and relevant further circumstances). So, if we entertain p as part of our engagement with fiction and p is based on truth, we might also imagine being motivated to act upon p, or we might actually be motivated to act upon p. Again, the above-mentioned example of the mass murderer may come to mind. If you think that what you read is true, then this may very well dispose you to act upon it in some appropriate ways. Without truth, we would expect your imaginings to be decoupled from reality, and hence there will be no immediate motivational pressure to interfere with things.
Second, the vividness of our imagining that p may be a function of the easiness or spontaneity with which we bring p to mind. It is hard to imagine p vividly if we have no idea what p is about, if we think it is unlikely or even impossible, or if we for other reasons have to invest a lot of effort into imagining it. In contrast, imagining that p is comparatively easy when it is based on what we take to be true anyway. Suppose a reader has never heard of Dave Scott and finds it hard to imagine that someone could win the Ironman World Championship six times. But suppose the reader believes that Scott did win the Ironman World Championship six times -then imagining it will be easy. 13 Note that the ease with which we imagine something may not constitute an aspect of vividness but rather only promote it. This would, of course, also provide an answer to the question why, in the context of reading literary fiction, imaginings based on what we take to be true tend to be more vivid than mere imaginings.
Third, beliefs are always embedded in other beliefs. If we believe that p, we will usually also believe many things implied by p; we will, following some minimal rationality requirement, believe some things that constitute evidence for p and that p is evidence for. We may, moreover, have certain higher-order attitudes (such as: I want to have this belief as long as it is true) and believe certain counterfactual conditions (I would give up this belief if I were convinced of its falsity). In short, a belief that p will be more or less firmly entrenched in our web of belief and, in this sense, be understood by us. 14 This may have two consequences for our corresponding imaginings. On the one hand, the vividness of an imaginative episode may be a function of the amount of detail of the imagining. 15 So, if, because of corresponding beliefs, there are many implications at hand, this may facilitate imagining something vividly in the sense of imagining something with many (imagined) implications. On the other hand, just as a belief that p would be more or less firmly entrenched in our web of belief and thereby understood, our imagining that p may be more vivid in the sense of both being entrenched in a 'web of imagining' and being understood. Consider Dave Scott again: if you believe that Scott won the Ironman World Championship six times, you will most likely believe many other things as well, including that there is a world-famous triathlon race located on Hawaii, that it consists of a swim, a bicycle ride, and a marathon, that it is thought of as one of the most difficult races in the world, that winning it is very hard, and so on. In this sense, your imagining that Scott won the Ironman World Championship six times will be both more detailed and (more fully) understood if it is based on further beliefs. 16 Fourth, if our imagining is based on truth, there will sometimes be some real object we may resort to in order to furnish our imaginings with phenomenal (visual, auditory, tactile, or gustatory) properties. For instance, imagining that some events take place on the Twin Towers in New York City may invoke mental images of the Twin Towers. Thus, imagining something of some real object may make our imagining categorically different from mere propositional imagining such that we imagine visual properties of something, or imagine seeing something. The imagining may therefore be more vivid in the sense that it has phenomenal properties that are retrieved from memory. 17 This is the sense in which it matters whether the reader has perceived the relevant actual object in the past (see the beginning of this section).
As the foregoing considerations should have shown, 'vividness', when applied to our engagement with works of fiction, is an umbrella term that covers a range of different phenomena, namely (comparatively high levels of) the attentiveness and focus we bring to our imaginative enterprise, its being affectively and motivationally charged, the ease and spontaneity of our imaginings, their amount of detail, the amount of understanding involved, and phenomenal properties of what we imagine. As we wish to understand it, there is a disjunctive set of individually sufficient and non-exclusive features that constitute the vividness of our engagement with works of fiction. What binds these different features together is that they are all sensitive to truth, as explained by the psychological dependency thesis. In a nutshell, given the appropriate background conditions are met, our imaginative engagement will (ceteris paribus) be more vivid in any, some, or all of the ways we have just explained, if we take what we imagine to be true.
