TOWARDS AFFECTIVE-EVALUATIVISM: THE INTENTIONAL STRUCTURE OF UNPLEASANT PAIN EXPERIENCE

Evaluativism about unpleasant pains offers one way to think about unpleasant pain experience. However, extant Evaluativist views do not pay enough attention to the affective dimension of pain experience and the complex relations between the affective, evaluative and sensory dimensions. This paper clariﬁes these relations and provides a view which more closely reﬂects the phenomenology of unpleasant pains. It argues that the intentional structure of paradigmatic unpleasant pain is as follows: unpleasant pains essentially involve a proprietary intentional mode—what I call affective-inter oception—and a distinctive kind of evaluatively qualiﬁed sensory content. The resulting view is Affective-Evaluativism.


I. INTRODUCTION
Pain is one of the most common experiences.If we are lucky, we might only ever undergo everyday pains like stomach cramps and headaches; if we are unlucky, we may suffer through migraines, labour pain, and sciatica.Pain experiences have a rich phenomenology, with sensory, affective, attentional, and motivational dimensions.Moreover, pains are unpleasant: pains, as experienced in normal circumstances by normal subjects, are painful .Reflecting this, typical pains are negative experiences which, ceteris paribus (given the survival value of some pains), we would gladly not have and seek to avoid.That pains are in some sense bad for us, and so have a negative dimension, is a truism.Although what their unpleasantness consists in is contested.
As should be evident from the above, claims about pain experience require qualifications.The caveats 'typical' and 'normal' are sensitive to experiences reported as pains by pain asymbolics, which lack, or have severely reduced, unpleasantness. 1 Recent work focuses on the question 'what makes pains unpleasant'; for those pains that are unpleasant, in what does their unpleasantness consist.One view is Evaluativism, which updates representationalist theories.The latter views claim that pain experiences represent tissue damage, and have a representational content specified in those terms. 2However, to capture the unpleasant character of paradigmatic pain experience, Evaluativism posits additional evaluative content.Pain experiences supposedly represent the relevant bodily states as bad-for-you , and it is their doing so that accounts for their unpleasantness. 3n this paper, I provide an account of the intentional structure of unpleasant pain experience, which builds on the Evaluativist claim that unpleasant pains have evaluative content, although I argue that this is not sufficient for their unpleasantness.On the account provided, the intentional structure of paradigmatic unpleasant pain is as follows: unpleasant pains essentially involve a proprietary intentional mode-what I label affective-interoceptionand a determinately evaluatively qualified sensory content.The resulting view is Affective-Evaluativism.
I take the structure of an experience to be the relations between its essential conscious components.As such, I identify the structural features of unpleasant pain, specifying what components are necessary and sufficient for such experiences.Such an account generates distinctiveness conditions (i.e., what makes an experience an unpleasant pain at all, rather than a different experience) and individuation conditions (i.e., grounds for differentiating between types of unpleasant pain).It is also worth emphasising that the account offered is of pain as a conscious experience with phenomenal character.
The roadmap is as follows.Section II frames pain experience in intentionalist terms.Section III discusses Evaluativism, raising the value-constitution problem.Section IV accounts for the affective dimension of unpleasant pains, explaining what the unpleasantness of pain consists in for Affective-Evaluativism.

II.1 The intentionalist framework
Here, I specify the basic intentionality of pain, detailing its sensory content and interoceptive mode.However, first I outline the theoretical notions I will work with, as part of an Intentionalist framework .Let me introduce two notions from theory of intentionality: (1) experiential content and (2) intentional mode.Experiential content can be specified in terms of the certain way the object is presented to the subject, as how the object seems .Intentional experiences have 'something to say' about their objects; they present their objects as having apparent properties and qualities, as being thus and so .Take visual experience: it is difficult to conceive of a visual experience of a particular object where that object fails to look someway (even if that way is indeterminate).These are the (relatively) determinate aspects under which the particular objects 'show up' in the relevant experiences. 4et me now say something about intentional modes.One motivation for introducing this notion is that intentional states can share the same content but differ in type.For example, judging that < p > is not the same type of mental state as desiring that < p > , even though they share the same content.Likewise, perceiving p is not the same type of intentional experience as remembering p , even though they (putatively) share the same content.So, to capture the intentional structure of an experience, we specify the experiential content and the mode that relates us to that content. 5Whether intentional modes partly determine the phenomenal character of experience is contentious; as we shall see in the case of pain, there are good reasons for thinking that intentional mode contributes to phenomenal character.

II.2 The object and content of pain experience
One of the most fundamental questions about an intentional experience is what its object is; i.e., to ask what the experience is directed towards.Say I have stubbed my toe, what is my experience directed towards?One answer is my toe .Alternatively, say I have a toothache.What is my experience about?One answer is my tooth .The object of pain experience can, therefore, be initially specified as a bodily location, as pointing towards what is going on in the body (or a part thereof).However, while toes and teeth specify relatively precise bodily locations, many pains are comparatively diffuse; consider a sciatic pain that radiates down one's leg.Yet, such cases don't undermine specifying the object of pain experience in terms of bodily locations since such locations can be more or less specific.Regardless of the specificity of the location, the object of pain experience can still be given in terms of a bodily location. 6ext, we should ask what we experience as occurring at the relevant bodily location.One answer is the pain itself .This response reveals an ambiguity.The term 'pain' can refer to both the type of experience and the place where it hurts (the bodily location).Within the intentionalist framework, this would reflect a distinction between (1) an experience of a subject that is directed at a state of the body, and (2) the state of the body itself. 7y interest is in providing a specification of what, in pain, we experience as occurring at the relevant bodily location .Consider Michael Tye's claim that 'whatever else pain is, at its core, it is a bodily sensation.Take away the characteristic sensory component, and no pain remains'.8While we needn't commit to metaphysical claims concerning what is required for a pain to exist (or take a stand on unconscious pains), the idea that pain experience has an essential sensory component is important.Indeed, what is experienced as occurring at a bodily location in pain experience-what is presented as occurring 'there'is a sensory state of the body .
What more precisely can be said about this?One proposal-defended in the following-is that the relevant bodily state is experienced as undergoing a disturbance, as 'something going on' at the relevant bodily location.So, what-it-is-like for a body part to be experienced as having 'a pain in it' is to have an intentional experience directed towards a disturbance occurring at the relevant bodily location.
However, strong representationalists about pain might respond that tissue damage is the more plausible candidate. 9It is important, however, to distinguish metaphysical and phenomenological levels of analysis.On a specific externalist psychosemantics of pain, the referent of the object of pain is specified in terms of co-variation with (more or less) localised tissue damage.Pain experiences would, therefore, 'represent' tissue damage in this externalist sense. 10Yet, regardless of whether that metaphysical view is correct, it is still possible that pains are experienced by their subjects as directed towards disturbances at bodily locations.Michael Tye emphasises that 'Pain O [the intentional object of pain] has as its nature tissue damage…but I am not aware of it as tissue damage'.11So regardless of whether the former metaphysics is correct, it leaves it open how to characterise pain's objectual phenomenology (i.e., what I am aware of the Pain O as ).And we might think it implausible that when I experience a bodily location as having a 'pain in it', I experience that as tissue damage.Rather, we are aware of the intentional object of pain in terms of a disturbance occurring at a bodily location. 12ases of phantom limb and referred pain complicate this analysis.However, in both cases, we don't withdraw the claim that subjects undergo pain experiences, and that the relevant disturbance is experienced as occurring at a (more or less) specific location.Instead, such cases motivate the claim that the bodily location is an intentional location, as an experienced location regardless of whether the relevant body part exists or is the cause of the disturbance. 13iven the analysis so far, we have a proposal concerning the intentional object and experiential content of pain in terms of a disturbance occurring at a bodily location . 14In the following sub-section, I consider objections to this proposal.

