1. The Intuition and The Puzzle

The aim of this paper is to clarify, and in so doing defend, an intuition about the role of emotional experience, and in particular its affective component, in our ability to learn value.Footnote ¹ The intuition is that just as in the absence of perceptual experience we lack an important kind of understanding of sensory properties like colour, in the absence of affective experience we lack an important kind of understanding of value.

This intuition drives several recent philosophical accounts of the relation between affective experience and value. For instance, Mark Johnston writes that ‘[j]ust as we need to sense cherry red to make a goodish range of judgments as to its nature, we need to encounter the determinate sensuous values in order to have them either as the topics or as the things predicated in our most basic evaluative judgements…If one has never been moved or affected by the determinate ways in which things are beautiful or charming or banal or sublime or horrific or appealing, then one is ignorant of the relevant determinate values’ (Johnston Reference Johnston2001, 182–183, Johnston's italics). Michelle Montague voices the same intuition when she writes that although Data, the emotionless alien in Star Trek, ‘can say what is of value and disvalue, and know what is of value and disvalue, he cannot really know value and disvalue. It may be similar to someone who is blind from birth who can say this is red, when told that it is a ripe tomato, and indeed know that it is red, but not know what red is’ (Montague Reference Montague2016, 231). In the final pages of their seminal work on the emotions, Julien Deonna and Fabrice Teroni (2012, 123–124) first invite us to ‘imagine a creature deprived of emotional responses who has been able to get a handle on our evaluative practices because, say, she has learned to recognize the responses of others. This creature is thus linking her application of evaluative concepts to responses she can discriminate correctly in others’. They then invite us to ask ‘[b]ut does she understand the evaluative judgements she makes?’. Deonna and Teroni answer by suggesting that although ‘[o]ur subject, devoid of emotional experiences, will by hypothesis come up with evaluative judgements that reliably fit the evaluative fact’, this ability is ‘completely divorced from the kind of understanding that is implied by our evaluative practices, one that is constitutively connected to our emotional engagement with the world’.Footnote ²

The analogy with perceptual experience helps us to a more positive articulation of the underlying intuition about affect and value. In recent work on perceptual experience, it is suggested that we should conceive the claim that the sensory, phenomenal character of perceptual experience provides us with a distinctive sort of understanding of sensory properties like colour in terms of the claim that the sensory, phenomenal character of perceptual experience grants us a distinctive sort of epistemic access to sensory properties like colour.Footnote ³ The intuition about affect and value, then, can be reformulated as the intuition that just as the phenomenal, sensory character of perceptual experience grants the subject a distinctive epistemic access to sensory properties like colour, the phenomenal, affective character of emotional experience grants the subject a distinctive epistemic access to evaluative properties. If so, then in the absence of an emotional experience with affective character, we lack a distinctive epistemic access to, and therefore understanding of, evaluative properties. In this paper, I employ the term ‘disclosure’ to refer to the structure of the relevant experience in virtue of which its phenomenal character grants the subject a distinctive epistemic access to the relevant property of the object of experience.Footnote ⁴ For this reason, I call the intuition that is the topic of this paper ‘affective disclosure of value’.

The problem is that although the intuition of the affective disclosure of value drives many recent philosophical accounts of the relation between affect and value, these accounts do not recognise a fundamental puzzle at the heart of the intuition that affective experience provides us with a distinctive sort of evaluative understanding. The puzzle is the following. As the quotes of the philosophers above suggest, the sort of understanding that putatively affect provides us with is an understanding of properties that are instantiated by objects distinct from our experiences. For instance, according to Deonna and Teroni (Reference Deonna and Teroni2012, 123), the experience of anger at someone's remark should be conceived as a response to, say, the remark's offensiveness. Assuming the plausibility of this example, the underlying idea is that the experience of anger provides us with an understanding of a property, namely offensiveness, instantiated by an object distinct from the experience itself, namely the remark. We might put the point by saying that the understanding provided to us by affect is objective in the sense that it is about properties instantiated by objects distinct from our experience. At the same time, affect is an experience that the subject undergoes. We might therefore want to say that affect is a subjective property in the sense that it is conceived as a property of the subject, in particular of her emotional experience. And here is the puzzle: how can a property pertaining to the experience of the subject provide us with a distinctive epistemic access to, and therefore an understanding of, properties that are instantiated by objects distinct from the experience? In other words, how can the affective component of emotional experience provide us with an understanding of properties of objects distinct from the experience itself, namely their value, rather than merely with an understanding of the experience itself?