To get clearer on our thesis, let us discuss some potential counterevidence. Suppose that we observe that some imaginings are especially vivid even if -or maybe in some sense because -they are very remote from the truth. Horror fiction may be especially exciting and gripping (indeed, all too gripping for some), and still it is abundantly clear to anybody that the dreadful monsters do not exist. What this shows is that truth is not necessary for imaginative vividness, or so the objection goes. 18 Now, while we certainly agree that truth is not necessary for imaginative vividness, pointing this out in no way threatens the psychological dependency thesis. Suppose that you watch a horror movie, and you find it exciting and gripping. The psychological dependency thesis predicts that you would find it even more exciting and gripping if you (for whatever reason) think that what you imagine is true. The psychological dependency thesis thus amounts to some sort of a (counterfactual) comparison between two imaginings of the very same content such that, in the one case, you will imagine some content (simpliciter), while in the other case you will imagine this content and also take it to be true. In this comparative scenario, the predicted effect is that your imaginings will ceteris paribus be more vivid in the second case. Note also that the psychological dependency thesis does not claim that truth will create vividness in the sense that it is causally sufficient. Rather, truth will (tend to) enhance vividness, given that all relevant circumstances are such that vividness is a possibility in the first place. 19 Is there any empirical evidence available that shows that the psychological dependency thesis might be true? We are aware of no studies that are directly pertinent to this thesis, but there is some interesting research in its vicinity that could be relevant. Nico Frijda hypothesizes that emotions ' emerge, wax, and wane according to rules in strictly determined fashion', 20 and he sets out to formulate 'laws' that capture these empirical regularities. Among them is the 'law of apparent reality', which reads as follows: 'Emotions are elicited by events appraised as real, and their intensity corresponds to the degree to which this is the case.' 21 The first thing to notice is that, since this 'law' is coined on the eliciting conditions of emotions specifically, it addresses only part of what we have identified as constituting the vividness of our engagement with works of fiction. However, it does fit the psychological dependency nicely since appraising events ' as real' seems to involve the belief that an emotion stimulus is real. 22 In a more recent study by Green, Chatham, and Sestir, Frijda's 'law' is dubbed as a '"common sense" view' that suggests 'that nonfiction (fact) would create heightened emotions as compared to fictional materials. […] All else equal, to the extent that an individual considers an actual person's death more real than a fictional death, it should evoke more grief.' 23 The study is of particular interest to us because the authors interpret their own results as showing that this ' common sense view' might be wrong. In their study, two experiments were conducted 18 Thanks to Peter Lamarque for providing this objection. 19 We should also not assume that a lack of truth generally leads to lack of vividness: Reading fiction may lead to vivid imaginings for reasons other than taking what we imagine to be true. However, suppose that you vividly imagine something you take to be true, only to learn then that it is not true after all. In such a case, your imagining will then arguably become less vivid in any, some, or maybe all of the sense we have explained. that measured the emotional reactions of readers depending on whether they read a text labelled as either 'fiction' or 'non-fiction'. What the authors found indicates that how a story is labelled does not have an effect on the individuals' emotional response. 24 At first sight, this seems to contradict the psychological dependency thesis with respect to the emotional aspects. However, here is why we think it does not. First, we must be careful not to equate the presentation of a story as 'fact' with the reader's believing what they have been told. Some readers may straightforwardly take as true what they are told due to such labelling alone, but others may be more careful. After all, all they get to convince them of the truth is a label in some experimental design. What is more, it may very well be that individuals 'forget' about the labelling of a story as fact or fiction when immersed in a story: they might process the information in various ways, including (perhaps automatically) checking its putative truth by drawing on their individual knowledge resources.
Second, inferences concerning the psychological dependency thesis will also be somewhat blurred by the fact that individuals who read the texts labelled as 'fiction' will presumably believe many things connected to the contents of the fiction. 25 In any case, we should not assume that labelling a text as 'fiction' will automatically result in readers' disbelief in what is said, let alone in what is implied or alluded to by the fiction. It is therefore not clear that any differences in emotional reaction can be traced back to the presence or absence of individuals' beliefs concerning the contents of the stories or concerning what is implied or alluded to.
Third, we have some qualms concerning some background assumptions of the study: Although there are some differences between emotions in response to fiction and emotions in response to real events (for example, being afraid of a monster in a movie does not generally prompt one to run for the exits or pull out a weapon; sadness over a character's death is rarely as long-lasting as grief over the loss of a real loved one), our premise is that the immediate experience of the emotion has the same essential qualities regardless of whether that emotion is sparked by narrative or by actual events. Therefore, our research question is not whether fiction can produce emotion, but whether the intensity of the emotion produced by a narrative varies depending on whether the narrative is described as factual or fictional. 26 Note, first, that the authors make two claims here that do not go well together. On the one hand, they acknowledge that emotions in response to fiction generally differ in their motivational power and duration from those produced by real stimuli; on the other hand, they hold that 'the immediate experience' of these emotions does not differ from their reality-directed counterparts. However, we believe that the motivational tendency of an emotion (whether one feels an urge to act) does count among the ' essential qualities' of 'the immediate experience' of an emotion. More importantly, in declaring that a motivational tendency is not among the ' essential qualities' of the 'immediate experience' of emotion, the authors simply gloss over what we take to be an important aspect of vividness (as explained above), namely the extent to which what is imagined is motivationally charged. If individuals in either the fiction or the non-fiction condition felt some tendency to act upon what they read, then this would count towards the vividness of the respective experience. In sum, we take it that empirically testing the psychological dependency thesis requires not only breaking it up into several theses concerning the different aspects of vividness identified above but also resolutely checking that the belief condition is satisfied as part of the testing procedure. This does not mean, however, that the experiments referred to above might not have an important lesson to teach for our case. In order to get to those lessons, let us assume (contrary to our worries articulated above) that individuals did believe the proposition expressed in the non-fiction condition of the experiment and did not take as true what was presented to them in the fiction condition. What could explain that there was no difference concerning the 'intensity' of the emotion experienced in either case?
