Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Extracting value : appreciative engagement as metacognition van der Berg, Servaas 2020

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


24-ubc_2020_may_vanderberg_servaas.pdf [ 1.11MB ]
JSON: 24-1.0389629.json
JSON-LD: 24-1.0389629-ld.json
RDF/XML (Pretty): 24-1.0389629-rdf.xml
RDF/JSON: 24-1.0389629-rdf.json
Turtle: 24-1.0389629-turtle.txt
N-Triples: 24-1.0389629-rdf-ntriples.txt
Original Record: 24-1.0389629-source.json
Full Text

Full Text

Extracting ValueAppreciative Engagement as MetacognitionbyServaas van der BergA THESIS SUBMITTED IN PARTIAL FULFILMENT OFTHE REQUIREMENT FOR THE DEGREE OFDoctor of Philosophyin The Faculty of Graduate and Postdoctoral Studies(Philosophy)THE UNIVERSITY OF BRITISH COLUMBIA(Vancouver)March, 2020© Servaas van der Berg, 2020The	following	individuals	cer3fy	that	they	have	read,	and	recommend	to	the	Faculty	of	Graduate	and	Postdoctoral	Studies	for	acceptance,	the	disserta3on	en3tled:	Examining	Commi+ee:	Addi0onal	Commi+ee	Member:	Extrac3ng	Value:	Apprecia3ve	Engagement	as	Metacogni3onsubmiDed	by Servaas	van	der	Berg in	par3al	fulfilment	of	the	requirements	forthe	degree	of Doctor	of	Philosophyin PhilosophyDominic	McIver	Lopes,	Professor,	Department	of	Philosophy,	UBCCo-supervisorChristopher	Mole,	Professor,	Department	of	Philosophy,	UBCCo-supervisorMaDhew	Bedke,	Professor,	Department	of	Philosophy,	UBCSupervisory	CommiDee	MemberMurat	Aydede,	Professor,	Department	of	Philosophy,	UBCUniversity	ExaminerAlexander	Dick,	Associate	Professor,	Department	of	English	Language	and	Literatures,	UBCUniversity	ExaminerRobert	Hopkins,	Professor,	Department	of	Philosophy,	New	York	UniversityExternal	ExamineriiAbstractA considerable part of human life is structured around appreciative pursuits, including but not limited to our dealings with the arts. While these pursuits are vastly heterogeneous, they share some general psychological underpinnings. This dissertation investigates those underpinnings.Firstly, I argue that theories of appreciation in aesthetics have been unduly constrained by the dominance of hedonist accounts of aesthetic value. Aesthetic hedonism posits a constitutive link between the value of appreciative experiences and the aesthetic values of their objects. But once alternatives to hedonism are taken seriously, and this link is no longer taken for granted, new avenues open up for theorizing about appreciative engagement, and its relevance to value theory beyond aesthetics comes into clearer view.Secondly, I revisit the aesthetic attitude theorists’ much maligned idea that appreciation involves a distinctive mode of attention. Drawing on recent work in the philosophy of games, I develop a novel account of this distinctive attentional mode, in terms of a nested hierarchy of goals by which attention is guided in appreciative episodes.Finally I argue that our thinking about how self-awareness figures in appreciation should be more thoroughly informed by empirical work on human metacognitive capacities. I review two bodies of empirical literature on the subject and use them to develop a proposal about the role of metacognition in our appreciative encounters with the world.iiiLay SummaryIt is part of human nature that we do not just blindly pursue things we value, to possess, use, or consume them. We spend a lot of time and cognitive effort appreciating them: we inspect, learn about, interpret, explore, compare, evaluate, reflect on, and savour them. But what kind of activity is appreciation, and how does it fit into our practices and a well-lived human life? This dissertation develops an account of appreciative engagement as a cognitive activity crucial not just to our interactions with art and the beautiful, but our dealing with things we value in general. I argue that, although appreciation often involves pleasure, it is not essentially aimed at pleasure; I use some tools from the philosophy of games to elucidate the appreciative mode of attention; and finally, I review work from psychology and cognitive science to learn about the role of self-awareness in appreciation.ivPrefaceThis dissertation is the original and independent work of the author, Servaas van der Berg.A version of Chapter 2 has been published as Van der Berg, S. (2020), ‘Aesthetic Hedonism and Its Critics’, Philosophy Compass 15/1: e12645, version of Chapter 3 has been published as Van der Berg, S. (2019), ‘The Motivational Structure of Appreciation’, Philosophical Quarterly 69/276: 445–466, iii..................................................................................................................................Lay Summary iv.........................................................................................................................Preface v.....................................................................................................................................Contents vi..................................................................................................................................List of Figures viii......................................................................................................................Acknowledgements ix................................................................................................................CHAPTER ONE   Introduction 1............................................................................................1.1  The preliminary identification of our topic 1...................................................................1.2  Aesthetic and non-aesthetic 4...........................................................................................1.3  A place for pleasure? 9.....................................................................................................1.4  A dynamic approach to appreciative attention 11............................................................1.5  A role for self-awareness 13.............................................................................................CHAPTER TWO   Aesthetic Hedonism and Its Critics 15....................................................2.1  Pleasure, aesthetic value, and appreciation 15.................................................................2.2  Fault lines in the debate 17..............................................................................................2.3  Objections to aesthetic hedonism 23................................................................................2.4  Alternatives to aesthetic hedonism 39..............................................................................CHAPTER THREE   The Motivational Structure of Appreciation 44................................3.1  Disinterest as mode of attention 44..................................................................................3.2  Dickie’s challenge 47.......................................................................................................3.3  Striving play 53................................................................................................................3.4  Qualifying the submersion requirement 61......................................................................3.5  From striving play to appreciation 65..............................................................................3.6  Disinterest revisited 75....................................................................................................CHAPTER FOUR   Appreciation and Metacognition 79......................................................4.1  Second-order troubles 79.................................................................................................4.2  The allure of cognitive ascent 82.....................................................................................4.3  The challenge from social psychology 87........................................................................vi4.4  Moving beyond reflective introspection 97.....................................................................4.5  A role for ‘procedural’ metacognition 104.......................................................................4.6  Moving beyond the merely procedural 110.....................................................................4.7  A proposal: cognitive agency under metarepresentational norms 115.............................CHAPTER FIVE   Conclusion 122..........................................................................................5.1  Assembling the pieces 122...............................................................................................5.2  Directions for future research 125...................................................................................Bibliography 135........................................................................................................................viiList of FiguresFigure 3.3: Four types of game play 56.......................................................................................viiiAcknowledgementsDom Lopes and Chris Mole deserve a great deal of credit for this dissertation’s merits and none of the blame for its shortcomings. Their guidance was judicious, practical, and extraordinarily efficient throughout, they were supportive to a fault, and they were always readily available despite an excess of other responsibilities. It is a rarer-than-acknowledged privilege to have a doctoral supervisor that one not only respects, but thoroughly likes and profoundly admires; in this regard, I have been privileged twice over.Thanks also to Matt Bedke, for graciously engaging with a project that veered further from his own research interests than initially planned, and to my other examiners, for their generous reading of my work: Rob Hopkins, for comments as astute, incisive, and challenging as ever; Alex Dick, for an historical perspective on my project that I wish I had taken up earlier; and Murat Aydede, for the characteristically collaborative spirit with which he approached the dissertation.UBC’s Philosophy faculty have almost all played a considerable part in my professional and personal development. Besides those already mentioned, thanks especially to Roberta Ballarin, John Beatty, Sylvia Berryman, Carrie Ichikawa Jenkins, Jonathan Jenkins Ichikawa, Eric Margolis, Alan Richardson, and Chris Stephens. The same goes for my former colleagues at Stellenbosch, for their continued interest despite the distance, especially Vasti Roodt, J. P. Smit, Willie van der Merwe, Anton van Niekerk, and Minka Vrba. For their friendly and competent administrative assistance over the years, thanks very much to Nissa Bell, Simone Dharmaratne, Hilary Geise, Jillian Read, Tim Son, Liesl van Kerwel, Erika Yep, and Mimi Yu.If there was one source that shaped my thinking in this dissertation most, it was probably the manuscript of Thi Nguyen’s riveting new book on the philosophy of games. I am not the only one to have been inspired by Thi’s invariably fresh and interesting perspectives on philosophical issues, nor will I be the last. He is a wellspring of worthwhile ideas. What luck to have had not just advance access to his work, but also his direct input, encouragement, and friendship throughout this project.Many other philosophers that I met along the way contributed ideas, opportunities, advice, and a sense of community for which I am deeply grateful. Thanks especially (but not only) to Julianne Chung, Anthony Cross, John Dyck, Anne Eaton, Dave Friedell, Keren Gorodeisky, Sarah Hegenbart, Zac Irving, Alex King, Irene Martínez Marín, Samantha Matherne, Derek Matravers, Jeremy Page, Antonia Peacocke, Nick Riggle, Guy Rohrbaugh, Elizabeth ixScarbrough, Elisabeth Schellekens, James Shelley, Brian Soucek, Matt Strohl, Rebecca Wallbank, Jonathan Weinberg, and Michel Xhignesse.I am indebted to my fellow graduate students at UBC for providing just about the ideal intellectual and social conditions for the pursuit of a PhD. They will know that a mere ‘thank you’ belies the depth and meaning of the connections we have forged, but nonetheless, thank you to Jiwon Byun, Irwin Chan, Kyle Da Silva, Taylor Davis, Tyler DesRoches, Emma Esmaili, Chris French, Joseph Frigault, Kinley Gillette, Jade Hadley, Julia Janczur, Curtis Kehler, Rebecca Livernois, Jonathan Lopez, Jelena Markovic, Phyllis Pearson, Madeleine Ransom, Aida Roige, Richard Sandlin, Matthew Smithdeal, Jerry Tsui, Jerry Viera, Jordan Wadden, and Kousaku Yui (and Laura Keith). It has been an exceptional joy to have the comradeship of Jasper Heaton for almost all of the years I spent in the programme, and I shall remember few activities from this part of my life as fondly as arguing with Aleksey Balotskiy, Kaitlin Graves, and Sina Fazelpour over drinks at Upstairs Campagnolo.Finally, thank you to my parents, Eunice and Servaas van der Berg, and my siblings, Ineke and Wimpie, for their unquestioning love and support, and to Sophia Sideris, without whom I might well have lost sight of what really matters. xCHAPTER ONE   IntroductionThe word we ought to talk about is ‘appreciated’. What does appreciation consist in?—Wittgenstein, Lectures on Aesthetics, §181.1  The preliminary identification of our topicAesthetic value, more so than value in other normative domains, is intimately tied to appreciation. Of all the ways we interact with aesthetic goods, appreciating them is among the most prominent. We spend immense amounts of time and cognitive resources examining, interpreting, learning about, evaluating, analysing, comparing, reflecting on, enjoying, and savouring the things we value aesthetically, whether they are artworks, features of the natural or urban environment, abstracta (like mathematical proofs), or mundane everyday artefacts. Many of the practices we build around aesthetic goods—ranging from large scale global institutions to informal conventions among friends with shared aesthetic interests—are aptly characterized as appreciative practices.It is no surprise, therefore, that appreciation also features centrally in philosophical aesthetics. The field’s core concepts and theoretical constructs often revolve around the appreciator’s perspective. Philosophers regularly take aesthetic properties, like beauty, gracefulness, stridency, or pizazz, to be prime contenders for sentimentalist analyses in terms 1of the responses involved in their appreciation (Goldman 1995, De Clercq 2002a, 2002b, Matravers 2005, Levinson 2005). And aesthetic judgements are widely thought to be governed by an acquaintance norm that ties their paradigmatic occurrences to contexts of direct appreciative engagement with the items judged (Wollheim 1980: 233, Hopkins 2000, 2011, Livingston 2003, Lopes 2014a: 169–76, Robson 2015, Nguyen 2017, 2019, Ransom 2019). Whatever one thinks of the details of such ideas, their prevalence in the field signals that the cognitive phenomena involved in appreciation are likely crucial to a comprehensive understanding of the aesthetic domain. But how should we explain the intimate connection between appreciation and the aesthetic?The currently orthodox view of aesthetic value offers a straightforward answer. According to aesthetic hedonism, the value of appreciative experience is the basic source of normativity in the aesthetic domain. To say that something has aesthetic value just is to say that it offers a certain kind of finally valuable experience to appreciators. Thus, for the hedonist, our aesthetic reasons are at base reasons to appreciate and we only derivatively have aesthetic reasons for other kinds of actions or activities (such as actions that promote opportunities for appreciation). In other words, on the hedonist picture, the connection between aesthetic value and the value of appreciation is direct and constitutive.Of course, orthodoxy is not without its detractors. Recent critics of aesthetic hedonism worry that it paints a specious picture of the place and importance of aesthetic value in our lives (see Chapter 2 for discussion). This dissertation presses the converse concern: that the hedonist construal of the link between aesthetic value and appreciation is detrimental to our understanding of the latter. The worry, broadly speaking, is that hedonism’s influence in aesthetics has made appreciation into a mere sidekick for aesthetic value. Aiming to give a 2general account, for any item x, of what x’s aesthetic value consists in, aesthetic hedonism points to the final value of appreciative experiences centred on x. A tacit correlate of this commitment is that, for any x, to aesthetically appreciate x is just to engage in whatever x-directed cognition engenders the experiences that ground x’s value. Call this the placeholder conception of aesthetic appreciation. It is a thin conception; it tells us little about the general features of token episodes of appreciative engagement. Instead, it just delineates the axiological role that hedonism takes such episodes to play—that is, the role of grounding aesthetic value.And yet, the placeholder conception has come to constrain, and even guide, substantive theorizing about appreciation within the hedonist paradigm. When hedonists characterize appreciation in terms like ‘valuing the experiencing of a state of affairs for its own sake’, or ‘the having of inherently worthwhile experiences’, or ‘attending for the sake of the very experience’ (see for example Iseminger 2005a: 99–100, 2005b: 99, Stecker 2006: 4, Levinson 2016: 31–2), they are—in some cases explicitly—offering us little more than alternate encodings of the placeholder conception. The work of the coming chapters reflects the belief that a more substantive grasp of the nature of appreciative engagement, and its place in human life, becomes available only once we stop shoehorning it into a hedonist axiology. The value of token appreciative episodes may well have some role to play in an accurate theory of aesthetic normativity, but that role need not be straightforward, and it is best not held fixed in advance of progress on an independent conception of what it is we do when we engage in appreciation.To that end, the main aim of this dissertation is to shed light on some of the basic cognitive capacities underlying our propensity for engaging with the world appreciatively. 3Focussing on three features traditionally prominent in discussions of aesthetic appreciation, the three main chapters provide accounts of how we should (or should not) conceive of these features and their role in appreciative engagement. This chapter provides the background to these three accounts. The first, discussed in Section 1.3, is largely negative in its aims, and concerns the reasons for rejecting the hedonist construal of appreciation purely in terms of aesthetic pleasure or value-grounding experience. The second, discussed in Section 1.4, concerns the notion of a distinctive appreciative mode of attention. The third, discussed in Section 1.5, is an account of the role that our metacognitive capacities play in episodes of appreciation. Before we proceed with the background to these chapters, however, the next section briefly flags an idiosyncrasy in this dissertation’s approach to its topic.1.2  Aesthetic and non-aestheticConsider the proposition that appreciation is as commonplace in our interactions with some non-aesthetic goods as it is in our dealings with the aesthetic. On the face of it, this claim appears uncomplicated, but it is tricky to assess for two reasons.Firstly, the concept of appreciation we are working with, especially this early in the inquiry, is at best an imprecise folk concept. The problem is not just that the term ‘appreciation’ is polysemous, because worries about its polysemy can mostly be cleared up by stipulation. For example, we can discount the purely epistemic sense of ‘appreciation’ on which it denotes knowledge or understanding—as in ‘I have come to appreciate that Black Friday is a scam’ or ‘I did not fully appreciate the implications of voting to leave the EU’. This narrow epistemic sense is not the one this dissertation is interested in. Similarly, the 4senses on which ‘appreciation’ denotes an attitude of gratefulness—as in ‘I appreciate your support’—or an increase in monetary value—as in ‘the appreciation of the Yen against the Rand’—are not our current concern.The notion of appreciation this inquiry targets is embedded in a particular use of the common phrase ‘art appreciation’. In one invocation of that phrase, ‘appreciation’ is meant in something like the purely epistemic sense we just discounted, as simply denoting aesthetic perception, cognition, or evaluation. We see this, for example, in David Davies’ (2004: 6–7) use of the term ‘appreciation’ to signify the epistemological aspect of a philosophical theory of art, in contradistinction to the definitional, metaphysical, and axiological components of such a theory (this use of the term is commonplace in aesthetics; see Dorsch 2007 for another example). The phrase ‘art appreciation’ is often used more broadly, however, to indicate not just the epistemic component of our engagement with artworks, but some larger concert between of all of the perceptual, cognitive and affective capacities brought to bear in our encounters with art. In this usage, the appreciation in ‘art appreciation’ is not just the recognition of value, but also its consumption. It is this same sense implicit in the observation attributed to Voltaire, that ‘Appreciation is a wonderful thing: it makes what is excellent in others belong to us as well.’ And Alexander Nehamas (2007) invokes the same notion in his stirring treatment of the encounter with beauty, which he conceives as the onset of a communing with and integrating of valuable things into our lives, in the face of the risk that they might change us. It is to this capacious notion of appreciation that the work of the coming chapters pertains—the notion of engaging with and extracting and partaking in, rather than merely recognizing, value.5These somewhat ornate observations about the pertinent sense of the term will perhaps  forestall some potential equivocations, but of course they still do not move us much beyond a vague working concept. This is deliberate, however. Target explananda cannot always be specified with full precision in advance of an inquiry, and the aim will be to narrow our focus to a clearer set of appreciative phenomena as we progress.Secondly, the claim that appreciative engagement occurs with respect to both aesthetic and non-aesthetic goods, presupposes a grasp of what separates the aesthetic from other normative domains. Luckily, readers of this dissertation likely have ample conceptual competence for distinguishing between paradigmatic examples of, say, aesthetic, moral, and prudential value. But competence with a concept is a far cry from having a theory of its application conditions, and, as a theoretical matter, the issue of how the aesthetic domain is demarcated is a perennial and intensely contested one (see Shelley 2017 for discussion). This is significant for current purposes, because there is a non-trivial question about whether appreciation as it is elucidated in the coming chapters should be thought of as a characteristically aesthetic phenomenon or not. Intuitions grounded in our basic conceptual competence with the pertinent notions of appreciation and the aesthetic will not decide the matter. A more promising tack, rather, is to venture a theoretical gambit.The approach favoured here will be to conceive appreciative engagement broadly—as a determinable, with episodes of aesthetic appreciation comprising a non-exhaustive subset of its determinate instances. Hence the claim this section opened with: we engage appreciatively with both aesthetic and non-aesthetic goods in the world. There is precedent for thinking of appreciation in these broader-than-aesthetic terms, especially in the work of philosophers who see parallels between our cognitive engagement with aesthetic goods and with other 6persons—just consider the Voltaire quotation for illustration. The idea, roughly, is that there can be significant similarities between, say, reflecting and savouring the aesthetic merits of a favourite piece of music, and reflecting on the moral character or endearing quirks of a close friend. While by no means new, the idea of such parallels between aesthetic and ‘personal’ appreciation has received renewed attention in aesthetics over the last decade (see for example Nehamas 2007, Moran 2012, Riggle 2016, Cross 2017b, Protasi 2017, Kubala 2018). A classic endorsement of this notion is also to be found in a famous passage of Principia Ethica, where G. E. Moore (1903: 188–90) declares the appreciation of beauty and of ‘personal affections’ to be the two greatest goods we can imagine. Moore proceeds to give a strikingly similar treatment of the value that accrues to both kinds of appreciation.  This 1dissertation follows his precedent, by conceiving of appreciation broadly and as a unified phenomenon across both its aesthetic and non-aesthetic occurrences.Having said that, a major part of the dialectical strategy in what follows will be to rework ideas that have their home in a tradition primarily interested in appreciation as it pertains to the aesthetic. This is inevitable in part because the field of aesthetics has, for obvious reasons, been the traditional headquarters for philosophical theorizing about appreciative phenomena. But readers may nonetheless have qualms about the move of relying on these ideas in the context of a project that conceives appreciation in broader-than-aesthetic terms. After all, the alleged features of appreciation investigated in coming chapters Moore treats token episodes of both kinds of appreciation as paradigm instances of his ‘principle of 1organic unities’, according to which the overall value of the episode cannot be reduced to any aggregate of the value inherent in either the appreciator’s cognitive and experiential states, or in the targets of appreciation considered on their own (for alternative versions of this view, see Parfit 1984: 500–502, Korsgaard 1983: 190–92, Hurka 1998). The Moorean picture thus also offers a refreshing contrast with aesthetic hedonism’s reduction of an item’s aesthetic value to the value of appreciative experience.7—its involving a distinctive kind of pleasure or experience (Chapter 2), a ‘disinterested’ mode of attention (Chapter 3), or a particular kind of self-awareness (Chapter 4)—are features that have traditionally been posited as criterial of aesthetic engagement or appreciation, considered as such.In response to these qualms it should be noted that, for the purposes of this dissertation, not a great deal is at stake in disputes about whether the notion of appreciation it elucidates is ultimately understood as characteristically aesthetic or not. The choice to cast it as a broader-than-aesthetic phenomenon reflects the conviction that many of the cognitive capacities traditional theories of aesthetic experience or appreciation track, are relevant to our cognitive engagement with value far beyond the boundaries imposed by traditional conceptions of the aesthetic domain. The fact that they are often treated as distinctively aesthetic can be explained as an artefact of their ubiquity and importance in the aesthetic domain, rather than as a result of their being coextensive with that domain.Readers who disagree with this diagnosis, however, are free to consider how the picture of appreciative phenomena painted here fits with their preferred conception of the aesthetic domain. Ultimately, the methodological gambit of not tying appreciation too closely to the aesthetic is intended primarily as a means to sidestep questions about demarcation. Yet perhaps, by the end of this dissertation, the incidental benefits of construing appreciation broadly in this way—benefits such as bringing some significant continuities between our aesthetic, moral, and personal lives into sharper focus—might help convince readers with strong views about the demarcation of the aesthetic, that joining in the gambit is worth the price of admission. Either way, none of the dissertation’s most central claims depend upon it.81.3  A place for pleasure?Recall that the placeholder account holds that aesthetic normativity is sourced in the value of appreciative experiences. More precisely, it ties the aesthetic value of an item, x, to the final value of pleasures or experiences involved in successful appreciative engagement with x. Thus, for all its shortcomings, the placeholder account comports with a compelling story about the value of appreciation—namely that it consists in experiential or hedonic value. That successful appreciation often or even typically coincides with pleasure, is a datum difficult to deny. Likewise, that pleasure or phenomenal experience can be a primitive source of value, is philosophical orthodoxy and common sense. The question is, should we come along for the further step of reducing appreciation’s significance to its capacity for engendering such pleasure or valuable experience?We should not agree to such a reduction. But to make the case that we should not, we cannot simply rely on intuitions grounded in our folk concept of appreciation (as defenders of the reduction sometimes do). Our working concept is too preliminary and vague to yield reliable support for either position, and, after the discussion in the last section, it is perhaps already a little theoretically loaded. Nor can appeals to precedents like Moore or Nehamas, who reject a thin conception of appreciation in terms of pleasure or experience alone, get us very far; they can too readily be countered with similarly venerable examples from the hedonist tradition. A better way to effect a shift away from the placeholder account’s emphasis on the experiential or hedonic value of appreciative episodes, is to subvert the hedonist picture of aesthetic value that lends weight to it.9Chapter 2 therefore addresses a challenge to hedonist orthodoxy, by collecting and organizing the main counterarguments to aesthetic hedonism from a growing recent body of work that is critical of the view. The chapter’s main aims are integrative, and it serves to systematize an active debate that has been steadily gaining steam but has been adversely affected by a shortage of shared theoretical vocabulary. In particular, the debate has been hindered by its relative isolation in the subfield of aesthetics, from similar debates in other parts of value theory (especially work on hedonism about prudential value and welfarism about value in general). The chapter draws out salient connections with those debates and shows that, with respect to several counterarguments, the hedonist about aesthetic value is not much better off than the prudential hedonist or welfarist.The upshot for the current inquiry into appreciation is mostly indirect, but nonetheless substantial. Whether or not the arguments against aesthetic hedonism are conclusive, they open up the space for considering alternative options, like objectivist and perfectionist theories of aesthetic value (see Section 2.4). And, once the constitutive tie between aesthetic value and the value of appreciative experience is rendered less axiomatic for work on aesthetic value, it also becomes less obvious to conceive of appreciation in terms of its role in producing value-grounding experiences. Thus the way is cleared to look beyond the thin, experience-centred placeholder account of appreciation, towards a thicker notion elucidated in subsequent chapters.101.4  A dynamic approach to appreciative attentionHaving set the reduction of appreciation to its capacity for affording pleasure to one side, we may approach a second perennial idea about aesthetic appreciation without prejudice. This is the idea that appreciation can be understood in terms of a distinctive, disinterested manner or mode of attending. The idea to connect disinterest specifically to attention has been around in aesthetics at least since Schopenhauer (1819/2010), and found a natural home in the aesthetic attitude theories of the early to mid-twentieth century (Bullough 1912, Stolnitz 1960). The potential relevance of differential patterns of attention to value theoretic concerns has also had a handful of champions outside aesthetics, notably in Simone Weil, who locates the source of ‘all authentic and pure values—truth, beauty and goodness—in a certain application of the full attention to the object’ (Weil 1986: 234), and Iris Murdoch (1985), who considers attention directed away from the self to be crucial to a morally good life (for recent treatments of these ideas, see Mole 2006, Vice 2006, Debus 2015, Bommarito 2013). The broad similarities and connections between the idea of disinterested attention in aesthetics and Murdoch’s notion of the moral-attentional activity of ‘unselfing’ provides cover for the gambit this dissertation engages in, of approaching the appreciative mode of attention as a broader-than-aesthetic phenomenon.An explosion of scientific work on attention since the rise of cognitive psychology in the mid-twentieth century, alongside the rediscovery of attention as a topic in philosophy over the last two decades (see Mole 2017 for discussion), has had a fairly noticeable impact on aesthetics. This is true especially in the fields of empirical and neuro-aesthetics, but philosophical aesthetics has also become more prone to exchanging ideas with attention 11research in the philosophy of mind and its cognate empirical disciplines. However, such exchange has been noticeably absent from philosophers’ engagement with the notion of a distinctively disinterested or appreciative mode of attention (although see Nanay 2016 for a notable exception). A partial explanation for this is that, in philosophical aesthetics, interest in the notion has typically begun and ended with its value theoretic implications within the aesthetic hedonist paradigm. As a result, what has been taken to matter is specifically the phenomenology and attendant value of the experience of attending disinterestedly or ‘for its own sake’. Characterizations of the appreciative mode of attention and its significance in aesthetics have therefore typically relied almost exclusively on introspective methods, focussing on what the experience of appreciative attention is like, rather than, say, on the kind of cognitive behaviours it involves or motivates, or on its cognitive underpinnings.This is unfortunate, not least because the philosophy of mind and cognitive science have been producing illuminating work on many dynamic mental phenomena aptly thought of as modes of attention, including for example creative and spontaneous thought, absorption, rumination, dreaming, and meditative states. An especially productive source of work on such phenomena has been Kalina Christoff’s Cognitive Neuroscience of Thought laboratory at UBC (see for example Christoff et al. 2009, Solomonova 2014, Christoff et al 2016, Mills et al. 2018, and for some other examples, see Metzinger 2013a, 2013b, Windt 2011, Irving 2016, and the entries in Christoff & Fox 2018).Chapter 3 capitalizes on this literature by borrowing an insight from recent work on mind-wandering: that modes of thought can be fruitfully understood in terms of how attention is guided as an episode of such thought unfolds over time. Using a minimal notion of guidance, the chapter develops an account of how one type of disinterested motivation 12might be understood within such framework. It then applies the account to episodes of deliberate thought to offer a new model of the appreciative mode of attention. 