Let me clarify further the puzzle. The puzzle articulated as in the previous paragraph might seem as a mere instance of a more general puzzle regarding how experience in general can grant us epistemic access to properties of objects distinct from the experience itself. Philosophers of perception grapple with this sort of puzzle, too. So, what is distinctive about the puzzle regarding affective experience and value? The philosophers of emotion mentioned above point out that emotional experience is the sort of psychological response that is open to rational assessment in a way that perceptual experience is not.Footnote ⁵ We can inquire into the justification of having an emotional experience in a way that would not make sense when asked about a perceptual experience. For instance, we can ask why someone felt angry at someone else's remark and in so doing ask for reasons in favour of or against having felt anger. It would not make sense to ask for that kind of justification in the case of perceptual experience. Inquiring for reasons for or against having a perceptual experience as of a red object would make little sense. All one could offer if asked for a reason why they enjoyed a perceptual experience is a causal story. So, one first constraint that distinguishes the puzzle when applied to affective experience and value is that the sort of experience that putatively provides us with an understanding of properties of objects distinct from the experience itself, namely their value, be conceived as open to rational assessment. In this sense, affective experience is conceived in the puzzle as rationally intelligible.

The puzzle becomes even more pronounced and peculiar to affect and value if we follow some philosophers in their claim that there is an intimate connection between the affective component of emotional experience and the formation of motivational states. For instance, Johnston articulates this intimate connection when he argues that affective experience has a normative authority not only over our evaluative judgements but also over the formation of motivational states. That is, Johnston argues that from a first-person perspective there is no felt need for immediate justification of one's evaluative judgments and motivational states when these arise when affectively engaged with the object of experience: affect can make these readily and rationally intelligible (Johnston Reference Johnston2001, 187). Presumably, according to Johnston, motivational states form during our affective engagement with the world and that is why affect renders them readily intelligible. Deonna and Teroni seem to take a step further and argue that we should conceive many emotional experiences, and in particular their affective component, as ‘feelings of action readiness’ (Deonna and Teroni Reference Deonna and Teroni2012, 79ff).Footnote ⁶ If we conceive the affective component of emotional experiences as intimately connected with the formation of motivational states, then the puzzle takes the following shape: how can properties pertaining to the experience of the subject, such as the formation of motivational states, provide us with a distinctive sort of epistemic access to, and therefore understanding of, properties that are instantiated by objects distinct from the experience itself, namely their value? So, a second constraint, at least for some philosophers, is that we conceive affective experience not only as rationally intelligible but also as involving the formation of motivational states.

The puzzle goes right at the heart of the intuition of the affective disclosure of value. For the intuition is that it is in virtue of properties of emotional experience, and in particular its rationally intelligible affective, and for some, motivational component, that we gain epistemic access to, and therefore understanding of, properties of objects distinct from the experience, namely their value. Crucially, the puzzle relies on a conception of value as a property of an object of experience distinct from the experience itself. If so, then unless we provide a solution to the puzzle, we are not in a position to explain how affect provides us with a distinctive sort of epistemic access to evaluative properties, properties of objects distinct from the experience itself. And if so, then the intuition is vacuous. In this paper, I propose a solution to the puzzle and in so doing both clarify and provide support for the intuition. I argue that if we want to solve the puzzle and thus understand how the affective component of emotional experience provides us with epistemic access to properties instantiated by objects distinct from the experience itself, namely value, then we have to commit to a specific metaphysical account of value. The metaphysical account I have in mind derives from the work of David Wiggins. In particular, I argue that a solution to the puzzle should commit us to a relation of explanatory circularity between affective experience and evaluative properties. As we shall see in more detail below, what this means is that we cannot provide an adequate account of affective experience without reference to evaluative properties and vice versa, we cannot provide an adequate account of evaluative properties without reference to affective experience. In other words, our conceptions of affective experience and of evaluative properties are interdependent. Importantly, I do not take the claim that affective experience involves the disclosure of value to follow directly from the claim that our conceptions of affective experience and of evaluative properties are interdependent. Rather, I will argue that, when properly understood, the claim that our conceptions of affective experience and of evaluative properties are interdependent provides us with the conceptual tools to solve the puzzle and thus provides a stable account of the affective disclosure of value. By relying on a metaphysical account of value to solve a puzzle about the epistemic role of affective experience, this paper effectively argues that a key step in explaining how affective experience provides us with understanding of evaluative properties is to show that evaluative properties are the sort of properties that need affective experience to be epistemically accessed in the first place.

The argument of this paper is distributed in the following five sections. I begin in Section 2 by unpacking the notion of disclosure employed in this paper. In Section 3, I set out the key claims from Wiggins’ metaphysics of value i.e. Wiggins’ circular interpretation of the neo-sentimentalist biconditional. In Section 4, I employ Wiggins’ notion of a relation of explanatory circularity between affect and value to solve the puzzle. In Section 5, I develop further the account of the affective disclosure of value by replying to a form of scepticism about affect voiced in the literature. I argue that the upshot of the solution provided to the puzzle is that the sort of epistemic access provided by affective experience into evaluative properties is the sort that leads to the generation of evaluative concepts.

