In the old Indian parable, three blind men encounter an elephant for the first time. The first, seizing hold of the trunk, declares that elephants are a kind of snake. The second, upon touching the massive leg, demurs. Elephants, he avers, are a kind of tree. The third touches the ear and declares that elephants are a kind of fan. Each held part of the truth. Part of the elephant is like a snake. But they were all wrong.
The proponents of the traditional accounts of pain's intrinsic badness are like these blind men. The dislike theory claims pain is intrinsically bad because we dislike the painful sensation; on the mental state theory it is because the sensation is unpleasant. I believe each holds part of the truth, but they are all blinded by a mistaken assumption about what pain is.Footnote 1
I shall ask two intimately related questions:
Q1: What is pain insofar as it is normatively significant?
and
Q2: What about pain makes it intrinsically bad?
An answer to Q1 delimits the normatively significant intrinsic and essential properties of a pain. An answer to Q2 is a substantive axiological claim. The mental state theory's claim that pain is intrinsically bad because it's unpleasant is one such answer; as are the claims of the other traditional accounts.
I believe that all existing accounts of pain's intrinsic badness assume the same answer to Q1: pains, insofar as they are normatively significant, are just sensations which hurt. This answer is false. The correct answer to Q1 will open the way to answers to Q2 which are new in the history of philosophy.
Having shown that the putative conception of pain is false in section 1, I shall argue for two theses. First, all of the traditional accounts of pain's intrinsic badness are false; but on the correct answer to Q1, pain's intrinsic badness lies in a conjunction of all of the traditional accounts' candidates. Pain is intrinsically bad because inter alia it is both unpleasant and disliked. Second, I shall argue that some pains have two distinct and normatively significant intrinsic evils. I argue for these theses in sections 2 and 3 respectively.
Before I begin, four notes about terminology. First, I shall argue that some pains are intrinsically good. Thus I shall use ‘value’ to cover all three evaluative valences: goodness, neutrality and badness. Second, I am only concerned with pain's intrinsic value. Thus I shall often abbreviate ‘intrinsic value’ with ‘value’. Third, I shall sometimes use ‘evil’ synonymously with ‘intrinsic badness’. Fourth, I am only concerned with pain as it is normatively significant. Since very few of those who conceive of their pains as bad are aware of the underlying physiology, I shall ignore intrinsic properties of a pain such as the firing of C and Aδ nociceptive afferents, and reticular formation activity. I shall therefore abbreviate ‘normatively significant intrinsic properties’ with ‘intrinsic properties’.
1. THE KERNEL VIEW
In answer to Q1, most people believe that a pain is just a painful sensation. Its nature and intrinsic badness lie solely in the way it hurts. When I stub my toe, the pain is wholly before my mind in the way it stings and throbs. Insofar as it affects our lives, there is nothing else to pain. This is the kernel view of what pains are.
1.1. The kernel view
The kernel view holds that pains are the atoms of experience which hurt. As an experiential atom, a pain is necessarily distinct from the other elements of one's experiential milieu. The arthritic pain in my hand as I type this sentence is distinct from my experience of the cat draped drooling across my forearms, though I am simultaneously conscious of both. Thus my reaction to a painful sensation is not part of the pain; it is a reaction to the kernel.
On this view, the character of the painful sensation exhausts the properties in virtue of which a pain is intrinsically bad. Stubbed toes throb; cuts sting and burn; migraines pound and crush. Hence if pains are bad because they are unpleasant, these properties constitute a pain's unpleasantness. If pains are bad because we dislike them, the kernel composed of these properties is what we dislike.Footnote 2
1.2. Against kernels
The kernel view is false. I shall now argue that pain's intrinsic value cannot depend solely on the kernel's intrinsic properties. When embedded in the right context, some pains are intrinsically good. If the same painful sensation can have different intrinsic values in different contexts, there is no hope for the kernel view.
To be a genuine intrinsically good pain, a pain must not satisfy any of the following.
(i) The pain is good solely in virtue of some instrumental purpose it serves.Footnote 3
(ii) The pain is good solely in virtue of its being an ineliminable part of a positively valued activity.
(iii) The sufferer wrongly believes the pain to be intrinsically good. She is deluded by some sort of sickness or psychosis.
All cases of allegedly good pain can be described so that they satisfy some of (i)–(iii). For example, an ascetic may whip herself to atone for her and humanity's sins through the pain: (i). A weightlifter may endure ‘the burn’ only because of her commitment to building muscle: (ii). And, in many sad cases, past abuse and psychological trauma are manifested in self-destructive desires and practices: (iii). But the fact that we could recast hypothetical cases doesn't show how we must describe them. The brute assertion that a case is impossible does not answer an argument from possibility.
The following involves an intrinsically good pain that need not be ruled out by (i)–(iii).