1.5  A role for self-awarenessIn an engaging recent entry in pop philosophy of art, Alva Noë (2015) presents a picture of our artistic practices as tools we use to investigate and reorganize ourselves and our own lives. Art and its appreciation, on Noë’s picture, are essentially second-order enterprises in which we (both individually and collectively) represent and engage with representations of ourselves, to ourselves, in order to make sense of and rearrange our first-order activities. Details of Noë’s account aside, there is something appealing in the thought that our capacity for self-awareness must surely be central to a broad explanation of why creatures like us engage with value in the complicated ways that we do—not least in our encounters with artworks and other objects of appreciative engagement. Unsurprisingly, philosophers picking up on this idea (or perhaps on some other ideas in the vicinity) have been characteristically quick to simply posit second-order mental states as essential to appreciation, ranging from second-order pleasures (Walton 1993) to second-order valuings (Iseminger 2005b) to complex, self-referential experiences that represent themselves as merited (Gorodeisky 2018), to mention just a few examples.Taking seriously the proposal from Chapter 3—that we can fruitfully think of appreciation in perdurant terms, as dynamic, unfolding episodes of cognitive engagement—Chapter 4 asks what role our metacognitive capacities might properly play in such episodes. The chapter argues that a good answer to this question will require bringing our best 13understanding of human metacognitive capacities to bear on our theorizing from the outset. To meet that requirement, the chapter looks to empirical work on metacognition.The empirical literature on metacognition is a fast growing field spread across various disciplines in the sciences of the mind. At least two major research programmes relevant to metacognitive phenomena have crystallized over the past few decades. One of them, in social psychology, focusses especially on first-personal reports and conceptual reasoning about experimental subjects’ own mental states and behaviours (see Section 4.3 for discussion of a sample of this research programme). The other, at the intersection of cognitive and comparative psychology, focusses on evidence of the differences that a subject’s low-level awareness of the relative fluency of their own perceptual and cognitive processing can make to their epistemic behaviours (see for example the papers in Beran et al. 2012, and see Section 4.5 for a broader sample).While both these research programmes focus on cognitive capacities that have come to be called ‘metacognitive’, there is consensus that the only the former targets conceptual metacognition—that is, the abilities contingent on the human mindreading system’s metarepresentational resources (Carruthers 2016, Proust 2012). Chapter 4 uses insights gleaned from the consideration of work in both these research programmes, first, to refine the question about the role of metacognition in appreciation in terms of the metarepresentational resources of the mind-reading system, and second, to develop a proposal about how we should conceive that role.14CHAPTER TWO   Aesthetic Hedonism and Its Critics2.1  Pleasure, aesthetic value, and appreciationAesthetic hedonism holds that aesthetic value is a special kind of hedonic value—that is, an item’s aesthetic value is simply its power to please us in a certain way.  If we construe 2hedonism broadly, to include views that go under the labels of aesthetic value empiricism or experientialism (see Section 2.2.2 for discussion), it would be fair to say that contemporary aesthetic value theory has been thoroughly dominated by a hedonist consensus.  This is 3hardly surprising. Boasting a generous share of intuitive plausibility, hedonism could arguably lay claim to being the common sense view of aesthetic value. After all, who would deny that our encounters with the aesthetic are typically a source of great enjoyment, and, occasionally, of transcendent delight? From here it is a short step to the hedonist doctrine that an item’s aesthetic value is constituted by its relation to such pleasure or valuable experience. Some propose hedonism as an account of artistic rather than aesthetic value. For ease of exposition, 2both are treated here under the banner of aesthetic hedonism and the difference is marked only where needed. For discussion of the artistic/aesthetic value distinction, see Lopes (2011), Huddleston (2012), Stecker (2012), Hanson (2013), Dodd (2014), and Forsey (2017). For contemporary articulations of aesthetic hedonism, construed broadly, see Dickie (1988), 3Mothersill (1989), Levinson (1992, 2002, 2016), Walton (1993), Stephen Davies (1994), Budd (1985, 1995, 2008), Iseminger (2004, 2005b), Goldman (1995, 2006), Stecker (1997, 2005, 2006, 2019), Stang (2012), and Matthen (2017, 2018). Many others working in aesthetics endorse some form of the view. For a more extensive though probably still not exhaustive list of its proponents, see Lopes (2018: 9).15Lately, however, resistance has been on the uptick. Among a growing list of dissenters, James Shelley (2010, 2011, 2013, 2017, 2019) and Dominic Lopes (2015, 2018) have proven especially persistent and methodical in their opposition to aesthetic hedonism.  But while the 4challenges mount, replies from within the hedonist camp have so far been scarce and, at best, perfunctory. A real debate about the view’s strengths and shortcomings, and about what the alternatives look like, is only now getting properly under way. This chapter’s primary aim is to aid in that debate, by cataloguing the main extant arguments against aesthetic hedonism and, thereby, mapping the territory for its defenders and detractors alike.A subsidiary aim—more central to this dissertation’s broader dialectic and already flagged in Chapter 1 (Section 1.3)—is to dislodge the dominance of the hedonist research programme in setting parameters for theorizing about aesthetic phenomena, including but not limited to the phenomenon of (aesthetic) appreciation. The picture of appreciation that falls naturally out of aesthetic hedonism entails that appreciating simply amounts to taking proper pleasure in aesthetic goods—that is, it amounts to tokening some subset of the pleasurable (or otherwise valuable) experiential states that serve to ground the aesthetic value of the goods in question (this the placeholder conception discussed in Chapter 1). It would be a fool’s errand to deny that pleasure and valuable experience can, and do, feature prominently in our aesthetic lives and in many, or even most, episodes of appreciation. But the move from recognizing pleasure’s prevalence in appreciative episodes to granting it an essential role in appreciation, is just as problematic as the move from pleasure’s prevalence in our interactions with the aesthetic, to making it constitutive of aesthetic value. The counterarguments to  Other critics of aesthetic hedonism include Sharpe (2000), David Davies (2004), Kieran (2005, 42008), Wolf (2010), Watkins and Shelley (2012), Riggle (2013, 2015), Gorodeisky (2019), Matherne & Riggle (ms.), and Peacocke (ms.a).16hedonism reviewed in this chapter suggest that, at worst, it is an error, and, at best, premature, to simply default to the hedonist conception of appreciation couched in terms of valuable experiential states. Thus, by spelling out the challenges to aesthetic hedonism, this chapter clears the way for the alternative picture of appreciative phenomena developed in the chapters that follow.The roadmap for this chapter is simple. Section 2.2 homes in on some key distinctions, setting up the review of counterarguments to hedonism in Section 2.3. Section 2.4 asks: if not hedonism, then where might work on aesthetic value go next?2.2  Fault lines in the debateThree preliminary distinctions can help shed light on features of aesthetic hedonism that have been targeted by the recent criticisms. The way in which particular hedonist theories situate themselves with respect to the distinctions will affect which of the counterarguments reviewed in the next section they are most vulnerable to.2.2.1 Demarcation versus normativityFirst, it is fast becoming standard among hedonism’s critics to flag a distinction between two questions that a complete theory of aesthetic value should answer (Shelley 2019: 1, Lopes 2018: 41–3, 2019, King 2020, Gorodeisky 2019, Matherne ms., Matherne & Riggle ms., Peacocke ms.a). The ‘demarcation question’ asks what makes aesthetic values aesthetic—what distinguishes them from values in other domains? The ‘normative question’ asks what makes aesthetic values values? One might understand the normative question in 17terms of reasons: what makes it the case that aesthetic values give us reasons for anything? Part of aesthetic hedonism’s appeal lies in its promise to bridge intuitive answers to both questions. It answers the normative question by reducing aesthetic value to hedonic value: aesthetic values generate reasons because we have reason to pursue pleasure, and the experiences said to ground aesthetic values are pleasures, or, at the very least, they are finally valuable like pleasures, such that we have non-derivative reason to pursue them (see Section 2.2.2). With respect to the demarcation question, aesthetic hedonism does not entail an answer, but it nonetheless recommends an approach by pointing to a theory of aesthetic experience. Whatever distinguishes aesthetic pleasures or experiences from non-aesthetic ones, the thought goes, will determine which of the values grounded in pleasant or finally valuable experience (that is, which hedonic values) are distinctively aesthetic.Unfortunately, agreement on the nature of aesthetic experience has proven elusive, with the result that this has become a main focal point for work on aesthetic value (see for example Iseminger 2005a, 2005b, Stecker 2006, Goldman 2006, Carroll 2002, 2012, Levinson 2016). The move to flag the demarcation/normativity distinction serves as a corrective for this tendency, by reminding us that hedonism’s success turns on more than just a consensus account of aesthetic experience. Just as important is its answer to the normative question. To borrow Roger Crisp’s (2006: 622–23) terminology: aesthetic hedonism should be evaluated not just as an enumerative theory, that pinpoints which sorts of things have aesthetic value—namely the things that offer valuable aesthetic experience. It should be evaluated also as an explanatory theory, that singles out the fundamental good-making or reason-giving features of aesthetic goods—namely the pleasantness (or some other value-grounding feature) of the experiences they offer. As we shall see, hedonism’s critics have 18found various reasons to take issue with this answer to the normative question (see especially Sections 2.3.1, 2.3.2, and 2.3.4).2.2.2 Narrow versus preference hedonismA second distinction key to evaluating aesthetic hedonism concerns the conception of aesthetic pleasure or experience to which hedonists appeal. To get a handle on the distinction, first consider the case of theories of pleasure simpliciter. Philosophical accounts of pleasure typically fall into either of two camps (see Aydede 2014, Crisp 2006: 623–30, Sumner 1996: 87–91). On the one hand, ‘felt-quality’ or ‘internalist’ accounts take pleasures to be typified by some feature of their phenomenology: either a distinctive feeling that all pleasant experiences include, or an hedonic ‘tone’ they all share. In value theory beyond aesthetics, hedonists who rely on a felt-quality notion of pleasure are known as narrow hedonists (see Parfit 1984: 492). On the other hand, motivated by the vast diversity of the experiences we find pleasant (the so-called heterogeneity problem for theories of pleasure), ‘attitudinal’ or ‘externalist’ accounts deny that there is a single phenomenological feature common to all pleasures. Instead, they analyse pleasure in terms of some conative or evaluative pro-attitude that a subject holds towards one of their own ongoing experiences. Thus, on an attitudinal account, pleasures are simply experiences occurrently preferred, desired, liked, or valued in the right way (the details of the pertinent pro-attitude vary across different attitudinal accounts). In contrast to narrow hedonism, hedonist theories that take an attitudinal view of pleasure are considered instances of preference hedonism (ibid.).19The distinction matters in the current context because it also applies to accounts of aesthetic pleasure (or aesthetic value-grounding experience, if you prefer). Contemporary aesthetic hedonism is particularly well matched with an attitudinal conception of aesthetic pleasure. Why is this the case? The heterogeneity problem remains as much a challenge for theories of aesthetic pleasure as for theories of pleasure simpliciter, but there is a second, aesthetic reason for going attitudinal, namely the problem of painful art. Some paradigmatic aesthetic goods—some artworks in particular—owe their aesthetic value to features that make them unsettling, jarring, emotionally taxing, or even painful to experience. There are various ways of coming to grips with this datum (see Strohl 2019 for an overview of research on the phenomenon), but one common strategy involves broadening the class of experiences that can ground aesthetic value to include some experiences that lack any positive hedonic tone, or even ones with decidedly negative felt quality. This move to evade the problem of painful art is among the main reasons why some contemporary aesthetic hedonists style their theories as value empiricism or experientialism instead of hedonism. For value empiricists, what ends up mattering to the value-grounding role of aesthetic experiences is not any pleasant-making feature of their phenomenology, but rather that we find them ‘worthwhile’ (Levinson 1992: 296), or that we ‘value them for their own sake’ (Stecker 2006, Iseminger 2004, 2005b: 99, Budd 2008: 45–47). In other words, value empiricists conceive of aesthetic experience or pleasure in terms of an evaluative attitude, as experience preferred or finally valued rather than strictly pleasant in terms of feeling tone. They are attitudinal theorists about aesthetic experience and, by extension, aesthetic preference hedonists. This is significant because preference hedonists are vulnerable to what we might call the normativity objection (Section 2.3.1), and maybe some others, that narrow hedonists may safely ignore.202.2.3 Basic versus standardized hedonismThe third distinction to play a crucial role in the objections to hedonism is the distinction between basic and standardized hedonist theories. Aesthetic hedonism faces a version of the problem of taste: propensities for pleasure vary—what gives one appreciator great aesthetic pleasure leaves another cold and makes yet another queasy. So, when two people disagree about something’s aesthetic value, how should the hedonist resolve the dispute? The problem runs especially deep for response-dependent theories of aesthetic value like hedonism, because for them it is not just an epistemic matter of deciding whose experience accurately reflects the aesthetic state of the world. Aesthetic hedonists reduce aesthetic value to the value of experience, so the question for them is: whose experience fixes or constitutes the aesthetic state of the world?Hedonists have a range of options to respond to the problem of taste. The limit case, on the most relativist side of a spectrum of possible views, settles for basic hedonism. In effect, this position indexes all aesthetic values to an individual at a time, thereby denying the intersubjective reality and temporal stability of aesthetic value: every appreciator sets their own standard; there is no beauty except in the beholder’s eye; the customer is always right. This is less a solution than a rejection of the problem of taste, and not many in aesthetics have found it an attractive position, though some (such as Melchionne 2010, Kölbel 2016) have dabbled with views in the vicinity. A much more popular option lies near the other, universalist end of the spectrum. It holds that there is an hedonically ideal set of propensities for aesthetic pleasure to which all should aspire, and this sets the standard for resolving 21disputes about taste. The ideal has come to be expressed in terms of Humean ‘true judges’, in large part owing to Mary Mothersill (1989) and Jerrold Levinson’s (2002) influential hedonist interpretation of Hume’s (1757) solution to the problem of taste. On their reading of Hume, the true judges are idealized creatures whose sensibilities are perfectly calibrated for the maximization of aesthetic pleasure, such that their hypothetical joint verdict on matters of aesthetic value fixes the aesthetic facts.The timeless and universal standard of Humean true judges has been extremely influential and, as a result, the middle of the spectrum of possible hedonisms is almost as sparsely populated as the relativist end. An interesting recent exception is due to Mohan Matthen (2017, 2018), who tries to balance the variance in our hedonic responses with a sophisticated account of how those responses are malleable and subject to cultural learning. Matthen defines aesthetic pleasure functionally, as a mental state that plays the role of facilitating the continuation of effortful perceptual or cognitive engagement with items of aesthetic interest. Given this definition, our propensities for feeling aesthetic pleasure will be partly determined by our perceptual and cognitive competencies. As these competencies are acquired and developed in response to constraints imposed by a cultural and historical context, Matthen’s account indexes aesthetic values not to a universal standard like the true judges, but rather to a standard set by the culture in whose artistic practices they feature.Whether or not to standardize and, if so, how, are questions pivotal to aesthetic hedonism’s success. The game for hedonism’s critics is to show that basic and standardized hedonism are two equally unacceptable horns of a dilemma. This explains why the true judges model, serving as proxy for the standardized horn of the dilemma, has become a 22regular target for counterarguments against hedonism (see especially Sections 2.3.5 and 2.3.6).2.3  Objections to aesthetic hedonismWith the demarcation/normativity distinction drawn, a working concept of preference hedonism in place, and a generic picture of the Humean true judges in the background, the main arguments against aesthetic hedonism can more readily be articulated. This section identifies and reviews six such major arguments, or, more precisely, six argumentative strategies that have emerged from the literature (Sections 2.3.1–2.3.6). It concludes with some brief thoughts on how these arguments should be evaluated (Section 2.3.7).2.3.1 The normativity objection to preference hedonismThe first counterargument targets aesthetic preference hedonists (read: value empiricists) in particular, by asking whether an attitudinal account of aesthetic pleasure can do the explanatory work for which they enlist it. The argument is most clearly articulated by Shelley (2019: 6–9), who attributes the core insight to Frank Sibley (1965).  Recall first that 5answering the normative question requires giving not just an enumerative but also an explanatory theory of aesthetic value—a theory that singles out the fundamental good-making or reason-giving features of aesthetic goods (see Section 2.2.1). Here narrow  A generic form of the argument applies equally to preference hedonists in other normative domains. 5In the context of prudential hedonism, for example, Crisp (2007: 128–134) levels a version of the objection against Sidgwick’s (1907: 126) appeal to an attitudinal account of pleasure.23aesthetic hedonists have no problem: they can point to the pleasantness of aesthetic experiences as their good-making feature. But preference hedonists do not have an equally satisfying answer. Recall that, in order to evade the problem of painful art, they opt for an attitudinal account on which some unsettling, jarring, and even painful experiences might ground their objects’ aesthetic values, provided that these experiences are preferred or finally valued (see Section 2.2.2). The problem is that it is unclear which feature of this broader, attitudinally defined class of experiences can serve as a good-making or reason-giving feature. Shelley (2019: 9) illustrates the worry concisely:I don’t mean to be saying that Guernica’s capacity for affording shocking, unsettling, dizzying, and despairing experiences cannot figure in an explanation of its value […]. I mean to be saying that Guernica’s capacity to afford such experiences cannot bring to completion an explanation of value in the way that the capacity to afford pleasure can.For aesthetic preference hedonists, some aesthetic experiences are ‘pleasant’ only in the sense of being preferred or finally valued, and not in the sense of having any transparently valuable phenomenological feature like a positive feeling tone. Do such experiences confer value on their objects simply in virtue of being preferred or valued? If so, then the question becomes why are they preferred or valued? Preference hedonists cannot answer that they are valued for their pleasantness, because they deny that aesthetic experiences have any phenomenological feature that could do the work of pleasantness. And with any other answer they give, they risk explaining aesthetic values with reference to some non-hedonic good-making feature and, thereby, giving up the ambition of an explanatory hedonism. The 24preference hedonist might double down and simply insist that the good-making feature of such experiences is the very fact that they are preferred or valued by the true judges. But this answer is only acceptable if one is antecedently committed to the rationality of the true judges’ experiential preferences and valuings—a commitment that faces some serious challenges of its own (see Sections 2.3.5 and 2.3.6). If the objection sticks, then aesthetic preference hedonism (value empiricism) fails to offer a principled answer to the normative question and can thus be at best an enumerative theory of aesthetic value.2.3.2 Motivational arguments: the disinterest and pleasure-paradox objectionsAmong the main attractions of a hedonist theory of value in any normative domain is that it comes with a built-in mechanism for explaining normative motivation. If the reasons we have in a given domain are hedonic, then it is easy to see how we might come to be motivated to act on them, because pleasure is intrinsically motivating. I buy a pass to the Vancouver International Film Festival; assume that, in buying it, I act on my aesthetic reasons. The hedonist analysis of what aesthetically rationalizes my buying the pass—say, that many of the films on this year’s programme offer some great aesthetic delights—can generally double as an explanation of what motivates my buying it. The second strategy for arguing against aesthetic hedonism targets its reliance on this kind of hedonic explanation of aesthetic motivation. The strategy has been implemented in two ways in the literature.First, some have argued that an hedonic account of aesthetic motivation is at odds with the disinterested nature of aesthetic pleasure or experience. An historically influential answer to the demarcation question, traceable to Kant’s influence, identifies aesthetic values as those 25that elicit disinterested pleasure. Unlike the pleasures we take in things we like, want, or desire—what Kant calls ‘the agreeable’—aesthetic pleasure is thought to be an elevated mental state divorced from its subject’s preferences, desires, and projects. In short, on a strong construal of what disinterest entails, aesthetic pleasure is motivationally inert. Suppose we grant this way of demarcating the aesthetic. Then aesthetic pleasure cannot explain how we are motivated to pursue aesthetic goods in the way that pleasure can explain our motivation for acting on hedonic reasons more generally. Hence, aesthetic hedonism cannot rely on an hedonic account of aesthetic motivation, and this fatally undermines the view’s appeal. Call this the disinterest objection. Compressed versions of this objection are articulated by Edward Bullough (1907: 108–9) and R. A. Sharpe (2000: 331), (although Sharpe does not use the term ‘disinterest’, instead describing the elevated mental state at issue as ‘absorption’ occasioned by ‘serious interest’). As it stands, however, the disinterest objection is too strong. Few in aesthetics now accept the restrictive kind of disinterest requirement on which the objection is premised. Disinterest-based conceptions of aesthetic response have been facing increasing pushback in aesthetics. The reasons for this pushback vary, but they reflect a general concern with how a disinterest requirement separates the aesthetic too sharply from other aspects of our lives, such as our everyday practical concerns (Wolterstorff 2015) and our desires and deeply personal commitments (Nehamas 2007; Riggle 2016). And even for those who retain some form of disinterest requirement, it has become common practice to conceive it in weaker terms than the objection requires (see for example Levinson 1992: 298–99, Carlson & Parsons 2008: 24–30, 105–6).More compelling is a second, weaker argument also centred on pleasure’s role in aesthetic motivation. Rather than deny that aesthetic experience is the right kind of state to 26motivate rational aesthetic agency (as in the disinterest objection), this second objection hones in on cases of aesthetic agency that defy explanation by hedonic motives. The argument, which is an application of the paradox of hedonism, is developed in detail by Lopes (2018: 83–6), who notes that some (perhaps many!) aesthetic pleasures are ‘essential byproducts’ of activities motivated by non-hedonic considerations (see also Elster 1983: 77, Nguyen 2019). I join the local Sunday evening drum circle on Vancouver’s Spanish Banks; assume that I play as I have aesthetic reason to—my acute focus on coordinating with the group’s beat is responsive to aesthetic values in the performance. Say that my focus pays off: I get into a nice groove and the result is a hit of pleasure. Caught off guard, I turn my attention to the pleasure, to savour it, but doing so breaks my focus and I lose my rhythm and the pleasure along with it. The pleasure is an essential byproduct of my well-executed drumming activity—activity that only yields pleasure when it is not executed in direct pursuit of pleasure. Note that the premise is not that all aesthetic pleasures are like this, only that some are. The problem for aesthetic hedonism is that the aesthetic reasons such pleasures generate—reasons to act in a manner that produces them and, thus, in a manner unconcerned with attaining them—paradoxically precludes these pleasures from playing a motivational role in agents acting on such reasons. Differently put, in such cases, hedonism implies that it is impossible for an agent to be motivated by the aesthetic reasons they are acting on. As Lopes (2018: 86) admits, this is a bullet that aesthetic hedonists could in principle decide to bite, but not without giving up a major source of their theory’s appeal: ‘Why be so sure that aesthetic values stand in constitutive relation to pleasures as long as we no longer think of aesthetic agents as just those agents who are moved to seek pleasure?’272.3.3 The instrumentality and fungibility objectionsTradition has it that aesthetic value is non-instrumental and final—that its bearers are good for their own sake. Yet, on its face, aesthetic hedonism seems to imply that aesthetic value is purely instrumental: if an item’s aesthetic value reduces to the final value of some pleasure or experience it affords, then surely it bears that value as a means to the experience. So, either aesthetic hedonists must find a way to deny that their view has this implication, and the prospects for doing so look limited, or they must concede that tradition is misguided and aesthetic value is purely instrumental and, thus, not final. Call this the instrumentality objection to aesthetic hedonism.Why would it be bad to concede that aesthetic value is instrumental? At least two hedonists think that it need not be. Robert Stecker (1997: 254–6, 2006: 5) positively embraces the implication that hedonism makes aesthetic value instrumental. Nick Stang (2012) is more circumspect. He first argues that aesthetic value is not final,  but that this need 6not make it instrumental unless the final/instrumental value distinction is exhaustive. But even if the distinction does turn out to be exhaustive, and thus aesthetic hedonism does make aesthetic value instrumental, this need not be a drawback of the view, according to Stang, unless it has the further upshot that aesthetic goods are fungible. Stang’s inventory of the logical options thus shows that what ultimately matters is not whether aesthetic value is instrumental or final or neither, but whether aesthetic goods can simply be traded for more convenient instrumental means to similar or better experiences. Call this refinement of the instrumentality worry the fungibility objection. To state the fanciful philosopher’s version:  Stang’s arguments are formulated with respect to artistic value but apply just as well to aesthetic 6value. Nothing important here rests on the difference (see footnote 1).28aesthetic hedonism implies that, if we could take a designer drug or put on a VR headset providing exactly the same experience as that of engaging with some aesthetically great artwork, we would have identical aesthetic reasons to opt for the drug or headset as we would to travel to visit the museum. (It is worth noting that the fungibility objection is a close cousin of Robert Nozick’s (1974: 43) famous experience machine argument that many have taken to be fatal for hedonist theories of well-being. Taking the drug or putting on the VR headset is the aesthetic analogue of plugging into the experience machine.) For a less fanciful recent statement of the fungibility objection, see King (ms.).But aesthetic hedonists mostly agree on a response to the fungibility objection. They claim that aesthetic goods are not fungible, because the valuable experiences they offer cannot, as a matter of principle, be separated from their objects. By their very definition, these experiences are what they are, and have the value that they have, in virtue of being experiences of the items whose value they explain. In Stang’s (2012: 274) words:The experientialist can consistently maintain that artworks are essential constituents of the finally valuable experiences they afford. Experientialism is not committed to the fungibility of works of art.Others who implement versions of this response include Malcolm Budd (1985: 123–4), Levinson (1992: 304), Stephen Davies (1994: 315–16), and Alan Goldman (2006: 339). Whether we should accept it, however, is a matter on which hedonism’s critics come apart. Lopes (2018: 57) concedes the response and opts to look for hedonism’s shortcomings elsewhere, but Shelley (2010) is less convinced. As he sees it, once the hedonist makes the item whose aesthetic value they want to explain an essential constituent of the experience, 29they cannot avoid explaining the experience’s value in terms of value that the item has independently of the experience. Thus, he argues, aesthetic hedonists manage to evade the fungibility objection only at the cost of rendering their answer to the normative question viciously circular: they explain the value of an item by appeal to the value of an experience of it, and the value of the experience by appeal to the value of the item (Shelley 2010: 711, see also Shelley 2017: Section 2.4, and Watkins & Shelley 2012: 343–5).2.3.4 The under-articulation objectionIn the course of a programmatic attack on welfarist theories of value—that is, theories that aim to reduce all values to considerations of well-being—Susan Wolf (2015: 76) makes the following observation about comparative value judgements (her example is a comparison of the novels Middlemarch and The Da Vinci Code):The complexity of the novel’s structure, the quality of the prose, the depth and subtlety of the character development, the insights into civil society, all go into explaining why Middlemarch is a better novel. But why is it better for us to read a novel that is better in these ways?The problem Wolf is onto is that a theory of value must, at least in principle, be able to account for every value difference in the domain it aims to explain. If, for example, welfarism is true of a normative domain, then every difference pertinent to something’s value in that domain should show up as a corresponding difference in some hypothetical agent’s well-being. Part of the reason this is a challenging requirement in the aesthetic domain is that 30our aesthetic thought and discourse recognizes fine grained value differences across many dimensions of variation in items of aesthetic interest—differences, for example, in a novel’s complexity, depth, subtlety, insightfulness, etc. As Wolf points out, there do not always seem to be obvious corresponding differences in how the items in question are better or worse for anyone—in the sense of contributing to or detracting from their well-being. This leaves the welfarist unable to account for how such differences in the items under consideration could matter to their value.Although Wolf’s complaint targets welfarism about value in general, it can be recast in doubly restricted form as a challenge to hedonism about aesthetic value. Her choice of example makes clear that the worry applies to theories of aesthetic value as much as theories of other normative domains. And limiting the objection’s target to hedonism (as opposed to welfarism more generally) only increases its bite, because unlike welfarism, hedonic explanations of value can appeal only to differences in what is good for someone to experience, and not what is good for them in some other way. Thus, restricted to aesthetic hedonism, the objection may be parsed as follows: not every difference in an item’s aesthetic value recognized by our discourse and thought seems to show up as an independent difference in value-conferring properties of someone’s experience of the item. The aesthetic domain is densely articulated with value differences; by contrast, our experiences do not exemplify sufficiently fine grained, aptly ordered, and independent differences in value to map cleanly onto the aesthetic value differences they are meant to explain. Call this the under-articulation objection against aesthetic hedonism.Statements of the objection have tended to take a similar dialectical form to Wolf’s challenge to the welfarist: pointing to an aesthetic value difference and then demanding that 31the hedonist provide an explanation of that difference in terms of their theory. This makes the force of the argument apparent, as any explanation that the hedonist might attempt is likely to flirt with circularity. Here, for example, is David Davies (2001: 258–9) on a difference in his experience of a Turner painting upon coming to understand it:It is certainly conceivable that my ‘informed’ experience differs in certain respects from my relatively ‘uninformed’ experience. But surely this is because I am now aware of a value that the picture has. The difference in experience is to be explained in terms of a recognition of a value ascribable to the work. This value does not itself consist in the difference of experience […], but itself accounts for the difference in experience.And here is Shelley (2019: 8), making a version of the same point by noting thatthe property that purports to explain the intrinsic value of the experience to which the empiricist appeals must be a property of the experience itself and not merely a property of the object that affords the experience […]. Gracefulness is a value we experience a graceful dance as having, not a value of the experience that a graceful dance affords.The challenge for the hedonist is to point out specific valuable properties in experience without appealing to the object-attributed properties whose value they are meant to explain. The hedonist’s inability to do so seems to show that the only candidate experiential properties in the vicinity depend for their value on antecedent values of the objects experienced. Such properties cannot do the necessary explanatory work; experience lacks the 32articulation in value-conferring properties necessary for serving as explanans in a theory of aesthetic value.2.3.5 The overvaluation argumentAny feasible theory of aesthetic value should be able to account for mistaken attributions of aesthetic value. To that end, aesthetic hedonists standardly append a cognitive rider to their view: the experiences or pleasures that ground an item’s value should be rooted in a correct understanding of the item. This allows hedonists to chalk up mistaken value attributions to an appreciator’s defective grasp