The key to answering this question could lie in the fact that in neither of the experimental conditions were the individuals in any way affected by what they were told. The stories in both the fiction and non-fiction conditions simply did not concern them, because (let us assume) they were taken either to be purely fictional or to deal with what happened to some unknown people some time ago. The 'truths' they conveyed were thus practically irrelevant. Above, we mentioned that among the background conditions that need to hold for the psychological dependency thesis are conditions pertaining to the content of what is believed. Specifically, we noted that the truths in question must not be regarded as dull or abstract. Now, in the present case, this may be relevant in the following sense: the truths in question have to somehow concern the individual. The proposition that 'man is mortal' will be abstract and dull for you for as long as you do not realize that it applies to you. This hypothesis is consistent with how Frijda explains the 'law of apparent reality'. According to him, emotional stimuli are especially effective ('real') if they are based on sense experience, because the perceiver is personally affected by them. Similarly, in our explanation of the notion of vividness, we referred to the example of believing that a mass murderer might be in the vicinity -which is supposed to account for a heightened affective and motivational tone of the experience. In such cases, individuals may develop beliefs that are egocentric: they in some sense involve the individuals themselves. We think that it is quite plausible to expect such beliefs in the context of reading literary works of fiction. As Martha Nussbaum emphasizes, literature 'speaks about us, about our lives and choices and emotions, about our social existence and the totality of our connections'. 27 We thus conclude our discussion of the connection between truth and vividness by assuming that the following holds: if we take what we imagine to be true, then this will (given that appropriate background conditions are met) ceteris paribus heighten the probability that our imaginings are more vivid in any, some, or all of the senses explained above.

III. Vividness, Aesthetic Value, and Literary Value
In this section, we discuss the thesis that, when engaged with a literary work of fiction, vivid imaginings bear on the aesthetic value of that work. We do so, first, by considering a more general notion of aesthetic value and, second, by considering the rather specific notion of literary value as developed by Lamarque and Olsen. While we touch upon the more general notion of aesthetic value only in passing, our main aim is to argue that vividness and hence truth matters for the specific notion of literary value.
In his The Opacity of Narrative, Lamarque uses 'vivid' to refer to a quality of the writing of a novel: 'novels are praised or denigrated for the quality of the writing, its vividness, power, resonance, expressiveness, clarity, wit and so on, through the familiar multiplicity of positive attributes […].' 28 In the previous section, we have explained 'vividness' as a qualitative feature of a reader's engagement with a literary work. We think that this explanation can easily, and indeed should, be used in an explanation of the notion of 'vivid writing'. Quite intuitively, the 'writing' of a novel is vivid if something is rendered vividly, and something is rendered vividly if it evokes vivid imaginings in the reader. 29 As Lamarque indicates, vividness has often been thought of as a positive aesthetic value. 30 Let us call something a value-constituting property of a piece of discourse if it constitutes a value of discourse of that type. For instance, the expression of truths is a value-constituting property of a newspaper article. Newspaper articles are valued, inter alia, for exactly this property. Call something a value-contributing property of a piece of discourse if its presence establishes, or tends to establish, a value of a piece of discourse. For instance, a clearly arranged wording will be value-contributing in a newspaper article, since it will help readers to grasp the truths expressed. This distinction brings us to the first thesis we wish to argue in this section: if vividness is a value-constituting property of a literary representation, the truth of what is expressed, implied, or alluded to is a value-contributing property of that representation. If the work (somehow) gets the reader to imagine what they take to be true, then this will (tend to) enhance the vividness of what they imagine (along the lines of the dimensions outlined above); the work's representation will thereby qualify as 'vivid', which constitutes an aesthetic value. Now, vivid descriptions may have aesthetic value in themselves, but this might not be literary value, that is, not value unique to literature, as explained by Lamarque and Olsen. For Lamarque and Olsen, what is 'integral to works of imaginative literature', and hence central in critical practice, includes the 'mimetic aspect' of literature. This means that works develop a theme by arranging things on the subject level. According to Lamarque, '[a]ppreciation and consequent evaluation of the individual literary work is a matter of eliciting and supporting the identification and development of a "perennial theme"'. 31 And: 'Literary appreciation is the appreciation of how a work interprets and develops the general themes which the reader identifies through the application of thematic concepts.' 32 A reader who fails to pursue the thematic significance of the subject level fails to appreciate the work as a literary work of art. But if ' a work interprets and develops the general themes' in an original, powerful way, thus presenting an 'imaginatively profound interpretation' or 'vision' of the theme, 33 this constitutes the work's value as literature or, for short, its literary value.