Two caveats. First, I restrict my account of the affective disclosure of value to the sort of value we ascribe to objects by deploying thick evaluative concepts (Williams Reference Williams1985, 140–141; for recent discussions on the subject, see Kirchin Reference Kirchin2017). Evaluative concepts are ‘thick’ insofar as their content provides us with a substantive description of the person, action or event of which such concepts are predicated. Examples of thick evaluative concepts are ‘coward’, ‘brutality’, ‘admirable’, ‘offensive’, and so on. Thick evaluative concepts contrast with ‘thin’ evaluative concepts, such as ‘good’, ‘bad’, ‘right’, ‘wrong’ the content of which does not provide us with a substantive description of the person or action of which such concepts are predicated. As Williams observes, since the content of thick evaluative concepts is in part descriptive, their deployment is guided by how the relevant person or event seems to one to be. At the same time, thick evaluative concepts are action-guiding. The deployment of a thick evaluative concept normally provides one with a reason for action. In addition to excluding thin evaluative concepts from my account, I will also not be dealing with evaluative concepts the deployment of which is not obviously action-guiding, such as aesthetic concepts.

Second, it might be wondered from the start whether there is any such puzzle by simply denying the intuition that is the topic of this paper. For instance, such denial might come from philosophers of a ‘buck-passing’ persuasion.Footnote ⁷ According to ‘buck-passers’, value is the property of an object of experience in virtue of which it provides one with reasons for action. If one's intuitions lie with the ‘buck-passers’, then affective experience becomes entirely superfluous in one's epistemic access to evaluative properties since rational reflection can do all the relevant work. Now, although this paper will not engage critically with ‘buck-passing’ accounts of valueFootnote ⁸, I do want to make the following two points. First, this paper should be seen as speaking primarily to those who do hold this intuition and want to understand it further rather than being a direct attempt to change contrary minds. But second, by clarifying exactly what is the role of affective experience in our gaining epistemic access to evaluative properties, this paper should also be seen as potentially answering fundamental doubts of philosophers who do not feel the pull of the intuition. Indeed, in Section 5 I will show how relying on Wiggins’ conception of the circularity between affect and value explains why many of us also have the intuition that we can learn about value through rational reflection without any emotional engagement with the relevant object of experience. At the very least, then, this paper provides deniers of the intuition with a better understanding of the object of their denial.

2. Disclosure

The best way to tease out what is at stake with the notion of disclosure is to focus on some intuitions regarding the relation between the phenomenal character of perceptual experience and our epistemic access to sensory properties like colour. Several philosophers have pointed out that there is a strong intuition behind the claim that phenomenally conscious perceptual experience grants us a distinctive sort of epistemic access to sensory properties such as colour and thus provides us with a distinctive conception of colour.Footnote ⁹ For instance, we find it intuitive that a congenitally blind person lacks an understanding of, say, the colour red which normally sighted individuals possess. To be sure, the congenitally blind person can still refer successfully to, and acquire knowledge about, the colour red. For example, congenitally blind people can still know that apples are red and refer to apples as objects that are often red coloured. Nevertheless, because congenitally blind people are unable to enjoy a phenomenally conscious visual experience as of a red object, we take it that they are unable to acquire that distinctive understanding of redness that subjects who can enjoy a phenomenally conscious visual experience normally possess.

If congenitally blind people possess some sort of understanding about the colour red, what is the sort of understanding that they lack? It is imperative to appreciate that the answer to this question depends just as much on how one understands the constitution of the phenomenal character of perceptual experience as on how one understands the metaphysical status of sensory properties like the colour red. Most philosophers would agree that congenitally blind people lack an understanding of what the colour red looks like in experience. This is a claim that most theories of experience can accommodate. For example, qualia theories of experience argue that the phenomenal character of perceptual experience is constituted by qualia, that is, intrinsic, non-intentional features of experience associated with the relevant colour.Footnote ¹⁰ According to qualia theories, congenitally blind people lack epistemic access to features of experience associated with the colour red. Therefore, they lack an understanding of what it is like to experience the colour red. But some philosophers might want to go further and argue that congenitally blind people lack an understanding not merely of what the colour red looks like in experience but of what redness is like. The thought here is that the phenomenal character of experience grants us epistemic access not merely to features of the experience but to features of the object of experience distinct from the experience itself. Can qualia theories accommodate this claim? It depends on the metaphysical account of sensory properties like the colour red that one opts for. If one opts for an account according to which the colour red is a categorical, mind-independent property of objects in our environment (see, e.g., Campbell Reference Campbell, Haldane and Wright1993), then qualia theories are not able to capture the claim that the phenomenal character of experience grants us epistemic access to features of the object of experience rather than merely of the experience itself.Footnote ¹¹ If the phenomenal character of experience is constituted by non-intentional, intrinsic features of experience, then enjoying a phenomenally conscious experience gets us epistemically only as far as properties of experience associated with the colour red and not as far as the colour red itself. But now suppose we opt for a dispositionalist account of colour (see, e.g, Johnston Reference Johnston, Byrne and Hilbert1992). Dispositionalists about colour argue that what the colour red is is the disposition of the object to cause an experience as of redness in a suitable observer under normal conditions. If so, then qualia theories are better positioned to capture the claim that the phenomenal character of experience grants us epistemic access not merely to features of experience but also to features of the object of experience. For if one argues that what the colour red is is the disposition of the object to cause an experience as of redness in a suitable observer under normal conditions, then one can argue that epistemic access to a feature of experience associated with the colour red counts as epistemic access to the colour red.