of an item’s nature or properties. A fifth line of argument against aesthetic hedonism asks how this cognitive rider can be squared with hedonism’s answer to the normative question. If pleasure is what ultimately matters, then why should the cognitive rider (or, for that matter, any non-hedonic constraint on which pleasures are the right ones) be allowed to get in the way of our taking the most or greatest pleasure possible from mediocre or even inferior aesthetic goods? Lopes (2018: 77–78) points out that this is the aesthetic analogue of a more general puzzle for hedonic theories of value: pure hedonism implies that we should alter or even delude ourselves so as to get maximum pleasure from almost everything. Goldman (2006: 336, 339) preemptively considers and rejects a summary version of this objection, but its most developed statement is again due to Shelley (2011).Shelley (2011: 215) first distinguishes two types of mistaken value attribution—overvaluations and undervaluations. Both are commonplace; fallible as we are, we routinely take great pleasure in some mediocre aesthetic goods and fail to take pleasure in goods of 33great aesthetic worth. But how, asks Shelley (2011: 216–17), is hedonism to capture the aesthetic harm in overvaluation? If aesthetic reasons are ultimately reasons to pursue pleasure or valuable experience, then hedonism has all the means necessary to explain the harm in our undervaluation of the excellent aesthetic goods that the Humean true judges prefer: when we undervalue, we fail to take pleasure in items we have aesthetic reason to enjoy. But by the same token, the true judges, by failing to overvalue the mediocre goods that we fallible aesthetic agents rate highly, fail to take pleasure in items that they have aesthetic reason to enjoy. After all, the hedonist answer to the normative question recommends taking all the aesthetic pleasure one can get. The problem is that this rationalizes the error of overvaluation and thereby contravenes the cognitive rider that hedonists rightly endorse. As Shelley (2011: 217) puts it, the hedonist ‘has one mechanism for explaining value and another for explaining mistaken value-attributions. If one is functioning, the other is not.’2.3.6 Against true judgesThe sixth line of attack on aesthetic hedonism consists in undermining the model of idealized appreciators on which the most sophisticated current forms of hedonism depend (see Section 2.3). To be sure, there are possible hedonisms that forego any appeal to ideal critics or Humean true judges, but none is as fully developed or widely influential as Hume-inspired, ideal-critic-centred hedonism. Subverting the ideal appreciator model would thus go a long way towards levelling the playing field between hedonism and its competitors.34Hedonism’s critics have raised three complaints against the ideal appreciator model. The first is epistemic. Matthew Kieran (2008: 280-83) argues that, given the delicate nature of the sensibilities that mark the true judges, there is no way for non-ideal appreciators—even ones who approximate the true judges’ sensibilities quite closely—to know which items they would recommend for appreciation. If Kieran is right, this implies that the true judges model is poorly suited to the task of providing ordinary appreciators guidance in the business of finding the best aesthetic goods. But this is only bad news for hedonism if providing such guidance is the point of the true judges model. Arguably, it is not. As the hedonist construes them, the true judges are an idealization for metaphysical rather than epistemic purposes. Nothing in principle prevents aesthetic hedonists from appealing to an idealization like the true judges to fix the aesthetic state of the world, while leaving the epistemic and practical challenge of navigating that world to real, fallible, flesh-and-blood critics and appreciators. A second complaint is stickier. It trades on the fact that the true judges, besides being an idealization, also represent an ideal: the ideal of possessing sensibilities that afford blanket access to the whole world of aesthetic value. In an exchange with Levinson (2010, 2013), who anticipates a version of the complaint, Nick Riggle (2013, 2015) takes issue with this picture of the aesthetic ideal. He argues that the project of cultivating the true judges’ generalist sensibilities in ourselves—which the hedonist ideal recommends—is at odds with the maintenance and cultivation of our meaningful personal attachments to particular aesthetic goods. For Riggle, this amounts to a reductio of the true judges model, given the obvious importance of personal aesthetic attachments to our aesthetic lives. (For alternative articulations of the objection, see Kieran 2008: 286–93, and Lopes 2018: 81–83. See also Cross 2017, Kubala 2018).35Finally, Lopes (2015, 2018) develops a third complaint against the true judges model by arguing that standardized hedonism lacks resources to explain what it should. Central to its explanatory failures is, once again, the model’s universalist or generalist scope. This time, however, its universalism is faulted not for posing a threat to our personal aesthetic commitments, but rather for its mismatch with the specificities of real world aesthetic action. Real world aesthetic agency (agency sensitive to aesthetic value) is deeply socially embedded, is specialized by aesthetic domain and activity, and draws on traits that are stable enough to be reliable across differences in context, but also flexible enough to adapt to novel situations (Lopes 2018: 25–31). By contrast, the expert agency modelled by the true judges floats free of social dependencies, is domain-general and inflexible, and is insufficiently grounded in specialized cognitive and practical skills (ibid.: 71–76). Thus, the true judges model is ill-suited to capturing the aesthetic doings of real world agents and, by extension, the values that shape those doings.2.3.7 Assessing the objectionsStepping back from the minutiae for a moment brings the overall threat the objections pose into clearer view. Each of the six lines of argument reviewed targets a different set of hedonist commitments, and each raises an independent challenge for those commitments (or multiple challenges, in the case of the true judges model). This seems to suggest that the objections are best assessed separately, by considering them one at a time. But an entirely piecemeal approach runs the risk missing ways in which the the objections reinforce each other at various junctures. We see such reinforcement, for example, when the preference 36hedonist responds to the normativity objection by locating aesthetic experiences’ good-making feature in their relation to the true judges’ experiential preferences, as opposed to their phenomenology (see the end of Section 2.3.1). While this move might appear to defuse the normativity objection, in effect it simply kicks the can down the road, to questions about whether the true judges model can carry the explanatory burden (Section 2.3.6). And we see it again, when hedonists try to head off the fungibility objection by making aesthetic goods essential constituents of the experiences that ground their value (end of Section 2.3.3). By making this move, they incur the burden of giving a non-circular account of the good-making or reason-giving features of experiences so constituted (a burden reminiscent of the one raised by the normativity objection).Thus, although the objections are logically independent, the threat they pose is more than merely cumulative. This should not come entirely as a surprise. Aesthetic hedonism is perhaps best thought of not as a single, self-contained theory, but as a theoretical framework, research programme, or paradigm. As such, it calls for evaluation on more holistic grounds than whether it can answer or accommodate individual objections considered in isolation.Thinking of hedonism in this way, as theoretical framework or research programme, foregrounds a significant difference among the objections. On the one hand, some of them—what we might call undermining objections—agitate from a perspective compatible with the hedonist programme’s basic presuppositions. Such objections take hedonism’s target explananda and starting assumptions as given, and function by showing how hedonist theories run into trouble by their own lights. Revolutionary objections, on the other hand, attack hedonism from an external perspective, contesting its starting assumptions and advocating for a shift in the common ground for theorizing about aesthetic value. Although 37the distinction is admittedly a blurry one, the first five lines of argument reviewed (Sections 3.1–3.5) can roughly be categorized as undermining objections. They work within the constraints the hedonist programme sets for itself, and their effective force is to compel dyed-in-the-wool hedonists to supplement, adjust, clarify, or give up some part of their theories. In principle, hedonists can continue to answer or deflect undermining arguments indefinitely, with countermoves or partial concessions designed especially for each objection. But such targeted replies usually come at a cost to a theory’s simplicity, parsimony, and explanatory power, and, as we saw, they may complicate the hedonist’s responses to other objections or introduce new vulnerabilities into the theory.By contrast, the objections against the true judges (Section 2.3.6) provide the clearest example of revolutionary impetus. Instead of just trying to prove hedonism wrong by its own lights, these criticisms strike at the core of the hedonist programme by challenging its methodological assumptions and its pre-theoretical construal of the target explananda. At the level of methodology, they suggest that aesthetic value theory should privilege non-ideal theorizing, aiming in the first instance to offer guidance to flesh-and-blood aesthetic agents navigating the messiness of aesthetic reality (see especially Kieran 2008: 280–83 but also Lopes 2018: 78–81; for a discussion of the ideal/non-ideal theory distinction, see Mills 2005). In addition, these criticisms contest the hedonist programme’s traditional approach of taking the nature and aptness conditions of aesthetic judgement as starting point for theorizing about aesthetic value (see Lopes 2018: 32–36). At the level of the hedonist programme’s construal of the target explananda, the criticisms of the true judges model contend that individual variance and idiosyncrasies in our aesthetic tastes an commitments (Riggle 2015), as well as specialization by aesthetic domain and activity (Lopes 2018), are 38central facts of aesthetic life that should be positively explained rather than merely accommodated by a theory of aesthetic value.Thus, by attacking not just the contents of hedonist theories, but also the hedonist’s methods and starting assumptions, the arguments against true judges raise a more fundamental challenge for aesthetic hedonism than the other five lines of argument reviewed. That is not to say that those arguments pose lesser problems, especially when taken together. But to respond to the arguments against true judges, hedonists must either defend their whole theoretical approach against alternatives, or show that proposed alternative approaches to aesthetic value theory still support a broadly hedonic answer to the normative question. Neither of these responses can be executed by merely supplementing or making minor changes to existing versions of hedonism, and, ultimately, both require reconsidering the hedonist programme as a whole in light of other options.2.4  Alternatives to aesthetic hedonismPhilosophical paradigm shifts seldom happen overnight, and although the objections reviewed pose considerable challenges, it would be overhasty to rule out future versions of hedonism capable of evading them all. In fact, there is room for good work within the hedonist paradigm with an eye to overcoming the objections. Such work is important especially where it amounts to more than ad hoc patches and tweaks to existing hedonist theories. In this regard, Matthen’s (2017) recent functionalist account of aesthetic pleasure, along with the theory of aesthetic value he builds on it (Matthen 2018), provides striking illustration of how one might blaze a fresh trail for the hedonist cause.39But there is also reason to worry that hedonism’s continued dominance in aesthetics unduly constrains our thinking about ancillary issues. The hedonist answer to the normative question has long been so prominent that alternatives have become difficult to envision. This has had the unfortunate upshot that the bulk of research on phenomena as various as aesthetic perception, judgement, appreciation, disagreement, testimony, personality, and motivation, to name but a few, often simply proceeds against the backdrop of hedonist assumptions. As Shelley (2019) puts it, hedonism is embedded in the ‘default settings’ for work on aesthetic value, and if the last section’s objections show anything, it is that this default status has not been earned. Thus, until answered, the objections serve to license and motivate work on non-hedonic theories of aesthetic value.Luckily, the search for alternatives is already gaining steam and the avenues to explore are many. It should come as no surprise that hedonism’s biggest critics are also at the forefront of this search. Enmeshed in his case against aesthetic hedonism, Lopes’s (2018) own network theory couches aesthetic normativity as a type of performance normativity, with the value of achievement displacing pleasure in the fundamental explanatory role. Central to Lopes’s network theory is the idea that aesthetic values are deeply embedded in social practices that serve as the ‘scaffolding’ for valuable aesthetic achievements. In as yet unpublished work, Shelley (ms.) develops the robustly realist account of aesthetic value that his published criticisms of hedonism have sometimes hinted at (see for example Shelley 2010: 715–20). His account—which we might call the Auburn view of aesthetic value (see also Gorodeisky 2019: §6, Watkins & Shelley 2012: 349–50)—makes aesthetic value into a normative primitive that cannot be analysed in terms of further normative concepts. The 40thought is that aesthetic value’s normative status is on a par with the basic normative status of truth in the epistemic domain.Other recent proposals for non-hedonic theories have been floated in skeletal form and await fleshing out. Peter Goldie (2007, 2008) and Matthew Kieran (2009) both advocate for an approach on which aesthetic normativity is cashed out on the model of a virtue-based conception of aesthetic character. Thi Nguyen (2019) gestures towards a practice-centred view in which the norms of aesthetic practices are contingently constructed around the value of active and effortful engagement with aesthetic goods, while the final value of such engagement accrues in turn from its expression of individual autonomy. Riggle (2015: 444–47) sketches the outlines of an account in which the hedonic ideal of the Humean true judges is supplanted by the ideal of having style. The normativity of this ideal, as Riggle understands it, is grounded in style’s role in building and maintaining ‘communities of individuals’ (see Riggle 2017).Leaning on Parfit’s (1984) taxonomy of three major types of theories of value in the prudential domain, Robbie Kubala (2019: 261) has suggested in passing that, instead of hedonism or an objective list theory, a desire-satisfaction theory of normativity might be especially well suited to aesthetic value. Having made the suggestion, however, Kubala goes on to ask whether we really have to choose: could we not perhaps accept a pluralist account on which the normativity of aesthetic value is sourced, on different occasions, in different kinds of basic goods, whether those goods are experiential or objective or constitutively tied to our desires? In recent work, Antonia Peacocke (ms.a) has also endorsed a kind of pluralism about the normativity of aesthetic value (or ‘liberalism’, in her terminology). Peacocke grants a version of the hedonist answer to the demarcation question, claiming that 41aesthetic value must be tied to the value of perceptual experiences in some important sense. But, unlike the hedonist, she denies that it must in all cases be grounded in the intrinsic or final value of those experiences. Instead, she allows cases in which aesthetic values are grounded in the instrumental value of certain experiences, such that they derive their normativity from non-hedonic final goods (such as, for example, special kinds of insight or understanding) to which the experiences are essential or ineliminable means. In other words, while Peacocke’s account shares much with hedonism, it permits the normativity of aesthetic values to be sourced in goods beyond the aesthetic experiences that give us access to those values.The call to look beyond aesthetic hedonism also provides impetus to re-examine historical figures and attend to theorists outside the mainstream of the Euro tradition whose work on aesthetic value does not fit cleanly within the hedonist paradigm. Samantha Matherne and Nick Riggle (ms.), for example, revisit Schiller’s aesthetics to extract a broadly ‘communitarian’ view that resonates with Riggle’s own and casts aesthetic value as primarily a social good. Lopes (2019) provides an example from South Asian aesthetics, by reading K.C. Bhattacharyya’s (2011) rasa theory as a kind of hybrid view that answers the demarcation question in terms of aesthetic pleasure, but departs from hedonism by answering the normative question by appeal to the value of the freedom characteristic of such pleasure. Julianne Chung (2018, 2019) considers the picture of aesthetic value that emerges from a family of views in East Asian philosophy—views that emphasize the deep interconnectedness of individuals with others and with their natural surroundings. On this picture, the normativity of (some) aesthetic values is grounded in their capacity to engender 42awareness of how we are connected with aesthetic objects and, by extension, with things in general.Once the outlines of some alternative answers to the normative question come into view, hedonism’s hegemony in aesthetics starts to look less inevitable, perhaps even surprising. Why is it that, in philosophical work on other normative domains, hedonic or experiential theories represent just one among several major theoretical strands, but in contemporary aesthetics, they have largely monopolized the field? Where are the aesthetic perfectionists, the desire-satisfaction and objective list theorists about aesthetic value? For a long time their absence from the field seemed to indicate the implausibility of any theory other than hedonism. It is fast becoming clear that this appearance was misleading: viable non-hedonic theories were not impossible, they had just not been articulated yet. With hedonism’s dominance a little less secure, we might well be entering an unusually auspicious time for new work on aesthetic value.43CHAPTER THREE   The Motivational Structure of Appreciation3.1  Disinterest as mode of attentionAccording to a contemporary version of the disinterest thesis in aesthetics, appreciation requires disinterested or non-instrumentally motivated attention. That is, to engage with an item appreciatively, one must attend to it ‘without ulterior purpose’, ‘for its own sake’, or ‘for the sake of the very experience of attending to it’ (see for example Stolnitz 1960; Iseminger 2005b; Stecker 2006; Levinson 2016). The thought that disinterest qualifies a way of attending instead of, as Kant (1790/2000) had it, a type of pleasure or judgment, originated with the aesthetic attitude theorists. Schopenhauer (1819/2010: 231) provides an early articulation when he locates the value of the aesthetic attitude in how it ‘lifts us out of the endless stream of willing, tearing cognition from its slavery to the will’, such that ‘attention is no longer directed to the motives of willing.’ Later aesthetic attitude theorists took up this thread. For Bullough (1912: 89), for example, aesthetic appreciation involves ‘putting the phenomenon, so to speak, out of gear with our practical, actual self’ and ‘allowing it to stand outside the context of our personal needs and ends’. For Stolnitz (1960: 32–36), the aesthetic attitude requires both disinterest, which he understood to mean a lack of any motive other than just to attend, and what he called ‘sympathy’, which entails not imposing one’s own preconceptions on the object of attention.44Aesthetic attitude theory’s heyday around the mid-twentieth century was short lived, however, and many now credit George Dickie’s (1964) paper, ‘The Myth of the Aesthetic Attitude’, with sounding its death knell. But despite Dickie’s role in curtailing the attitude theorists’ influence, his attack on the very existence of a disinterested mode of attention has failed to engender consensus. A chorus of dissenting voices, several of them recent (Lind 1980; Zangwill 1992; Kemp 1999; Shelley 2017; Levinson 2016: 28–31; Nanay 2016: 20–21; Matthen 2017, 2018), has helped the notion of an appreciative mode of attention, marked by its own peculiar motivational profile, to outlive its theoretical beginnings.Dickie’s critics are right to point out that his attack on the existence of a disinterested mode of attention was inconclusive, but pointing this out falls short of meeting the challenge his arguments pose. In particular, there is still no informative positive analysis in the literature of what ‘attending for its own sake’ amounts to, capable of bearing the explanatory weight routinely thrust upon the notion. The current chapter takes up this challenge. It does so by revisiting Dickie’s arguments and extracting from them an overlooked desideratum for accounts that seek to understand appreciation in terms of a distinctive motivational profile. Once this desideratum is clearly articulated (in Section 3.2), the aim will be to develop a model of appreciative motivation that meets it (Sections 3.3–3.5).Two caveats are in order at the outset. What follows should not be construed as an attempt at a complete theory of appreciation, and even less so a theory of aesthetic appreciation. First, even if we were to grant the attitude theorists that a non-instrumental mode of attention is necessary for engaging in appreciation, this does not yet imply that it would be sufficient. Hence, the account this chapter develops will provide a model only of the appreciative mode of attention as put forward by the aesthetic attitude theorists. Some 45hints and suggestions will be offered about how the mental kind picked out by the account should feature in a complete picture of the set of cognitive phenomena that properly belong under the label of the appreciative. Implicit in the background, however, is the assumption that this mental kind is indeed central to such a picture, but the task of providing the full picture is beyond this chapter—and indeed, this dissertation’s—remit.7Second, even if we find, contra Dickie, that something in the vicinity of a non-instrumental mode of attention is a real and commonplace mental phenomenon, why think that its occurrence would track the aesthetic domain? Attending to an item non-instrumentally, whatever we take that to entail, plausibly (or likely!) comes apart from ascribing aesthetic value to the item, and from attending to all or any of its aesthetic features. In other words, the approach here will be to agree with Dickie by not assuming, as the aesthetic attitude theorists arguably did, that an account of the appreciative mode of attention could furnish an easy answer to the question of how the domain of the aesthetic is to be demarcated from other domains. In keeping with the methodological assumptions set our in the introduction (see Section 1.2), the account developed in this chapter intentionally leaves open—indeed, it invites the conclusion—that there is a double dissociation between the appreciative mode of attention on the one hand, and aesthetic cognition on the other. We can engage appreciatively with non-aesthetic features of the world, and we can experientially register aesthetic values and properties without thereby realizing the motivational profile underlying the appreciative mode of attention.  There is of course also a rich history of positive accounts of appreciation that do not invoke a 7distinctive non-instrumental mode of attention at all. These are not targets of this chapter’s critical arguments.46Despite these caveats, and notwithstanding appearances to the contrary, this chapter’s aims retain a deep continuity with the aesthetic attitude theorists’ views. Broadly speaking, they believed that there is a distinctive mode or manner of attending that is pivotal to our aesthetic lives, that this mode supervenes on how our attention is motivated, and that it is possible to give a general account of the motivational profile that grounds it. What follows agrees with these commitments, at least in this broad formulation.But where the attitude theorists construed appreciative attention in terms of a subject’s ascription of final (that is, non-instrumental) value to the object of their attention, or to their own attentional engagement with it, the proposal in this chapter will be that no such final valuing is required. Instead, the sense in which appreciative attention is non-instrumentally motivated resides in how the appreciating subject selects cognitive goals not merely for the sake of the broader ends those goals’ attainment serves, but rather for the sake of ends served by the process of being engaged in the cognitive activity that results from those goals’ pursuit. Unlike in everyday cognition, the motivational relationship of means to ends is inverted in the appreciating subject’s cognitive engagement with the world. This proposal will be thoroughly unpacked below, but the argument starts with Dickie’s challenge to the attitude theorists.3.2  Dickie’s challengeCrucial to the aesthetic attitude theorists’ project is the feasibility of distinguishing in a principled way between disinterested (or non-instrumental) and interested (or instrumental) modes of attention. The gist of Dickie’s complaint is that they conflate this distinction, 47between different species of attention, with a distinction between different kinds of motives for which one may attend. Take one of Dickie’s examples: Smith and Jones both listen to a piece of music, Jones with the aim of remembering its details for an exam the next day, and Smith with no such ulterior purpose. The thought is that, on the aesthetic attitude theorists’ criterion, Jones’s practical interest in the exam would get in the way of his simply savouring the music. Smith, on the other hand, can listen unencumbered and, provided he attends closely, his attention can rise to the level of aesthetic appreciation. Of this example Dickie (1964: 58) writes that ‘what initially appears to be a perceptual distinction—listening in a certain way (interestedly or disinterestedly)—turns out to be a motivational or intentional distinction—listening for or with a certain purpose.’ Disparate motives for attending, he argues, are not sufficient for disparate modes of attention: ‘There is only one way to listen to (to attend to) music, although the listening may be more or less attentive and there may be a variety of motives, intentions, and reasons for doing so’ (ibid.).Dickie’s argument for this conclusion consists of a series of intuition pumps, each involving a subject whose attention would, like that of Jones, putatively count as interested on the attitude theorists’ criterion. In each case, Dickie proclaims, there is either no difference between the subject’s attention and that of a disinterested subject at all, or, if there is a difference, it is in what they are attending to rather than the manner of their attention. The impresario at the theatre, taking pleasure in the size of the audience, for example, is not attending to the play interestedly, but is attending to the till instead (ibid.: 58–9). The reader approaching a poem solely for diagnostic evidence of the author’s neuroses, is at worst a case of attention to the author instead of the poem, and at best, a case of partial attention to the poem (ibid.: 60). Thus, the argument goes, the distinction the attitude theorists are tracking, 48in so far as they are tracking any distinction at all, is not one between disinterested and interested modes of attention, but rather one between attention and distraction.Whether we buy this line of argument against the existence of a disinterested mode of attention depends on whether we share Dickie’s intuitions in the pertinent cases, about who is attending to what, and how. The problem is that questions about who is attending to what are not always straightforward. In work on the nature of attention, Mole (2011) makes this point by emphasizing that questions about an agent’s attention are connected to questions about which perceptual, cognitive, and practical tasks they are engaged in. In laboratory conditions, honing in on which task a group of experimental subjects is engaged in is a matter of experimental design. But the same precision is not available in thought experiments involving agents in ordinary circumstances:Outside of the lab an agent’s tasks will tend to be more numerous, complex, nested, and overlapping. As a result the question ‘What task is this agent performing?’, when asked outside the lab, may admit of many answers, some of which are only vaguely true. […] [A]nswers to the question ‘To what is this agent attending?’ will inherit some of this vagueness. (Mole 2011: 52)In as much as Dickie’s arguments depend, then, on clear verdicts about what the subjects in his hypothetical cases are attending to, there is room for disagreement. And if vagueness besets our intuitions concerning what an agent in ordinary circumstances is attending to, we at least need an argument for thinking that our intuitions about whether two agents are attending in the same way are any more probative. Dickie’s critics have not been remiss in pointing this out. About the Smith and Jones example, Shelley (2017: §2.3) writes:49The contention that Jones and Smith are attending in the same way appears to be question-begging, as it evidently depends on a principle of individuation that the attitude theorist rejects: if Jones’s attention is governed by some ulterior purpose and Smith’s is not, and we individuate attention according to the purpose that governs it, their attention is not the same.To accept Dickie’s argument, in other words, we must first accept his contention that the mode of attention exemplified in an attentional episode is not determined by the purpose that motivates the attending that occurs in that episode. But this is precisely what is under contention in his dispute with the attitude theorists. Besides his discussion of the sample of cases, Dickie offers no argument for his restrictive standard for differentiating between attentional episodes.What arguments could he offer? To answer this question, we need an account of the relation between an attentional episode, and the purpose that motivates it. We can understand this relation in one of two ways, each corresponding to a major conception of how purposeful behaviours in general relate to their goals. Davidson (1963) famously holds that what distinguishes purposeful behaviours (actions) from mere bodily movements is that they are caused in the right way by their agents’ beliefs and desires. Frankfurt (1978) criticizes Davidson’s belief-desire account of action. On the alternative he proposes, calling a behaviour purposeful entails nothing about how it is initiated or caused, but rather that it is guided or kept on track by its purpose as it unfolds.Much has been written on this dispute in action theory and, in retrospect, it seems clear that one could make a Davidsonian account of action compatible with Frankfurt’s by 50incorporating a more sophisticated notion of causation than the one Frankfurt’s criticisms assume (see Setiya 2003: 348, n.11). But the point here is not to take sides or resolve the dispute; it is to highlight the difference in emphasis between the two views—one on causation, the other on guidance. We can use the contrast between the two views to articulate two different ways an attentional episode could be understood to relate to its goal or purpose. On a Davidsonian picture, to attend for a particular purpose would be for one’s attending to have been caused by the appropriate beliefs and desires regarding that purpose. On a Frankfurtian picture, it would be for one’s attention to be guided by that purpose as it unfolds.8Return now to Dickie’s standard for individuating modes of attention. His insistence that they cannot be differentiated by the purposes that motivate an attentional episode makes perfect sense on the Davidsonian picture. If motivation concerns just how an attentional episode is initiated or caused (in the sense of a triggering cause), then one episode can be intrinsically identical—that is, identical in contents to which the subject attends—to another episode with very different motivation (read: causal antecedents). For two differently motivated episodes to differ in mode, we need to understand their motivation on the Frankfurtian picture instead, in terms of what guides the course of an episode as it unfolds. If motivation is a matter of guidance, then a difference in type of motivation can yield a difference in the mode of attention an episode instantiates.Just as an instance of purposeful physical behaviour may be identical in all its intrinsic properties to a set of mere bodily movements not guided by the agent, so two attentional  The idea of applying a Frankfurtian notion of guidance to episodes of attention to capture common 8sense distinctions between mental kinds is due to Irving (2016: 563–4), who uses it in an account of the nature of mind-wandering.51episodes may in principle be identical in all of their contents while only one has the contents it does because of the right kind of guidance. As in the former case, of purposeful behaviour versus mere bodily movement, we surely want to maintain the distinction between types of attentional episodes that are differently guided. Thus the notion of a distinctive way of attending, corresponding to a distinctive motivation—or, more specifically, a distinctive kind of attentional guidance—remains a live possibility despite Dickie’s insistence to the contrary.None of this is to say that Dickie’s arguments are inconsequential. In fact, once understood in this way, they pose a challenge to any feasible analysis of the appreciative motivational profile. They show that such an analysis will have to specify the distinctive motivation of appreciative episodes in the Frankfurtian guidance sense. It will not do to specify a characteristic type of reason for which appreciators attend to the objects of appreciation. Rather, what is required is a characteristic type of purpose by which, or manner in which, attention to an object is guided as appreciative episodes unfold over time. Appreciative motivation, if there is such a thing, should help explain why appreciative episodes proceed as they do, rather than merely why they are engaged in at all. Until now Dickie’s critics have failed to recognizes this challenge, let alone attempted to meet it.9One reason the challenge has some real traction is that while the standard characterizations of appreciative motivation—using locutions like attending ‘disinterestedly’, ‘non-instrumentally’, or ‘without ulterior purpose’—admit of quite natural Davidsonian  A possible exception is Matthen (2017, 2018), whose account of appreciative (or aesthetic) 9motivation does imply a link between an attentional episode’s motivation and how it unfolds. There is much to like about Matthen’s account, but the account developed in this paper differs from his in several respects: it gives a non-hedonic reading of appreciative motivation; it shifts the focus to guidance implemented by deliberate attentional constraints (see Section 3.5); and, unlike Matthen’s view, it does not tie appreciative motivation to fluent cognitive processing (Matthen 2018: 24).52readings, they do not admit of equally obvious Frankfurtian readings. This is in part because they are contrastive rather than positive, prohibiting certain kinds of motivations or purposes (interested, instrumental, and ulterior ones).  