II.3 Objections
Why it is more convincing to assume that we experience the relevant body part as disturbed rather than as damaged, unusually affected, or 'not being as it should be'?First, phrases like 'not being as it should be' and 'unusually affected', as applied to bodily states, are alternative terms for (bodily) disturbance (or more profligate descriptions).This is evident from the way they imply a phenomenal registering of disruption or deviation from the 'bodily norm'.Indeed, to experience a body part as unusually affected is plausibly to experience it as undergoing a disturbance (even if such terms pick out specific forms of bodily disturbance).
Such considerations, however, doesn't apply to 'damage' as an alternative to 'disturbance', as the qualification of what is occurring at the bodily location.However, one consideration in favour of the current proposal is as follows.Whereas 'damage' is often represented in non-bodily, non-sensory, and third-personal ways (e.g.I can see my toenail as damaged; I can see that the lock is damaged)-and so have clear applications beyond pain and bodily sensation-disturbance is much more tightly connected to bodily sensation.
Sharpening the above, arguably the following holds.In its primary sense , the concept of disturbance applies to sensory-experiential states, and more specifically to bodily sensations.Put otherwise, we get our primary fix or understanding of the concept of disturbance by reference to how it figures in sensory-experiential states.There are no doubt non-sensory-experiential uses of the concept of disturbance, but these would be extended senses , where the primary sense serves as a model use.The same isn't true of 'damage': the primary sense of the concept of damage applies to something like (im)proper functioning, such that we get our primary fix or understanding of the concept of damage by reference to contexts pertaining to what something is for (e.g.insofar as I see that the lock is damaged it can't serve its proper function), rather than in reference to sensory-experiential states.Perhaps there are sensory-experiential instances of the concept of 'damage' (e.g.I feel damaged by that comment) but these would be extended senses .
If the above is correct, then it is more plausible that disturbance rather than damage would characterise the experiential content of pain experience: the alternative 'damage proposal' would entail that the sensory content of pain experience represents a property instance (damage) whose primary sense isn't sensory-experiential.We would be acquainted with damage first and foremost in non-experiential functional contexts, and only then come to 'apply it' or represent it in pain experience 'derivatively', which is a more complicated and so potentially unattractive view.
Moving on, we can note the variability in the way the bodily disturbance is presented, as varied forms of sensory disturbance.We can do this by reference to intensity, temporality, and quality.Indeed, the disturbance involved in the majority of pain experience involves further determinations relating to intensity (e.g.mild, moderate, and extreme), temporality (e.g.pulsing, throbbing, and beating), and quality (e.g.pricking, stabbing, cramping, and tugging).By identifying these determinations, we recognise that pain's sensory content is multi-faceted and finely-grained; given this, specifications will be approximations, deploying coarsely-grained descriptors which do not fully capture the rich and varied sensational phenomenology. 15ith objections responded to it is plausible to claim that the experiential content of pain experience as follows: a (more or less) specific bodily location, as undergoing a disturbance with a determinate intensity, temporality, and quality.

II.4 The mode of pain
Given what we have said about pain's experiential content, why is it necessary to introduce an intentional mode for pain?First, consider the difference in phenomenal character-used as a diagnostic tool by physicians to check sensitivity after suspected spinal injury-between seeing the doctor prick your toe while 'feeling nothing', and 'feeling' that same prick.Both experiences are about, and so have as part of their intentional content, the body, but in latter case we feel the body .
Such examples motivate the following thought.There are important differences between the way we 'relate to objects' or 'how we represent objects' in paradigmatic exteroceptive sense-perception vs various forms of bodily feeling, regardless of content (i.e., regardless of what we represent).Put otherwise, our awareness of our bodily states is a different mode of awareness from that characteristic of exteroceptive sense perception (even if there are similarities), and certainly, modes of bodily awareness are phenomenally distinct from cognitive modes like judgement and belief. 16o, what is the intentional mode of pain experience?One answer is interoception.Following A.D. Craig, interoception can be defined as the mode of awareness of the physiological condition of the body, also covering tissues of the body, and so including personal level feelings of pain, temperature, itch, sensual touch (i.e., the feeling of being touched), and other related bodily feelings.17On this, admittedly inclusive , definition it would be incorrect to think of experiences in the interoceptive mode as only covering experiential presentations 'from the inside', in the sense of internal sensations (e.g.visceral feelings), since interoception would also cover experiential presentations relating to the surface of the skin. 18t is also worth noting that interoception (inclusively understood) is preferable to nociception as the relevant intentional mode.As it is commonly understood in the psychological literature nociception refers to the subpersonal causal mechanisms which subtend or at that level causally contribute to the subjective experience of pain, and are usually framed in terms of neural encoding of impending or actual tissue damage. 19As such, not only would there be a problem of circularity-pain is a bodily disturbance represented in the mode of pain -but nociception is not obviously something that occurs at the experiential level, and so would be inappropriate for our analysis of the intentional structure of pain experience .