By contrast, suppose we reject qualia theories. One common objection to qualia theories is that if we introspect our perceptual experience as of a red object, it seems like we are unable to shift our attention from the property of an object i.e. its redness, to a property of experience associated with redness.Footnote ¹² If so, then it is unclear why we should think that the phenomenal character of experience is constituted by non-intentional, intrinsic features of experience. In light of this objection, many philosophers opt for theories of experience that understand differently the constitution of the phenomenal character of experience. For instance, take naïve realism.Footnote ¹³ According to naïve realists, if the experience is veridical, the phenomenal character of perceptual experience is in part constituted by the features of the object of experience. For example, if veridical, the phenomenal character of my perceptual experience of a red object is in part constituted by the object's redness. If so, then in contrast to qualia theories, naïve realists are not forced to opt for a dispositionalist account of colour if they want to capture the claim that the phenomenal character of experience grants us epistemic access to features of the object of experience and not merely of the experience itself. Naïve realists can opt for an account of colour as a categorical, mind-independent property of objects in our environment. That is because, if veridical, the phenomenal character of experience is in part constituted by the colour of the object of experience and therefore the phenomenal character of experience, in virtue of its constitution, grants us epistemic access to feature of the object of experience, namely its colour.

In principle, then, both qualia theorists and naïve realists are able to capture the claim that the phenomenal character of experience grants us epistemic access to features of the object of experience and not merely to features of the experience itself. Nevertheless, notice that although they would both agree that specifying the conception of the property of the object of experience acquired by means of phenomenally conscious experience ought to make reference to the way the property appears in experience, their way of cashing out this claim takes very different shapes. In the case of qualia theories that marry a dispositionalist account of colour, experience of the colour red is conceived as providing us with a conception of the colour red that ought to be specified by making reference to non-intentional, intrinsic properties of experience i.e. qualia. By contrast, in the case of naïve realists that marry an account of colour as a categorical, mind-independent property of objects in our environment, experience is conceived as providing us with a conception of the colour red that ought to be specified by making reference to a categorical, mind-independent property of objects and not in terms of intrinsic properties of our experience.

The point is that the account we give of the sort of conception of the colour red we are provided with in virtue of phenomenally conscious perceptual experience will depend both on our account of the constitution of the phenomenal character of experience and on the metaphysical status of the property experience grants us epistemic access to. To put it the other way around, the conception of the colour red we take congenitally blind people to lack will depend on the relation between our conception of the constitution of the phenomenal character of experience and our conception of the metaphysical status of the properties experience grants us epistemic access to.

In addition, the discussion above clearly spells out the distinction between the claim that the phenomenal character of experience grants us epistemic access merely to features of experience and the claim that the phenomenal character of experience grants us epistemic access not merely to features of experience but also to features of the object of experience. As pointed out in Section 1, the notion of ‘disclosure’ used in this paper is meant to capture the structural feature of the phenomenal character of experience in virtue of which experience grants us epistemic access not merely to features of experience but also to features of the object of experience. In sum, this section displayed how our ability to capture the claim that the phenomenal character of experience is characterised by the disclosure of features of the object of experience is inextricably tied to the relation between our account of the constitution of the phenomenal character of experience and our metaphysical account of the sort of properties disclosed by experience.

3. Wiggins and Explanatory Circularity

In the previous section, I highlighted the tight connection between our conceptions of the epistemic role of the phenomenology of experience and the metaphysical status of the properties epistemically accessed by experience. In this section, I set out the metaphysical commitments regarding evaluative properties that, in the next section, I will argue are needed to solve the puzzle at the heart of the intuition of the affective disclosure of value.

The needed metaphysical commitments are found in the work of David Wiggins. At the core of Wiggins’ account of value is the following biconditional: ‘x is good/right/beautiful if and only if x is such as to make a certain sentiment of approbation appropriate’ (Wiggins Reference Wiggins and Wiggins1987b, 187, Wiggins’ italics). Wiggins does not speak in terms of affective experience, affective response or phenomenal character but I want nevertheless to paraphrase the biconditional as follows: an object or event has value if and only if the object is such as to make a certain affective experience appropriate.Footnote ¹⁴ I don't think the plausibility of my argument depends much on this paraphrasing. Assuming this, there are two points I want to stress regarding the biconditional. First, following Wiggins, I read the internal link between affect and value embodied by the biconditional in terms of specification, or elucidation: we cannot adequately specify, or elucidate, evaluative properties without making reference to affective experience. And vice versa, we cannot adequately specify, or elucidate, affective experience without reference to value. Second, the biconditional can be interpreted in at least two conflicting ways, namely through a projectivist or an objectivist interpretation. According to a projectivist neo-sentimentalist, evaluative properties are reductively analysable in terms of affective responses and are the mere projection of our affective dispositions onto a value neutral world (see, e.g., Blackburn Reference Blackburn and Honderich1985). By contrast, while an objectivist neo-sentimentalist agrees with the projectivist that evaluative properties have an internal link to affective responses, an objectivist argues that values are not reductively analysable in terms of affective responses and, in this sense, preserve a distinctive kind of objectivity (Wiggins Reference David Wiggins and Wiggins1987a,Reference Wiggins and Wigginsb; cf. McDowell Reference McDowell and McDowell1998b). If we want to capture the affective disclosure of value, we need to commit to an objectivist reading of the biconditional in order to give sufficient substance to the notion of disclosure at stake. For if evaluative properties are mere figments of our mind, then the notion of disclosure is either mere metaphor or mere illusion, or both.