But a Frankfurtian account of the motivational 10profile these locutions pick out should specify how attention is guided; settling on which goals or purposes it may not be guided by, does not yet much constrain how attention must be guided to count as appreciative.The challenge, then, is to give a positive account of appreciative motivation which could help explain why cognition unfolds as it does in appreciation. The rest of this chapter develops such an account via an analogy with a practical activity that we often do engage in ‘for its own sake’: playing games. The next section introduces the notion of striving play, a special form of game play that has been the focus of recent work in the philosophy of games (Section 3.3). Looking further ahead, the plan will be to show that striving play amounts to a genuinely distinctive practical mode of engaging in games (Section 3.4) and then to argue that striving play’s practical motivational structure is isomorphic to the cognitive motivational structure by which attention is guided in episodes of appreciative engagement (Section 3.5).3.3  Striving playIn what is by now a classic work in the philosophy of games, Bernard Suits (1978: 55) defies Wittgenstein’s famous dictum on the impossibility of giving necessary and sufficient  Note that the formulation ‘attending for its own sake’ is also naturally read contrastively: as ruling 10out cases of attending for something else’s sake. That is, to attend for its own sake is to attend because one attributes final (rather than instrumental) value to doing so.53conditions for the concept ‘game’, and offers a definition of game play as ‘the voluntary attempt to overcome unnecessary obstacles’. The point of our efforts in playing a game is not just to achieve the goal the game prescribes, but to achieve it under prescribed limitations. Suits expresses the thought in terms of a distinction between a game’s lusory and prelusory goals (ibid.: 36–7). The in-game (lusory) goal of a bicycle race is not just to reach the finishing line first (the prelusory goal), but to reach it using only the permitted means (in this case, cycling along a prescribed route). Note that the prelusory goal can be specified independently of the game’s existence, but the lusory goal owes its existence to the game as constituted by the combination of the prelusory goal and the pertinent restrictions on means. On Suits’s definition, all games have this structure, of pursuit of a lusory goal that is constituted by the combination of a prelusory goal and rules restricting efficient means.One may doubt the definition’s extensional adequacy, but the model of game play Suits develops in the process of defending it generates illuminating descriptions of many game-related phenomena. In recent work inspired by Suits, Nguyen (2020: ch. 1) takes advantage of this fact by suggesting that the set of games that satisfy Suits’s definition—call them ‘Suitsian games’—can be approached as an independently interesting topic of inquiry, whether or not it coincides with the set that ordinary usage categorizes under ‘games’. For Nguyen (ibid.), the ‘most interesting possibility raised by the Suitsian analysis’ is a form of engagement in Suitsian games he calls striving play.543.3.1 Two distinctions in game playTo distinguish striving play from other forms of play, consider the diversity of reasons for which one might engage in a Suitsian game. Sometimes we play sudoku just for the sake of distracting ourselves on the morning commute. Wagering games and professional sports show that often money and social prestige figure into players’ reasons for playing a game. We might play a sport for the sake of our physical health. Or perhaps the feeling of winning is all the reward we need. In fact, we can be maximally pluralist about potential reasons for engaging in games. Taking up a game’s prelusory goal (the game-independent state to be achieved) and constitutive rules (the restrictions on means) is done for the same proximal reason across all cases: to make possible the activity of playing the game. But, as constituted, a game’s (in-game) lusory goal can be pursued for any number of reasons.Nguyen (ibid.: ch. 3) argues that these reasons vary along two dimensions. Firstly he distinguishes between intrinsic and extrinsic play, a distinction pertaining to whether a player ultimately engages in the game for goods internal or external to it. Secondly, both intrinsic and extrinsic players may differ with regard to the locus within game play to which the values they pursue, adhere. Nguyen dubs this the distinction between achievement and striving play: a player may care either for goods associated with the achieving of the game’s lusory goal (winning) or for goods associated with the activity or process of attempting to achieve it. The two distinctions crosscut, yielding four types of play.55Figure	3.3:	Four	types	of	game	play	Intrinsic achievement players care non-instrumentally for winning; they are pure competitors. Extrinsic achievement players care for the instrumental benefits of winning, like money or social prestige. Extrinsic striving players care for game-independent benefits that result from being engaged in the in-game activity, such as the fitness or health that result from playing sport, cognitive skills acquired by playing chess, or the distraction to be had by working at the crossword. Finally, intrinsic striving players care non-instrumentally for engagement in the in-game activity.11Now consider how Nguyen’s two motivational distinctions may be used to give two different readings of what it means to play a game for its own sake. On one reading, to play a   Nguyen (ibid.) rightly notes that these categories are not exclusive. In real world cases our reasons 11for engaging in games are typically mixed.56game for its own sake is to engage in intrinsic play: to play for the sake of goods internal to the game. The distinction between extrinsic and intrinsic play presupposes a working conception of what it means to participate in a game for non-instrumental reasons. Elsewhere in value theory this kind of non-instrumental motivation is couched in the terminology of final value (Korsgaard 1983, Kagan 1998). Intrinsic players ascribe final value to the game, whereas extrinsic players value it instrumentally. As final value is typically equated with value something has ‘for its own sake’, this makes it natural to understand intrinsic play as the paradigm case of game play for its own sake.But notice that whether we are intrinsic or extrinsic players of a game—whether or not we attribute final value to it—is a matter of our motives for playing the game at all. The distinction pertains to why we take up the prelusory goal and rules of the game voluntarily, to whether we do so in pursuit of goods internal or external to the game we thereby constitute. Thus, the explanatory purchase that the distinction between intrinsic and extrinsic play gives us, is limited to our actions with respect to the game, as opposed to our actions within the game. It answers questions such as why we play this game rather than that one, or why we spend as much time on a certain game as we do, but not questions about why our in-game pursuit of the lusory goal proceeds as it does.What the discussion of Dickie’s challenge in the last section showed is that an account of appreciative motivation should provide explanatory purchase not just on why appreciative episodes are initiated (how they are caused), but on why they unfold as they do (Section 3.2). We want an account of appreciative motivation in the Frankfurtian sense. But an analogy with intrinsic play would be ill suited to delivering such explanation. The notion of intrinsic play gives us only a Davidsonian account of play for its own sake—it explains why we play 57the games that we do rather than why our playing goes as it does. This chapter’s central hypothesis is that the sense in which appreciative attention is non-instrumentally motivated, is better captured by an analogy with striving play than intrinsic play. Ascribing final value to an episode of cognitive engagement (in the way the intrinsic player ascribes final value to the game) turns out to be neither necessary nor sufficient for realizing the appreciative motivational profile.3.3.2 Striving play as play for its own sakePart of what makes striving play interesting is that it demands a kind of reversal of our ordinary means-ends engagement in practical life. Nguyen (2020) calls this a ‘motivational inversion’, but the motivational structure underlying striving play is perhaps better understood as involving a motivational feedback loop. Suitsian game play in general (including both striving and achievement play) already involves a kind of inversion of means and ends. To constitute a game, one must take up the prelusory goal as an instrument. In a bicycle race, being at the finishing line is worth pursuing only in as much as pursuing it helps to bring about the game. If being at the finishing line (the prelusory goal) were too highly valued independently of the race itself (the restricted pursuit of the prelusory goal), then we would take the shortcut by taxi instead of cycling the long way around. The prelusory goal is thus, paradoxically, an ‘end’ that a Suitsian game player takes up as a means—it is an instrument used to constitute the game.But not all participation in Suitsian games is striving play. The achievement player’s reasons for taking up the prelusory goal (and thereby constituting the game) involve the 58potential benefits of winning, whereas the striving player’s reasons involve benefits that accrue from being engaged in the in-game activity, independently of success within the game. This does not mean that she can be entirely indifferent to winning, because the activity she cares to be engaged in just is the sincere pursuit of the game’s lusory goal. Like the achievement player, she must be submerged in the game to the extent that the lusory goal takes on motivational salience prompting real attempts to achieve it (more on this ‘submersion requirement’ in the next section). But her overall motivational relation to the lusory goal differs from the achievement player’s. Whereas the achievement player’s pursuit of the lusory goal is nested in a broader, game-independent interest in maximizing achievement, the striving player’s pursuit of the lusory goal is nested in a broader, game-independent interest in activities like the one constituted by the very attempt to achieve the game’s lusory goal.This, then, is the sense in which the striving player engages in a game non-instrumentally: it is not that she ascribes final value to either the in-game activity or in-game achievement (like the intrinsic player), but rather that her interest in achievement within the game is motivationally grounded in a more fundamental interest in being engaged in the in-game activity constituted by the pursuit of that achievement. The pursuit of achievement thus has a role both as a means and as an end in the striving player’s global motivational set. This yields a motivational feedback loop. The in-game activity can still be characterized in perfectly vanilla means-ends instrumental terms, as activity in service of winning, but the attempt to win, to achieve the lusory goal, can in turn be understood as a means to the further, game-independent end of being engaged in the in-game activity of its pursuit. In a sense then, 59the striving player’s in-game activity, while proximally aimed at winning, is ultimately activity enacted in its own pursuit (or, if you will, for its own sake!).Crucially, this feedback loop conception of the non-instrumental motivation that marks striving play (if ‘non-instrumental’ is indeed an apt term for it) differs from the non-instrumental motivation of intrinsic play. Intrinsic play requires that the in-game goods for which a player plays are final goods. No further rational grounds external to the game may be invoked to explain why an intrinsic player pursues either the in-game activity or the in-game achievement. But striving players can be intrinsic or extrinsic players. Their reasons for pursuing a particular game’s in-game activity may involve valuing that activity intrinsically, but it may also be thoroughly integrated with the rest of their practical projects outside of the game. As long as their interest in winning is motivationally less fundamental than their interest in the in-game activity that is the pursuit of winning, nothing prevents that latter interest from being further grounded in some other, more fundamental global goals (like getting fit, escaping boredom, gaining skills, etc.).Nguyen’s account of striving play thus illuminates an interesting sense of engaging in an activity ‘for its own sake’—one which allows but does not require ascribing final value to the activity. Moreover, nothing in his account is inimical to the possibility of the same kind of motivational inversion or feedback loop involved in striving play occurring outside of the artificial practical context of Suitsian game play (for instance, in everyday cognitive activities picked out at a more basic level). The question that concerns us now is whether the motivational model provided by striving play could perhaps be better suited to the current aim of meeting Dickie’s challenge than the model provided by intrinsic play. Plausibly, unlike intrinsic play, striving play is not just a matter of what explains our actions with 60respect to the game, but also of how our in-game pursuit of the lusory goal is guided. Hence, it holds promise as a model for how attention is guided in appreciation. But there is a complication: the submersion requirement on Suitsian game play puts pressure on the idea that the striving player’s interest in striving makes a difference to how her in-game actions are guided.3.4  Qualifying the submersion requirementTo be engaged in a game—to have accepted its prelusory goal and constitutive rules—is what Suits (1971: 38–41, 142–146) refers to as being in the lusory attitude. Sustaining the lusory attitude is required of achievement and striving players alike. Nguyen (2020: ch. 1) offers a strong reading of what this entails:To play a game is to behave, during the game, as if the pre-lusory goal were a final end. To be gripped by the game, to be absorbed by it, we must be able to enter the phenomenal state of holding the pre-lusory goal as a final end.It is not that the prelusory goal should actually figure as a final end in the player’s global motivational set, but that she must temporarily induce in herself a state in which she is practically motivated by it as if it were, so that the activity that constitutes the game can emerge. Let us stick with the terminology Nguyen introduces for entering this phenomenal state—being ‘submerged’ in the practical perspective the game prescribes—and call this the submersion requirement on Suitsian game play.61The submersion requirement arguably imposes a greater demand on the striving player than the achievement player. When an achievement player enters a game, the prelusory goal presumably takes on and retains motivational salience for them directly and automatically, as means to the achievement goods they are after. Parsed in Nguyen’s terminology: the prelusory goal automatically takes on the guise of a final end with respect to an achievement player’s in-game actions. By contrast, the goods that the striving player is after attend the lusory goal’s pursuit regardless of whether that pursuit succeeds. The problem is that a complete lack of interest in goods attending the lusory goal’s successful achievement threatens to undermine the normative traction of the prelusory goal on the player’s actions.Nguyen (ibid.) illustrates the problem with an anecdote of a child playing Monopoly with his father and enjoying each time he wins game money so much that, whenever he approaches victory, he hands his father some of his own game money in order to extend the game. Although a parent might humour their child in this situation, there is an important sense in which they have then given up playing Monopoly. This is an instance of a more general predicament—call it the Monopoly trap—in which a player’s interest in striving undermines their wholehearted pursuit of the prelusory goal.The Monopoly trap is Nguyen’s reason for construing the lusory attitude as a comprehensive submersion in the practical perspective the game prescribes. On his strong construal, submersion, for the striving player, requires completely disengaging from the motivational pull of her broader interest in striving, even while her ultimate reason for doing so just is to find herself engaged in striving. Notice, however, that this rules out any principled difference between how the achievement and striving players’ in-game actions are guided. What differentiates them—their respective interests in achievement and striving 62goods—is relegated by a strong submersion requirement to figuring only in their reasons for entering and exiting the lusory attitude, and hence, to explaining their actions with respect to rather than within the game. But this would make striving play, like intrinsic play, an unsuitable model for building a Frankfurtian account of the appreciative motivational profile in terms of how attention is guided in appreciative episodes.The crucial question, then, is whether the Monopoly trap really necessitates such a strong construal of the submersion requirement. There is reason to think that it does not. To see why, first consider achievement play. The achievement player faces no Monopoly trap, no tension between their pursuit of achievement goods on the one hand, and their experience of the game’s prelusory goal as motivationally akin to a final end, on the other. The hazard for the achievement player is in fact the converse of the Monopoly trap; it is that their interest in achievement might undermine the normative traction not of the prelusory goal, but of the game’s restrictions on permitted means. In this vicinity lies the temptation to cheat: to pursue the game’s prelusory goal by flouting the game’s constitutive rules.The hazard of cheating does not, however, give us reason to construe the submersion requirement strongly, at least not as a constitutive condition on achievement play. As a matter of course, we do not expect achievement players to disengage completely from the guiding influence of their game-independent interest in achievement goods. On the contrary, the particulars of expected achievement goods typically do (and should!) make a difference to how achievement players play. When the stakes are high, skilled achievement players play to the percentages; when the stakes are low, they might opt for more risky but showy play. Game play does not break down until the interest in achievement puts sufficient pressure on a player’s capacity to support the lusory attitude. Only such complete breakdowns of the game63—when guidance by the aim to achieve renders a player’s in-game actions incompatible with simultaneous guidance by the lusory goal—should be ruled out by the submersion requirement.Much the same goes for the striving player. The Monopoly trap should not lead us to require a complete disengagement from the broader interest in striving. Striving play only breaks down when guidance by that interest defeats the motivational traction of the prelusory goal and thereby renders the player’s in-game actions incompatible with simultaneous guidance by the game’s lusory goal. A strong submersion requirement rules out too much. We should temper the submersion requirement, in other words: game-independent interests, whether in achievement or striving, may make a difference to how a player’s in-game pursuit of that goal unfolds, as long as this does not interfere excessively with the motivational traction that the constitutive rules and prelusory goal have on their actions.This more moderate reading of the submersion requirement leaves us with a picture of the difference between achievement and striving play as not merely a matter of what motivates a player to take up the lusory attitude, but also of how their in-game pursuit of the lusory goal unfolds. As long as the achievement and striving players’ divergent interests do not lead them too far astray from their common commitment to the lusory goal—to playing the game—we should expect those interests to make a difference to how they respectively play. Tempering the submersion requirement thus clears the way for using striving play as analogy for building a Frankfurtian account of appreciative motivation.643.5  From striving play to appreciationThe last two sections showed that striving play amounts to an interesting sense of game play ‘for its own sake’ (Section 3.3) and that, unlike intrinsic play, it finds expression in how episodes of game play unfold (Section 3.4). What remains, in this section, is to generalize the motivational structure underlying striving play to how attention is guided in cognitive activities centred on objects of appreciation.Attentional guidance is best thought of in terms of the constraints on how cognition unfolds over time. Such constraints come in broadly two kinds (Christoff et al. 2016): automatic constraints, such as habitual cues and the mechanisms of perceptual and affective salience, influence where our attention is directed without our executive control; deliberate constraints, by contrast, are implemented by our executive functions in service of our goals. It is the presence of deliberate attentional constraints in appreciation—its involving goal-directed thought—that makes it amenable to the analogy with striving play. We can start by spelling out striving play’s side of the analogy.The crux of striving play lies in the player’s complex motivational relation to the game’s lusory goal. Striving players differ from achievement players with regard to the focus of the broader goals that rationalize their having and being guided by the intention to achieve the lusory goal. It is a difference, in other words, between the hierarchical goal structures within which their respective commitments to achieving the lusory goal is nested.Goal-guided endeavours of almost any level of complexity involves some hierarchical nesting of tasks. If I am rational, the intention to have friends for dinner would typically lead to further intentions to complete tasks that I consider ways of accomplishing the goal of 65dinner with friends. I have to invite them, check my refrigerator, decide what to serve, go to the grocery store, etc. Commitment to each subtask counts as a move guided by the broader intention, and the subtasks figure in ways of accomplishing that intention—they are motivationally nested in the intention. Subtasks can be further subdivided as the ways of completing them correspond to more nested intentions formed lower down the hierarchy (to invite my friends, I have to first get up off the couch to go get my phone, etc.). The initial overarching intention of dinner with friends can also be formed under guidance of yet broader personal goals (seeing my friends more often, introducing two of them to each other, etc.). Goals higher up in the hierarchy serve to guide, and, in good cases, rationally justify, the tasks of achieving nested goals lower down.This gives us a general way to characterize achievement and striving play as modes of Suitsian game play. Both kinds of players share a commitment to achieving the game’s lusory goal, which in Suitsian game play is represented by a personal level mental state (an intention). Both probably have multiple broader goals higher up their motivational hierarchies served by engaging in the game, some of which perhaps guided their initial decision to play the game. The discussion of the submersion requirement in the last section showed that, in order for there to be a difference in their mode of play, they cannot differ just with respect to which goals higher up the hierarchy secure their commitment to the lusory goal; there must also be a difference in which goals higher up influence how they go about pursuing the lusory goal. Thus the striving player’s in-game moves and decisions must, for some part of their pursuit of the lusory goal, be under active guidance not just of commitment to the lusory goal, but also of a goal higher up their motivational hierarchy that stands to be 66promoted by some in-game behaviours rather than others, irrespective of their contribution to efficient or successful attainment of the game’s lusory goal.Kumiko and Siviwe are playing a friendly set of tennis after school. Both are trying to win, so they are occupied in the same Suitsian game.  Siviwe is due to play in a tournament 12in a month or two, so for him the match is an opportunity to hone his craft. Unbeknownst to him, Kumiko has accepted a wager with a mutual friend, that she can win against Siviwe. In her hierarchy of goals, higher up than winning the match is the aim of winning the bet, to which the match win serves as a means. As a result, she plays in achievement mode, choosing only strategies she thinks likely to produce the win, which pushes her play towards efficiency and risk avoidance.Siviwe also cares about the win. But his reasons for caring include the aim of improving his slice backhand before the tournament. His play is guided by that aim, and he looks for every opportunity to use his slice to good effect. This makes for riskier, less efficient play, as his topspin backhand is more secure. He loses the set, to Kumiko’s delight, but Siviwe does not leave empty handed. Towards the end of the set his slice backhand starts clicking into place, and the satisfaction of hitting a few with growing confidence compensates for the disappointment of losing the match. Siviwe plays in striving mode because his in-game activity in pursuit of winning is actively guided by a goal higher up his motivational hierarchy that is satisfied by his pursuing the win in a certain way. How he plays is sensitive to features of the in-game activity that he values, independently of whether those features are success-conducive in the particular match.  To see that tennis satisfies the Suitsian analysis, start with the prelusory goal of making a tennis 12ball bounce twice without interruption and add materials and restrictions stepwise until you have the constitutive rules for a tennis point.67At this point we have all the necessary materials in hand for spelling out this chapter’s proposal. Here is a first gloss: appreciative engagement is the striving mode of goal-directed cognition.In fleshing out the analogy, it should be kept in mind that the cognitive activities involved in goal-directed thought are of a different order than the practical activities of game play. Game play features explicit goals and rules, intentionally taken up by the players. In goal-directed cognition, however, cognitive goals need not always be represented by personal level mental states (see Bargh 1990) and, even when they are, they need not be within reach of consciousness to continue guiding an agent’s cognition (Bos et al. 2008). We can abstract away from such differences by thinking of what Suitsian game play and goal-directed cognition have in common: both have a task-like structure, proceeding under constraints set by the agent’s goals. To capture this common feature, it will be useful to have a minimal model of what guidance by a goal consists in. We can borrow such a model from work by Imogen Dickie (2015: 95–98). Minimally, on her model, a system realizes guidance when it includes the following components:(A) a state of the system that represents a goal;(B) a feedback mechanism generating reports on current status; and(C) moves made by the system that are jointly determined by the goal at (A) and the feedback about current status at (B).68Guidance is apt when it leads to goal-fulfilment, which happens when(D) the function described at (C) tends towards outputs that minimize the difference between the goal represented in (A) and the current status represented at (B).Dickie’s model is fully general. It captures the basic features of guidance as it is realized across all levels of system complexity, and across both living and inanimate systems. Just as it applies to an automated heating system, so also to an agent’s goal-directed actions in Suitsian game play.13In a Suitsian game, at a broad level of description, the in-game activity consists of behaviours put out at (C)—for instance: turn shoulders!, adjust grip on racquet!, etc.—in response to intentional commitment to the game’s lusory goal and its nested goals at (A)—for instance: I will keep the ball in play, I will foil my opponent’s attempts to approach the net by placing my shots deep into the court, etc.—and perceptual feedback about the state of the game at (B)—my opponent is off balance, the ball is headed for the left sideline, etc. Behaviours at (C) are proximally determined by (A) and (B), but need not be fully determined by them.In the striving player, another higher-order system realizes guidance relative to a goal higher up in the player’s motivational hierarchy. Call the goal represented in this second   Dickie’s guidance system is similar to what Miller, Galanter & Pribram (1960) call a ‘test-operate-13test-exit (TOTE) unit’. The strategy to be implemented here, of using nested iterations of such guidance systems to capture complex goal-guided behaviours, is reminiscent of the approach they take in their seminal work in cybernetics.69system (A*)—for instance: I will practice my slice backhand. It has contents satisfied by some (A)-guided behaviours rather than others.  Thus the feedback mechanism in the 14higher-order system, (B*), makes the overall system sensitive to features of how the (A)-subsystem realizes guidance, by reporting on features of the behaviours output at (C)—for instance: are there relatively many slice backhands compared to topspin ones?, how successful are they?, how does their success correlate with features of their execution?, etc. A new, complex guidance system arises in which subsequent behaviours output at (C) are jointly determined by (A), (B), (A*) and (B*). Crucially, in striving play, whether outputs at (C) render the guidance realized by the (A*)-subsystem apt is independent of whether those same outputs at (C) render the (A)-subsystem’s guidance apt.The last paragraph uses the Suitsian game of tennis to illustrate the model of non-instrumental guidance that striving play instantiates. But with the model in hand, it should be clear that it need not be specific to practical activities dependent on the lusory attitude. Generalizing to the case of goal-directed cognition is straightforward: to attend in the appreciative mode is to be engaged in perdurant cognitive activity during which attention is guided by commitment to a goal, and the activity constituted by that guidance is modified by second-order guidance towards manifestations we find worthwhile for reasons independent of their efficiency in attaining our first-order goal. Thus, a cognitive agent engaged in appreciation has their cognition under guidance realized by a complex motivational structure including:  Note that (A*)’s satisfaction conditions will be constrained, minimally, by whether the outputs of 14the overall complex system are compatible with guidance by (A), otherwise guidance by (A*) would make the lower-order, (A)-guided system break down. This builds the modest submersion requirement endorsed at the end of the last section into the overall system.70(A) a mental state of either personal or sub-personal level representing a cognitive goal;(B) a feedback mechanism generating reports on progress towards the goal in (A);(A*) a mental state representing a second-order goal that favours some (A)-guided cognitive behaviours of the agent over none, and some particular (A)-guided cognitive behaviours over others irrespective of whether they are the best available minimizers of the difference between (A) and (B);(B*) a feedback mechanism reporting on (A*)-relevant features of ongoing (A)-guided cognitive behaviours;(C) cognitive behaviours implemented by the subject’s executive functions, jointly determined by (A), (B), (A*) and (B*).This application of the model to goal-directed cognition provides a picture of the appreciative motivational profile as involving attention guided by proximal cognitive goals that are taken up by the appreciating agent instrumentally, for the sake of cognitive activities that result from attending under the guidance of those goals. In appreciation our motivation is, in a manner of speaking, inverted: the first-order cognitive goals guiding our attention become means to the very attention they guide. We carry out cognitive tasks not primarily for the knowledge or information contingent on their completion, but to be engaged in the cognitive activities that constitute our attempts to complete those tasks. Crucially, this biases 71our strategies for completing the pertinent tasks towards attempts consisting of cognitive activities that we care (either intrinsically or instrumentally) to be engaged in.Consider again George Dickie’s Smith and Jones example from Section 3.2, in which Jones listens to remember a piece of music’s details for an exam, and Smith listens with no such purpose. Dickie’s claim was that they differ in their motives for attending and perhaps the degree of their attentiveness, but not with respect to the mode their attention to the music exemplifies. What does the motivational inversion model just outlined allow us to say about this case?Recall first that intuitions about such cases are complicated by the fact that agents in ordinary circumstances are engaged in ‘numerous, complex, nested, and overlapping’ tasks (Mole 2012: 52), and hence, over the course of their attention to the piece, Smith and Jones may each well instantiate a mixture of appreciative and non-appreciative motivation (see footnote 10). We can simplify by focussing only on pertinent differences between them. In as much as Jones is guided by the overarching goal of doing well on the exam, his attempts to latch onto, classify, memorize, compare, and assess features of the music will tend to emerge from and unfold in service of that overarching goal. This does not mean that his exam-focussed listening is a case of inattention or distraction from the music; on the contrary, the cognitive tasks Jones engages in will pertain directly to features of the piece and, if he cares about the exam, we should expect his listening to be attentive to a high degree. But nor does it mean that he is attending to the music in just the same way as Smith. Jones’s aim of doing well in the exam makes the payoff of his cognitive efforts contingent on the efficient selection and attainment of his proximal cognitive goals. He gets what he wants when he gleans information he deems exam-relevant, irrespective of how he gleans it, and he will 72likely avoid expending cognitive resources on features of the music that fall outside exam-relevant range, or on inefficient ways of gleaning exam-relevant information. Jones attends like Kumiko plays in the tennis example: in achievement mode.Smith’s attention, on the other hand, is meant to qualify as appreciative on the aesthetic attitude theorists’ criterion because, unlike Jones, he has ‘no ulterior motive’ for listening. We have seen, however, that this kind of contrastive gloss on appreciative motivation provides little actual purchase on what the mode of Smith’s attention to the music consists in. Here the motivational inversion account steps into the gap. If Smith really is engaging with the piece in the appreciative mode, the account suggests that it must be in virtue of some ongoing, goal-directed cognitive activities pertaining to the piece, guided by an overarching second-order interest of a different sort than that of Jones.Let us for a moment imagine that the piece was a favourite of Smith’s late grandmother and, as he hears it now, it activates in him the tacitly held goal to commemorate her. Inadvertently he finds himself actively trying to latch onto, classify, memorize, compare and assess subtle features