II.5 Distinctiveness and individuation
We now have a first approximation to the intentional structure of pain experience.Pain experience is an intentional experience, in the mode of interoception, where the experiential content is that of a (more or less) specific bodily location as undergoing a disturbance with a determinate intensity, temporality, and quality (I refer to this complex as bodily disturbance ).Pain experiences, so understood, are first-order intentional experiences; their intentionality is not directed towards the experience itself but to the extramental world.
An account of the intentional structure of pain experience provides distinctiveness and individuation conditions.Taking distinctiveness conditions first, an experience being in the interoceptive mode will not suffice for being a pain (although it is necessary) since the interoceptive mode covers bodily feelings that are not pains.Likewise, the relevant sensory content does not help if itches and tickles can have similar sensory content to pain experience. 20oncerning individuation of pains, interoception will also not be much help since it is the mode of all pain experience.However, the sensory content does more work; the variable intensity, temporality, and quality of the disturbance are essential to the individuation of different types of pain.Consider an experience of a moderate, deep and tender throbbing in the bicep (i.e., a muscle strain), compared with an extreme, sharp pulsating feeling behind the eye (i.e., a migraine).Variation in these sensory parameters, as phenomenal properties of bodily disturbance, accounts for the variation we experience in pain's sensory dimension.
The partiality of this Intentionalist account is not surprising.As noted by Murat Aydede and Matthew Fulkerson, pain experiences also include an affective-motivational phenomenology, and 'experiencing affective qualities doesn't have the same kind of phenomenology as that of experiencing standard sensible qualities'. 21We need to go beyond the account so far presented, even if its specification of the basic intentionality of pains is accurate.

III.1 Framing evaluativism
The account provided so far does not account for the affective, evaluative, and motivational dimensions of pain experience, which are often claimed to constitute its unpleasantness.Evaluativism is an Intentionalist view of unpleasant pains, which claims that paradigmatic pain experiences have evaluative content, which explains their unpleasantness: pain experiences present relevant bodily disturbances as bad-for-you , and it is in virtue of doing so that they are unpleasant. 22On this view, there is only one intentional presentation, namely of a bodily disturbance as bad-for-one; subjects don't undergo any further second-order representation (evaluative or otherwise), which takes the experience itself as an intentional object.Evaluativism frames unpleasant pains as first-order evaluative intentional experiences.
Linking to the above, Evaluativism explains the difference between unpleasant and asymbolic pains as follows: asymbolic pains are not unpleasant because they lack the unpleasantness-constituting layer of evaluative content. 23Note, though, in paradigmatic unpleasant pains, the additional unpleasantness-constituting layer of evaluative content is not experienced as additional.Paradigmatic unpleasant pains are phenomenologically unified: subjects enjoy one intentional presentation, namely an interoceptive experience, which presents a bodily disturbance as bad-for-one .
Given our interest in the intentional structure of unpleasant pains questions need addressing.First, is the posited evaluative content experiential content?Given that evaluative content is said to constitute the experience being unpleasant, and we can only make sense of a mental state being unpleasant as manifest in its phenomenology, then the posited evaluative content is experiential content which reflects how things evaluatively seem (and it is their seeming so that constitutes their unpleasantness).
So clarified, the critic may ask what in the phenomenology attests to the presence of the evaluative property of badness-for-one, as qualifying the bodily disturbance.Part of the problem is due to the determinable character of the evaluative content.In the McGill Pain Questionnaire, which is used to report the character of different pain experiences and specifies descriptors, the term 'bad' does not figure in response to questions like 'what does your pain feel like'. 24Instead, subjects use evaluative anchor words, such as distracting , blinding, annoying, troublesome, miserable, unbearable, wretched, nagging, agonising, dreadful,  torturing, and savage .Now, we should be cautious in moving too quickly from the use of adjectives in first-person reports to the positing of phenomenal properties.However, we are justified in thinking that a nagging bodily disturbance, for example, feels evaluatively different from a dreadful bodily disturbance, or indeed from an unbearable bodily disturbance-that bodily disturbances can seem evaluatively different in different pains.
So, if we are looking for a way things evaluatively seem in unpleasant pains, as motivating positing evaluative experiential content, then it is more plausible that unpleasant pains present felt bodily disturbances under these pain-specific determinate evaluative properties.There is no folk psychological term that classifies these together, so I refer to them as p-evaluative properties.They are all negatively valenced, so the idea of an evaluative content of badness-for-one is not wrong; it is just more phenomenologically accurate to claim that pain experiences present determinate p -evaluative properties.
The best way to think of the relation between 'badness-for-one' and p -evaluative properties is along the lines of the thick-thin distinction which is often applied to evaluative properties (examples of thin evaluative properties are the good , bad, valuable , and disvaluable ).The thick-thin distinction is an instance of the deter minate-deter minable relation; thick evaluative properties are determinates of the determinable (dis)value.In our case p -evaluative properties-such as the blinding and unbearable -are 'thick' determinates of the determinable 'bad-for-one'.
Such properties are relational in that a bodily disturbance that is agonisingfor-me needn't be agonising for a 'hard-nut'.However, that relationality, while revealed in reflection on the nature of such properties, isn't usually phenomenologically manifest.In the grip of a headache, the terribleness of that specific disturbance isn't experienced as relational, but rather as a monadic property of the bodily disturbance.So, it is important to emphasise that these p -evaluative properties are to be understood as the way things evaluatively seem; they are properties that seem to qualify bodily disturbances, as reflected in the way the terms for them are phenomenal-recognitional concepts.