While Lamarque and Olsen's Truth, Fiction, and Literature only hints at the notion of a vision primarily by discussing examples, Lamarque's Philosophy of Literature provides some more detail on how to understand it. A 'vision' is, for instance, taken to involve ' attention to  Sibley's list of aesthetic concepts: unified, balanced, integrated, lifeless, serene, sombre, dynamic, powerful, vivid, delicate, moving. Frank Sibley, 'Aesthetic Concepts', Philosophical Review 68 (1959) Lamarque and Olsen,Truth,Fiction,and Literature, fine-grained particularities'. 34 These particularities, we take it, consist of subject-level existents (characters, incidents, places, and so on). Since our access to subject-level existents is through the imagination, the 'vision' that is said to be expressed by a work seems to involve, in the first instance, a particular way of imaginatively relating to the subject level. The work, as it were, brings us to imaginatively 'see' characters, situations, and events on the subject level in a particular manner. As an example, Lamarque and Olsen provide a detailed interpretation of George Eliot's Middlemarch. A central part of this interpretation reads as follows: The passage [from Middlemarch] renders the inner workings of Lydgate's mind in such a way as to make the reader see how Lydgate is trapped by the combination of his own weaknesses of personality and the limitations of Middlemarch society as represented by Rosamond Vincy. This imaginative realization of the theme is complex and nuanced and other parts of the Lydgate story add to this complexity and these nuances. 35 The manner in which the storyworld existents are to be seen certainly involves the notion of a theme, such that there is ' a unifying thread that bids together incident and character in an illuminating way', or a thematic 'prism through which to view the various couplings and uncouplings that make up the subject'. 36 But this does not mean that the value of literature lies in some thematic statements that can be abstracted from the work. Rather, a 'vision' is something that can only be experienced while reading a work, and especially by attending to its subject level. Indeed, Lamarque seems to suggest that a 'vision' might not be something that is ' distinct from experience' and among the ' consequences of experience'. Rather, ' employing a notion of "vision" […] might bring awareness and insight closer toward intrinsic experiential qualities'. 37 A 'vision', therefore, centrally involves an activity of the reader; this activity has to do with imaginatively relating to the work's subject level. Moreover, this imaginative activity can be more or less powerful or intense, or more or less 'satisfying and engaging'. We take it that the notions of 'powerfulness' and 'intensity' can be cashed out in terms of the work's capacity to mandate a vivid imaginative experience as explained above. It is then plausible to claim that a powerful 'vision' centrally involves vividly imagining whatever takes place on the subject level of the work. Now, vivid imagining of the characters, situations, and events on the subject level of a work is surely not sufficient for a 'vision', because the subject level has to be seen in light of the theme. The theme must, as it were, direct our imaginative involvement. However, vivid imagining seems to be necessary for a 'vision' to be powerful or intense, or satisfying and engaging. For what would the 'powerfulness' and 'intensity' of an imaginative reading experience mean, if not what we have identified above as dimensions of vividness? Consequently, there is good reason to assume that truth plays a value-contributing part here too: as we have argued in the previous section, truth tends to promote the vividness of a reader's engagement with the work, and it thereby also tends to enhance the 'powerfulness' of a 'vision' as expressed by the work. 34 Peter Lamarque, The Philosophy of Literature (Malden: Blackwell, 2009), 240. 35 Lamarque and Olsen,Truth,Fiction,and Literature,338. 36 Lamarque,Philosophy of Literature, Ibid., 267. Lamarque and Olsen also note that 'literary appreciation lies as much in the route taken as in the final destination' (Lamarque and Olsen,Truth,Fiction,and Literature,403), which also suggests that what constitutes appreciation is a particular reading experience. Note also that Lamarque (Philosophy of Literature, 168) speaks of ' a satisfying and engaging perspective or vision'. We take it that this is not supposed to mean that an abstracted 'vision' (say, a moral doctrine) is judged to be satisfying in some respect, but rather that (imaginatively) relating to the work such that its vision is appreciated can be satisfying and engaging.
