Jonathan Gilmore’s latest book collates, organizes and distils his thinking on fiction, art, imagination, and the connections among them – subjects that he has been researching for at least a decade. The book has much to offer to those interested in the three terms individually and collectively, and what’s on offer is presented in appealing style. The book’s argument is motivated by bright-eyed insight concerning the question of how affective, cognitive, evaluative, and conative reactions to fictions differ from reactions prompted by real-world objects. With the question in open view, and summary accounts of imagination and emotion offered, an argument is then pursued in favour of discontinuity: the view that the norms by which we assess the appropriateness of emotions, beliefs, judgements, and desires ought to differ across fictional and real-world contexts. Each of these kinds of reaction gets its own chapter, with relatively independent but structurally similar arguments. The four are brought together in a final chapter that makes explicit a thesis underpinning those arguments: the idea that norms of aptitude differ according to the function of the representation with which one is engaging.

There is a lot to say about the merits of all these elements. But the question of whether a reader will get along with a book perhaps turns less on matters of content and conviction, and more on style, tone, and argumentative proclivities. In general, the discussion is pursued in crisp, clear prose that manages to be neither austerely terse nor indulgently prolix: the arguments are allowed to make themselves comfortable, but don’t get to overstay their welcome. Gilmore has a nice ear for a wry example, and he engages genuinely with actual works of fiction, rather than treating them as fields of examples to be harvested. The tone of discussion is calm and judicious, never less than respectful of the strength of opposing arguments. Indeed, some might find the tone more even-handed than ideal; this reader, at least, would not have minded a little more decisive direction at times (for example, in the discussion of the nature of emotions, pp. 44–57).

As for argumentative proclivities, Gilmore ably pulls off a balancing act that is not easily done. The discussion is, as one might expect, philosophical – distinctions are made, conceptions are clarified, cases are examined. The discussion is enriched by frequent excursions into relevant literature from psychology, cognitive science, neuroscience, and suchlike. The risks of such excursions are that the philosopher is either overly deferential, treating the Word of The Scientist as the arbiter of truth, or else (or also) insufficiently thorough, glibly reporting the few results and hypotheses that suit them and ignoring all the rest. Gilmore avoids both risks. Naturally, he tends to favour those theories that support his positions, but he explains them with due caution, and treats them as important yet not incontrovertible. This aspect of the book is exemplary.

There is also a lot to like about the content of the discussion. The good stuff begins with the way in which Gilmore sets up the book’s motivating question. We can see here the benefits of prolonged thinking about a topic, resulting in the kind of mastery that allows one to see its central problems clearly and originally. Gilmore puts the question in the form of a dilemma. On the one hand, we have a line of thought from Plato to the present day according to which our reactions to fictions are, or should be, the same as our real-life ones. After all, if they weren’t, we wouldn’t worry about the contagious nature of immoral reactions to fictions. On the other hand, we have a tradition of engaging with art that treats fictions as insulated and independent from the real world: a realm in which reactions can differ from those we might have to similar events, characters or actions in real life. The dilemma is that the two cannot both be quite right, since one presumes the continuity of reactions across contexts and one denies it.

This is a neat dilemma, and perhaps not one whose force is always felt as strongly as it should be. Gilmore is particularly good at pointing out where philosophical theories appear to rest on some version of continuity (for example, pp. 102–8). It would be an interesting exercise to pursue more fully than Gilmore himself does the consequences of his arguments against continuity for those theories.

Another virtue of Gilmore’s framing is his smart separation of the descriptive and normative versions of the question and views. One might wonder whether our emotions, imaginings, and so on are descriptively distinct in kind in different contexts: whether, for example, reactions to fictions involve ‘quasi-emotions’ that aren’t quite the real things. Gilmore’s view is that the evidence on this descriptive question tells fairly conclusively for continuity: there is, for example, ‘a great deal of descriptive continuity between our experiences furnished by the imagination, and those supplied by beliefs and perceptions’ (p. 30). One might also wonder, descriptively, whether our emotions and judgements differ in content in different contexts, but one need not wonder for long: we manifestly do react differently to fictions. The real interest is in the question of whether we should do so: whether, that is, the norms that govern the aptitude of reactions differ. This is not a question of whether we are allowed (in whatever normative sense) to treat fictional worlds as holiday resorts for the mind where the usual rules don’t apply; it is a question of whether we ought to have different reactions, whether the aesthetic demands of fiction oblige certain emotions and evaluations.