So how do we obtain an objectivist reading of the neo-sentimentalist biconditional? The answer lies in the circular interpretation of the biconditional. The biconditional states that an object has value if and only if the object is such as to make a certain affective response appropriate. The circularity consists in that once we try to cash out what it means for an affective response to be appropriate, we revert to making reference to the object having value. For instance, a remark is offensive if and only if the remark is such as to make an angry response appropriate. At the same time, in explicating the appropriateness of an angry response we make reference to the remark being offensive. It is this circular explanation of the relation between affect and value that allows us to slip in objectivity. To appreciate this, consider two initial potential worries discussed by Wiggins. The first worry is that the resultant account of value is, precisely, circular and circular definitions are problematic since they use the terms to be defined as part of the definition. The source of this worry, though, is a misunderstanding of the commitments to such circularity. The circularity is not meant to be an attempt at defining value in terms of the appropriateness of our responses. The circularity would be vicious solely if we attempted to give a reductive definition of value in terms of the appropriateness of our affective responses and then found ourselves defining the latter in terms of the former. Since analytic reduction is no part of the agenda of this circular conception of value, the circularity is not vicious (Wiggins Reference David Wiggins and Wiggins1987, 188–189). The second worry consists in the fact that someone might argue that even if the explanatory circularity is not vicious insofar as it is not committed to a reductive account of value, it is nevertheless uninformative since the terms within which value is characterised, namely an appropriate affective response, is in turn characterised in terms of the object's value. But, as Wiggins points out, this worry is unfounded since the circular interpretation of the conception of value is informative insofar as it informs us regarding the way in which the subject learns about the evaluative import of the object, namely via her affective experience (ibid., 189). Indeed, in the rest of this paper I want to show that committing to explanatory circularity provides us with the conceptual material needed to give an account of the distinctive kind of disclosure at play in affective experience.

The key point that should be taken from Wiggins’ response to both worries about circularity is that what a circular interpretation of the internal link between value and affective responses delivers is precisely a denial of, and alternative to, the need to conceive the internal link in terms of reductive analysability. The key thought is that the claim according to which our conception of a property makes necessary reference to our experiential responses does not entail that therefore the property is reducible to our experiential response. For instance, one might argue, as McDowell (Reference McDowell and McDowell1998) does, that for an object to be, say, red is for the object to be disposed to look red to a subject under normal conditions. This entails that understanding what it is for an object to be red, and thus attributing redness to an object, is conceptually dependent on the subject understanding what it is for an object to look red. Yet this does not detract from our understanding of what it is for something to look red that we refer to redness as a property of the object i.e. the object being red. In other words, the claim that the understanding of what it is for an object to be red depends conceptually on the subject understanding the way the property looks in experience does not entail the claim that redness is a mere figment of our mind projected onto the object rather than being a property of the object itself.

The circular interpretation of the neo-sentimentalist biconditional shares with McDowell's account of colour the idea that referring to the relevant experience in order to specify what a property of the object is does not entail that the latter is reductively analysable to the former. At the same time, it differs in (at least) one crucial respect. That is, normally affective experience, unlike perceptual experience, is characterised as the sort of psychological response that can be assessed as appropriate or not. Let me elaborate on this point. ‘Appropriateness’ is a property of psychological responses that are open to rational assessment. As anticipated in Section 1, in the case of affect, we can provide reasons for and against having an affective response towards an object. For instance, we can provide reasons for or against having an angry response towards someone's remark. Broadly speaking, if we think that there is sufficient reason in favour of having an angry response towards someone's remark, we will think of the angry response as appropriate. If we think that there isn't sufficient reason, we will think of the angry response as inappropriate. The fact that affective experience is the sort of response that can be assessed for appropriateness points to the fact that affective experience is rationally intelligible. Rational intelligibility is contrasted with mere causal intelligibility. To illustrate. If I feel angry at the remark, my affective experience is rationally intelligible insofar as we can give reasons as to why the affective experience is appropriate or not. By contrast, suppose that you develop repulsion for sushi after a strong case of indigestion. In this case, your repulsion is causally intelligible because we can explain the cause behind its formation when confronted with sushi. But there is no sense in which we can provide reasons to justify your repulsion. Your repulsion towards sushi is simply caused by the visceral association you make between sushi and the nauseous state characteristic of indigestion. In this case, your repulsion is only causally intelligible and not rationally intelligible. Notice that the claim is not that rational and causal intelligibility are mutually exclusive. For instance, one can interpret McDowell (Reference McDowell and McDowell1998) as arguing that affective experiences are both rationally and causally intelligible. This is a claim I want to remain neutral on. The point is that, whatever other kind of intelligibility we might want to bestow on affective experience, we should conceive affective experience as rationally intelligible.