of the piece. Many of the cognitive tasks that Smith takes up overlap with those of Jones: determining how the composer achieves certain effects, picking out the number of time signature changes in one part, anticipating the recurrence of a musical theme, and so on. But unlike for Jones, the payoff of Smith’s cognitive efforts does not lie in his success at memorizing, predicting, or finding things out about the piece. Instead, Smith gets what he wants when his attempts to do so transpire in the right ways: when his listening proceeds as though through his grandmother’s ears; when his anticipation of a motif foregrounds it as representative of her taste; when the manner of his focus on the rhythm in a certain passage helps call to mind the memory of her voice humming along.73Alternatively we might envision Smith’s listening as guided by a less personal goal—one that is in principle available to anyone who listens: say, a goal that is mandated by an appreciative practice centred on the genre of music in question. Imagine, for instance, that the piece is a composition in counterpoint and Smith wishes simply to be taken in by its complexities and melodic interactions. As he listens, he tries, perhaps deliberately, or perhaps he simply finds himself trying, to track the individual melodic lines. Again his cognitive efforts overlap substantially with those of Jones, who might also be tracking the individual lines, say, to find out (for exam purposes) how a recurring motif is adapted in each voice. But again, the payoff of Smith’s efforts do not lie in successfully keeping track of the different voices or rightly identifying variations in a recurring motif. In fact, it is sometimes the manner of Smith’s failures to keep track of all the melodies simultaneously—his temporarily losing the thread of one voice as he tracks another, only for the former to reemerge delightfully a few bars later—that gets him the payoff he is after. Smith attends, unlike Jones, in striving mode.These examples suggest that it is not the specific contents of Smith’s higher-order goal that matter, but rather that, whatever they are, those contents are satisfied by certain first-order cognitive efforts irrespective of their relative efficiency or actual success. The assortment of higher-order goals that could render the mode of Smith’s listening appreciative can vary immensely. They could be goals narrowly focussed on the particular piece, or ones that connect Smith’s current listening, say, to his broader interest in the genre (as in the counterpoint example), or they could be goals not essentially tied to music or listening at all (as in the grandmother example). They could be impersonal goals that derive from formal 74listening practices (the counterpoint case), or they could be goals grounded in deeply subjective personal commitments (the grandmother case).Additionally, to count as attending in the appreciative mode, Smith need not have anything that amounts to an intention to appreciate the music or, for that matter, to attend to it in the way the motivational inversion account describes. The motivational inversion account picks out a mental capacity we have and exercise independently of reflective awareness of it or ability to activate it at will. Much like our capacity to see things in the clouds or in marked surfaces precedes and provides the foundation for practices of pictorial representation (see Wollheim 1986: 46), so our capacity to attend to things in the appreciative mode precedes and grounds the appreciative endeavours and practices that depend upon it.3.6  Disinterest revisitedIf the model developed in the last section is on the right track, then our cognitive efforts in appreciation instantiate the inverted motivation or motivational feedback loop identified in the discussion of striving play (section 3.3). When an appreciator’s cognition proceeds in pursuit of proximal cognitive goals (such as to find certain things out about the object of appreciation), those goals are ultimately pursued not for any purpose their attainment would serve, but rather for the sake of aspects of their attentive pursuit. In this narrow sense, then, appreciation does consist in attention ‘for its own sake’.The motivational inversion account meets Dickie’s challenge to the aesthetic attitude theorists. It is formulated in expressly ‘Frankfurtian’ terms pertaining to how attention is guided over the course of appreciative episodes. Moreover, it provides a positive rather than privative analysis of the appreciative motivational profile, committing to a determinate 75nested structure of goals by which attention is motivated, rather than just ruling out certain kinds of motivation. But the account is admittedly unorthodox in several respects. For a start, whereas appreciative motivation is traditionally made criterial for aesthetic mental states (such as pleasures, judgments, an attitude, or responses), it is presented here as qualifying a distinctive kind of process or activity instead. The perdurance of cognitive episodes—their unfolding over time—is not typically given explanatory import in general theorizing about appreciation. At most, it sometimes features in the claim that appreciation or aesthetic experience is ‘autotelic’ or ‘self-sustaining’ (for example in Schaeffer 2015, Dokic 2016). But on its own, the invocation of autotelicity is more notional than illuminating. One way to read the current account is as offering a determinate analysis of what the pertinent kind of autotelic motivation amounts to.The account further deviates from tradition by picking out a motivational profile that is ‘disinterested’ or ‘non-instrumental’ only in an attenuated sense. It eschews any requirement that a cognitive agent must attribute final value to the activity of attending to an object of appreciation. As in the example involving Smith’s grandmother in the last section, the second-order goal rationalizing and guiding an appreciator’s attention may be completely integrated in an idiosyncratic set of personal ends. Token episodes of appreciation may be engaged in for entirely instrumental purposes, as long as those purposes are served by being engaged in the cognitive activity comprising the episode rather than just by the attainment of that activity’s proximal cognitive goals.As briefly noted in the last chapter (Section 2.3.1), the abandonment of a full-fledged, restrictive notion of disinterest is a move with some recent precedent in aesthetics. Some philosophers have, for example, been advocating for the revival of an older, Platonic tradition 76in which aesthetic value is considered the object proper to loving (and, hence, deeply interested) attitudes (Nehamas 2007; Riggle 2016). Others have taken issue with how the disinterest requirement erroneously casts appreciative engagement as a passive affair (Berleant 2016; Lopes 2018: 159–163). Yet another line of criticism laments that the valorization of disinterest in aesthetics has severed the arts—and our appreciative engagement with them—from the broader human activities from which they arose in the first place, and from which they derive their meaning and significance (Wolterstorff 2015).The motivational inversion account is rooted in sympathy with these concerns, but ultimately offers a strategy for sidestepping the dispute about disinterest. On the one hand it accommodates the critics of disinterest on all fronts. It brooks no prohibition on personal concerns and passions (including loving attitudes) prompting and helping to guide an agent’s attention in token appreciative episodes; it conceives appreciation as involving active, task-like exercises of cognitive agency; and it integrates appreciative motivation with the broader motives underlying our practical projects outside the confines of the gallery and concert hall. Nonetheless, the account also allows the proponents of disinterest to maintain that appreciative episodes are non-instrumentally motivated in the substantive (if non-traditional) sense captured by the motivational inversion model. Live questions thus remain about the extent to which the model is compatible with, or even fitted to, the theories of some proponents of disinterest such as Schopenhauer and Bullough.Finally, as conceived in this chapter, the appreciative mode of attention is ubiquitous. The motivational profile picked out by the proposed account can be instantiated in attention to any manner of item or task, ranging from things of great aesthetic significance to the utterly mundane or even frivolous. This forestalls the temptation to draw an essential 77connection, as the aesthetic attitude theorists did, between appreciative motivation and the aesthetic. But this is as it should be. The aim of an account of appreciative attention should be to explain why the values of things—both aesthetic and otherwise—come to fruition in experience or in our lives more broadly, when we attend to them appreciatively.78CHAPTER FOUR   Appreciation and Metacognition4.1  Second-order troublesPrevailing accounts of aesthetic experience and appreciation frequently make speculative appeals to what appear to be metacognitive phenomena. In characterizing appreciative engagement, they implicate our capacities for becoming aware of, thinking about, evaluating, intervening in, and forming attitudes towards features of our own ongoing perception and cognition. Gary Iseminger (2005b: 99), for example, defines appreciation as ‘valuing the experiencing of [a] state of affairs for its own sake,’ where ‘valuing’ designates an occurrent attitude with an ongoing first-order experience as its intentional object. Noël Carroll (2002: 167), who defends a deflationary view of what is required for aesthetic experience, denies second-order cognition’s necessity, but nonetheless suggests that second-order attention to how something’s aesthetic and formal properties ‘engage our sensibilities and imagination’ constitutes one kind of aesthetic experience. Jerrold Levinson (2016: 39), similarly, endorses a disjunctive requirement on aesthetic experience, one disjunct of which comprises an apparently second-order state: ‘a positive hedonic, affective, or evaluative response to the perception itself or the content of that perception.’ Matthew Kieran (2005: 213) claims that ‘truly appreciating’ an artwork involves apprehending ‘the ways in which the artistry shapes 79and guides our responses.’ Kendall Walton (1993: 505) thinks that for pleasure to be aesthetic it must be taken ‘not only in the [object] or one’s experience of it, but also in one’s experience of admiring it, in one’s judging it to be good.’ Murray Smith (2012: 80) notes that ‘when [aesthetic] experiences go well, they are not merely had, but savored. They become the object of a particular kind of self-consciousness.’ In as yet unpublished work, James Shelley (ms.) argues for a view that makes appreciative judgements of aesthetic value ‘self-referential,’ in that they attribute to their objects the normative status to rationalize the very perceptual activities from which they issue. Keren Gorodeisky (2018) argues for a view of aesthetic pleasure as a complex mental state that represents itself as merited by the object that elicits it. And Nick Riggle (2016: 12), protesting against disinterest-based accounts of the appreciation of beauty, writes that ‘self-awareness of some sort is partly constitutive of the affective state in the experience of beauty, not simply a downstream effect.’Despite the prevalence of such apparent appeals to second-order cognition, their advocates seldom engage with broader empirical and theoretical debates about the nature and remit of the metacognitive. Indeed, typically these theorists’ appeals are not couched in the jargon of ‘meta-awareness’, ‘metacognition’, or ‘second-order response’ at all, and the folk-psychological categories they do employ (quite reasonably, given their aims) can make it difficult to gauge the precise cognitive underpinnings to which they are committed. Furthermore, in cases where the allusion to metacognition seems clear, it is not always obvious that it passes empirical muster. Take Iseminger’s use of the term ‘valuing’ to describe an apparently second-order attitude. Intuitively, ‘valuing’ seems like an apt enough term for capturing something about the motivational pull of some paradigmatic aesthetic experiences. But if work in moral psychology is anything to go by (see for example Scheffler 802010 and Kubala 2017), valuing is a complex psychological phenomenon even where it takes a first-order intentional object, so there are open questions about what exactly occurrent second-order valuing (of an ongoing experience rather than its contents) would amount to, and whether creatures with our cognitive abilities are even capable of it.The lesson is not that Iseminger, or any of the other theorists mentioned, is wrong, but rather that invocations of second-order phenomena in theorizing about appreciation are merely promissory until due attention has been paid to what psychology and cognitive science have to say about metacognition. It is this lesson that motivates the work of the current chapter. For a start, heeding it could help settle which speculative appeals to second-order phenomena in aesthetics are compatible with our best understanding of the mind. But an even more incisive approach asks whether work on the place of the metacognitive in our mental economies in general could hold untapped resources for thinking productively about the place of self-awareness and self-knowledge in appreciative engagement in particular. Invocations of the metacognitive can be lifted from the realm of speculation only once our best scientific understanding of metacognition informs and constrains theorizing about appreciation from the outset.To that end, this chapter investigates current developments in the empirical literature on metacognitive phenomena so as to leverage them into a proposal about metacognition’s role in appreciative engagement. An important first port of call in this literature is the body of work in social psychology that draws into question some of our introspective capacities. A review of the most pertinent of these findings (in Section 4.3) helps with working out their implications for aesthetics (Section 4.4). The challenge these findings pose is real, but mainly affects excessively intellectual theories that make conscious, reflective introspective 81reasoning about our own mental states a prerequisite for appreciative engagement. Subsequent sections therefore turn to accounts of appreciation that envision a role only for what has been called ‘implicit’ or ‘procedural’ metacognition—a low-level variety of self-monitoring and control thought to attend all exercises of mental agency, but not involving reflection or metarepresentation (Section 4.5).While the ‘procedural’ picture of metacognition’s role comfortably avoids the charge of over-intellectualizing appreciation, there is reason to think that it dismisses metarepresentation too hastily (Section 4.6). An alternative proposal finds middle ground by assigning metarepresentation a role in the monitoring and modulation of our responses as appreciative episodes unfold over time (Section 4.7). This role implicates none of the introspective capacities threatened by the social psychology results, and can be discharged in tandem with the non-metarepresentational processes that constitute procedural metacognition.First, however, the next section briefly reflects on why philosophers have been tempted to speculate about the connection between appreciation and metacognition in the first place.4.2  The allure of cognitive ascentAesthetic theorists’ failure to engage substantively with empirical work on metacognition is unfortunate but also unsurprising. Recourse to second-order mental phenomena has long been a popular theoretical strategy in philosophy, and has delivered some marked successes in other branches of value theory. For an obvious example, take Harry Frankfurt’s (1971) influential invocation of second-order volitions in his compatibilist account of free will. 82Alongside such precedents, the plain fact that human beings clearly do have considerable (if fallible) competence for reasoning about and forming attitudes towards aspects of our own mental lives, lends some license to speculation about second-order cognition, even on purely theoretical grounds. Employing cognitive ascent as a theoretical strategy, one might think, holds promise for addressing challenges in aesthetics that look intractable with reference to first-order perception and cognition alone. A quick tour of four such challenges serves to illustrate the point.The first and perhaps thorniest, familiar from earlier chapters (see Sections 1.2, 2.2.1 and 3.1), is the aesthetic demarcation problem—the perennial question of how the domain of the aesthetic is to be differentiated from the non-aesthetic. If aesthetic experience involves some kind of metacognitive component, then, whatever its particulars, it could perhaps be the ingredient distinguishing instances of aesthetic experience from instances of non-aesthetic. Attempting to solve the demarcation problem is a heroic task, however, as any answer to it will depend on a wide array of other issues in aesthetics. This perhaps explain why this strategy, of appealing to a distinctive kind of second order state to demarcate the aesthetic, is not explicitly adopted by too many of the theorists listed in the previous section (with the exceptions of Gorodeisky and, possibly, Shelley). One of the advantages of the current project’s approach is that it disarms the aesthetic demarcation problem, not by solving it, but by distinguishing between aesthetic experience and appreciation, and suggesting that the two phenomena crosscut (see Section 1.2). Thus, while most of the appeals to metacognition discussed in this chapter are made in connection with aesthetic experience, we may treat them as targeting appreciative instances of aesthetic experience (as they arguably do), and 83proceed to ask about whether metacognition belongs in an account of appreciation broadly construed, instead of in an account of aesthetic experience per se.The second challenge that an appeal to metacognition could plausibly address, is what we might call the heterogeneity problem.  Our aesthetic and appreciative encounters with 15the world centre on extremely diverse types of objects and features, and may draw on the full gamut of perceptual, cognitive, and affective capacities. But this means that the first-order cognitive operations and states implicated in appreciative engagement may vary so widely from one kind of object to another, that finding a common denominator that unifies appreciative phenomena looks like an impossible task. Walton (1995: 499) richly articulates a version of the problem:There is enormous variety among the works we take to be of high aesthetic quality, and our reasons for praising them, for pronouncing them aesthetically valuable, are astonishingly diverse. Some good or great works stimulate, some soothe, others are disturbingly provocative or upsetting. Some afford intellectual pleasures; others emotional experiences—fulfilling emotional experiences in some instances, distressing ones in others. Some works offer insight or illumination; others catharsis. Some provide escape from everyday cares; others help us to deal with them. Some require careful study and analysis; others wear their charms on their sleeves. Great works can be exuberant or gloomy; they can be intense, or serene, or painful, or funny. ‘Aesthetic value’ appears to be an incredible grab bag. What justification is there for speaking of a single kind of value in cases of all of these sorts? This is a close cousin of the heterogeneity problem for theories of aesthetic experience discussed in 15Chapter 2 (Section 2.2.2). The version discussed there, however, focussed exclusively on heterogeneity in the phenomenological features of aesthetic experiences.84Walton’s own solution locates aesthetic value in a particular kind of pleasure taken in acts of first-order evaluation across all their infinite variety. It is a tricky interpretive business to settle whether Walton intends the variety of pleasure he postulates to involve second-order intentionality (but see Shelley 2018 for one example of a second-order reading). Either way, by temporarily stipulating a second-order reading of Walton’s view for argument’s sake, we may take his account (so interpreted) as illustrative of the more generally available strategy: appealing to a unified second-order phenomenon to counteract heterogeneity as it occurs at the first-order level.Thirdly, second-order responses are sometimes invoked to address what might be called the valence problem. This concerns the question of why we so often willingly engage appreciatively with objects that elicit mainly negative emotion, sometimes very strong negative emotion. The question has a long history in aesthetics and has generated a large contemporary literature about artistic genres that trade specifically on eliciting negative affect in an audience, for example tragedy, horror, gore, and cringe comedy (see Strohl 2019 for discussion). The valence problem is particularly pressing for hedonists about aesthetic value, who hold that appropriate appreciative engagement with the aesthetically valuable must always include a degree of positively valenced affect. But even without antecedent commitment to hedonism, the puzzle of why we actively seek out unpleasant experiences in the aesthetic domain demands an answer.Susan Feagin’s (1983) treatment of the problem of tragedy exemplifies the cognitive ascent strategy for responding to the valence problem. The pleasures of tragedy, she suggests, reside in the second-order satisfaction we take in the fit of our own displeasurable first-order responses, to the villainy, treachery, and injustice portrayed onstage (Feagin 1983: 98). There 85are reasons to resist the particulars of this proposal but, again, it is the general strategy that is the current concern. We find different implementations of it in general theories of aesthetic experience too. Levinson’s (2016) hedonism is a case in point: where there is no positive affect in our first-order responses, his account requires positive affect at the second-order level for an experience to qualify as aesthetic.The fourth challenge is the aesthetic reasoning problem. A standing issue in aesthetics is that of accounting for the practice of citing reasons for our aesthetic responses (see Dorsch 2013 for discussion). One way to interpret what we do when we point to particular features of an object to justify our responses, is that we are providing causal explanations for those responses, thereby making them rationally intelligible. Relatedly, a cursory look at aesthetic behaviour shows that people in general succeed to some extent in securing a measure of consistency among their aesthetic pursuits. We manage to predict, at least sometimes, what will and will not appeal to us, and this ability plausibly depends on an understanding of how our own responses are caused (although see Melchionne 2010, and Section 4.4 below). A straightforward way to explain both these phenomena—our capacities to rationally account for and to predict our aesthetic responses—is that we have introspective access to the mental processes underlying those responses. Thus metacognition can provide a parsimonious and unified account of both interpersonal and individual aesthetic reasoning.Solutions to problems like the four just enumerated might make metacognitive theoretical gambits alluring, but gambits come with costs. One such cost pertains to whether the metacognitive capacities invoked are empirically well founded. Over the last several decades, work in social psychology has seen a profusion of results demonstrating systematic failures of some of the metacognitive abilities that theorists in aesthetics and elsewhere in 86philosophy have arguably taken to be beyond reproach. Philosophers of mind and empirically inclined moral psychologists have been processing the fallout of these results for some time (for example Dennett 1991; Gopnik 1993; Haidt 2001; Lawlor 2003; Goldman 2004; Schwitzgebel 2008; Carruthers 2011). But only quite recently have their implications started receiving due consideration in philosophical aesthetics. The next section reviews the most important of these empirical results, and the subsequent section discusses their upshot for aesthetics.4.3  The challenge from social psychologyAs we saw in discussing the aesthetic reasoning problem, one way that metacognition might potentially figure in appreciation is by providing us with conscious access to the rational grounds for our aesthetic responses. Introspecting on our responses, the thought goes, can reveal facts about their aetiology and thus put us in a position to justify them to others and to plan for future aesthetic pursuits. Additionally, in many of our appreciative practices, knowledge of how an object elicits specific reactions in its audience counts towards a full grasp the object’s aesthetic character. What better way to gain such knowledge than to turn our mind’s eye inwards to our own responses to investigate how they arise? The problem is that a large body of work in experimental psychology suggests that some varieties of conscious introspection are susceptible to systematic error.874.3.1 Nisbett, Wilson, and colleagues on the limits of introspectionThe seminal paper from this literature, Nisbett and Wilson’s ‘Telling more than we can know’ (1977a), did much to consolidate work available at the time of its publication that supported a view that was already gaining traction in cognitive psychology: that our subjective verbal reports about the sources of our own judgements and responses do not arise from direct introspective access to those sources, even while it seems to us that they do. Nisbett and Wilson found the existing evidence striking but also incomplete in a number of ways, so their paper’s second aim was to conduct a series of experiments to fill in gaps in the literature. Their primary hypotheses were, firstly, that people often cannot accurately identify the effects of particular stimuli on their own decisions, preferences, judgements, choices, and other responses; secondly, that when people are asked to report on the effect a stimulus had on them, their subjective reports are based on implicit, a priori theories about the causal connection between the stimulus and their response, rather than on memory of their own cognitive processes operating on the stimulus to produce the response; and thirdly, that even cases of accurate subjective reports are due to incidentally correct employment of an a priori causal theory (Nisbett and Wilson 1977a: 233). The experiments they conducted to test these hypotheses yielded some famous findings.In one set of studies, subjects were asked, under the guise of a market research survey, to inspect an array of four identical items of clothing (nightgowns in one experiment, pairs of stockings in a second) and to decide which of the four was of the highest quality (Nisbett and Wilson 1977a: 243–4). Their choices were found to show a substantial left to right position effect: the further to the right in the array an item, the more often it was chosen. Nonetheless, 88when asked to explain their choices, the position of the item the they chose was never spontaneously cited as a factor in their choice, and even when explicitly prompted to think about how its position affected their choice, most subjects strongly denied that it did. Despite its clear causal influence on their assessments, in other words, they were unaware of the effect of the stockings’ relative position on their choices.Another study exploited the halo effect to demonstrate subjects’ lack of awareness of how a stimulus affects them (Nisbett and Wilson 1977a: 244–5 and 1977b). Subjects were shown one of two recorded interviews with a college teacher. Those in the ‘warm’ experimental condition were shown a recording in which the teacher answered questions in a pleasant, agreeable, enthusiastic way. In the recording shown to those in the ‘cold’ condition, the same teacher answered in a rigid, intolerant, and distrustful way. Afterwards subjects in both conditions had to rate their liking of the teacher, as well as the attractiveness of three of the teacher’s particular attributes (accent, appearance, and mannerisms) that were kept as close as possible to constant across the two recordings. It was found that subjects in the warm condition liked the teacher much better than those in cold, and, consistent with the halo effect, also rated the teacher’s particular features much more attractive. Some subjects in each condition were asked about the effect of their general liking rating on their ratings of the teacher’s particular attributes. The rest were asked the converse: how their rating of the attributes influenced their general liking rating. Across both conditions, subjects strongly denied the influence of their global liking judgements on their ratings of particular features. In the cold condition, subjects also positively misjudged the direction of causation, claiming that their dislike of particular features influenced their general liking rating.89Experiments like these indicate that we are poor at identifying actual influences of effective stimuli on our judgements and preferences and, by extension, we lack introspective access to the cognitive processes that mediate those effects. A second set of studies confirmed that we also tend to erroneously attribute causal efficacy to causally ineffective stimulus features.The subjects in one of these (Nisbett and Wilson 1977a: 245–6) were shown a documentary film while divided into three conditions. In the two experimental conditions, the film was shown either slightly out of focus, or with noise produced by a power saw in the hall outside. A control group watched the film with no distractions. After the screening, subjects were asked to rate the film on three dimensions. Those in the two experimental conditions were then given an apology for the poor viewing conditions and asked to indicate which of their three ratings they thought were influenced by the distraction and how. Across the three conditions, subjects rated the film very similarly on all three dimensions, which indicated that the distractions were not actually causally influencing the subjects in the experimental conditions’ ratings. Interestingly, subjects in the poor focus condition correctly judged that their ratings were not much affected by the distraction, although they did show a significant tendency to think they were affected on at least one of their three ratings. (Here Nisbett and Wilson could invoke their hypothesis that incidentally correct implicit causal theories could lead to accurate subjective reports.) Most of the subjects in the noise condition, however, judged erroneously that their ratings were substantially affected, imputing causal efficacy to factors that had no influence on their responses.Some of Wilson’s later collaborations with other colleagues also deserve mention for how they go beyond just the issue of introspective access. Wilson and Schooler (1991) found 90that subjects’ reported preferences (when choosing between different brands of strawberry jam or between different college courses) diverged much more from experts’ ratings if they introspected on the reasons for their preference before choosing than if they chose without such introspection. In a similar vein, Wilson et al. (1993) found that when subjects introspected on their reasons before choosing between two types of posters, it increased their likelihood of choosing the type that rated lower in pretesting and, whichever poster they actually chose, they were likely to be less satisfied with their choice several weeks later than the control group who chose without introspecting. These studies suggest that conscious introspection is not only a poor method for identifying the causes of our own responses, but can sometimes also interfere with those responses’ capacity to guide practical decisions.Finally, a study by Wilson and Dunn (1985) illustrates how the results reviewed so far pertain only to a specific kind of introspection, namely conscious, verbalized, introspective judgements about the reasons for one’s own preferences, choices, judgements, responses, and behaviour. In a first experiment Wilson and Dunn asked students for feedback on a number of beverages under the guise that they were completing consumer surveys for market research. They randomly divided the subjects into three groups, one placed in the ‘analyse’ condition, one in the ‘focus’ condition, and one a control group. Subjects in analyse were asked to think carefully about and explain their reasons for liking or disliking different beverages. Those in focus were asked to just think carefully about how much they like or dislike the beverages and then to rate them accordingly. A second, similar, experiment was done in laboratory conditions on a different set of subjects. In this study the analyse, and focus conditions were given similar instructions with regard to a set of puzzles—either to analyse their reasons for 91preferring or focus on their attitudes towards puzzles with which they were given free play time.The contrast between analyse and focus in these experiments is crucial. The findings show that analysing and thinking about the reasons for one’s attitudes to a stimulus can significantly reduce the quality of those attitudes in several respects, such as their stability, their consistency with consequent behaviour involving the stimulus, and their correspondence with consensus attitudes. But just carefully focussing on one’s own attitudes to the stimulus, on the other hand, showed no similar adverse effect. In fact, doing so marginally increased attitude-behaviour consistency as compared to the control group subjects. Wilson and Dunn (1985: 250) point out that this consolidating effect of consciously focussing on one’s own attitudes had previously been confirmed by Fazio et al. (1982) and Snyder and Kendzierski (1982). What this is taken to show is that introspective focus on one’s own attitudes increases their accessibility and salience, and, thereby, makes them more likely to influence consequent behaviours.4.3.2 Johansson, Hall, and colleagues on choice blindnessNisbett and Wilson’s original paper overshadowed subsequent work on introspective reports for a considerable time. Part of this was due to the sheer breadth of its influence, but part was because the research programme it outlined eventually stalled. It stalled because the experimental protocol they recommended for further research, the ‘actor-observer paradigm’ (see Nisbett and Bellows 1977), relied heavily on the truth of their preferred explanation for their findings. They hypothesized that our verbal introspective reports result 92from implicit causal theories we apply to our own responses and behaviours post hoc. But their protocol turned out to be virtually impracticable (for discussion see White 1988, and Johansson, Hall et al. 2006), and, in time, the emergence of the debate between ‘simulation theory’ and ‘theory theory’ in the psychology of social cognition complicated the prospects for their hypothesis.This impasse was eventually broken by a series of papers from a laboratory led by Petter Johansson and Lars Hall. Their breakthrough came with the development of an alternative experimental protocol that could facilitate the study of introspective misattribution without relying on assumptions about such attributions’ origins. Their protocol is based on techniques from close-up card-magic that enables an experimenter to manipulate the relationship between what experimental subjects choose, and the actual outcomes of their choices. The results show that people very often fail to detect mismatches between what they choose and what they get, and, moreover, they confabulate reasons for why they did in fact choose the outcome they had not actually chosen. Johansson and Hall named this phenomenon ‘choice blindness’, after the perceptual effect of change blindness: our introspective explanations track features of the outcomes of our choices, but are blind to features targeted