Weightlifter: Kylie is a weightlifter. She enjoys going to the gym and looks forward to her workouts. One reason why she looks forward to her workouts is that she enjoys the burning sensation caused by the buildup of lactic acid and the microtearing of muscle the exercise involves. She readily admits that this sensation, the burn, hurts. However, Kylie looks forward to experiencing the burn. For her, the burn is not just an unavoidable byproduct of the exercise; nor does her liking it consist in the fact that its onset signals that she is nearing the end of a successful set. She enjoys the burn not just despite the fact that it hurts, but because of the way it feels. For Kylie, the burn is intrinsically good.
Imagine that Kylie and her workout partner Kyle are both given a drug which suppresses the burn without affecting their performance. Taking it does not allow them to lift more and they remain perfectly aware of their level of exertion. Kyle regards the burn as an unpleasant side-effect. He may occasionally say that he likes the burn but he really only means it in the extrinsic senses of (i) and (ii); or, if deluded by machismo, in a way explained by (iii). Kyle is enthusiastic about the drug. It affords him all the enjoyment without the pain. Kylie claims she enjoyed her workout less, and that she would not use the drug again.
Weightlifter need not be an isolated case. Let me sketch a few others; each can be sharpened as necessary.
Coffee Drinker: Natalie looks forward to her morning cup of coffee. She enjoys its aroma, its deep flavor, the gentle buzz it imparts, the warmth of the mug in her hands, and the way the first sip burns her lips. Like the rest of us, she attests that burning her lips hurts. Nonetheless, she enjoys that particular pain. When one morning she is given a cup of slightly cooler coffee which does not burn her lips, but which has all the same characteristics, she claims that she enjoys it less than one which burns her lips. Given the choice, she prefers to have her lips slightly scalded by the coffee.Footnote 4
Ascetic: Frances belongs to a religious order of ascetics. She acknowledges that being whipped hurts. However, she always volunteers to be scourged during ceremonies and whips herself during solitary prayer. The pain caused by whipping holds an important place in her religious asceticism. Her beliefs about the mortification of the flesh underlie its goodness for her but it is the pain per se which is good. She does not value it as a means for atoning for the sins of mankind. It is good because the pain in the context of religious ceremony is a religious experience.Footnote 5
Masochist: Melissa the masochist enjoys certain pains in certain contexts. She does not deny that they hurt or claim that they are pleasurable. She has not been abused in childhood nor does she have any ‘dark reasons’. She balks at any such suggestion, claiming instead that some pains can be good if they are suffered in the right contexts.
Spicy Food: Meena likes spicy food. When she goes to Korean restaurants she orders the spiciest dishes and requests them ‘aju mepke’ (extra hot). The heat of the food makes her mouth hurt, she readily admits, and she sips water and tea frequently. Nonetheless, it is precisely that burning sensation which she enjoys.
These cases likely strike you as variously more and less plausible. I hope some strike you as possible.Footnote 6
These pains are intrinsically good in virtue of the contexts in which they occur. Consider two new cases.
Weightlifter*: Walking to the gym, Kylie slips and suffers a minor tear of the biceps in her left arm as she grabs a railing to arrest her fall. The sensation is phenomenologically indistinguishable from the burn that a set of curls produces. Nonetheless, her cursing and complaining about the pain are evidence that it is intrinsically bad.
Ascetic*: Frances is kidnapped by the state police who whip her to extract information. The sensation is phenomenologically indistinguishable from that which she experiences during her religious ceremonies. Nonetheless, her pleas for mercy are evidence that the pain is intrinsically bad.
If Weightlifter and Ascetic are possible, these extensions should be as well. If both Weightlifter and Weightlifter* involve the same sensation, and the sensation is intrinsically good in Weightlifter and intrinsically bad in Weightlifter*, then the same pain has different intrinsic values in different contexts. That difference can only be explained by the change in context. But the properties of a context are not intrinsic properties of a pain on the kernel view. Therefore, the kernel view is committed to the intrinsic value of pain depending on non-intrinsic properties. The kernel view is false.Footnote 7
1.3. Phenomenologically indistinguishable sensations
I shall assume that the phenomenology suggests that the pain's intrinsic value really can be what is changing in these cases. Hopefully you will agree that this is possible – if only provisionally until you've seen its fruits. I do, however, want to argue briefly for another assumption: that these contrasting cases involve phenomenologically indistinguishable sensations.
It is empirically true that the character of a painful sensation can vary independently of the emotions and attitudes which accompany it.Footnote 8 Moreover, I am not alone in believing that a painful sensation can remain the same between contexts in which the pain seems to have different intrinsic values. For example, Korsgaard writes:
Pain really is less horrible if you can curb your inclination to fight it. This is why it helps, in dealing with pain, to take a tranquilizer or to lie down. Ask yourself how, if the painfulness of pain rested just in the character of the sensations, it could help to lie down? The sensations do not change. Pain wouldn't hurt if you could just relax and enjoy it.Footnote 9
Similarly, Hare imagines jumping repeatedly into cold water to generate an analogy to feeling pain without disliking it.