I want to stress how crucial it is that we appreciate that affective experience, as understood within Wiggins’ neo-sentimentalist bi-conditional, is rationally intelligible and therefore assessable for appropriateness. This is a defining property of affective experience as understood within the biconditional. It is clear that, according to Wiggins, it is in virtue of this property that we need to think of it as a response to value if we want to specify adequately affective experience. Our conception of what sort of response an object ought to evoke is dependent on our conception of the evaluative property that the object is taken to instantiate. This is in part why Wiggins writes, when commenting on our conception of amusement, that ‘[t[here is no object-independent and property-independent, ‘purely phenomenological’ or ‘purely introspective’ account of amusement’ (Wiggins Reference David Wiggins and Wiggins1987, 195). For us to specify our conception of amusement, we need to make reference to the property that makes an object funny. We find the same thought in Johnston when he warns us against the mistake of ‘mentalising’ affect (Johnston Reference Johnston2001, 203). If we sever the specification of our responses from our conception of value, then we end up distorting the nature of our original target, namely, our affective-cum-motivational responses, to such an extent that we risk ending up describing a completely different phenomenon from the original target, such as one's awareness of physiological changes. That is why Wiggins seems to intimate that there is something fishy in thinking that it is possible for a subject to tell in what experiential state she is without ‘looking outward’ (ibid., 208). The characterisation of affective experience as a response to value is also part of Wiggins’ argument against reductively analysing value in terms of affective experience. If we were to do so, then we would end up eliminating the element we need to adequately specify affective experience. Hence the need to conceive the relation between affect and value in terms of explanatory circularity.

4. Circularity and Disclosure

We have arrived at the crux of the matter. Recall that the puzzle I set out to solve asks how is it possible that properties pertaining to the subject's emotional experience can provide us with a distinctive sort of epistemic access to, and therefore understanding of, properties, namely value, instantiated by objects distinct from the experience itself. Recall also the constraints to resolving such a puzzle. These are, first of all, that the affective component of emotional experience is rationally intelligible, that is, open to rational assessment. Moreover, according to some philosophers, emotional experience involves the formation of motivational states. So, a solution to the puzzle needs to explain how the first, and for some also the second, of these properties of emotional experience play an essential role in granting us a distinctive sort of epistemic access to value i.e. a property of the object of experience distinct from the experience itself. What, then, is the solution to the puzzle?

The solution is the following. The neo-sentimentalist biconditional committed to above tells us that what value is is the feature of the object that renders appropriate the relevant affective experience. We have seen above that what it means for affective experience to be the sort of experience that is assessable for appropriateness is that one of its defining properties is rational intelligibility. It is because affective experience is open to rational assessment that it is assessable for appropriateness. If so, and if value just is the property of the object that renders appropriate the relevant affective experience, then rationally intelligible affective experience is our epistemic access to evaluative properties. Once value and affective experience are conceived as in Wiggins’ neo-sentimentalist biconditional, then we can see why it is in virtue of a property of our experience, namely rationally intelligible affective experience, that we epistemically access the property of an object distinct from our experience, namely value. Let me illustrate. According to the neo-sentimentalist bi-conditional, offensiveness is the feature of, say, a remark that renders appropriate the relevant affective experience towards it, say anger. If so, then our epistemic access to the offensiveness of someone's remark is granted by anger understood as open to rational assessment i.e. as rationally intelligible. And that is because offensiveness just is the property of the object that renders appropriate one's experience of anger. The puzzle is solved because we have explained how a property of the experience of anger grants us epistemic access to a property of the object of experience, namely the remark's offensiveness.

Notice how in this solution to the puzzle the rational intelligibility of affective experience is a key component because it is in virtue of its openness to rational assessment that we can explain how a property of our experience allows epistemic access to properties of objects of experience, namely their evaluative properties. That is because, again, evaluative properties are properties of the object that render appropriate affective experience. Notice also how positing a relation of explanatory circularity is all-important. If we were to reductively analyse value in terms of affective experience, then we would not be able to explain how a property of affective experience grants us epistemic access to a property of the object of experience, namely its value, because we would eliminate value from the picture. What circularity allows us to do is preserve the idea that we are allowed to conceive of value as a property of the object of experience distinct from the experience itself. Moreover, circularity allows us to understand how a property of the experience, namely the rational intelligibility of affective experience, grants us epistemic access to a property of the object of experience. And this is what we needed to solve the puzzle.