III.2 The phenomenological supervenience claim
If we accept the above qualifications to Evaluativism, then paradigmatic pain experience has determinate evaluative content.However, what is the relation between the sensory and evaluative contents?Evaluativists have not said, but one option-often appealed to for evaluative properties-is that p -evaluative properties are higher-order properties that stand in supervenience relations to conjunctive co-instantiations of the basal sensory properties of bodily disturbance (i.e., intensity, temporality, and quality).This is borne out phenomenologically in the way variation in the sensory content of pain can often lead to variation in the experienced p -evaluative qualification.For example, a mild, pulsating feeling spread across one's forehead may present as annoying .In contrast, if it develops into an intense, sharp stabbing behind the eye it may present as agonising.
Further to the above, in paradigmatic cases, the sensory features are manifest to the subject as resulting in the p -evaluative property.Let me explain this by way of contrast.Attending a social gathering, I end up in conversation with a repugnant individual.His demeanour offends me, and there is something about him that I cannot stand.The relevant evaluative property is consciously given.Yet during my emotional experience, I was aware of the individual's Downloaded from https://academic.oup.com/pq/advance-article/doi/10.1093/pq/pqae002/7588593 by guest on 06 February 2024 repugnance without being aware of which non-evaluative basal properties it supervenes on. 25So, in certain emotion cases, one can be consciously aware of a determinate higher-order evaluative property without a corresponding awareness of the specific conjunction of subtending properties on which it is resultant.
However, this is not the norm for pain experiences.In the case of a migraine, the relevant bodily disturbance seems agonising in virtue of its being an intense, sharp stabbing behind my eye, and I am all too aware of these sensory properties.Likewise, in case of sciatica, the relevant bodily disturbance seems intolerable in virtue of its being a constant tingling numbness down the side of my leg.So, in paradigm cases, the relevant basal properties are phenomenologically salient as resulting in the p -evaluative properties.Call this the phenomenological supervenience claim (PSC hereafter): the sensory features of unpleasant pains seem themselves to result in the relevant p -evaluative properties.
However, this might seem controversial, so let me consider putative cases in which the PSC does not hold.Consider chronic pains in which patients use less sensory and more evaluative and affective vocabulary to describe their experiences,26 such that perhaps what is awful isn't necessarily the sensory character but rather the pain's non-sensory debilitating effects.
However, it is important to distinguish between pains' instrumental and non-instrumental evaluative character or 'badness'.What I-and other Evaluativists-are trying to capture is the experienced non-instrumental unpleasantness of pain. 27And regardless of the negative evaluative standing of instrumental aspects of chronic pain, the pain experiences constitutive of chronic pain are themselves non-instrumentally unpleasant .Now, chronic pain may also be awful in a non-sensory way because of its awful effects (its instrumental badness), and in that respect, the PSC wouldn't hold, but this is orthogonal to whether it holds concerning the non-instrumental badness of the pain experiences constitutive of the chronic pain.And once we have got the relevant non-instrumental evaluative dimension in view, the PSC remains plausible: while my chronic sciatica is certainly awful in that it has a range of debilitating effects, within the episodes of sciatica themselves the relevant sensory features-the persistent, dull feeling that runs down my left side-seem to give rise to the relevant p -evaluative property of 'awfulness'.
Let me now note further worries about the PSC.At the start of the section, it was said that the PSC is borne out in how variation in the sensory content of pain 'can often' lead to variation in the experienced p -evaluative qualification of the relevant bodily disturbance.This qualification raises the issue of dissociation between sensory and evaluative aspects.After all, couldn't two subjects' pains be sensorily identical but vary as to 'how bad' the relevant bodily disturbance feels?The answer is yes, but this doesn't undermine the PSC: for both subjects it could still be that the sensory features seem themselves to result in the way the relevant bodily disturbance evaluatively seems, even if the way it evaluatively seems varies.
However, consider a case in which the sensory content changes without any change to the p -evaluative qualities.Say I am experiencing a bodily disturbance that turns from burning (sensory) to pulsing (sensory) but nonetheless both disturbances seem 'equally awful'.However, again we don't have a counterexample to the PSC.After all, while the burning sensation vs the pulsing sensation may seem 'equally awful', they can both still seem equally awful in virtue of their different sensory features.Further to this, we might say they are equally awful but in different ways, where the reference to 'different ways' would be framed in terms of sensory variation (highlighting the tight connection between the sensory and evaluative features of unpleasant pain).