Suppose . . . that I do this diving act many times in the hope of getting not to mind this degree of cold; and that in the end I succeed. It is not necessary to suppose that there is any change in the degree of cold that I feel (even subjectively); there might be, but that would spoil the example. It may be merely that through habituation I stop minding my skin feeling like that. We do not even need to suppose any course of habituation. Whether I found the cold unpleasant or invigorating might depend on my general state of mind – on whether I was feeling depressed or elated.Footnote 10
This suggestion that the sensation itself (the cold kernel) can remain invariant between cases where its value differs is an analogue of my assumption.
But perhaps the assumption that the phenomenology does not change is implausible where the pains allegedly differ in valence. A more streamlined case will bolster the intuition supporting my assumption. Consider:
Weightlifter**: Kylie is carrying her friend's couch up five flights of stairs. She really dislikes the attendant burning sensation of the exertion. At the third flight it dawns on her that this is just another form of weightlifting. With that realization she comes to like the sensation.
Indeed, we can imagine that, after banging into a wall, Kylie stops thinking of the job as a form of weightlifting and the burn becomes bad again (we can iterate so that the burn flips back and forth between bad and good). If this is possible, we should accept the possibility that the sensations in my cases are phenomenologically indistinguishable. That is enough to undermine the kernel view.
2. THE COMPOSITE VIEW
Pain therefore cannot be merely a sensation. To avoid the problems of the kernel view, a pain must have some intrinsic aspect which can be affected by context while its sensory quality remains unchanged. The composite view holds that a pain is a composite of a painful sensation and a reaction. In Weightlifter, Kylie likes the burn; in Weightlifter* she hates it. Since the two burn tokens each include different reactions, they are tokens of different pain types. The fact that they involve the same sensation but have different intrinsic values is thus unproblematic.
This section sets out the composite view. I shall expand on the composite view's answer to Q1 in section 2.1 by explaining the relationship between the context, the sensation and the reaction-component of a pain. In section 2.2 I shall turn to the substantive content of the reaction-component. I shall explain how the traditional answers to Q2 fit with the composite view, and then argue that we should reject these accounts of pain's evil. I then set out an alternative account on which pain's intrinsic badness lies in a conjunction of all the traditional candidate answers to Q2 – dislike, unpleasantness, etc. – as well as some affective, desiderative, conative and cognitive states. I shall argue in section 3 that this is not pain's only intrinsic evil.
2.1. The metaphysical structure of the composite view
The composite view's answer to Q1 is straightforward. Considering two cases will help bring out the formal relationship between the painful sensation, the reaction-component, and the context.
Normal Day: My day so far isn't either especially good or bad. While putting some papers away, I accidentally slam my finger in a drawer. On an arbitrary scale of 0–100, the resulting pain is bad to degree 12 (bad12).
and
Bad Day: I'm having a bad day. I'm feeling downtrodden, anxious, irritable, and generally on edge. I accidentally slam my finger in a drawer. The sensation coupled with my preexisting affect and the explosion of negative emotions makes the pain bad20.
In the second case, the reaction-component of my pain is influenced by my dispositions to think negatively and to react explosively. Hence my pain in Bad Day is intrinsically worse than my pain in Normal Day, even though they involve the same sensation. Similarly, the fact that a cancer patient's headache throbs memento mori causes her to have a very strong negative reaction to the otherwise merely irritating sensation. More outré contextual elements can also affect the reaction-component. For example, the gender of those present can cause one's reaction to be more positive or negative than it would be otherwise. These influences are no more metaphysically mysterious than the way the presence of sour cream causes me to decline the avocado served alone but partake of the proffered guacamole.Footnote 11
2.2. The aversion theory
I have argued that, given that the kernel view is false, the intrinsic properties of a pain must contain some additional component which, by being affected by the context, can be responsible for a difference in intrinsic value between two pains with identical painful sensations. That completes my answer to Q1.
Given the composite view of pain, Q2 now becomes: What does the reaction-component contain? Different substantive theories of intrinsic value explain the nature of the reaction differently. Coupled with the composite view, the dislike theory entails that the reaction-component of pain is the dislike of the sensation. The mental state theory entails that the reaction-component is the sensation's appearing unpleasant.
I shall now argue that, while formally compatible with these substantive theories, the composite view suggests (but does not entail) an alternative account of pain's intrinsic badness which was logically unavailable on the kernel view. Section 2.2.1 argues that the reaction-component can contain a substantially more diverse array of affective, desiderative, conative and cognitive states than the traditional accounts have supposed. Sections 2.3.2 and 2.3.3 argue that we should reject the traditional answers to Q2 in favor of a more catholic account – the aversion theory – which is motivated by the diversity of the reaction-component.