by the preferences and intentions that causally explain those choices.In the landmark study that launched the choice blindness paradigm, they showed subjects fifteen pairs of photographs of female faces and asked them to decide, for each pair, which face they found the more attractive (Johansson, Hall, et al. 2005). Immediately after some of the fifteen trials, subjects were given the photograph they chose and asked to give reasons for their choice. On three manipulated trials per subject, however, the experimenter would use a card trick to switch the photograph chosen with a copy of the photograph not 93chosen before it was handed to the subject, who then had to explain their choice. Remarkably, the mismatch between the photograph chosen and the one received was detected in fewer than twenty percent of manipulated trials. This figure dropped to a meagre thirteen per cent detection rate, once they discounted the subsequent manipulated trials with those subjects whose initial detections came in the first or second manipulated trial. Just as remarkably, subjects who failed to detect the mismatches in manipulated trials, explained their choices in those trials with introspective reports that closely resembled the character of their own and other subjects’ reports in non-manipulated trials. In other words, they confabulated reasons for having chosen the photograph they are given, when they had actually chosen the other one.A follow-up paper submitted the verbal reports in the photograph experiment to a battery of tests and analysis to corroborate the initial findings (Johansson, Hall, et al. 2006). The similarities between introspective reports in the manipulated and non-manipulated trials strongly support the hypothesis that introspective reports are often confabulatory across both conditions. That is, when we have to give reasons for why we make a certain choice, we do not base those reasons on our introspective memory of the actual choice, we base those reasons on features of the perceived outcome of the choice, whether or not those features played any role in our actual choice or not.The same laboratory has since also established similar choice blindness effects with choices involving other sense modalities—people similarly fail to detect mismatches when presented with samples of jam they did not choose after tasting, and teas they did not choose after smelling (Hall, Johansson, et al. 2010). In a clever recent adaptation of their protocol, Lind, Hall, et al. (2014) studied the use of auditory feedback in speech production. They 94found that when subjects utter a word while simultaneously hearing feedback of themselves saying something other than what they actually say, they regularly fail to detect the mismatch and are prone to interpret themselves as having said what they hear rather than what they actually say. In other words, just as we interpret the causes of our own choices in the photograph experiment with reference to feedback about their outcomes, we interpret our own speech acts with reference to how we perceive their sonic outcomes (through concurrent auditory feedback) rather than with reference to our memory of the intention that produced those sounds.By facilitating the study of introspective error and confabulation in individual subjects, the choice blindness paradigm improves on earlier approaches like Nisbett and Wilson’s, where such errors could only be pinpointed at the group level.4.3.3 Cutting on mere exposureRecall that several of Nisbett and Wilson’s experiments worked by exploiting incidental group level effects (like the position effect in the stockings study) or known cognitive biases (like the halo effect in the teacher interview study). James Cutting’s (2003) investigation of another known cognitive bias—the mere exposure effect—shows that similar mischief is afoot outside the laboratory context in our interaction with art. The mere exposure effect, first explored in detail in work by Robert Zajonc (1968, 1980), entails that encounters with an object or event can lead to unconscious acquisition of information and attitudes towards it. All things equal, prior exposure to a stimulus typically leads to an increase in our liking or other positive attitudes towards it. This happens automatically, irrespective of 95whether we attend to a stimulus when we encounter it; in fact, stimuli perceived without awareness tend to produce larger mere exposure effects (Bornstein and D’Agostino 1992).Cutting did a series of studies to examine how mere exposure plays out in our responses to art. He started by selecting a sizeable sample of Impressionist paintings determined by careful analysis to be representative of the Impressionist canon in important respects, and counted the frequency with which each work in the sample appeared in books in the Cornell University library (on the assumption that relative frequency of appearance in a large sample of publicly available publications would be a fair proxy for how likely people are to encounter the works). Standardized images of the works were then used to measure experimental subjects’ preferences. Cutting found that their preferences tracked none of a number of potentially explanatory factors—like a work’s display in the Musee D’Orsay, whether it is a prototypical sample of Impressionism, subjects’ recent museum visits, the extent of their art history education, their recognition of the work, or their judgements of its relative complexity. But their preferences did reliably track the simple frequency of the images’ appearance.This result is generally taken to show that our responses to works of art have less to do with features relevant to their artistic quality, and more with arbitrary factors like how familiar they seem to us. Cutting’s study has prompted concerned responses from philosophers looking to defend the integrity of the mechanisms that regulate entry to the artistic canon (see for example Meskin et al. 2013), and in one case Cutting has even been accused of endorsing antirealism about artistic or aesthetic value (Nanay 2017). But his avowed conclusions are actually quite modest. He claims only that his experiment shows mere exposure to be one important factor in canon maintenance (as opposed to canon 96formation), and remains agnostic about which further theoretical and metaphysical conclusions his findings may or may not sanction (see Cutting 2017).Our current concern, of course, is not with what the findings tell us about the sources of aesthetic response in general, but with the implications for whether we have first-personal introspective access to those sources when we respond to art. On that score, the results are telling. Unlike Nisbett and Wilson, Cutting did not explicitly aim to examine subjects’ conscious access to the sources of their responses, and accordingly he did not ask them for introspective reports about those responses. But the mere exposure effect is known to do its work through unconscious avenues. Laboratory studies of the effect have demonstrated that subliminal exposure, where subjects are exposed to a stimulus for intervals of less than five milliseconds, leads to a more pronounced effect than when subjects are exposed to the stimulus for long enough to be aware of the stimulus (Bornstein and D’Agostino 1992). Thus, the mere exposure effect is a clear example of a causal factor in our aesthetic responses that, in its prototypical instances, is unavailable for introspective scrutiny.4.4  Moving beyond reflective introspectionUntil recently, philosophers working in analytic aesthetics have had surprisingly little to say about the repercussions of the social psychology literature on the limits and fallibility introspection. Three papers, by Kevin Melchionne (2010), Sherri Irvin (2014), and Dominic Lopes (2014b) break the silence, each taking stock of the challenge this literature poses and each drawing slightly different lessons from it. Of the three, Irvin is the only one to ask specifically after its implications for the place of metacognitive phenomena in accounts of the 97nature of aesthetic experience or appreciation. Melchionne and Lopes both focus on the broader question of how we should think about the role of conscious reasoning in our aesthetic lives.4.4.1 Aesthetic reasoning without introspective accessThe research reviewed challenges our conscious, reflective, verbalized, second-order judgements about the aetiology of our own responses. The main lesson for aesthetics is that we will have to look elsewhere for a solution to the aesthetic reasoning problem. Our practice of citing reasons for our aesthetic responses cannot be epistemically vindicated by first-person authority that we derive from privileged introspective access to those responses’ causes, because we do not have such access. Ditto for direct introspection’s presumed role in our ability to plan for future aesthetic satisfactions.Melchionne’s paper emphasizes this latter point, concluding that ‘by embracing anti-introspectivism, we move toward a more naturalistic view of taste, with more room for confusion, indifference, and transience in our aesthetic preferences’ (Melchionne 2010: 140). This should not strike us as an entirely unintuitive result. It is a familiar experiential datum that some of our most intense aesthetic responses can be attended by (perhaps even enhanced by) a sense of mystery about their psychological provenance. But the challenge runs deeper than this. The results not only show that we are typically unable to introspectively access our responses’ causal origins, but that we are often systematically wrong about what we are confident we do know about those responses.98Fortunately, the accuracy of conscious introspection is not the only avenue along which one might look for a solution to the aesthetic reasoning problem. On the question of what explains our practice of citing aesthetic reasons, a tradition running back to Arnold Isenberg’s work on criticism takes aesthetic reason-giving to be a forward- rather than backward-looking activity (see for example Isenberg 1949; Sibley 1965; Mothersill 1961; Ziff 1966; Hampshire 1970; Strawson 1974; Hopkins 2004; Cross 2017). This makes available the following response to the aesthetic reasoning problem: when we cite reasons, we are not causally explaining our own responses but rather attempting to influence others’ responses and attitudes. Lopes (2014b: 34) endorses a version of this move to address the empirical challenge, arguing that we should take care to distinguish reasons verbalized in critical discourse from causal factors in aesthetic response. He notes that this separation accords with work on the evolution of reflective reasoning that suggests ‘the main function of reflective inference is to produce and evaluate arguments occurring in interpersonal communication (rather than to help individual ratiocination)’ (Mercier and Sperber 2009: 150).On the question of how we navigate the task of planning for future aesthetic pursuits, the same empirical literature that rules out conscious introspection as a source of data for deliberate reflection offers alternative resources. Recall Wilson’s later studies (Wilson and Schooler 1991; Wilson et al. 1993), in which choice quality and post-choice satisfaction were found to be better among subjects who chose without introspective reasoning. These findings suggest that unconscious and unreflective mechanisms may do a better job of steering our aesthetic pursuits, whereas conscious means-ends reasoning based on introspective judgement tends to lead us astray. Lopes (2014b: 33, 2018) advances a version of this idea by pointing out that behavioural sensitivity to aesthetic reasons does not require the ability to 99verbalize those reasons. We can, and often do, act on aesthetic reasons that we cannot quite articulate. This is because conscious first-order awareness of aesthetically relevant features of a stimulus can play a causal role in reasons-sensitive behaviour without the need for that first-order awareness to be metarepresented in conscious, reflective inferences.There is also evidence that goal-driven unconscious reasoning can be a beneficial component in preference formation and decision making. A study by Bos et al. (2008), for example, compared the effects of conscious reflection and unconscious reasoning on subjects’ performance in decision and impression formation problems. It involved giving a group of subjects in a conscious reflection condition information pertaining to a problem and then giving them a fixed time to consciously think about it before giving them the problem to solve. Subjects in a mere distraction condition were given the same information, then told that the first experiment was over and given a distraction task to occupy them before eventually being given the same problem as the first group to solve. A third group, in the unconscious thought condition, was given the information but, before being given the same distraction task as the second group, were told they would later use the information to attempt the problem, thus giving them the goal to continue processing information while their conscious attention was occupied by the distraction task. The results show that goal-driven unconscious processing improved subjects’ performance relative to both the conscious reflection and mere distraction groups. This suggests that goal-driven unconscious processing is well suited to help guide aesthetic preference formation, choice, and the prospective pursuit of aesthetic satisfactions, even while conscious introspective reasoning about our own aesthetic responses yields poor results.1004.4.2 Irvin and ‘deep appreciation’Where does this leave us with regard to the place of metacognition in an account of aesthetic experience or appreciation? Irvin (2014: 37) zeroes in on whether theories that require ‘second-order awareness of one’s own mental states or processes’ are viable given the findings. For a large part, her conclusions dovetail with what Lopes and Melchionne have to say about aesthetic reasoning. She starts by distinguishing between second-order awareness of occurrent mental states, and second-order awareness of the processes that cause aesthetic response, arguing that only the latter is plausibly threatened by the social psychology results (Irvin, 2014: 50). This distinction sits nicely with Wilson and Dunn’s (1985) findings (see Section 4.3.1) that while introspective analysis of our reasons is detrimental to choice quality, mere introspective focussing on our attitudes, without reflecting on their origins or rationality, improves both choice quality and attitude-behaviour consistency. In other words, there are some forms of second-order awareness not concerned with our first-order attitudes’ origins or rationality that can have favourable rather than deleterious effects on our behaviour. Irvin (2014: 48–49) further corroborates this idea by citing a long list of studies that demonstrate performance improvements in several cognitive domains as a result of participation in mindfulness meditation practices. Clearly not all varieties of second-order awareness are affected by the challenge from social psychology.But Irvin goes further by also trying to maintain a place for introspecting on the sources of our responses in some elevated cases of appreciation. To do so, she proposes distinguishing between three degrees of appreciative success that she dubs ‘aesthetic experience’, ‘mere aesthetic appreciation’, and ‘deep appreciation’ (Irvin 2014: 45–47). 101Aesthetic experience simpliciter is cheap and quotidian, requiring just first-order perceptual experience with aesthetic contents, whether veridical or not. Mere appreciation adds a cognitive constraint: in order to aesthetically appreciate, our understanding of the object of appreciation must meet some threshold of accuracy, although no second-order awareness of how the object elicits our responses is required. Deep appreciation is that most rare of beasts, sighted only where expertise in an aesthetic domain affords us an accurate grasp of how the object of appreciation has elicited our response. This kind of grasp, Irvin concedes, need not come by means of direct introspective access alone, but can be achieved with the help of reflective third-personal theorizing about aesthetic responses in the relevant domain.Assigning this kind of hierarchical priority among different types of aesthetic response gives away the game. Even while conceding that the social psychology results make trouble for direct introspective access to our responses’ aetiology, Irvin nonetheless attempts to salvage pride of place for introspective awareness of our reasons by making it a prerequisite for achieving the rarified state of ‘deep appreciation’. But why privilege deep appreciation in this way? It may be true, as a matter of historical fact, that many western appreciative practices have converged on norms that valorize intellectual understanding of the causal workings of the objects of appreciation. If what Irvin wants is an account on which achieving the normative standards set by such practices is the ultimate aim of the cognitive efforts we exert in appreciation, then she is right to privilege deep appreciation. But if what one is after is a descriptive account of the central features of appreciative engagement considered as a cognitive phenomenon, as is the case in the current project, then one should be careful not to simply codify the contingent norms of the social practices in which that phenomenon plays an important role.102By conceding that the empirical challenge undermines conscious introspection’s capacity to play the functional roles of justifying claims in interpersonal critical reasoning and of securing consistency in individual planning of future aesthetic pursuits, we give up the only plausible rationale for construing the phenomenon of appreciative engagement in terms of the aim of a second-order grasp of the causes of our responses. Thus, the lesson from the social psychology results is not, as Irvin’s discussion of deep appreciation seems to suggest, that the type of second-order awareness those results show to be rare and difficult to achieve should be understood as the ultimate aim of the cognitive efforts we exert in appreciative engagement. The lesson is rather that, whatever value there might be in such hard-won second-order awareness, and even if some appreciative episodes do happen to involve attempts at achieving such awareness, it is inessential to the nature of appreciative engagement—whether as a minimal requirement or as a constitutive normative ideal.Thus, if there is an essential role for metacognition in appreciation, it is likely to be played instead by less intellectually demanding varieties (perhaps, for example, like the introspective focussing in Wilson and Dunn’s beverage experiment). But how should we understand these less intellectual forms of metacognition, and, if not aesthetic reasoning, what role (if any) should we envision for them in appreciative engagement? One place to look for empirical insights on these questions, is in work on the epistemic feelings and their role in what has been called ‘implicit’ or ‘procedural’ metacognition.1034.5  A role for ‘procedural’ metacognitionMuch metacognition research in recent cognitive science has centred not on introspection, but on the epistemic feelings (sometimes also called noetic or metacognitive feelings). These are feelings that carry information about the subject’s epistemic situation. The feeling of uncertainty, for example, can tell us that our belief that we turned off the stove might be mistaken and we should check again before leaving the house; a feeling of familiarity can tell us that we may have come across a person before even if we do not quite recognize them; the tip-of-the-tongue experience suggests we are close to recalling some piece of information currently just eluding us; curiosity points us to our own ignorance about something; and surprise alerts us that we had not been expecting what just happened. These feelings’ capacity for carrying such information has led researchers to accord them a central role in the online monitoring and control of cognitive action, and some researchers to consider them ‘procedurally’ metacognitive. In this capacity they have also been implicated in work on aesthetic experience in empirical aesthetics.4.5.1 The metacognitive aspirations of the epistemic feelingsThe fact that epistemic feelings carry information about our epistemic situation makes it tempting to construe them as being, at least in part, about the epistemic states they disclose at the personal level. If this were right, the feeling of surprise would, for example, consist partly in a metarepresentation of one’s contradicted expectation, and the feeling of curiosity, in a metarepresentation of one’s current ignorance. But, depending on what is meant by 104‘metarepresentation’ here, this picture is at risk of over-intellectualizing the epistemic feelings.The weight of the evidence shows that the epistemic feelings fulfil their function without implicating anything like the human mindreading system (see Perner 2012; Kornell 2014; Carruthers 2016). Creatures without a concept of a belief, for instance, can nonetheless feel uncertain they buried the acorn under this tree rather than that, and no sophisticated mindreading system is required for feeling surprise at finding it buried where they did not expect. (This is of particular importance in comparative psychology, where some animals have been found capable of decisions based on uncertainty-monitoring. Some researchers have wanted to interpret these findings as clues to the origins of human self-awareness (Couchman et al. 2012), but others argue that that the basic self-monitoring capacities mediated by the epistemic feelings are independent of the mindreading system, both in terms of their function (Proust 2012) and their phylogeny (Carruthers, Fletcher, and Ritchie 2012).)Although they are thus not about our own minds (in the sense of metarepresenting our mental states as mental states), the epistemic feelings nonetheless can serve as cues for reliable metarepresentational judgements about our epistemic predicament. This is because they are responsive to features of our cognitive processing that covary with features of that predicament. In particular, they are responsive to processing fluency. Processing fluency has been widely studied as a factor impacting a range of evaluative judgements, both about our own epistemic states and external stimulus features (see Alter & Oppenheimer 2009 for a review). In general, high fluency is associated with positive evaluation, and low fluency with negative. As high fluency has been shown to be hedonically marked (Winkielman and Cacioppo 2001), its positive-leaning effects on our judgements are generally understood to 105be mediated by mild positive affect, and vice versa for the negative-leaning effects of low fluency (Reber, Winkielman, and Schwarz 2004: 367–368).The epistemic feelings are sensitive to—and perhaps in some cases partly or wholly constituted by—the affective signal arising out of the dynamics of their subject’s perceptual and cognitive processing. Processing dynamics are in turn affected by our epistemic relations to stimulus properties and other contents we process. This allows the epistemic feelings, through their sensitivity to processing dynamics, to function as cues for reliable judgements about their subjects’ epistemic situation. The metacognitive judgement ‘I know that person’ can, for example, result from a feeling of familiarity that arises out of a discrepancy between actual and expected fluency of the visual processing of their facial features, where that discrepancy is sometimes a result of having processed those features before (Whittlesea and Williams 2000).Based on this picture of the epistemic feelings and their connection to processing dynamics, Proust (2012; 2014) has developed an account of metacognition in which they feature prominently. Metacognition, on her definition, comprises the ‘set of capacities through which an operational cognitive subsystem is evaluated or represented by another in a context-sensitive way’ (Proust 2014: 4). She dubs this the evaluativist conception of metacognition, in contradistinction to what she calls the attributivist view. Whereas attributivists associate metacognition with self-ascriptions of epistemic states mediated by the conceptual resources of the mindreading system, the evaluativist conception broadens the remit of the metacognitive to also include ‘experience-based’ epistemic self-evaluation. Experience-based self-evaluation occurs when a cognitive subsystem uses the affective signal arising from the processing dynamics of another subsystem (i.e. an epistemic feeling), as 106feedback in order to adjust or keep track of the epistemic tasks that the latter subsystem’s processing serves.In other words, epistemic feelings have the functional role of a feedback mechanism. They sometimes feed forward, as we have seen, into the mindreading system, to cue metarepresentational judgements about our epistemic states and affordances. But most of the time the information they carry feeds straight into other cognitive subsystems without passing through the mindreading system, and these other subsystems use the information to implement executive control over the subsystems from which the feelings arise. This is how rhesus macaques and dolphins can have basic capacities to adapt their cognitive behaviours to their epistemic context, even while they do not have sophisticated mindreading systems (Hampton 2001; Smith, Shields, and Washburn 2003): through decisions informed by affective feedback.The epistemic feelings’ functional role as feedback signals mediating executive control, prompts Proust to call them implicitly or procedurally metacognitive. While they do not constitutively involve conceptual metarepresentation, they provide their agents with procedural guidance for the exercise of cognitive agency in pursuit of epistemic aims.4.5.2 Processing fluency accounts of aesthetic experienceA recently influential view in empirical aesthetics, the processing fluency theory, takes aesthetic experience to be a species of epistemic feeling. Given the evidence of high processing fluency’s positive-leaning influence on preference and judgement across domains (especially judgements of liking) and the apparent mediation of this influence by the 107experience of positive affect, Reber, Winkielman, and Schwarz (2004) hypothesize that our aesthetic responses are largely a function of processing fluency. They note that perceptual fluency is facilitated by a variety of perceiver-independent stimulus features—like symmetry, rounded shapes, and figure-ground contrast—that have traditionally been associated with beauty. But individual perceiver history also impacts processing fluency. Processing ease is facilitated by prior familiarity with a stimulus and stimulus prototypicality, for example. (For this reason, fluency is also generally thought to be the mechanism behind the mere exposure effect discussed in section 4.3 above (see Bornstein and D’Agostino 1992).) Reber (2012: 238–241) argues that the effects of perceiver history on processing fluency provide an elegant means for explaining how individual aesthetic preferences are acquired through habituation and how cultural patterns of taste develop over time.One of the standard challenges to the processing fluency theory is the problem of boredom: while high fluency is indeed well suited to explain at least some part of the pleasurable aspect of aesthetic experience, it does not correlate equally well with continued interest in objects of aesthetic value (Silvia 2012: 259–60; Schaeffer 2015: 161–163). Reber (2012: 231–232) makes a compelling case that the processing fluency theory has resources for answering this challenge, but a recent proposal by Dokic (2016) takes the problem of boredom as an occasion for adjusting the theory instead. Dokic first argues for a constraint on an adequate account of aesthetic experience: that it must capture the autotelic or self-sustaining motivational profile traditionally associated with appreciative engagement (Dokic 2016: 74–76). The processing fluency theory goes only part of the way to satisfying this constraint, because high fluency might cause passing pleasure and yet fail to motivate continued engagement with the stimulus being processed (Dokic 2016: 78).108Dokic’s proposed improvement is to conceive of aesthetic experience not as a single epistemic feeling, but rather as a mode of arrangement of more than one intentional attitude: ‘aesthetic experience is a combination of non-aesthetic attitudes unified by the aesthetic motivational profile’ (Dokic, 2016: 76). More specifically, he envisions a dynamic tension at the heart of aesthetic experience between feelings of liking and familiarity (grounded in processing fluency) on the one hand, and feelings of interest and curiosity (grounded in disfluency) on the other. His hypothesis is that the autotelic motivational profile of aesthetic experience might ‘arise from a suitable dynamic combination of familiarity and novelty or, at the phenomenological level, pleasure and interest.’ (Dokic 2016: 80)If Dokic’s proposal is a departure from the processing fluency theory as expounded by Reber and colleagues, it is not entirely clear how substantive a departure it is. He remains vague about what exactly would suffice for the ‘suitable dynamic combination’ of disparate epistemic feelings to constitute the autotelic motivational profile he has in mind. And saying that the feelings are ‘unified by the aesthetic motivational profile’ does not help, because the aesthetic motivational profile is the phenomenon his theory is attempting to adequately account for. Nonetheless, his proposal brings about a productive shift in focus from processing fluency’s mere causal effects on aesthetic judgement and preference formation (Reber’s focus), towards its role in how episodes of aesthetic experience unfold. The epistemic feelings, through their sensitivity to processing fluency, function as affective cues mediating the perceiver’s continued epistemic engagement with the object of aesthetic interest. In other words their role in aesthetic experience is implicitly or procedurally metacognitive, in Proust’s terminology. They are the procedurally metacognitive residue of 109ongoing processing, serving as a feedback signal for the perceiver’s exercises of cognitive control in guiding their attention to the object.4.6  Moving beyond the merely proceduralAt the end of Section 4.4 it was suggested that if there is a metacognitive component essential to aesthetic experience or appreciative engagement, it would have to be contributed by less intellectual sorts of metacognition that serve some purpose other than analysing the reasons for our responses. Proust’s notion of implicit or procedural metacognition fits the bill. Hence, by implicating the procedural, cognition-guiding function of the epistemic feelings, Dokic’s adapted version of the processing fluency theory finds a role in appreciation for putatively metacognitive phenomena that do not fall prey to the empirical challenge to our conscious introspective capacities.But the notion of implicit or procedural metacognition is controversial. Nagel (2014) and Carruthers (2016) have both argued against the distinction between implicit and explicit metacognition on the grounds that it obscures a more fundamental distinction between cognitive phenomena that respectively do and do not involve metarepresentation. The worry is that, in the attempt to vindicate ordinary usage of the term ‘metacognition’ among experimental psychologists that study processing fluency and the epistemic feelings, Proust specially tailors her definition so as to count the epistemic feelings as metacognitive and, in the process, lets too much in the door.The sceptics of the implicit/explicit metacognition distinction point out that the epistemic feelings provide their feedback for the procedural guidance of epistemic acts on the 110basis of mechanisms closely analogous to processes underlying uncontroversially first-order cognitive operations like, for example, multi-sensory integration (Nagel, 2014: 711–713) and informational matching between distinct streams in visual processing (Carruthers, 2016: 64–65). It is hard to see how multi-sensory integration, for example, could happen without any kind of ‘evaluation of one cognitive subsystem by another in a context sensitive way’ (cf. Proust 2014: 4), and the feedback mechanisms this requires are relevantly similar to the feedback constituted by the affective signal arising from processing fluency. Thus, the argument goes, if epistemic feelings that arise from the dynamics of first-order processing are constitutively metacognitive for the reasons Proust gives, it turns out that a vast proportion of both human and animal cognition is constitutively metacognitive by the same token. This broadening of what we mean by metacognition threatens to rob the category of its theoretical interest and, Carruthers (2014) argues, muddies the waters for research about the roles played at various levels of human (and possibly animal) cognition by fully metarepresentational capacities.Carruthers is right to put a premium on conceptual clarity here, especially in view of the potential implications of a definition of metacognition for how we interpret findings in comparative psychology. But the dispute over Proust’s definition also has a tinge of the verbal about it. Proust (2012) clearly recognizes the functional and phylogenetic independence of ‘procedurally’ metacognitive phenomena from the full-blown metarepresentational capacities of the mindreading system. Moreover, her definition tracks a standing practice in experimental psychology of calling phenomena related to processing fluency and the epistemic feelings metacognitive, and this practice is unlikely to change at Carruthers’ insistence.111The dispute does speak to our current concerns, however, by throwing in sharp relief the difference between merely finding a role for procedural metacognition in appreciation, and finding a role for full-blown metarepresentation. Dokic’s version of the processing fluency theory does only the former (if that). He commits to the involvement of epistemic feelings arranged in a suitably dynamic tension. Taken together with his defence, elsewhere, of Proust’s view of the procedurally metacognitive function of the epistemic feelings (Dokic 2012), and the fact that he takes aesthetic experience to ‘[require] a dynamic interaction between the subject and the aesthetic object’ (Dokic 2016: 83), this suggests that he takes aesthetic experience to constitutively involve not only the feelings themselves, but also the exercises of cognitive agency they mediate. This is already a substantive position to stake out, because it makes aesthetic experience something more than just passive, stimulus-driven response. If we proceed on the assumption that what Dokic is tracking here is, in this dissertation’s terminology, appreciative instances of aesthetic experience (an assumption his discussion seems to licence), it should be clear from earlier chapters that this is a welcome result. But there is a sense in which denying that appreciation is passive and stimulus-driven is also not the most interesting result.Just to say that we exercise cognitive agency in appreciation is not yet to say all that much. If such agency, mediated by the epistemic feelings underlying it, suffices to make appreciation metacognitive, it makes it so only in a manner that all exercises of cognitive agency are metacognitive. This limits the role that metacognition plays to that of a mere auxiliary to the real business of appreciation, which is to exercise cognitive agency in some further specified way (presumably whatever way is motivated by an autotelic motivational profile). Thus, to the question whether appreciation constitutively involves metacognition, 112Dokic’s view responds with an apparently affirmative but somewhat unsatisfying answer: yes, appreciation does involve feelings of liking, familiarity, curiosity and interest. These do not really seem like the kind of metacognitive phenomena the question was targeting