To appreciate further this point, compare, for instance, if we were to commit to a dispositionalist conception of value. According to an, admittedly crude, dispositionalist conception of value, the value of an object is to cause the relevant affective experience in a suitable subject under certain conditions. One might then wonder why such a view shouldn't be able to solve the puzzle. After all, if affective experience is caused by an evaluative property conceived as a property of the object distinct from the experience itself, then affective experience should be a good indicator of where to find value and in this sense, provide us with knowledge of value. But here is the problem. Since a dispositionalist view of value conceives the relation between the object's evaluative property and the subject's affective experience in causal terms, it does not conceive affective experience as rationally intelligible. If the relation between an affective response and the evaluative property is merely causal, then the affective response is not open to rational assessment. It would rather be, as mentioned in Section 3, conceiving the relation between affective response and evaluative property analogously to one's developing repulsion for sushi after a strong case of indigestion. In light of this, a dispositionalist view of value is unable to capture the notion that the evaluative property renders appropriate the relevant affective experience. For these reasons, although a dispositionalist view of value might be able to capture the thought that affective experience discloses value conceived as a property of the object distinct from the experience itself, it ends up providing us with a picture where affective experience is not open to rational assessment.

Let me now make two points of clarification. First, although both ‘appropriateness’ and ‘rational intelligibility’ are properties of affective experience associated with their openness to rational assessment, they should be thought of as constituting two different stages. An affective experience is rationally intelligible insofar as we are able to give reasons in favour or against having it, and therefore establishing whether or not an object has the relevant value. Whether the evaluative property is indeed instantiated by the object, according to the neo-sentimentalist biconditional, is then a matter of establishing whether the object of experience has features that render the relevant affective experience appropriate. An affective experience is judged appropriate when one establishes that indeed the object of experience has the relevant value. Compare with perceptual experience of colour. When I visually experience a red chair in front of me, the deployment of the concept ‘red’ is rationally intelligible to me in virtue of enjoying the visual experience as of a red object (notice: the deployment of the concept ‘red’ is rationally intelligible, not having the experience as of a red object). Yet whether the visual experience is veridical and not the result of an illusion, and therefore my deployment of the concept ‘red’ justified, is a matter of further investigation. (Granted, such investigation is in most cases needless). Therefore, strictly speaking, it is only when the affective experience is in fact appropriate that we can speak of the affective disclosure of value. That is because it is only when the relevant affective experience is in fact appropriate that the object has the relevant value.

Second, let me clarify the relation between disclosure and circularity. On the one hand, circularity is a conceptual tool by means of which to understand the relation between affective experience and evaluative properties if we want to conceive evaluative properties as properties of objects of experience; if we want to conceive affect and value as inextricably linked; and if we want to capture the normative relation between evaluative properties and affective experience. As we have seen, these are all desiderata of anyone who wants to argue that affective experience is characterised by the disclosure of value. On the other hand, disclosure is a structural feature of the phenomenal character of experience. The phenomenal character of experience involves disclosure if it grants us epistemic access to features of the object of experience rather than merely of the experience itself. In the case of affective experience, an adequate specification of the disclosure characterising its phenomenal character, in virtue of being constituted by the formation of rationally intelligible motivational states, necessitates reference to our conception of evaluative properties. Circularity, then, is a conceptual tool that allows us to conceive of the sort of disclosure characterising affective experience. While disclosure is a structural feature of affective experience, circularity is not. Circularity is a conceptual tool for us theorists to understand how affective experience involves the disclosure of value.

Importantly, the account given in this section can also serve the purposes of those philosophers who argue that there is an intimate connection between affective experience and motivational states. If the formation of motivational states is conceived as intimately related to affective experience, and if we want to argue that value is disclosed by affect, then we might be tempted to argue that the formation of motivational states discloses value. But how can properties of the subject's experience, namely the formation of motivational states, grant us epistemic access to properties of the object of experience, namely its value? The answer, again, lies in conceiving the explanatory relation between affective-cum-motivational experience and value as circular and the formation of affective-cum-motivational experiences as rationally intelligible. Once value is conceived as the property of the object that renders appropriate affective experience, and therefore the formation of its motivational component, then rationally intelligible affective-cum-motivational experience can be conceived as our epistemic access to value in virtue of the sort of property value is. In other words, the formation of rationally intelligible motivational states, and the affective experiences they are part of, when appropriate, can be conceived as constituting the disclosure of value. Notice that I haven't argued for the claim that motivational states are intimately connected to affective experience. All I have argued is that my solution to the puzzle can potentially serve the purposes of philosophers who do argue for an intimate connection of this sort.