III.3 The value-constitution problem
In bringing together the analysis of Evaluativism so far, there emerges a problem, which I call the value-constitution problem.Consider the following steps: ativism claims about asymbolic pain, the asymbolic should be able to enjoy an experience with identical sensory content to the normal subject.All the asymbolic (putatively) lacks is the additional (unpleasantness-constituting) evaluative content.(Step 5) However, (2) and (4) are inconsistent.Why is it that the asymbolic's sensory content, which is identical to the normal subject's, does not give rise to the relevant evaluative content that it does for the normal subject?
Take Abe, who is a pain asymbolic, and Norm, who is a normal subject.Both Abe and Norm are currently experiencing an intense, sharp, pulsating feeling behind the eye.Yet ex hypothesi for Evaluativism, only Norm's experience has the evaluative content which constitutes its unpleasantness-Norm experiences his bodily disturbance as intolerable .Contrastingly, Abe's bodily disturbance lacks this layer of evaluative content and hence per Evaluativism is not unpleasant. 28Why is it that Abe's sensory content, which is identical to Norm's, does not give rise to the relevant evaluative content that it does for Norm?Evaluativism should explain why the value-constituting relation posited in Step 2 does not obtain in the asymbolic case.This is the value-constitution problem .Now there might be an easy way for the Evaluativist to respond.Namely, that since the PSC was only said to hold in paradigmatic cases, and since the pains of asymbolics are non-paradigmatic, then we reject Step 2-there is no need to assume for asymbolic pains that the relevant supervenience relation should hold.However, while this is part of a response to the valueconstitution problem what it misses is any substantive explanation of why that is the case.To say that the sensory content of asymbolics pains don't give rise to any p -evaluative content because they are non-paradigmatic qua the PSC not holding of them just serves to highlight the question of what it is about asymbolic pains such that the PSC doesn't hold of them; we still need an explanation of what it is about asymobolic pains such that they are non-paradigmatic in this respect.
One suggestion is that there are other factors that explain why we don't get the supervenient p -evaluative properties in pain asymbolia.One option discussed in the literature is bodily care. 29As Bain puts it, 'your interoceptive experience won't represent a condition of your body as bad for you-hence won't be unpleasant-if you don't care about your body'. 30Note, bodily care is put forward as a necessary enabling condition on pains having the evaluative content they do, where that concerns pain's non-instrumental unpleasantness, rather that something that might contribute to pain's instrumental 'badness'.
How convincing is this appeal?Consider the following thoughtexperiment.Say we accept that bodily care is a necessary condition on pain's having evaluative content (thus being unpleasant as per Evaluativism).Isn't it possible that an asmybolic subject could be habituated to 'care for her body' by internalising, over a sufficiently long period and with help from medical professionals and psychologists, pain specific 'care-manifesting' actions and behaviours (e.g.drawing one's hand immediately away from the fire, rather than requiring a process of inferential reasoning to appreciate why doing so is in their interest)?If this habituation and internalisation of bodily care is possible, then it should allow the asymbolic to overcome their condition.However, that is counter-intuitive-it is implausible that unpleasantness can emerge in this way.Now, the defender of the appeal to bodily care could reply that the notion of care they are adverting to is a primitive, biologically hard-wired capacity, or mechanism, in which case such possibility would be ruled out in advance.
Given the above response let me mention a further issue with the appeal to bodily care.Frederique de Vignemont draws attention to problematic cases.First, she notes that the notion of bodily care in play is first-personal, in that one should be thought to care about specifically the body represented as one's own .In that sense, bodily care requires some sense of bodily ownership.Given transitivity, if unpleasant pain requires bodily care and bodily care requires bodily ownership, then unpleasant pain requires bodily ownership.So we get a prediction: 'if the body part that is hurt feels alien, one should not be able to feel pain…' 31 However, this prediction is falsified by cases of somatoparaphrenia in which patients experience a lack of ownership over a body part after a lesion of the right parietal lobe.As Vignemont describes this case, 'patients seem to experience "normal" pain, no matter what this is, and to react normally to their painful experience, but they have a sense of dis-ownership of the body part that is in pain… If their "alien" hand is hurt, they wince and spontaneously verbally complain….'32So, it is possible to feel unpleasant pain in the absence of bodily ownership concerning the relevant bodily part.Yet, this falsifies the aforementioned prediction.And then it looks like bodily care isn't required for unpleasant pain: as per Evaluativism (with the bodily care condition on evaluative content), the pains of subjects with somatoparaphrenia should be like asymbolic pain in the relevant respect, but they are not.Naturally this turns on accepting the idea that bodily care implies bodily ownership, but if that is plausible, then the above case puts empirical pressure on the appeal to 'care-lack' as an explanation of why asymbolic pains are not unpleasant.

III.4 A final issue
There is a further issue for Evaluativism once we consider in more detail the type of experiences that unpleasant pains are.To frame this, let me outline a feature of perceptual experiences.In a visual experience of my garden, I enjoy a non-doxastic direct presentation of various trees and shrubs, and their colours, shapes, and spatial relations-I seemingly encounter the relevant part of the environment, or put otherwise the relevant part of the environment is impressed on the senses. 33ow, Evaluativists appeal to the perceptuality of pains to explain their unpleasantness when faced with counter-examples of states with the same evaluative content (e.g.evaluative beliefs) but which it is counter-intuitive to claim are unpleasant. 34The thought is that analogously to sense-perceptual experience, interoceptive experience is an 'impression': in unpleasant pain, we have an interoceptive 'impression' of a disturbance as bad-for-one.So, unpleasant pain experiences are claimed to be analogous to sense-perceptual experiences in having a non-doxastic presentational phenomenology. 35owever, in contrast to sense-perception, pain experience is not merely a matter of undergoing an 'impression' of a bodily disturbance, even evaluatively qualified.Rather there is a fundamentally reactive or responsive character to paradigmatic unpleasant pain, which is missing in talk of 'impressions' and 'encounters'-or at least so I hope to show this is the case in the following section.Yet, if there is a distinctive affective dimension to unpleasant pains, which cannot be captured in terms of evaluative content and the interoceptive mode, then perhaps we need to posit something additional to fully capture unpleasant pains.In the final section, I show that affectivity of a specific sort is salient in unpleasant pain experiences, is essential to capturing their unpleasantness, and provides a better solution to the value-constitution problem than standard Evaluativism.