in the first place.The issue then becomes whether or not there are reasons to go further, beyond just the procedural variety of metacognition, in characterizing appreciation. Dokic is unequivocal in opposing a necessary role for full fledged metarepresentation. Interestingly, his motivation for this is that he wants to distance himself from overly intellectual views that ‘picture the subject as explicitly raising questions about her epistemic relation to the aesthetic object’ (Dokic 2016: 83). As our discussion in Section 4.4 showed, one indeed does well to avoid over-intellectualizing appreciation by connecting it, say, to conscious introspective reasoning about our own responses. But there is a lot of room in between views requiring just procedural metacognition, and views requiring the kind of intellectual introspection ruled out by the social psychology literature.The charge of intellectualism paints Dokic’s potential competition with too undiscriminating a brush. It assumes that doxastically driven question-raising and conscious, reflective reasoning about the aetiology of our own responses are the only forms of cognition that could implicate metarepresentational resources. This assumption is clearly mistaken. As Carruthers (2016: 61–63) has been at pains to emphasize, full fledged metarepresentation can be very much embedded in an agent’s ongoing mental processes, and crosscuts the divides between conscious and unconscious states, and reflective and intuitive processes. It is worth keeping in mind that much third-personal mindreading occurs without the involvement of conscious, reflective, and verbalized reasoning about others’ mental states (as evidenced in 113preverbal infant mindreading—see for example Hamlin, Wynn, & Bloom 2007 and Song & Baillargeon 2007). Equally revealing is the evidence suggesting that our basic metarepresentational capacities are phylogenetically older than, and thus in principle independent from, our linguistic capabilities, even if the two did subsequently co-evolve (Sperber 2000: 121–127). There is no prima facie reason to suppose that self-directed metarepresentation should be any different than third-personal mindreading in these regards. Fully metarepresentational forms of metacognition can in some instances be, but need not be of an analytic or intellectual character, nor need it be available for verbalization in order to fulfil its monitoring and control functions in our own cognition.If this is right, then not all uses of metarepresentation are conscious, reflective, verbally mediated, and intellectually demanding, and thus not every theory of aesthetic experience or appreciation that invokes metarepresentation thereby succumbs to Dokic’s charge of excessive intellectualism. There is room for an account that takes seriously the hypothesis that our capacity for appreciative engagement is built on our metarepresentational competences. The challenges such an account faces are, firstly, to specify the function that metarepresentational capacities supposedly fulfil in appreciative engagement. This function cannot be the one ruled out by the empirical challenge to introspection—of giving us conscious, verbally reportable access to the aetiology of our own responses. Secondly, an explanation must be given for why the function thus specified requires metarepresentation and could not be fulfilled by merely procedural metacognition or other, manifestly first-order, cognitive processes. The next section sketches an account that meets these desiderata.1144.7  A proposal: cognitive agency under metarepresentational normsDokic’s view rightly suggests that appreciation (or in his terminology, aesthetic experience) constitutively involves cognitive agency. Not any cognitive agency counts, but, as we have seen, what he has to say about the particulars of the epistemic feelings involved is not enough to substantially constrain which exercises of cognitive agency might feature constitutively in token episodes of appreciation. Perhaps more could be said here.The place to start is to note that many of our paradigmatic appreciative engagements with objects of aesthetic interest happen against the backdrop of appreciative practices. Art practices are a case in point, but appreciative practices comprise a much broader category, ranging over both institutional and informal social arrangements that normatively constrain interactions with objects of practice-associated value (for a theory of appreciative practices see Lopes 2014a: Chapters 7 and 8; see also Lamarque 2010 and Wolterstorff 2015: Chapter 11). Getting a fix on ways in which cognitive agency is exercised in appreciative engagement in contexts where there are socially mandated normative constraints on our actions, evaluations, and responses, can help provide a springboard for generalizing beyond just such contexts.The locus classicus on how the normative constraints underlying artistic practices impinge on aesthetic response is Walton’s ‘Categories of Art’ (1970). Walton famously argues that the aesthetic properties an artwork seems to have, depends on the category in which it is experienced or, differently put, the contrast class we bring to bear in our experiential engagement with it. In Walton’s terms, aesthetic appearances depend systematically on which non-aesthetic properties of a work are standard, variable, and 115contra-standard with respect to the artistic category or categories the work is perceived in. To illustrate the point, Walton imagines a society with a practice of producing guernicas, ‘versions of Picasso’s “Guernica” done in various bas-relief dimensions’ (Walton 1970: 347). In their practice, the aesthetic properties of a guernica supervenes, as one might imagine, on the way in which their marked surfaces are moulded to protrude from the two dimensional plane. Imagine being a member of this culture, Walton suggests, and seeing Picasso’s ‘Guernica’ not as a painting but as a guernica. ‘Guernica’ seen as a painting seems vivid and jarring; ‘Guernica’ seen as a guernica, he proposes, would seem serene and peaceful. The point does not hold for far-fetched categories like guernicas only, but for our own artistic categories as well. Eliot’s ‘Preludes’ seems rich when read as a poem, but rather sparse when read as a novel.As any artwork can in principle belong to indefinitely many categories and, consequently, be experienced as having almost any set of aesthetic properties, Walton further argues for a correctness condition constraining appreciation: an artwork only actually has those properties that it seems to have when experienced in a correct category. His account, in other words, really comprises two separate components (see Laetz 2010 for discussion): a psychological component—pertaining to the mechanisms by which a perceiver’s operative background assumptions (whether explicit or tacit) and, hence, their approach to an artwork, causally impacts their perceptual and affective responses to it—and a metaphysical component—pertaining to the role of the correct-category condition in determining which aesthetic seemings constitute the actual aesthetic state of affairs. It is the psychological component of the account that interests us here.116Given the psychological fact that our responses to artworks are sensitive to the categories in which we experience them, and given that categories in which we experience a work (in the broad sense encompassing both the ‘correct’ and the ‘incorrect’ categories we might experience it in) can in principle be constructed in indefinitely many ways, there is room for us as perceivers to exercise some control over the aesthetic appearances that are realized in our experiences of artworks. That is, by wilfully manipulating the categories or contrast classes we bring to bear in our perception of an artwork, we can indirectly manipulate our own responses to it. Such control over our own responses is subject to some real constraints (after all, it is difficult for us to bring ourselves to actually see ‘Guernica’ as a guernica), but it is not negligible. What it illustrates is the broader point that our affective responses have a surprising measure of flexibility, which can be exploited in order to deliberately modulate them as an episode of appreciative engagement unfolds. Walton (1970: 362) hints at this flexibility, and the wealth of possibilities it implies for artists and appreciators able to exploit it, when he notes that ‘[works] may be fascinating precisely because of shifts between equally permissible ways of perceiving them.’With a little reflection it becomes apparent that the capacity for modulating our own responses is not limited to just what is possible through varying the categories we bring to bear in our perceptual engagement with a stimulus. All manner of exercises of cognitive control have affective upshots—the phenomenon of processing fluency illustrates this (see Section 4.4), but there are many other mechanisms with similar and sometimes more pronounced effects. We often exploit these mechanisms as means for intervening in our own responses. Take for example a case in which one actively recalls a memory that has salient similarities with an ongoing aesthetic encounter (similarities, for example, in terms of 117content, or in affective valence or arousal properties) in order to consolidate or extend the occurrent affective states elicited in the encounter. Other strategies are more direct. Recall Wilson and Dunn’s (1985) study that showed deliberate introspective focussing on an attitude (rather than reflective reasoning about it) can have positive downstream effects on its capacity to influence our choices by making it more salient (see Section 4.3.1). Think, for another example, of cases where shifts in attention to an object are conjoined with deliberate monitoring of the effects these shifts have on our perceptual and affective responses, with some of the shifts chosen precisely on the basis of sensitivity to its effects (as with our capacity to sometimes manipulate what we see when we look at a bistable perceptual stimulus like the duck-rabbit).These cases are all examples of a particular type of exercise of cognitive agency that is commonplace in appreciative practices. It is a type in which agency is not in the first place exercised for the sake of the direct epistemic outcome of the cognitive act undertaken, with the resulting affective modulations mere incidental side-effects. It is an exercise of agency, rather, in order to effect changes in the affective responses that attend a cognitive act. That is, it is under the normative guidance of an aim to modulate one’s own responses.It is not difficult to see how such an aim could arise in the context of an appreciative practice. Appreciative practices are constituted by norms that determine the values of the products of the practice. But by determining those items’ values, the norms also determine the aptness conditions for affective and other evaluative responses to them. Thus, the practice imposes on participants the ideal of responding affectively to those items in certain specified ways. The main way to approximate this ideal is, through painstaking habituation, to condition our affective dispositions so that over time our automatic responses come to align 118with the value profile of the practice. But we can also take up the ideal more locally, as a normative aim in episodes of more direct wilful engagement with the flexibility in our occurrent affective responses. When we do this, we exercise cognitive agency appreciatively.Arguably such direct wilful engagement with our responses often forms part of the longer term process of conditioning our affective dispositions. This is one of the reasons why the capacity for appreciative engagement can play such a central role in many of our appreciative practices—by contributing to the shaping of participants’ affective dispositions, in which a practice’s value profile is partially grounded. But episodes of goal-guided modulation of our own responses are not always and only instrumental in longer term projects of affective habituation. We can, and do, engage in such episodes for any number of reasons. Moreover, while they neatly find a place within the context of the normative social bulwark of appreciative practices, the basic phenomenon they exemplify is by no means limited to this context. We likely often actively engage with our affective responses for reasons that do not derive from socially mandated normative constraints. It is a commitment of the current account that appreciative engagement as a psychological type is a widespread cognitive phenomenon that, while having an important role in our appreciative practices, precedes and outstrips them. The cognitive capacity for appreciation is a precondition for, rather than a result of, full participation in appreciative practices.At this point it is worth pausing to consider how this proposal meets the desiderata we set for a metacognition-invoking theory of appreciation at the end of the previous section. A first question is what function the account assigns to metarepresentation. The answer is straightforward: the type of exercise of cognitive agency that the account makes constitutive of appreciation is specifiable only in terms of the instrumental norm under which cognitive 119agency is exercised, and this norm has metarepresentational contents. Cognitive agency in appreciation is exercised in order to modulate our own responses—in other words, it is guided by intentions that make reference to cognitive-affective states of the subject represented as such.A second question is whether this kind of metarepresentation-mediated monitoring and guided modulation of our responses falls prey to the empirical challenge to our introspective capacities. To this the answer is no. Recall that the challenge pertains in particular to our ability to accurately identify the causes of our own responses, and to verbally articulate these causes. Consider now the task of using cognitive acts (like the ones our examples contained: deliberate recall, introspective focussing, varying categories brought to bear in perception, etc.) in order to influence our affective responses. No theory about how our responses are caused is implicated, no causal judgements required, only the monitoring of ongoing affective states as such and exercises of executive control in a way that is sensitive to the results of that metarepresentational monitoring. Notice also that none of this requires verbally mediated reasoning (in the previous section we noted the independence of mindreading and language capacities). Nor does the involvement of goal states of the agent with metarepresentational contents imply anything overly intellectual. Nothing in the current account even limits appreciative episodes to ones in which those goals are represented at the personal level as conscious intentions.In addition, the account does not need our control of our own responses to be particularly precise or efficient. Our responses may be malleable, but their malleability also has substantial limits, and our means for taking advantage of this flexibility (means such as deliberate changes in attention, manipulating categories brought to bear in perception, etc.), 120are indirect and imprecise. This is not a strike against the proposal, however, because our own affective responses’ opacity and seeming intransigence are among the very reasons why we engage with them actively. Thus the current picture resonates with Melchionne’s suggestion (Section 4.4) that an adequate account of our aesthetic lives leaves ‘room for confusion, indifference, and transience in our aesthetic preferences.’ There is no reason to think, then, that the current account falls to Dokic’s charge that it would over-intellectualize appreciation.It is in the active tracking and the deployment of cognitive strategies in order to exercise influence over, modulate, and produce downstream responses (often, but not only, in accordance with internalized normative commitments derived from appreciative practices), that metacognition finds its role in appreciation. The measure of control we exercise over our own responses forms a natural part of a story about why they are suited to the normative regulation to which they become subject in appreciative practices. When we engage in appreciation, our responses are moulded by sensitivity to norms—sometimes socially mandated, sometimes taken up for our own idiosyncratic reasons. These norms play their regulative role by coopting metarepresentational resources in our mental economies.121CHAPTER FIVE   Conclusion5.1  Assembling the piecesWe began, in the introduction, with the question of how to understand the especially close connection between appreciation and aesthetic value—a question, as should now be apparent, that this dissertation does not intend to answer. It was posed there, rather, as prelude to a case for how not to answer it, namely in terms of a reductive link between the values of appreciative experiences and the aesthetic values of their objects. Rejecting this overly rigid and straightforward construal of the link implicit in aesthetic hedonism has allowed us, in the intervening chapters, to develop a richer picture of (some of) the sophisticated cognitive capacities underlying our appreciative encounters with the world.In Chapter 2, the stage was cleared for the work of later chapters by laying out the arguments opposing aesthetic hedonism dispersed throughout the current aesthetics literature. The dispute between hedonists (or value empiricists) and their detractors is very much a live one, but variance in how different participants interpret the minutiae, as well as variety among hedonist views and their self-conceptions—especially with regard to how aesthetic value empiricism fits into the broader hedonist research programme—can make it challenging to assess the overall state of play in the debate. Chapter 2 addressed this problem by, firstly, clarifying the important but overlooked connection between current versions of 122value empiricism in aesthetics, and preference hedonism elsewhere in value theory, and secondly, systematizing extant arguments against aesthetic hedonism in a taxonomy of six major argumentative strategies. In so doing, the chapter developed the most comprehensive overview available of the challenges aesthetic hedonism faces, and of what might be required to overcome those challenges. More pertinently, in the context of this dissertation, by making clear the scope of aesthetic hedonism’s difficulties, the chapter undermined a large part of the motivation to simply default to a hedonic account of (aesthetic) appreciation.Chapter 3 set out to show that there are untapped resources in recent work on dynamic mental phenomena for understanding the notion of an appreciative mode of attention. It was argued that George Dickie’s famous criticisms of the disinterested mode of attention fall short of the mark, but that they offer the lesson that a change in motivational profile can be sufficient for a change in mode of attention only if it makes a difference to how an attentional episode unfolds. Based on this lesson from Dickie, the chapter then imported tools from the philosophy of games, and argued that the striving mode of game play offers a better motivational analogy for the disinterested mode of attention than intrinsic play, which is the natural match for standard construals of disinterested attention in aesthetics. An analysis of the structure of action guidance operative in striving play was then used to develop a new model of appreciative attention.Chapter 4 argued that a common practice in philosophical aesthetics, of appealing to complex second-order mental phenomena on merely intuitive grounds or for the normative functions such phenomena might fulfil in a theory of aesthetic appreciation, can be improved upon if we let empirical work on human metacognitive capacities inform our theorizing from the outset instead. The chapter illustrated this approach by, firstly, arguing that the social 123psychology literature on the limits of our capacities for direct introspective insights into our own reasons rules out a role for metacognition in aesthetic reasoning. Secondly, the literature on the ‘metacognitive’ functions of processing fluency was used to refine the question about metacognition’s role in appreciative engagement: the kinds of metacognitive phenomena that a substantive hypothesis about the role of second-order thought in appreciation should target, are the conceptual metacognitive capacities grounded in the mindreading system. Although much more can likely be said about the place of self-directed mindreading in our aesthetic lives and in appreciative practices, the chapter developed a proposal about one role it likely plays in many appreciative episodes—the role of exercising a measure of deliberate influence over the shape of our own perceptual and affective responses as our engagement with objects of appreciation transpires.Instead of attempting a complete, self-contained general theory of appreciation, the strategy in this dissertation has been to use tools from adjacent disciplines and sub-disciplines to illuminate two perennial but imperfectly understood ideas in aesthetics, about specific cognitive phenomena that underpin our capacity for appreciative engagement. By not letting fixed preconceptions about the axiological roles of these phenomena constrain the investigation, and by construing their potential relevance to our lives broadly—as transcending the aesthetic domain—Chapters 3 and 4 were able to approach the appreciative mode of attention and the role of metacognition in appreciation, as basic human capacities that we can learn about unimpeded by arcane concerns with the demarcation of the aesthetic. Moreover, this approach allowed this dissertation to draw attention to previously untapped connections between enduring ideas in aesthetics and recent developments in empirical and theoretical work on dynamic mental phenomena.1245.2  Directions for future researchThe picture of appreciative engagement sketched in the preceding chapters dovetails with a broader research agenda at the intersection of philosophy of mind and action that stresses the importance of active and perdurant mental phenomena for our overall understanding of the mind (for recent examples of work that pushes this agenda, see the volumes edited by O’Brien & Soteriou 2009, Stout 2018, Brent forthcoming, as well as a monograph by Mole 2016). Indeed, this dissertation’s focus on how appreciative episodes unfold over time, and on how exercises of cognitive control help shape those episodes, is partly motivated by sympathy with that agenda’s emphasis on the dynamics of cognition. Consequently, several promising avenues for future research in the extension of the ideas developed here involve leveraging connections with ongoing work on dynamic mental phenomena and mental action. This section briefly outlines three of these avenues.5.2.1 Aesthetic inferenceThere is a longstanding puzzle in aesthetics concerning how to square our critical practice of citing lower-level, non-aesthetic features of items as justification for our aesthetic judgements, with the widely accepted view that aesthetic judgements are deeply perceptual in an important sense and, thus, non-inferential (see Isenberg 1949, Sibley 1965, 2001). Rationalists claim that we should allow inference to play a role in how we arrive at our aesthetic judgements (e.g. Dorsch 2013); perceptualists claim that we should explain our practice of giving reasons for our judgements in some other way (e.g. Hopkins 2007, Lord 125forthcoming). One central motivation behind the perceptualist position is the datum that aesthetic judgement is psychologically immediate—that is, despite all of the cognitive work we go through in the run-up to some aesthetic judgements, they ultimately occur instantaneously, like sensory judgements, without being preceded by any kind of ‘movement in thought’ from consciously entertained premises to a conclusion.The proposal outlined in Chapter 4, on which our appreciative encounters with aesthetic goods involve deliberate exercises of cognitive agency aimed at modulating and exploring our own aesthetic responses, taken together with recent work in philosophy of mind on the dynamic and agential nature of conscious inference (Mole 2018, Peacocke ms.b), suggests a new way of thinking about the place of inference in aesthetic cognition. The aim of a first of three upcoming research projects will be to flesh out the details of this account of ‘aesthetic inference’.For a start, notice that perceptualism’s prohibition on inferences from lower-level, non-aesthetic features of items judged does not take into account a different source of data that might feature as premises in inferences to aesthetic judgement, namely, features of our own affective responses. The idea is that, in many contexts, the process of deciding one’s evaluative commitments with regard to an item of aesthetic interest could be a matter of inferring the contents of aesthetic judgements—not from lower-level features of the item—but from introspectable features one’s own responses to the item.Antonia Peacocke’s (ms.b) recent account of conscious inference provides a useful piece of the puzzle for making good on this proposal. Peacocke considers the struggles of standard views of conscious inference with capturing the nature of the inferential connection between the judgements that have the premises as contents, and the judgement with the 126conclusion as its content. Her diagnosis is that these struggles are rooted in a failure to recognize that token mental actions can have multiple contents. Thus, the same token mental action (a particular judgement) can instantiate two mental action types (the judgment that p and the judgement that q). This allows her to say that conscious inferences are single token judgements executed under complex intentions of the form ‘determining whether p as a means to determining whether q’, such that if the intention succeeds, the very token judgement that p also counts as a judgement that q.Setting aside the intricacies of the Peacocke’s account, notice how it looks when applied to the kind of inferences we are considering—that is, inferences from our own affective responses to aesthetic judgements of the items that elicit those responses. Firstly, the account represents such inferences as the upshot of a complex intention to figure something out about the object of aesthetic interest by means of figuring something out about our own responses. This fits well with the proposal in Chapter 4, that much of what we do in appreciative contexts is to actively engage with and modulate our own responses. Such engagement appears well suited to the execution of the part of the inferential intention under the description of ‘figuring something out about our own responses’. Secondly, Peacocke’s account of inference does not require any kind of ‘movement in thought’ between two token judgements, but allows inferences to occur instantaneously, as the upshot of an arrival at one token judgement with dual contents—say, some non-aesthetic contents pertaining to my affective responses, and an aesthetic content pertaining to the object that elicited those responses. This allows for the type of psychological immediacy in inference that could help capture the aesthetic perceptualist’s intuition that aesthetic 127judgement is never a matter of a movement from consideration of premises to a consecutive acceptance of a conclusion.Thirdly, one of the major benefits of Peacocke’s move to think of inference as an intentional action is that this secures the minimal link between conclusion and premises necessary for the occurrence of an inference in terms that require no reference to rules or general principles. All of the heavy lifting in the account is done by a cognitive agent’s practical conception of their own inquiry, such that the complex intention—to figure out whether p as a means to figuring out whether q—renders the judgement that p an inference. Of course, in many cases, the inferential intention might be formed under guidance of some general rule or principle that connects the content p with the content q. But this is not required for the intention to ground an actual inference—we might have completely situation-specific, non-general reasons for believing p to be good evidence for q in a particular context, and infer q on that basis. In Peacocke’s own words, the account allows for cases of ‘particularist inference’.Finally, the proposal that aesthetic inferences operate on data from our own affective responses makes for an interesting new reading of the Sibleyan idea that, when we cite an object’s low-level features as reasons for our aesthetic judgements, we are providing ‘perceptual proofs’ for our judgements, in the sense of guiding others to perceive as we do (Sibley 1965: 145). When a critic cites the lower-level features of an item in conjunction with a certain description of the aesthetic judgement they are looking to support, they are (1) giving an appreciator reasons to inspect their own responses centred on the features cited, and (2) proposing that they do so under some complex intention that, by the appreciator’s own lights, could suffice to ground an inference to the aesthetic judgement proposed. This 128reading of the notion of a ‘perceptual proof’ has several merits. Most notably, it offers a resolution of the tension between perceptualism and rationalism. It makes the activity of criticism rational in the sense of guiding rationally evaluable inferences. And it also makes it deeply perceptual—in the sense of allowing aesthetic inferences to be grounded only in context-specific data from the appreciator’s own perceptual and affective responses to the object of appreciation, considered in light of their own autonomously formed aesthetic commitments.5.2.2 Action in aesthetic perceptionTo be clear, the proposal just canvassed, that we sometimes infer aesthetic judgements, does not rule out that at other times (perhaps most of the time!) we directly perceive aesthetic properties in some meaningful sense of ‘perceive’, without the involvement of conscious inference. In other words the current perceptualist programme, of elucidating what aesthetic perception consists in, remains a live one even if inference does turn out to sometimes play a role in aesthetic judgement. Over the last decade, debates in this research programme have focussed especially on high-level perceptual contents and the question of whether aesthetic properties can be represented in perception (Siegel 2006, Stokes 2014, 2018, Logue 2018, Majeed 2018, Oddie 2018, Lord forthcoming) and, more recently, on the role of perceptual learning in making aesthetic perception possible (Connolly 2019, Stokes forthcoming, Ransom ms.). A second project for upcoming research extends this dissertation’s focus on the active nature of appreciative engagement by asking about potential roles for action in aesthetic perception.129One possible way to conceive the relation between action and aesthetic perception is purely instrumental. We act to gain more information about a target of appreciation or to better situate ourselves with respect to it, clearing the way for our aesthetic sensibilities to do the real work of generating the aesthetic contents of experience independently and automatically. Deciding whether this instrumental picture is the right way to construe action’s role in aesthetic experience is an analogous question to the question in philosophy of perception, of whether action ever plays a non-instrumental role in low-level perceptual experience. On the ‘traditional view’ of perception—what Susan Hurley (1998: 342) calls the ‘Input-Output Picture’—action’s role in perception is merely causal and instrumental: exploratory actions change the inputs to our perceptual systems and thereby cause changes in what we perceive, but such actions play no part in how the perceptual systems fulfil their proprietary functions. Action-based theories, on the other hand, claim that perceptual experience can depend on action in non-instrumental or constitutive ways (for an overview of action-based theories of visual perception, for example, see Grush & Briscoe 2017).Given this structural analogy with the debate about the action-perception relation, we should expect to find lessons in philosophy of perception for investigating hypotheses about action’s role in aesthetic experience. Care should be taken with extracting such lessons, however, both because there are real limits to the analogy and because theorists in the philosophy and cognitive science of perception still remain divided on a wide array of issues. One potential point of disanalogy is that no functionally unified cognitive system likely undergirds aesthetic perception in the way that, say, the visual system undergirds visual perception. For this reason, it will not work to transplant general claims about action’s role in a sense modality, or even in perception at large, uncritically to aesthetic perception. Useful 130clues will more likely come from detailed work on the full variety of connections between particular types of action and specific functions of the individual sense modalities.In recent work on the modality of haptic touch, Matthew Fulkerson (2013: 76) advocates a similarly piecemeal approach in the philosophy of perception:The kinds of action-involving relations that occur in touch may not occur in the sense of taste, for instance. The best way to move forward on the debate, I think, is to better understand the many detailed relations between action and perception in the individual modalities […].His own account of touch offers a rich picture of how specific features of tactual experience depend on particular exploratory actions. The sense modality of touch is also a natural place to look for ways in which actions might contribute constitutively to the generation of low-level perceptual contents. This makes Fulkerson’s work a good point of departure for thinking about specific roles that particular kinds of action might play in higher-level aesthetic perception. Central among the actions that play mediating roles in tactual experience, for instance, Fulkerson (2013: 54–58) discusses a set of standardized hand-movements called exploratory procedures (or EP’s for short). Lederman & Klatzky (1987) and Jones & Lederman (2006) have shown that different EP’s are attuned to the detection of different tangible properties and Fulkerson further suggests that EP’s may play differential roles in perceptual functions like haptic feature binding, perceptual grouping, segmentation, and object recognition.One promising way to flesh out the idea that action might play a constitutive role in some aesthetic perception is in terms of a notion of aesthetic EP’s. The thought is that, just as 131EP’s structure haptic experience, the exploratory aspects of our engagement with objects of aesthetic interest will likely display some systematic structure, and we can get at this structure by identifying and analysing types of activities and actions typically performed in episodes of aesthetic appraisal and appreciation. We might also expect that, if there are activities with such roles, many of them may well be specific to the types of objects of aesthetic appreciation they target. Thus, the plan is to focus specifically on artistic media that prompt bodily involvement in audiences, such as music (Judge 2015, 2019) and dance (Montero 2013). Our bodily responses to rhythm (such as foot-tapping and swaying) look like exactly the sort of active engagement that could yield access to aesthetically valuable features of music in a way analogous to how haptic EP’s yield access to the tactile properties of the world. Ultimately, however, the hope is to draw lessons that might apply not only to to gross motor actions, but also instances of cognitive action of the kind discussed in Chapter 4, such as covert shifts in attention to explore the aesthetic affordances of an object of appreciation.5.2.3 Appreciation and curiosityA recurring theme in work on aesthetic experience and appreciation, and foregrounded in parts of this dissertation (see Chapters 3 and 4), is the idea that it involves autotelic or self-sustaining motivation (Schaeffer 2015; Dokic 2016, Matthen 2017, 2018, Shelley ms.) or that it is fundamentally ‘open-ended’ in some other way (Nehamas 2007, Kulvicki 2015). A third line of inquiry for future research emerges from the question of how the open-ended 132character of aesthetic appreciation relates to other cognitive phenomena that might aptly be characterized as self-sustaining or open-ended.One potential example, briefly touched on in Chapter 3, is the phenomenon of mind-wandering, which has been receiving increased scrutiny in philosophy of mind in recent years (Christoff 2016, Irving 2016, Mills 2018, Metzinger 2013a, 2013b). But while interestingly open-ended, mind-wandering’s dispersed and passive aspects make for substantive disanalogies with the focussed and active character of appreciative engagement. A more promising ‘open-ended’ mental kind to target for comparison and joint investigation alongside appreciation, is curiosity and the related inquisitive attitudes involved in motivating and guiding exploratory cognitive behaviours.There has been a flurry of recent interest in curiosity-related phenomena across disciplinary boundaries. Philosophers often approach questions about the nature and value of curiosity by focussing on its manifestations in cognitively sophisticated, language-mediated, and socially complex behaviours (for example Inan 2013, Ross 2018). There are important questions about curiosity’s role in such behaviours: how does it feature, for example, in responsible epistemic agency, in self-regulated learning, and in endeavours like scientific inquiry? But an alternative approach has also recently been gaining traction—one that treats curiosity as a primitive, pre-conceptual cognitive capacity that we share with many other sentient creatures (see Whitcomb 2010, Friedman 2013, Dickie 2015: Chapter 3, Carruthers 2018). This approach dovetails with work on curiosity across several other disciplines (see for example Schulz 2012 on curiosity in preverbal infants, Kidd & Hayden 2015 for a review of work in psychology and neuroscience, as well as the literature that conceives of curiosity 133and other ‘noetic’ or epistemic feelings in terms of processing fluency, as discussed in Chapter 4).In addition, a particularly interesting new source of insight is work in artificial intelligence research, which has been making exciting advances on the task of designing artificial agents that act in ways analogous to natural agents’ exploratory (i.e. curiosity-induced) behaviour. Hester & Stone (2017), for example, recently published a paper on a reinforcement learning algorithm that succeeds at ‘curiously’ exploring a domain without having been given any task in that domain, and is then able to effectively use the information gleaned through exploration to perform tasks assigned in the domain later. For other recent examples of breakthroughs in ‘artificial curiosity’, see Pathak et al. (2017) and Savinov et al. (2018). What makes the AI literature especially interesting is that it emphasizes the need for improved ways to model and operationalize inquisitive motivation and behaviours.These developments present potential opportunities for learning about the motivational profile underlying the autotelic or open-ended appreciative engagement common in our interactions with aesthetic goods. Surprisingly, while related cognitive attitudes like wonder and amazement have received some attention in aesthetics (for a recent example, see Fingerhut & Prinz 2018), a cursory search suggests that not much has been written on connections between aesthetic experience or appreciation and curiosity (although Dokic’s 2016 epistemic feeling account of aesthetic experience, discussed in Chapter 4, is a notable exception). This third trajectory for future research will explore connections between appreciation, aesthetic cognition, and the various developments in the emerging cross-disciplinary literature on curiosity.134BibliographyAlter, A. L., and Oppenheimer, D. M. 2009. Uniting the Tribes of Fluency to Form a Metacognitive Nation, Personality and Social Psychology Review 13/3: 219–35.Aydede, M. 2014. How to Unify Theories of Sensory Pleasure: An Adverbialist Proposal, Review of Philosophy and Psychology 5:119–133.Bargh, J. A. 1990. Goal ≠ Intent: Goal-Directed Thought and Behavior Are Often Unintentional, Psychological Inquiry 1/3: 248–77.Beran, M., Brandl, J., Perner, J., and Proust, J. (eds.). 