5. Learning About Value and Learning To Value

The argument above is meant to give substance to the intuition that in the absence of affective experience we lack a distinctive sort of understanding of value, a sort of understanding we can gain only in the presence of affective experience. In this final section I want to tie together the argument solving the puzzle with the intuition we began with by answering a criticism levelled against the intuition itself. Against the intuition, some philosophers have expressed scepticism at the thought that there is any difference in the content of our conception of value when acquired by means of affective experience and when acquired in the absence of affective experience, say by testimony (see, e.g., Brady Reference Brady2013, 67–68; see also Whiting Reference Whiting2012, 96–97). For instance, according to these philosophers, if we argue that offensiveness is the property of a remark that renders appropriate an experience of anger, then it might be thought that the content of the understanding of offensiveness we acquire by means of anger is such as to include whatever renders our response of anger appropriate. Yet why should we think that there is any need to form rationally intelligible affective experiences like anger in order to learn what renders appropriate a response of anger and thus learn what offensiveness is? Can we not simply learn this simply by rational reflection, or observation, or by testimony? If so, why should we think that there is any difference between the sort of understanding of value acquired by means of affective experience and the sort acquired in its absence? After all, as we have seen in Section 1, even philosophers like Deonna and Teroni admit that we can conceive of an emotionless creature that learns all that there is to learn about value by learning what the relevant appropriate responses are. So how does the account above solving the puzzle help us to explain what such an emotionless creature is missing?

The solution to the puzzle relies on Wiggins’ insight regarding the interdependence of affective response and evaluative concepts. One of the most important yet neglected lessons to be learned from Wiggins’ work on the interdependence of affective response and evaluative concepts is that it is very much doubtful that the generation of such evaluative concepts could occur in a context populated by creatures bereft of any affective capability. This thought comes out in particular in Wiggins’ ‘speculative’ account of the generation of evaluative concepts (Reference Wiggins and WigginsWiggins 1987b, 194–199; see also Wiggins Reference David Wiggins and Wiggins1987a, sec.5; cf. Pettit Reference Pettit1991, 600–602). According to Wiggins, we might imagine this process as starting from the observation that we respond affectively to objects and persons we confront. We then group the objects and persons that seem to elicit similar affective responses and label them with names: the offensive, the repulsive, the shameful, the respectable. This leads us to seek to find the features that allow the objects to fit in the categories we have devised. This entails arguing as to why and whether our actual responses are appropriate, or to put it the other way around, whether the object or person merit that sort of affective response. In so doing, we specify the relation of the relevant ‘evaluative concept-affective response’ pair. Simultaneously, we refine our understanding of the sort of affective response that an object renders appropriate. The process is one of mutual adjustment between the evaluative concept and the affective response. Wiggins characterises this circular yet informative activity as a ‘process of interpersonal education, instruction and mutual enlightenment’ (Wiggins Reference Wiggins and Wiggins1987b, 196). Circularity between affective responses and value is therefore deeply involved in the generation of evaluative concepts. For our purposes, the main point to take from Wiggins’ speculative account is the idea that the most plausible scenario in which creatures like us would generate evaluative concepts is a scenario characterised by the interest that we would feel towards the objects and situations we apprehend in need of being categorised evaluatively. To put the same thought the other way around, Wiggins argues that it is hard to imagine that evaluative concepts such as ‘cowardice’, ‘brutality, ‘heroism’, and so on, could be generated in situations within which subjects were not affectively engaged with the objects they learnt to categorise through those evaluative concepts.

Now, we should view the sort of process that Wiggins speculatively ascribes to the generation of evaluative concepts as a process where the boundaries between learning the value of an object and valuing that same object are blurred. The creatures in Wiggins’ story find themselves in need to categorise objects evaluatively because they find themselves valuing those objects in a way that impacts their interests, needs and concerns. That sort of valuing just is their responding affectively to those objects. I shall call acquiring an understanding of the value of an object by means of valuing that object learning to value. By contrast, I shall call learning about value the acquisition of one's understanding of the value of an object where valuing that very object, and therefore being affectively engaged with it, is not involved. For example, we can learn about the value of things distant from us both historically and culturally by reading a book or by testimony. In learning about value, our learning is greatly enhanced by learning what sort of affective responses those distant cultures might have found appropriate towards the relevant objects. But there can be no serious attempt at learning how to make those objects impact our needs and concerns. Once we draw the distinction between learning about value and learning to value, we can see that affective experience is needed to generate evaluative concepts because their generation is motivated by our valuing the objects instantiating the relevant values. The scepticism above takes for granted that evaluative concepts have already been generated, and only then asks why should we think that affective experience has anything to contribute to our understanding of value. What the solution to the puzzle shows is how rationally intelligible affective experience, and for some the formation of rationally intelligible motivational states, grants us a distinctive epistemic access to evaluative properties, the sort of epistemic access that is needed to generate in the first place the corresponding evaluative concepts. Notice that the argument solving the puzzle is not claiming that the only way in which we can learn value is via affective experience. Indeed, learning about value is a form of learning value. Rather, the solution to the puzzle explains how learning to value works, the sort of learning value that motivates the generation of evaluative concepts in the first place. It is this sort of learning value that an emotionless creature lacks. We can fully appreciate what is at stake in the intuition that in the absence of affective experience we lack a distinctive sort of understanding of value once we have appreciated the distinctive sort of property that value is. Or so I argued in this paper.Footnote ¹⁵