IV.1 Pain's affective attitude
Let me start this section by making clear its structure.First, I summarise Affective-Evaluativism , then I specify the phenomenal profile of the affective attitude I take to be constitutive of unpleasant pains (4.1).I go on to explain pain's unpleasantness on this account, claiming that the relevant affective-attitude of strong aversion is the proprietary mode of unpleasant pain, also showing how Affective-Evaluativism solves the value-constitution problem (4.2 and 4.3).
Here is a summary of the view that is explicated and defended.According to Affective-Evaluativism, unpleasant pain experience includes a negatively valenced affective attitude.More specifically, it includes a (typically strong) felt aversion towards the relevant bodily disturbance (evaluatively qualified).This affective attitude of felt aversion is the mechanism through which attention is focused in unpleasant pain.Further to this, in typical cases, this attitude of aversion has an action-ready character, involving a potentiation of one's body as poised to act.This attitude of felt aversion is a first-order intentional attitude.Felt aversion is not a separate reaction to an experience that could be characterised as an unpleasant pain independently of it, but rather is a constituent of unpleasant pain experience.This first-order affective attitude serves to present the p -evaluatively qualified bodily disturbance; put otherwise, this aversive attitude is the affective mode in which we experience a bodily disturbance as awful, intolerable, generally unpleasant (etc.).This characterisation points towards a distinctive feature, namely a negatively valenced, affective attitude, which is part of unpleasant pains, and which has a strong connection to motivating avoidance behaviour.
In this context, it is important to be clear that the relevant attitude concerns what is sometimes called primary affect, as a first-order attitude intrinsic to unpleasant pain experience itself, rather than a form of 'secondary affect', as some form of aversion to the pain experience itself .For example, it is sometimes claimed that patients under the influence of morphine still experience their pains as unpleasant, but that what is 'neutralised' are their attitudes towards the pain experience itself , so 'neutralising' secondary affect.By the lights of Affective-Evaluativism insofar as such patients still experience their pains as unpleasant this requires the presence of 'primary affect', which is theorised in terms of the relevant first-order aversive attitude.
Let me unpack this account.By specifying this affective-attitude as one of (typically strong) felt aversion, we capture the idea-broached at the end of the previous section-that we need to account for unpleasant pain's reactive character of feeling affected.The existence of such a component in affective experience is well-attested.Anthony Marcel and John Lambie talk similarly of an experiential correspondence of significance, as a felt orientation towards or away from relevant features of our environment, as a valenced component of experience which connects us with, and makes us aware of, significant features of that environment. 36In the context of pain, this experiential correspondence of significance is a negatively valenced, affective attitude of aversion.
Building on the above, let me detail the connection between this affective dimension and attention.A central part of our affect-laden responses to bodily disturbances is the interruption of attention. 37For example, chatting to my partner, I am paying attention to what they are saying.I then stub my toe.My attention is interrupted and diverted to the searing, sharp throbbing in my toe.While this interruption of attention is a characteristic of unpleasant pains, the bodily disturbance captures attention in a peremptory way.
First, note that it is not a deliberate paying of attention; the diversion of attention is a passive capturing .Think again of attention being directed to my searing toe; there is no need for any conscious effort-attention is automatically and immediately diverted.Second, once attention is so passively captured, it is very difficult to redirect: it can be incredibly difficult to concentrate on anything other than the relevant bodily disturbance (evaluatively qualified) when undergoing unpleasant pain.Once attention is automatically and immediately captured, attention is consumed in a primitively compelling way.And while we may develop pain-specific coping mechanisms which involve (more or less successful) attempts to divert attention, the need to develop them attests to the typical situation being one of a primitively compelling consumption of attention.This is borne out in the advice given by medical practitioners that pain control is more successful when analgesics are taken prophylactically; once 'the pain' sets in, it becomes more difficult to ignore since attention is consumed in a peremptory way. 38he conscious mechanism through which attention is focused in this peremptory way is the first-order attitude of aversion.Importantly, the experienced strength of this aversive response varies in relation to the experienced strength of the p -evaluative property.This is reflected in the way a bodily disturbance which presents as merely annoying, will not capture and consume attention to the same extent as one which presents as excruciating-the first-order affective response will not be as strong.
I now emphasise the connection between the attitude of aversion and motivation.It is misleading to think that first-order pain experience is, or involves, an action in any simple sense-pain experiences are one thing, actions consequent on them, as further responses to them, are another.Nonetheless, unpleasant pain experience is typically motivational.We can gloss this by saying that it non-inferentially and immediately prompts intentional action (e.g.avoidance behaviours) and provides us with reasons for action. 39Reflex withdrawals do not have this character-rather the stimulus is registered (at least partially) non-consciously at a rapid speed.
One way to capture this motivational dimension is to say that the attitude of aversion typically involves action-readiness, as preparedness to do something.We can clarify this by appealing to proprioceptive phenomenology, as one's body being poised to act, with accompanying muscle tension (as often reflected in wincing).However, since the affective and motivational components of unpleasant pain experience can dissociate in atypical cases, it may be too strong to claim that the affective attitude of strong aversion is necessarily an action-ready attitude, although this is the norm (see Section IV.4).