2012. Foundations of Metacognition. New York: Oxford University PressBerleant, A. 2016. Re-thinking Aesthetics. New York: Routledge.Bhattacharyya, K. C. 2011. The Concept of Rasa, in Indian Philosophy in English: From Renaissance to Independence, eds. N. Bhushan and J. L. Garfield. Oxford: Oxford University Press, 195–206.Bommarito, N. 2013. Modesty as a Virtue of Attention, Philosophical Review, 122/1: 93–117.Bornstein, R. F., and D’Agostino, P. R. 1992. Stimulus Recognition and the Mere Exposure Effect, Journal of Personality and Social Psychology 63/4: 545–52.Bos, M. W., Dijksterhuis, A and Van Baaren, R. B. 2008. On the goal-dependency of unconscious thought, Journal of Experimental Social Psychology 44/4: 1114–20.Brent, M. (ed.). Forthcoming. Mental Action and the Conscious Mind. Routledge.Budd, M. 1985. Music and the Emotions: The Philosophical Theories. London: Routledge.Budd, M. 1995. Values of Art: Painting, Poetry, and Music. London: Penguin.135Budd, M. 2008. Aesthetic Essence, in his Aesthetic Essays. Oxford: Oxford University Press, 31–47.Bullough, E. 1912. ‘Psychical Distance’ as a Factor in Art and an Aesthetic Principle, British Journal of Psychology 5: 87–117. Carlson, A., and Parsons, G. 2008. Functional Beauty. Oxford: Clarendon Press.Carroll, N. 2002. Aesthetic Experience Revisited, British Journal of Aesthetics 42: 145–68.Carroll, N. 2012. Recent Approaches to Aesthetic Experience, Journal of Aesthetics and Art Criticism 70/2:165–177.Carruthers, P. 2011. The Opacity of Mind. An integrative theory of self-knowledge. Oxford: Oxford University Press.Carruthers, P. 2014. Two Concepts of Metacognition, Journal of Comparative Psychology 128/2: 138–9.Carruthers, P. 2016. Are epistemic emotions metacognitive? Philosophical Psychology 30/1-2: 58–78.Carruthers, P. 2018. Basic Questions, Mind and Language 33:130–147.Carruthers, P., Fletcher, L., and Ritchie, J. B. 2012. The evolution of self-knowledge, Philosophical Topics 40: 13–37.Chevalier, N.,  Martis, S. B., Curran, T., and Munakata, Y. 2015. Metacognitive Processes in Executive Control Development: the case of reactive and proactive control, Journal of Cognitive Neuroscience 27/6: 1125–36.Christoff, K., and Fox, K. C. R. (eds.). 2018. The Oxford Handbook of Spontaneous Thought: Mind-Wandering, Creativity, and Dreaming. Oxford: Oxford University Press,Christoff, K., Gordon, A. M., Smallwood, J., Smith, R., & Schooler, J. W. 2009. Experience sampling during fMRI reveals default network and executive system contributions to 136mind wandering, Proceedings of the National Academy of Sciences USA 106/21, 8719–24.Christoff, K., Irving, Z. C., Fox, K. C. R., Spreng, R. N., and Andrews-Hanna, J. R. 2016. Mind-wandering as spontaneous thought: a dynamic framework, Neuroscience 17: 718–31.Chung, J. 2018. Moral Cultivation: Japanese Gardens, Personal Ideals, and Ecological Citizenship, Journal of Aesthetics and Art Criticism 76/4: 507–18.Chung, J. 2019. The Oneness Hypothesis and Aesthetic Obligation, Philosophy and Phenomenological Research 99/2: 501–7.Connolly, K. 2019. Perceptual Learning: The Flexibility of the Senses. Oxford: Oxford University Press.Couchman, J. J., Beran, M. J., Coutinho, M. V. C., Boomer, J., and Smith, J. D. 2012. Evidence for animal metaminds, in Foundations of Metacognition, eds. M. J. Beran, J. Brandl, J. Perner, and J. Proust. New York: Oxford University Press, 21–5.Crisp, R. 2006. Hedonism Reconsidered, Philosophy and Phenomenological Research 73/3: 619–45.Crisp, R. 2007. Sidgwick’s Hedonism, in Proceedings of the World Congress on Henry Sidgwick: Happiness and Religion, ed. P. Buccolo, R. Crisp, and B. Schultz. Catania: University of Catania, 104–57.Cross, A. 2017a. Art Criticism as Practical Reasoning, British Journal of Aesthetics 57/3: 299–317.Cross, A. 2017b. Obligations to Artworks as Duties of Love, Estetika 54/1: 85–101.Cutting, J. E. 2003. Gustave Caillebotte, French Impressionism, and mere exposure, Psychonomic Bulletin and Review 10/2: 319–43.137Cutting, J. E. 2017. Mere Exposure and Aesthetic Realism: a response to Bence Nanay, Leonardo 50(1): 64–6.Davidson, D. 1963. Actions, reasons, and causes, Journal of Philosophy 60/23: 685–700.Davies, D. 2004. Art as Performance. Oxford: Blackwell.Davies, S. 1994. The Evaluation of Music, in What is Music?, ed. Philip Alperson, University Park, PA: Pennsylvania State University Press, 307–25.De Clercq, R. 2002a. The Concept of an Aesthetic Property, Journal of Aesthetics and Art Criticism 60/2: 167–76.De Clercq, R. 2002b. Two Conceptions of Response-Dependence, Philosophical Studies 107/2: 159–77.Debus, D. 2015. Losing Oneself: On the Value of Full Attention, European Journal of Philosophy 23/4: 1174–91.Dickie, G. 1964. The Myth of the Aesthetic Attitude, American Philosophical Quarterly 1/1: 56–65.Dickie, G. 1988. Evaluating Art. Philadelphia: Temple University Press.Dickie, I. 2015. Fixing Reference. Oxford: OUP.Dodd, J. 2014. On a Proposed Test for Artistic Value. British Journal of Aesthetics 54/4: 395–407.Dokic, J. 2012. Seeds of self-knowledge: noetic feelings and metacognition, in Foundations of Metacognition, eds. M. J. Beran, J. Brandl, J. Perner, and J. Proust. New York: Oxford University Press, 302–19. Dokic, J. 2016. Aesthetic Experience as a Metacognitive Feeling? A Dual-Aspect View, Proceedings of the Aristotelean Society 116/1: 69–88.Dorsch, F. 2007. Sentimentalism and the Intersubjectivity of Aesthetic Evaluations, Dialectica 61/3: 417–46.138Dorsch, F. 2013. Non-Inferentialism About Justification - the case of aesthetic judgements, Philosophical Quarterly, 63: 660–82.Elster, J. 1983. Sour Grapes: Studies in the Subversion of Rationality. Cambridge: Cambridge University Press.Fazio, R. H.; Chen J.; McDonel, E.; and Sherman, S. J. 1982. Attitude accessibility, attitude-behaviour consistency, and the strength of object-evaluation association, Journal of Experimental Social Psychology 18: 339–57.Feagin, S. 1983. The Pleasures of Tragedy, American Philosophical Quarterly 20/1: 95–104.Fingerhut, J. and Prinz, J. J. 2018. Wonder, Appreciation, and the Value of Art, Progress in Brain Research, 237: 107–128.Forsey, J. 2017. Aesthetic Experience, Aesthetic Value, Estetika 54/2: 175–88.Frankfurt, H. 1971. Freedom of the Will and the Concept of a Person, Journal of Philosophy 68: 5–20.Frankfurt, H. 1978. The problem of action, American Philosophical Quarterly 15/2: 157–62.Fridland, E. 2017. Automatically Minded. Synthese 194/11: 4337–63.Friedman, J. 2013. Question-directed attitudes, Philosophical Perspectives 27: 145–174.Fulkerson, M. 2013. The First Sense: A Philosophical Study of Human Touch. Cambridge, MA: MIT Press.Goldie, P. 2007. Towards a Virtue Theory of Art, British Journal of Aesthetics 47/4: 372–87.Goldie, P. 2008. Virtues of Art and Human Well-Being, Aristotelian Society Supplementary Volume 82/1: 179–95.Goldman, A. 1995. Aesthetic Value. Boulder: Westview Press.Goldman, A. 2006. The Experiential Account of Aesthetic Value, Journal of Aesthetics and Art Criticism 64/3: 333–42.139Goldman, A. I. 2004. Epistemology and the evidential status of introspective reports, Journal of Consciousness Studies 11/7: 1–16.Gopnik, A. 1993. How We Know Our Own Minds: The Illusion of First-Person Knowledge of Intentionality, Behavioral and Brain Sciences 16: 1–14.Gorodeisky, K. 2018. Rationally Agential Pleasure: A Kantian Proposal, in Pleasure: A History, ed. L. Shapiro. Oxford: Oxford University Press, 167–194.Gorodeisky, K. 2019. On Liking Aesthetic Value, Philosophy and Phenomenological Research. Advance online publication,, R. and Briscoe, R. 2017. Action-Based Theories of Perception, in The Stanford Encyclopedia of Philosophy, Spring 2017 Edition, ed. E. N. Zalta, (First version published in Fall 2015, accessed 4 December 2019),, J. 2001. The emotional dog and its rational tail: A social intuitionist approach to moral judgment, Psychological Review 108/4: 814–34.Hall, L., and Johansson, P. 2008. Using Choice Blindness to Study Decision Making and Introspection, in Cognition: A Smorgasbord of Cognitive Science, eds. P. Gärdenfors, and A. Wallin. Nora: Nya Doxa, 267–83.Hall, L., Johansson, P., Tärning, B., Sikström, S., and Deutgen, T. 2010. Magic at the marketplace: Choice blindness for the taste of jam and the smell of tea, Cognition 117: 54–61.Hamlin, J. K.; Wynn, K.; and Bloom, P. 2007. Social evaluation by preverbal infants, Nature 450: 557–9.Hampshire, S. 1970. Logic and Appreciation, in Problems in Aesthetics, ed. M. Weitz. London: Macmillan.Hampton, R. R. 2001. Rhesus Monkeys Know When They Remember, Proceedings of the National Academy of Sciences U.S.A 98: 5359–62.140Hanson, L. 2013. The Reality of (Non-aesthetic) Artistic Value, Philosophical Quarterly 63/252: 492–508.Hester, T. and Stone, P. 2017. Intrinsically motivated model learning for developing curious robots, Artificial Intelligence 247: 170–86.Hopkins, R. 2000. Beauty and Testimony, Royal Institute of Philosophy Supplement 47: 209-236.Hopkins, R. 2004. Critical Reasoning and Critical Perception, in Knowing Art, ed. M. Kieran, and D. M. Lopes. Dordrecht: Springer, 137–153.Hopkins, R. 2011. How to Be a Pessimist about Aesthetic Testimony, Journal of Philosophy, 108/3: 138-157.Huddleston, A. 2012. In Defense of Artistic Value, Philosophical Quarterly 62/249: 705–14.Hume, D. 1777. Of the Standard of Taste. Four Dissertations. London: Millar, 227–49.Hurka, T. 1998. Two Kinds of Organic Unity, Journal of Ethics 2: 299–320.Hurka, T. 2001. Virtue, Vice, and Value. Oxford: Oxford University Press.Hurley, S. L. 1998. Consciousness in Action. Cambridge, MA: Harvard University Press.Inan, I. 2013. The Philosophy of Curiosity. New York: Routledge.Irvin, S. 2014. Is aesthetic experience possible? in Aesthetics and the Sciences of the Mind, eds. G. Currie, M. Kieran, A. Meskin, and J. Robson. Oxford: Oxford University Press, 37–57.Irving, Z. C. 2016. Mind-wandering is unguided attention: accounting for the ‘purposeful’ wanderer, Philosophical Studies 173: 547–71.Iseminger, G. 2004. The Aesthetic Function of Art. Ithaca, NY: Cornell University Press.Iseminger, G. 2005a. Aesthetic Experience, in The Oxford Handbook of Aesthetics, ed. J. Levinson. Oxford: Oxford University Press, 99–116.141Iseminger, G. 2005b. The Aesthetic State of Mind, in Contemporary Debates in Aesthetics and the Philosophy of Art, ed.Matthew Kieran. Oxford: Blackwell, 98–112.Isenberg, A. 1949. Critical Communication, reprinted in Aesthetics and the Philosophy of Criticism, eds. Callaghan, W., Cauman, L., Hempel, C., Morgenbesser, S., Mothersill, M., Nagel, E., and Norman, T. [1973], 156–71.Johansson, P., Hall, L., Sikström, S., and Olsson, A. 2005. Failure to Detect Mismatches between Intention and Outcome in a Simple Decision Task, Science 310: 116–19.Johansson, P., Hall, L., Sikström, S., Tärning, B., and Lind, A. 2006. How something can be said about telling more than we can know: On choice blindness and introspection, Consciousness and Cognition 15: 673–9.Jones, L. A., & Lederman, S. J. 2006. Human hand function. New York: Oxford University Press.Judge, J. A. 2015. Looking at Sound: Reconciling philosophical and psychological approaches to musical experience (Unpublished doctoral dissertation). University of Cambridge, Cambridge, England.Judge, J. 2019. ‘Feeling the Beat’: Multimodal Perception and the Experience of Musical Movement, in The Philosophy of Rhythm: Aesthetics, Music, Poetics, eds. P. Cheyne, A. Hamilton, and M. Paddison. Oxford Oxford University Press.Kagan, S. 1998. Rethinking Intrinsic Value, Journal of Ethics 2: 277–97.Kahneman, D. 2011. Thinking, fast and slow. New York: Farrar, Straus and Giroux.Kant, I. 1790/2000. Critique of the Power of Judgement. Trans. P. Guyer and E. Matthews. New York: Cambridge University Press.Kemp, G. .1999. The Aesthetic Attitude, British Journal of Aesthetics 39/4: 392–99.Kidd, C. and Hayden, B. Y. 2015. The Psychology and Neuroscience of Curiosity, Neuron 88/3: 449–60.142Kieran, M. 2005. Revealing Art. London: Routledge.Kieran, M. 2008. Why Ideal Critics Are Not Ideal: Aesthetic Character, Motivation, and Value, British Journal of Aesthetics 48/3: 278–94.Kieran, M. 2009. The Vice of Snobbery: Aesthetic Knowledge, Justification, and Virtue in Art Appreciation, Philosophical Quarterly 60/239: 243–63.King, A. 2020. [Review of Being for Beauty: Aesthetic Agency and Value, by D. M. Lopes], British Journal of Aesthetics 60/1: 99–102.King, A. ms. Normativity and the Case of Aesthetics. Article manuscript in preparation.Kölbel, M. 2016. Aesthetic Judge-Dependence and Expertise, Inquiry 59/6: 589–617.Kornell, N. 2014. Where to draw the line on metacognition: a taxonomy of meta-cognitive cues, Journal of Comparative Psychology, 128/2: 160–62.Korsgaard, C. 1983. Two Distinctions in Goodness, Philosophical Review 92: 169–95.Kubala, R. 2017. Valuing and believing valuable, Analysis 77/1: 59–65.Kubala, R. 2018. Grounding Aesthetic Obligations, British Journal of Aesthetics 58/3: 271-85.Kubala, R. 2019. [Review of Being for Beauty: Aesthetic Agency and Value, by D. M. Lopes], Estetika 56/2: 250–62.Kulvicki, J. 2015. Analog Representation and the Parts Principle, Review of Philosophy and Psychology 6: 165–80.Laetz, B. 2010. Kendall Walton’s ‘Categories of Art’: A Critical Commentary, British Journal of Aesthetics 50/3: 287–306.Lakatos, I. 1970. Falsification and the methodology of scientific research programmes, in Criticism and the Growth of Knowledge, eds. Lakatos, I. and A. Musgrave. Cambridge: Cambridge University Press, 91–196.143Lamarque, P. 2010. Wittgenstein, Literature, and the Idea of a Practice, British Journal of Aesthetics 50/3: 375–88.Lawlor, K. 2003. Elusive Reasons: A Problem for First-Person Authority, Philosophical Psychology 16: 549–74.Lederman, S. J., & Klatzky, R. L. 1987. Hand movements: A window into haptic object recognition. Cognitive Psychology 19/3: 342–68.Levinson, J. 1992. Pleasure and the Value of Works of Art, British Journal of Aesthetics 32/4: 295–306.Levinson, J. 2002. Hume’s Standard of Taste: The Real Problem, Journal of Aesthetics and Art Criticism 60/3: 227–38.Levinson, J. 2005. Aesthetic Properties II, Proceedings of the Aristotelian Society. Supplementary Volumes 79: 211– 27.Levinson, J. 2010. Artistic Worth and Personal Taste, Journal of Aesthetics and Art Criticism 68/3: 225–33.Levinson, J. 2013. Reply to Riggle: Aesthetic History, Personality, and Profile, Journal of Aesthetics and Art Criticism 71/3: 281–82.Levinson, J. 2016. Toward an Adequate Conception of Aesthetic Experience, in Aesthetic Pursuits: Essays in Philosophy of Art. Oxford: Oxford University Press, 28–46.Lind, A., Hall, L., Breidegard, B., Balkenius, C., and Johansson, P. 2014. Speakers’ Acceptance of Real-Time Speech Exchange Indicates That We Use Auditory Feedback to Specify the Meaning of What We Say, Psychological Science 25/6: 1198–205.Lind, R. W. 1980. Attention and the Aesthetic Object, Journal of Aesthetics and Art Criticism 39/2: 131–42.Livingston, P. 2003. On an Apparent Truism in Aesthetics, British Journal of Aesthetics 43/3: 260-278.144Logue, H. 2018. Can We Visually Experience Aesthetic Properties? in Evaluative Perception, eds. A. Bergqvist and R. Cowan. Oxford: Oxford University Press, 42–55.Lopes, D. M. 2011. The Myth of (Non-Aesthetic) Artistic Value, Philosophical Quarterly 61/244: 518–36.Lopes, D. M. 2014a. Beyond Art. Oxford: Oxford University Press. Lopes, D. M. 2014b. Feckless Reason, in Aesthetics and the Sciences of the Mind, eds. G. Currie, M. Kieran, A. Meskin, and J. Robson. Oxford: Oxford University Press, 21–36.Lopes, D. M. 2015. Aesthetic Experts: Guides to Value, Journal of Aesthetics and Art Criticism 73/3: 235–246.Lopes, D. M. 2018. Being for Beauty: Aesthetic Agency and Value. Oxford: Oxford University Press.Lopes, D. M. 2019. Feeling for Freedom: K. C. Bhattacharyya on Rasa, British Journal of Aesthetics 59/4: 465–477.Lord, E. Forthcoming. The Nature of Perceptual Experience and the Rationality of Criticism, Ergo.Majeed, R. 2018. Do gestalt effects show that we perceive high-level aesthetic properties? Analysis 78/3: 440–450.Matthen, M. 2017. The Pleasure of Art, Australasian Philosophical Review 1/1: 6–28.Matthen, M. 2018. New Prospects for Aesthetic Hedonism, in Social Aesthetics and Moral Judgment: Pleasure, Reflection and Accountability, ed. J. A. McMahon. New York: Routledge, 13–33.Matherne, S. ms. Edith Landmann-Kalischer on Aesthetic Demarcation and Normativity. Article manuscript in preparation.Matherne, S., and Riggle, N. ms. Schiller on Freedom and Aesthetic Value, Part 1 and Part 2. Article manuscripts in preparation.145Matravers, D. 2005. Aesthetic Properties II, Proceedings of the Aristotelian Society. Supplementary Volumes 79: 191– 210.Melchionne, K. 2010. On the Old Saw ‘I know nothing about art but I know what I like’, Journal of Aesthetics and Art Criticism 68/2: 131–41.Mercier, H. and Sperber, D. 2009. Intuitive and reflective inferences, in In two minds: Dual processes and beyond, eds. J. Evans and K. Frankish. Oxford: Oxford University Press, 149–70.Meskin, A., Phelan, M., Moore, M., and Kieran, M. 2013. Mere Exposure to Bad Art, British Journal of Aesthetics 53/2: 139–64.Metzinger, T. 2013a. The myth of cognitive agency: subpersonal thinking as a cyclically recurring loss of mental autonomy, Frontiers of Psychology 4: Article 931.Metzinger, T. 2013b. Why are dreams interesting for philosophers? The example of minimal phenomenal selfhood plus an agenda for future research, Frontiers in Psychology 4: Article 746.Miller, G. A., Galanter, E., and Pribram, K. H. 1960. Plans and the Structure of Behavior. New York: Holt, Rinehart and Winston.Mills, C., Raffaelli, Q., Irving, Z. C., Stan, D., and Christoff, K. 2018. Is an off-task mind a freely-moving mind? Examining the relationship between different dimensions of thought. Consciousness and Cognition 58: 20–33.Mills, C. W. 2005. ‘Ideal Theory’ as Ideology, Hypatia 20/3: 165–84.Mole, C. 2006. Attention, Self, and The Sovereignty of Good, in Iris Murdoch: A Reassessment, ed. A. Rowe. Basingstoke: Palgrave Macmillan, 72–84.Mole, C. 2011. Attention is Cognitive Unison. An essay in cognitive psychology. Oxford: Oxford University Press.146Mole, C. 2017. Attention, in The Stanford Encyclopedia of Philosophy, ed. E. N. Zalta, (First version published in Fall 2009, accessed 2 December 2019),, C. 2016. The Unexplained Intellect: Complexity, time, and the metaphysics of embodied thought. London: Routledge.Mole, C. 2018. The Process of Inference, in Process, Action, and Experience, ed. Stout Oxford: Oxford University Press, 149–167.Montero, B. G. 2013. The Artist as Critic: Dance Training, Neuroscience, and Aesthetic Evaluation, Journal of Aesthetics and Art Criticism 71/2: 169–175.Moore, G. E. 1903. Principia Ethica. Cambridge: Cambridge University Press.Moran, R. 2012. Kant, Proust, and the appeal of beauty. Critical Inquiry 38/2: 398–329.Mothersill, M. 1961. Critical Reasons, Philosophical Quarterly, 11: 74–8.Mothersill, M. 1989. Hume and the Paradox of Taste, in Aesthetics: A Critical Anthology, ed. G. Dickie, R. Sclafani, and R. Roblin. New York: St Martin’s, 269–86.Murdoch, I. 1985. The Sovereignty of Good. London: Ark Paperbacks.Nagel, J. 2014. The Meanings of Metacognition, Philosophy and Phenomenological Research 89/3: 710–18.Nanay, B. 2016. Aesthetics as Philosophy of Perception. Oxford: Oxford University Press.Nanay, B. 2017. Perceptual Learning, the Mere Exposure Effect and Aesthetic Antirealism, Leonardo 50/1: 58–63.Nehamas, A. 2007. Only a Promise of Happiness: The Place of Beauty in a World of Art. Princeton: Princeton University Press.Nguyen, C. T. 2017. The Uses of Aesthetic Testimony, British Journal of Aesthetics 57/1: 19–36147Nguyen, C. T. 2019. Autonomy and Aesthetic Engagement, Mind. Advance online publication,, C. T. 2020 (forthcoming). Games: Agency as Art. Oxford University Press.Nisbett, R. E. and Bellows, N. 1977. Verbal reports about causal influences on social judgments: private access versus public theories, Journal of Personality and Social Psychology, 35/9: 613–24.Nisbett, R. E. and Wilson, T. D. 1977a. Telling More than We Can Know: Verbal Reports of Mental Processes, Psychological Review 84: 231–59.Nisbett, R. E. and Wilson, T. D. 1977b. The Halo Effect: evidence for unconscious alteration of judgments, Journal of Personality and Social Psychology 35/4: 250–56.Noë, A. 2015. Strange Tools: Art and Human Nature. New York: Hill and Wang.Nozick, R. 1974. Anarchy, State, and Utopia. Oxford: Blackwell.O’Brien, L., and Soteriou, M. (eds.). 2009. Mental Actions. Oxford: Oxford University Press.Oddie, G. 2018. Value Perception, Properties, and the Primary Bearers of Value, in Evaluative Perception, eds. A. Bergqvist and R. Cowan. Oxford: Oxford University Press, 240–257.Parfit, D. 1984. Reasons and Persons. Oxford: Clarendon Press.Pathak, D. et al. 2017. Curiosity-driven Exploration by Self-supervised Prediction, Proceedings of the 34th International Conference on Machine Learning, Sydney, Australia.Peacocke, A. Forthcoming. How to Think Several Thoughts at Once: Content Plurality in Mental Action, in Mental Action and the Conscious Mind, ed. M. Brent. Routledge.Peacocke, A. ms.a. Let’s Be Liberal. Article manuscript in preparation.Peacocke, A. ms.b. What You Do When You Infer. Article manuscript in preparation.148Perner, J. 2012. MiniMeta: In search of minimal criteria for metacognition, in Foundations of Metacognition, eds. M. Beran, J. Brandl, J. Perner, & J. Proust. New York: Oxford University Press, 94–116.Protasi, S. 2017. The Perfect Bikini Body: Can We All Really Have It? Loving Gaze as an Antioppressive Beauty Ideal, Thought 6/2:93–101.Proust, J. 2012. Metacognition and mindreading: one or two functions?, in Foundations of Metacognition, eds. M. Beran, J. Brandl, J. Perner, & J. Proust. New York: Oxford University Press, 234–51.Proust, J. 2014. The Philosophy of Metacognition: Mental Agency and Self-Awareness. Oxford: Oxford University Press.Ransom, M. 2019. Frauds, Posers And Sheep: A Virtue Theoretic Solution To The Acquaintance Debate, Philosophy and Phenomenological Research 98/2: 417–34.Ransom, M. ms. Perceptual Learning for Expertise (Unpublished doctoral dissertation, under examination). University of British Columbia, Vancouver, Canada.Reber, R. 2012. Processing Fluency, Aesthetic Pleasure, and Culturally Shared Taste, in Aesthetic Science: connecting minds, brains, and experience, eds. A. P. Shimamura and S. E. Palmer. Oxford: Oxford University Press, 223–49.Reber, R., Schwarz, N., and Winkielman, P. 1998. Effects of Perceptual Fluency on Affective Judgments, Psychological Science 9: 45–8.Reber, R. Winkielman, P., and Schwarz, N. 2004. Processing Fluency and Aesthetic Pleasure: is beauty in the perceiver’s processing experience?, Personality and Social Psychology Review 8/4: 364–82.Riggle, N. 2013. Levinson on the Aesthetic Ideal, Journal of Aesthetics and Art Criticism 71/3: 277–81.Riggle, N. 2015. On the Aesthetic Ideal, British Journal of Aesthetics 55/4: 433–47.149Riggle, N. 2016. On the Interest in Beauty and Disinterest, Philosopher’s Imprint 16/9: 1–14.Riggle, N. 2017. On Being Awesome: A Unified Theory of How Not to Suck. London: Penguin.Robson, J. 2012. Aesthetic Testimony, Philosophy Compass 7/1: 1-10.Ross, L. 2018. The Virtue of Curiosity, Episteme. Advance online publication,, G. 1949. The Concept of Mind. New York: Barnes & Noble.Savinov, N. et al. 2018. Episodic Curiosity through Reachability, (working paper, v5, 1 August 2019, accessed 2 December 2019),, J. 2015. Aesthetic Relationship, Cognition, and the Pleasures of Art, in Investigations into the Phenomenology and the Ontology of the Work of Art, eds. P. F. Bundgaard and F. Stjernfelt. New York: Springer, 145–167.Scheffler, S. 2010. Valuing, in Equality and Tradition: Questions of Value in Moral and Political Philosophy. Oxford: Oxford University Press, 15–40.Schopenhauer, A. 1819/2010. The World as Will and Representation, Vol. 1. Trans. J. Norman and A. Welchman. New York: Cambridge University Press.Schulz, L.E. (2012) The origins of inquiry: Inductive inference and exploration in early childhood, Trends in Cognitive Science 16: 382–89.Schwitzgebel, E. 2008. The Unreliability of Naive Introspection, Philosophical Review 117: 245–73.Setiya, K. 2003. Explaining Action, Philosophical Review 112/3: 339–393.Sharpe, R. A. 2000. The Empiricist Theory of Artistic Value, Journal of Aesthetics and Art Criticism 58/4: 321–32. Shelley, J. 2002. The Character and Role of Principles in the Evaluation of Art, British Journal of Aesthetics 42/1: 37–51.150Shelley, J. 2006. Critical Compatibilism, in Knowing Art: Essays in Epistemology and Aesthetics, ed. M. Kieran and D. M. Lopes. Dordrecht: Springer, 125–36.Shelley, J. 2010. Against Value Empiricism in Aesthetics, Australasian Journal of Philosophy 88/4: 707–20. Shelley, J. 2011. Hume and the Value of the Beautiful, British Journal of Aesthetics 51/2: 213–22. Shelley, J. 2013. Hume and the Joint Verdict of True Judges, Journal of Aesthetics and Art Criticism 71/2: 146–53.Shelley, J. 2017. The Concept of the Aesthetic, in The Stanford Encyclopedia of Philosophy, Winter 2017 Edition., ed. E. N. Zalta, (First version published in Fall 2009, accessed 1 December 2019),, J. 2019. The Default Theory of Aesthetic Value, British Journal of Aesthetics 59/1: 1–12.Shelley, J. ms. The Aesthetic Value Question. Book manuscript in preparation.Sibley, F. 1965. Aesthetic and nonaesthetic, Philosophical Review 74/2: 135–159. Sibley, F. 2001. Approach to Aesthetics: Collected Papers on Philosophical Aesthetics, eds. J. Benson, H. B Redfern, and J. R. Cox. Oxford: Clarendon Press.Sidgwick, H. 1907. The Methods of Ethics, 7th edn. London: MacMillan.Siegel, S. 2006. Which properties are represented in perception? in Perceptual Experience, eds. T.S. Gendler and J. Hawthorne. Oxford: Oxford University Press, 481–503.Silvia, P. J. 2012. Human Emotions and Aesthetic Experience: an overview of empirical aesthetics, in Aesthetic Science: Connecting Minds, Brains, and Experience, eds. A. P. Shimamura and S. E. Palmer. Oxford: Oxford University Press, 250–75.151Smith, J. D., Shields, W. E., and Washburn, D. A. 2003. The comparative psychology of uncertainty monitoring and metacognition, Behavioral and Brain Sciences 26: 317–73.Smith, M. 2012. Triangulating Aesthetic Experience, in Aesthetic Science: Connecting Minds, Brains, and Experience, eds. A. P. Shimamura and S. E. Palmer. Oxford: Oxford University Press, 80–106.Snyder, M. and Kendzierski, D. 1982. Acting on one’s attitudes: Procedures for linking attitude and behavior, Journal of Experimental Social Psychology 18: 165–83.Solomonova, E., Fox, K.C.R., & Nielsen, T.A. 2014. Methodological considerations for the neurophenomenology of dreaming: Commentary on Windt’s ‘Reporting dream experience’, Frontiers in Human Neuroscience 8/317: 1–3.Song, H. and Baillargeon, R. 2007. Can 9.5-month-old infants attribute to an actor a disposition to perform a particular action on objects?, Acta Psychologica 124: 79–105.Sperber, D. 2000. Metarepresentations in an Evolutionary Perspective, in Metarepresentations: A Multidisciplinary Perspective, ed. D. Sperber. Oxford: Oxford University Press. pp. 117–38.Stang, N. F. 2012. Artworks Are Not Valuable for Their Own Sakes, Journal of Aesthetics and Art Criticism 70/3: 271–80.Stecker, R. 1997. Artworks: Definition, Meaning, Value. University Park: Penn State University Press.Stecker, R. 2005. The Interaction of Ethical and Aesthetic Value, British Journal of Aesthetics 45/2: 138–50.Stecker, R. 2006. Aesthetic Experience and Aesthetic Value, Philosophy Compass 1/1: 1–10.Stecker, R. 2012. Artistic Value Defended, Journal of Aesthetics and Art Criticism 70/4: 355–62.152Stecker, R. 2019. Intersections of Value: Art, Nature, and the Everyday. Oxford: Oxford University Press.Stokes, D. 2014. Cognitive Penetration and the Perception of Art, Dialectica 68/1: 1-34.Stokes, D. 2018. Rich perceptual content and aesthetic properties, in Evaluative Perception, eds. A. Bergqvist and R. Cowan. Oxford: Oxford University Press, 19–41.Stokes, D. Forthcoming. On Perceptual Expertise, Mind and Language.Stolnitz, J. 1960. Aesthetics and Philosophy of Art Criticism. New York: Houghton Mifflin.Stout, R. (ed.) 2018. Process, Action, and Experience. Oxford: Oxford University Press.Strawson, P. 1974. Aesthetic Appraisal and Works of Art, in Freedom and Resentment and Other Essays. London: Methuen, 196–207.Strohl, M. 2019. Art and Painful Emotion, Philosophy Compass 14/1: e12558.Suits, B. 1978. The Grasshopper: Games, Life, and Utopia. Toronto: University of Toronto Press.Sumner, L. W. 1996. Welfare, Happiness, and Ethics. Oxford: Clarendon Press.Thomson, V. A. 2009. Dual-process theories: a metacognitive perspective, in In two minds: Dual processes and beyond, eds. J. Evans and K. Frankish. Oxford: Oxford University Press, 171–95.Vice, S. 2006. The Ethics of Self-Concern, in Iris Murdoch: A Reassessment, ed. A. Rowe. Basingstoke: Palgrave Macmillan, 60–71.Walton, K. L. 1970. Categories of Art, Philosophical Review 79/3: 334–67. Walton, K. L. 1993. How Marvelous! Towards a Theory of Aesthetic Value, Journal of Aesthetics and Art Criticism 51/3: 499–510. Walton, K. L. 2007. Aesthetics—What? Why? And Wherefore?, Journal of Aesthetics and Art Criticism 65/2: 147–61.153Watkins, M., and Shelley, J. 2012. Response-Dependence about Aesthetic Value, Pacific Philosophical Quarterly 93/3: 338–52.Weil, S., 1986. Simone Weil: An Anthology, ed. S. Miles. New York: Grove Press.Whitcomb, D. 2010. Curiosity was framed, Philosophy and Phenomenological Research 81: 664–87.White, P. A. 1988. Knowing more about what we can tell: introspective access and causal report accuracy 10 years later, British Journal of Psychology 79: 13–45.Whittlesea, B. W. A., and Williams, L. 2000. The Source of Feelings of Familiarity: the discrepancy-attribution hypothesis, Journal of Experimental Psychology 26/3: 547–65.Windt, J. M. 2011. Altered Consciousness in Philosophy, in Altering Consciousness. Multidisciplinary Perspectives, Volume 1: History, Culture, and the Humanities, eds. E. Cardeña and M. Winkelman, Oxford: Praeger, 229–54.Wilson, T. D. 2002. Strangers to Ourselves: Discovering the adaptive unconscious. London: Bellknap.Wilson, T. D. and Dunn, D. S. 1985. Effects of Introspection on Attitude–Behavior Consistency: Analysing Reasons Versus Focusing on Feelings, Journal of Experimental Social Psychology, 22: 249–63.Wilson, T. D. and Schooler J. W. 1991. Thinking Too Much: Introspection Can Reduce the Quality of Preferences and Decisions,  Journal of Personality and Social Psychology 60/2: 181–92.Wilson, T. D., Lisle, D. J., Schooler, J. W., Hodges, S. D., Klaaren, K. J., and LaFleur, S. J. 1993. Introspecting About Reasons can Reduce Post-Choice Satisfaction, Personality and Social Psychology Bulletin 19: 331–339.Winkielman, P. and Cacioppo, J. T. 2001. Mind at ease puts a smile on the face: Psychophysiological evidence that processing facilitation elicits positive affect, Journal of Personality and Social Psychology 81/6: 989–1000.154Wolf, S. 2015. Good-for-Nothings, in The Variety of Values: Essays on Morality, Meaning, and Love. Oxford: Oxford University Press, 67–86.Wollheim, R. 1980. Art and Its Objects, 2nd edition. Cambridge: Cambridge University Press.Wollheim, R. 1986. Imagination and Pictorial Understanding, Proceedings of the Aristotelian Society, Supplementary Volumes 60/1: 45-60.Wolterstorff, N. 2015. Art Rethought: the Social Practices of Art. Oxford: Oxford University Press.Zajonc, R. B. 1968. Attitudinal effects of mere exposure, Journal of Personality and Social Psychology 9: 1–27.Zajonc, R. B. 1980. Feeling and thinking: Preferences need no inferences, American Psychologist 35: 151–175.Zangwill, N. 1992. Unkantian notions of disinterest, British Journal of Aesthetics 32/2: 149–52.Ziff, R. P. 1966. Reasons in Art Criticism, in Philosophic Turnings. Ithaca, NY: Cornell University Press, 47–74.155


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items