Gilmore argues that they indeed do so, at least sometimes. His case for discontinuity begins with accounts of imagining (Chap. 2) and emotion (Chap. 3). Imagining is conceptualized as a cognitive attitude attached to a propositional content, descriptively continuous with but ultimately distinct from ‘truth-apt attitudes such as belief’ (p. 22). It is identified and distinguished by its ‘causal-functional role’ in nexuses of other mental states, perceptions, and behaviour (p. 18). This role implicates imagining in all sorts of engagements with the counterfactual. There is a general question, possibly too general a question, about such engagements: How can it be established whether the imaginings involved are accurate, or reliable, or similarly fitting? More tractably, there is a specific question concerning engagements with fictions: how is it established what is true in a fiction? Gilmore rejects answers that rely only on ‘internal’ determinants, such as the propositional contents of the sentences in a book, and argues that we should allow also a role for ‘external’ determinants, such as genre (pp. 30–40). He also argues that artistic fictions tend to also shape how we imagine the truths of the fiction, besides telling us what those truths are (pp. 40–42). Here, the theme of the relevance of function to aptitude of reaction starts to develop.

The account of imagination is founded on the majority or consensus view in that field, but becomes more opinionated as the focus narrows from imagination in general to imagination and fictions. A similar pattern is evident in the chapter on emotion, although the consensus is perhaps less clear in that field. This means that the chapter provides a useful summary of work on emotion, which could stand alone as an overview. The position that Gilmore adopts is that emotions consist in a cognitive representation of some kind that presents an object and its qualities, and an affective ‘appraisal’ or evaluation of the object so represented. Emotions so conceived can be genuinely held towards objects of any metaphysical status one wishes, since the subtending cognitive representations need not be perceptual – they could, for example, be imaginings. Thus, we have descriptive continuity concerning the nature of emotional reactions to fictions and to the real world.

Chapter 4 establishes that the question of whether emotional reactions to fictions are normatively continuous or discontinuous can only arise against a shared assumption that there are such things as apt, fitting, or rational emotions, and begins a thread of discussion that continues into Chapter 5 concerning the appeal of normative continuity with respect to emotions: roughly, that the assumption of continuity makes sense of our practices of treating people’s reactions to fiction as indicative of their real-world attitudes. Despite that appeal, Chapter 5 makes a compelling case for normative discontinuity with respect to emotions: the argument, again roughly, is that aesthetic considerations can make emotions apt towards fictions that would be inapt in related real-world situations.

Chapters 6, 7, and 8 make similar cases with respect to epistemic, conative, and evaluative norms. What we take to be true in a fiction, what we desire concerning its events and characters, our moral attitudes towards them: in each case, what is appropriate can legitimately differ from what would be appropriate in the real world, and can do so owing to aesthetic considerations. These considerations essentially concern the function of engagements with fictions, or more precisely with those fictions that merit aesthetic engagement. Accordingly, Chapter 9 addresses and explains the question of what that function is, why we should think that artworks do indeed have such a function and how the function legitimates the application of different norms to engagements with fictions. The argument, once more roughly, is that artworks essentially have functions because they are intentionally made, and generally have the functions their creators intended; that anything that has an essential function is, normatively, supposed to fulfil that function; and thus that, if the creator of a work intends it to function such that certain reactions are apt, those reactions are indeed apt if one wishes to engage with the work as one is supposed to, even if the reactions are not those one would have in the real world. (Of course, one can refuse to engage with works, perhaps because one does not want to have the apt reaction.)