IV.2 Pain's overall intentional structure, unpleasantness, and distinctiveness
According to Affective-Evaluativism, the affective dimension of pain experience is not part of its intentional content.Rather, the attitude of strong aversion serves as a further intentional mode.Put otherwise, this aversive attitude is an affective mode to experience a bodily disturbance as awful , intolerable , excruciating , (etc.).If this is the correct understanding of the affective dimension of unpleasant pain experience, then its intentional structure includes a distinctive kind of affective intentionality.
Based on what we have said so far, unpleasant pain experience involves two types of content and mode: we have sensory content, and the intentional mode which relates us to this is interoception; and we have p -evaluative content, and the intentional mode which relates us to this is an affective-attitude.It is a further question how this all hangs together.It is implausible pain experience is such that we have an interoceptive experience with the relevant sensory content, and then a separate (phenomenologically distinguishable) affective experience with p -evaluative content.Rather, unpleasant pain experience is phenomenologically unified.To reflect the phenomenological unity of unpleasant pain experience, while building in that attitude of aversion, Affective-Evaluativism specifies the intentional structure as follows: unpleasant pain experience is an experience in a proprietary intentional mode , what we can call affective-interoception, which is phenomenologically manifest to the subject of those experiences as an attitude of strong aversion directed 'inward' (interoceptively) towards its p -evaluatively qualified bodily disturbance.
Let me explain this picture in more detail by explaining how pain's unpleasantness is to be understood.We can phrase the question as follows: given unpleasant pain experience has this intentional structure what does its unpleasantness consist in?At least part of the answer (see discussion below) is that both the additional affective mode and p -evaluative content constitute the overall phenomenal character of a negatively valenced affective-interoceptive experience-undergoing an experience, so characterised, is to experience unpleasant pain.Although what is distinctive about unpleasant pains is that part of the mode through which they present p -evaluative properties is a conscious (often action-ready) attitude of aversion.
Tying into the distinctiveness point, enjoying an affective-interoceptive experience with a content spelt out in terms of a p -evaluatively qualified bodily disturbance is both necessary and sufficient for unpleasant pain experience (although see below on the importance of getting the PSC in view).To stress again, undergoing an interoceptive experience would not suffice for being an unpleasant pain given that bodily feelings are interoceptive experiences with the relevant sensory content (or something sufficiently similar).Likewise, exteroceptive affective experiences (e.g.world-directed emotions) are different affective states than unpleasant pains.It is promising that Affective-Evaluativism provides criteria for distinguishing unpleasant pains from states similar to them (including other affective states).So, the account generates the following condition: any intentional experience which can be characterised as an affective-interoceptive one, whereby a conscious attitude of strong aversion is directed towards a p -evaluatively qualified bodily disturbance, will be an unpleasant pain.However, don't we find experiences with this intentional structure but which we don't ordinarily categorise as unpleasant pains, say unpleasant itch, and hunger and thirst?If so, then arguably Affective-Evaluativism merely provides an account of unpleasant bodily sensations, thus failing to provide an account of the distinctiveness of unpleasant pain. 40And given our phenomenological focus this is pressing; after all, some unpleasant bodily sensations have a different phenomenal profile-they non-trivially feel different-from paradigmatic pains, and so adverting to non-phenomenal differences won't help.
Affective-Evaluativism can be maintained as a bone-fide theory of unpleasant pains , rather than merely of unpleasant bodily sensations, insofar as it is committed to the PSC.Remember the PSC: the sensory features of unpleasant pains seem themselves to result in the relevant p -evaluative properties.Now Section III.2 gave reasons for thinking the PSC true of paradigmatic unpleasant pains.However, note if the PSC is part of the phenomenal profile of unpleasant pains, then we now have a criterion for distinguishing unpleasant pains from unpleasant bodily sensations.Let me explain.If unpleasant pains are experiences with the above specified intentional structure and in which the PSC holds, then pain's unpleasantness would be phenomenologically different from the putative unpleasantness of any other unpleasant bodily sensation in which the PSC didn't hold.Now is it the case that the PSC doesn't hold of experiences we classify as unpleasant bodily sensations but not as pains?Take unpleasant itch: it is more plausible that the 'awful' evaluative character of an itch doesn't supervene on the relevant sensory aspects, but rather is more closely connected to intrusive instrumental effects, for example, that the bodily disturbance characteristic of an itch is distracting, or gets (predictably) worse when one scratches it.In that sense, the more specific phenomenal profile of pain qua unpleasantness turns out to be phenomenally different from the unpleasantness of unpleasant itch in virtue of the tight relation that obtains between the sensory and evaluative contents in the former case.
However, while this might rule out unpleasant itch as an instance of pain experience, arguably it won't work for cases of unpleasant hunger and thirst.It is certainly not implausible that the negative evaluative character of the bodily disturbance characteristic of intense hunger, say, and so hunger which we want to say is unpleasant, seems to supervene on the relevant sensory aspects, that is the gnawing empty feeling in the pit of one's stomach; likewise, for intense thirst.Now, there is one obvious response that Affective-Evaluativism can make in light of the above considerations.Namely, that at least intense unpleasant hunger and thirst-if not necessarily so-called regular hunger and thirst, which are not obviously unpleasant anyway (see below)-are in fact instances of unpleasant pain.After all, we are happy to talk of hunger pains , and there is a philosophical reason why: namely because they have the same intentionalphenomenal structure as more paradigmatic unpleasant pains (they are of the same phenomenal kind).Indeed, as Michele Ombrato and Edgar Phillips put it, 'in cases of extreme deprivation, hunger is viewed as a form of abject suffering'. 41As such, this isn't a particularly bad place for Affective-Evaluativism to land with respect to intense and unpleasant hunger and thirst.
Note though, for this explanation to work, we would have to make a distinction between regular hunger and thirst, and intense hunger and thirst.It is only the latter that are to count as unpleasant pains.The view then would have to add that regular hunger, say, is either not unpleasant or has an unpleasantness that is different from that of unpleasant pain; perhaps we could here provide the same analysis as we did above for unpleasant itch: the 'annoying' evaluative character of regular hunger doesn't obviously supervene on the relevant sensory aspects, but rather turns on its intrusive instrumental effects, e.g. that the bodily disturbance characteristic of regular hunger is distracting.While this is not implausible, more work needs to be done concerning these cases of supposed unpleasant bodily sensations.
Moving on, sensory content was previously seen to be important to individuating pains.We can complete this picture by adding that p -evaluative content and the related affective response are also essential to such individuation.Remember, the evaluative content of unpleasant pains was specified in terms of a range of pain-specific determinate evaluative properties, and the strength of these p -evaluative properties is mirrored in the strength of the attitude of aversion.When we individuate pains, we refer to pains that were dreadful, or unbearable, or those which we could tolerate.Indeed, a medical professional will not just ask about the location of the bodily disturbance, or its duration, for example, but also 'how bad is the pain', 'how much does it hurt'.In answers to such questions, we specify, more or less precisely, the p -evaluative content of our unpleasant pains and the related attitudes of aversion.So, Affective-Evaluativism provides criteria for individuation within the class of unpleasant pains.