Lamarque and Olsen at times speak as if they think that what makes for the quality of a 'vision' is that it makes ' demands on either the intellectual, emotional, or moral nature', 38 or that it involves taking into account several aspects or 'layers' of the work, including stylistic devices. Hence, a 'vision' presented by a work could be said to be powerful simply because it is morally or intellectually demanding, or because it is highly integrative regarding many of the work's different features. But note that we have located a way of cashing out the notion of a 'powerful vision' that fits imaginative literature especially well. We may get a powerful vision of the 'morally demanding' kind by, say, arguments and explanations as presented in philosophy. But this does not seem to have much to do with the specifics of imaginative literature, that is, literary fiction. (Note that Lamarque and Olsen are at pains to reserve a special place for literature when it comes to revealing a 'humanly interesting content' for us.) And thinking that the 'vision' presented by a work could be said to be powerful simply because it has a highly integrative character concerning the work's diverse properties fails to do justice to the fact that a 'vision', as explained, is first and foremost a vision of what the work represents on the subject level. So, the predicates 'powerful', 'intense', and so on should be spelled out as ways of imaginatively 'seeing' that content, or as characteristics of the imaginative experience through which the work's subject level is presented to us.
In this section, we have argued that truth is a value-contributing property in two ways. First, it tends to enhance the vividness of a (literary) representation, which in turn is a valueconstituting property in literature; and, second, it tends to enhance the 'powerfulness' of the 'vision' as expressed by a literary work. Hence, if truth enhances vividness, it also bears on a work's literary value in Lamarque and Olsen's very own terms.

IV. Truth and Literary Criticism
Lamarque and Olsen make much of the observation that the truth of the statements implied by a work of literary fiction does not seem to matter in criticism. In particular, they hold that critics do not argue for the truths implied in literature. 39 Now, if what we have argued is correct, it might take us some steps towards explaining both why truth should matter in criticism and why there is a notable absence of arguing or debate concerning these truths in criticism. If critics want their readers to truly realize the powerful 'vision' implied by the work, they may very well want to point to these truths in order to make the reader's involvement with the work more vivid. However, this does not mean that the critic is well advised to argue for the truth of the statements expressed, implied, or alluded to by the work. There are several reasons for this. First, arguing is something that we do if we want to establish some truth that our interlocutor is prone to deny. But, very often, this may simply not be the position of the critic. As we have pointed out above, the critic may often feel that their readers could simply be reminded of what they already know. Think about how odd it is to argue for something if you believe that your interlocuter already believes the propositions in question.
Second, truth is not a value-constituting property but merely a value-contributing property of literary works. If a literary critic spends too much time and energy arguing for the truths expressed, implied, or alluded to by the work, then their readers may simply lose sight of this fact. Rather, they may come to think that what the critic is after is some didactic end, while in fact they are in the business of enhancing literary appreciation. Compare this to philosophy: here, truth is a value-constituting property, so we would expect a philosopher to do what they can to establish some truth. A literary critic, in contrast, really wants their reader to realize the vividness of some literary 'vision'. For this, it is not necessary to know what counts in favour of some proposition, what would be evidence against it, and so on. If a critic talks their readers through these things at great length, then the readers may be right in turning their backs on them.
Third, recall that the psychological dependency thesis does not imply that we need to ponder the truth of some proposition for it to affect the vividness of our engagement with the work. Simply taking the proposition to be true will do. Pondering its truth may actually stand in the way of vividly imagining something. It may very well amount to one of the very distractions we have alluded to above that may keep one from vividly imagining something in the first place.
We conclude that there is some reason to assume that it would be odd if critics argued at length for the truth implied in literature. What we have suggested in this paper may thus go some way towards explaining why arguing for the truths implied in literature does not seem to play a prominent place in critical practice.

V. Conclusion
The aim of the paper has been to defend a version of aesthetic cognitivism: the truth of statements expressed, implied, or alluded to by a work of fiction matters aesthetically, and bears upon the work's aesthetic value. We first gave some reasons to think that a reader's imaginative engagement with a literary work of fiction tends to be more vivid if it is based on truth. We then argued that truth is a value-contributing property, which contributes to the literary value of a work in the specific sense developed by Lamarque and Olsen. If both claims are true, this has consequences for critical practice: pointing to, but not arguing for, truths should play an important part in critical practice.