Two brief comments on all this. First, from my summary, it might look as though the arguments in Chapters 5–8 are repetitive. It is worth emphasizing that, while they are structurally similar, there is a good deal of difference in their details, and each is worth consideration independently. Second, it will be apparent that the aptitude of emotions get the majority of the attention: we are given a full account of emotion, for example, but not similar accounts of belief, desire, or moral judgement. This is a mild curiosity, and one wonders whether it belies an implicit commitment to the primacy of emotion in aesthetic engagement.

That’s enough summary. What you really want to know is, should you read this book? Yes, of course you should. Several of its parts have appeared in some form beforehand, but there is an inevitable bittiness about journal articles and the like, and none but Gilmore’s most assiduous fans will have traced his thought across the various pieces from which elements of the book are drawn. The book is rewarding and engaging.

Of course, part of the reward of a philosophy book is the puzzles it raises. Readers with different preoccupations will no doubt find their own puzzles, but I will mention one with which I was left.

The puzzle concerns the role of imagination, or rather imaginings, in the argument. As I mentioned, Gilmore argues that imaginings are descriptively continuous with beliefs. Both, qua attitudes, are typified by their functional role, and their functional roles are sufficiently similar to support descriptive continuity. The experiences they prompt are descriptively continuous, for example, and they do not (as one might think) play sharply distinguished roles in inferential processes (Gilmore details studies that seem to show the breakdown of the ‘quarantine’ between imaginings and beliefs that this would require, pp. 24–27). But he does not, as he does with emotion, draw the conclusion that imaginings are simply the same things as their reality-directed counterparts. He is keen to maintain the distinction, and offers reasons to think that imaginings cannot be ‘reduced’ to beliefs (pp. 18–22). But hold on. When it comes to emotion, the argument is not that quasi-emotions reduce to emotions: it is that so-called quasi-emotions just are emotions, only with different objects, owing to differences in the subtending representations they involve. As Gilmore repeatedly says, psychology does not respect metaphysics: irreal beasts can be as scary as real ones. But then: Why should we not say, similarly, that imaginings just are beliefs, with different kinds of object?

The more I wonder about this, the more I also wonder whether the belief-imagining distinction can only really be sustained by reference to the metaphysical status of the objects of these states. If their functional roles are overlapping to the extent that they are continuous, and if they have much the same kinds of content, it seems that the case for avoiding assimilation (if not reduction) of one to the other can only be premised on differences in the kinds of object they take. What defines and distinguishes imaginings from beliefs is just that imaginings stand to irreal entities and worlds as beliefs do to real ones. And if that is so, one might further wonder whether imaginings themselves, rather than the metaphysical statuses of their objects, are doing much real explanatory work when it comes to discontinuity. The normative legitimation of differing reactions to fictions and reality is grounded in the irreal status of the former, not in anything specifically to do with imagining.

This point is not at all fatal to Gilmore’s position; rather, it underscores the importance of function therein. The appropriateness of our reactions to fictions might be grounded in their irreality, but it is not determined by them. Exactly what reaction is appropriate will depend on what fiction in general, and each individual fiction, is for. Gilmore does acknowledge the importance of these considerations to his argument, dedicating an entire chapter to its functional foundations. But I think there is a lot more to be said here: in particular, a lot more to be said about the distinction implied in the subtitle between feelings for fictions and for ‘other creatures of the mind’. This distinction is not really addressed in the text, which concentrates almost exclusively on fictions; those other creatures remain mostly shadowy. Presumably, a full story about the aptness of imaginings would have to tell us about the normative status of our reactions to irreal entities with no aesthetic value, and that status would have to be tied to accounts of the various functions of various entities (as Gilmore briefly acknowledges, p. 203). Whether this is work for us or him is debatable. It is no doubt immensely satisfying to bring together all one’s long thinking on a topic, as this book does, and it must be tempting to leave things just as they are and move on to something new. Perhaps, then, the reader ought not to wait for Gilmore to say more, and instead consider whether they ought to say something themselves. That, I suppose, is the function of engagements with philosophies.