IV.3 Resolving the value-constitution problem
With this account in hand, we can offer a more plausible solution to the value-constitution problem than (non-affective) Evaluativism.Remember, this is the problem of explaining why there is a block on sensory content-which 41 Ombrato and Phillips (2021 : 517).As they note (2021: 517), however, there are feelings of hunger (and plausibly thirst also) that 'are often inextricably intertwined with a host of positive affective phenomenon: feelings connected to the anticipation of satisfaction', but in that case, we are not trying to account for unpleasant bodily sensations and their differentiation (or not) from unpleasant pains.Downloaded from https://academic.oup.com/pq/advance-article/doi/10.1093/pq/pqae002/7588593 by guest on 06 February 2024 can be identical across Abe's and Norm's pain experiences-giving rise to evaluative content in Abe's case.
Claiming asymbolics lack the relevant layer of p -evaluative content is only a part of the story.On one interpretation of asymbolics reports of not being 'bothered' by the relevant sensory disturbances, they also lack affective attitudes.Understood along the lines of Affective-Evaluativism, what they lack is the possibility of having experiences in the relevant affective mode: since asymbolics lack the relevant affective attitudes of strong aversion, they lack the personal level intentional mechanism through which evaluative content is 'taken up' in unpleasant pain.This is borne out in significant differences in the motivational profile and phenomenology of attention.The intentional objects of Abe's pains-the bodily disturbances-will not capture and consume attention in the primitively compelling way that Norm's do, and certainly don't motivate the normal range of pain behaviours.And the reason for this is that Abe's experience lacks the relevant affective-attitudes, which are the intentional mechanisms through which attention is focused in a peremptory way.
However, in solving the value-constitution problem in this way, we have pushed the problem one rung back.We still need to answer why it is that Abe lacks the relevant affective-attitude, as an additional personal level explanation, rather than one in terms of a-typical neurological features of asymbolics (e.g.damaged insular cortex).Given we don't want to appeal to the absence of bodily care something else is required.
One option is to appeal to a subject's (1) network of dispositional psychological attitudes concerning matters relevant to pain and pain experiences, which in Norm's case, include things as diverse as standing-beliefs about object-types that are closely connected to pain for humans (e.g.sharp and hot things) to internalised cultural norms surrounding pain (e.g.what is an appropriate level of pain for which to seek medical assistance), where this affective network is constantly updating and revisable (often on the basis of occurrent pain experiences), and (2) a background of pain-related pre-intentional capacities, some of which are common to all normal humans in virtue of their biology, examples of which often take the form of know-how, say knowing how to move one's limbs without injuring oneself, or kno wing ho w to (attempt to) redirect attention when in pain.This pain-related affective network and background constitute an individual's personal level pain psychology, and as such plausibly act as an enabling condition for affective representation in pain experience, also contributing to the precise character that those experiential pain-representations have in any given case.Describing this pain psychology in detail is a significant undertaking, and parts of it may resist articulation since many of the relevant states of the affective network , for example, are unconscious.Nevertheless, its import in the present context should be clear.
What is impaired in the asymbolic case-such that they lack the relevant affective-attitudes-is precisely this enabling personal level pain psychology, which due to neurological damage or abnormality doesn't function in the way it does for a normal subject.Indeed, Stengel's and Berthier's asymbolic patients were noted to be bad at learning what circumstances require avoidance behaviour, and sometimes engaged in self-harm behaviours. 42These behaviours suggest malfunctioning of the affective network.In the case of the affective background, it has been noted that individuals-similar to asymbolics in many respects-who suffer from congenital pain insensitivity often use their hands and feet in such a way that they develop severe injuries to those limbs due to repeated activity, ostensibly lacking pain-relevant pre-intentional capacities.Importantly, the relative complexity of these deficiencies and impairments to asymbolic pain psychology go well-beyond 'care-lack'.What is being appealed to is impairments of a more complex and multi-faceted pain psychology.No doubt more needs to be said about this, but by drawing on these ideas Affective-Evaluativism has the resources to resolve the valueconstitution problem in a distinctive way.

IV.4 A contrast with FOD theories
In closing let me contrast Affective-Evaluativism with a similar approach.A critic might ask whether the first-order affective dimension in unpleasant pain experience is not better construed as a first-order intrinsic desire towards the intentional object.Namely, that the bodily disturbance cease.Building on this, the unpleasantness of pain experience would consist in the subjective-desire frustration attendant to the non-satisfaction of the relevant first-order conative attitude-this is the first-order-desire view (FOD theories hereafter). 43here are disanalogies between FOD theories and Affective-Evaluativism, principally that Affective-Evaluativism appeals to evaluative content, but let me here focus on contrasts that pertain to the character of the relevant affective attitude.
First, aversiveness towards P is not identical with a conative attitude towards P, such that P should stop or cease ; aversion is not the same thing as wanting something to cease.Say I am listening to a speech which strikes me as repugnant, and to which I have an aversive response.I need not, during that response, desire that the speech immediately cease or stop; I need not have a conative attitude with a content approximating to 'I wish they would (immediately) stop speaking'.And while my aversive response may precipitate some such conative attitude, I can have the former without the latter.If the attitudinal dimension Affective-Evaluativism is trying to capture were identical with a conative attitude (that the relevant intentional object stop occurring), then it should be very difficult to imagine cases where we have the relevant aversive 42 See Berthier et al. (1988 : 41-3). 43See Aydede (2014 : 119-33), Jacobson (2018 : 1-27).Downloaded from https://academic.oup.com/pq/advance-article/doi/10.1093/pq/pqae002/7588593 by guest on 06 February 2024 attitude but lack the conative attitude.Since that is not the case, this shows that Affective-Evaluativism and the FOD view mark out different proposals.
Affective-Evaluativism should, nonetheless, concede that when it comes to pain, aversion is closely connected to conative attitudes, even if they are not identical.It can accept that a pain experience that experientially presents p -determinate evaluative properties on the basis of an affective-attitude of aversion normally brings about a conative-attitude to the effect that the object of the pain experience immediately ceases or stops .This would explain why we encounter some imaginative resistance when trying to imagine a pain that is unpleasant but in which the subject has no wish whatsoever for the pain to stop (although note the cases of morphine patients discussed in IV.1).However, note the dispute: Affective-Evaluativism accepts that unpleasant pains paradigmatically cause conative states.What is contested is whether the affective attitude constitutive of unpleasant pain is itself a conative state with the character of 'wanting' the relevant felt bodily disturbance to stop (and further, that it is the frustration of some such desire that makes pains unpleasant ).Affective-Evaluativism can accept that it is hard to imagine an unpleasant pain in which the subject has no wish whatsoever for that pain to stop , but not concede that the primary affective-attitude constitutive of (rather than merely caused by) unpleasant pain is as the FOD says it is, rather than as the Affective-Evaluativist says it is.
Turning to criticism of FOD views, consider that conative-attitudes are necessarily motivational: insofar as one has a conative attitude of wanting the circumstances represented by the pain experience to immediately stop or cease, then a necessary part of conative phenomenology is a categorical pull towards that valued end-we feel action needs to be taken to alter the situation; a conative attitude that was not motivational would not warrant the label.Yet, unpleasant pains and their 'motivational oomph' can dissociate in atypical cases, suggesting that unpleasant pains are only typically motivational . 44f such cases are possible, then unpleasant pains don't necessarily have the motivational profile that the FOD views entails.The burden is on the defender of the FOD view to explain away these cases.Contrastingly, it is a benefit of Affective-Evaluativism that it can accommodate them, insofar as the attitude of strong aversion is not essentially tied to motivation.

V. CONCLUSION
This paper has developed Affective-Evaluativism, as an account of the intentional structure of unpleasant pain experiences.While it was not possible to compare Affective-Evaluativism with all rival theories, I hope to have demonstrated that the view is plausible and should be taken seriously.