2.2.1. The diverse reaction-component
I believe that the logical relationship between affective, desiderative, cognitive and conative attitudes, and a painful sensation is identical to the relationship between the putative reaction-component (e.g. dislike) and the sensation. That will suggest that these attitudes are also part of the reaction-component. Consider:
Operation: You must undergo a painful operation without anesthetic. The intense pain you feel at the first incision elicits a heavy dose of fear. You've been told that the pain will only get worse. The fear thus makes the present pain much worse than it would be otherwise.
On the composite view coupled with the dislike theory, the fear is not part of the pain. Its contribution to the pain's intrinsic badness is mediated by the reaction-component. In Operation, the sensation and context cause the fear which, in turn, causes a greater dislike of the painful sensation. Because this is a change in the pain's intrinsic properties, the pain has become intrinsically worse – it doesn't matter that the change was caused by a non-intrinsic property.
I think this is a mistake. I shall now argue that fear's contribution to the pain's intrinsic badness need not be mediated by a separate attitude such as dislike. Instead, fear stands in the same relationship to the painful sensation as the putative reaction-component. We should, I think, take the fear to be part of the reaction.
In Operation, the fear's badness need not depend on the reaction which it influences. This can even be true on the dislike theory. The victim of a serious accident could have two separate, but causally interacting, attitudes: a dislike of her fear of dying (which the pain arouses), and a dislike of the painful sensation. Morphine would alleviate one but not the other. This is compatible with the fear influencing her dislike of the sensation (and it does not entail the problematic conclusion that she dislikes the sensation in virtue of the fear's badness). Thus the fear itself can be bad in Operation.
The sensation may cause the accompanying fear. But the painful sensation does not on its own determine the level of fear. For example, if you knew that the forthcoming pains will be no worse, you may fear them much less. Thus the context's effect on the degree of the fear's evil is determined independently of the painful sensation.
Your fear in Operation need not be restricted to future pains. When I am running from the axe-wielding psychopath and hit a dead-end, I certainly fear my impending death. But I can also fear her as she slowly approaches. Similarly, it is possible to fear the present painful sensation in Operation.Footnote 12
Therefore, in Operation, your fear is a response to the painful sensation; it is itself bad; its badness is influenced by the context; and it (partially) determines the badness of the pain. That should sound familiar. Fear and the putative reaction-component can stand in the same relationship to the painful sensation. That is, both attitudes:
(1) Arise because of the painful sensation.
(2) Have the painful sensation as their object.
(3) Can be bad per se when accompanying the painful sensation.Footnote 13
(4) (Partially) determine the pain's intrinsic badness.
I think (1)–(4) are jointly sufficient conditions for an attitude being a constituent of a pain's reaction-component. There are myriad and interrelated, affective, motivational and cognitive attitudes which satisfy these conditions. Anger, despair, the impulse to escape and feelings of helplessness, among many others, are parts of the reaction-component. They are therefore intrinsic properties of the pain.
2.2.2. Rejecting the traditional accounts
As we saw above, the proponents of the traditional answers to Q2 hold that states like fear influence the intrinsic badness of pain by influencing the dislike which wholly composes the reaction-component. Thus (4) seems to beg the question against this view. By adding ‘partially’ to (4) I claim that dislike and fear affect the value in the same unmediated way. But that is precisely what is at issue.
However, I think we have shifted the burden onto this opponent. We know that when the fear of the sensation is greater, the pain is intrinsically worse. Coupled with the claim that fear is an intrinsic property of the pain, this seems to be a complete explanation of fear's contribution to pain's intrinsic badness. Thus, given the metaphysical capaciousness of the composite view, the dislike theorist owes us an account of the alleged gap between fear and the pain's value which dislike must bridge.
More importantly, to claim that we need dislike to fill this alleged gap, she owes us a substantive account of the normatively significant form of ‘dislike’. It cannot simply be a negative attitude toward the painful sensation. Fear satisfies that requirement; and we've seen that the relationship between fear and the sensation is very tight as it stands. There must be something more to the substantive conception of dislike. But this will be hard to come by. One of the perennial strengths – and most frustrating aspects – of dislike theories is that the attitude is so non-specific and thin. I shall return to this in a moment.Footnote 14
2.2.3. The aversion theory
Thus once we accept that the reaction-component of a pain can contain fear and other attitudes which satisfy (1)–(4), we must abandon the traditional accounts of pain's evil. Fortunately, the composite view's metaphysics suggest (but do not entail) an alternative answer to Q2. On the composite view, a pain is a complex of a particular kind of sensation and a reaction to it. I have argued that this reaction can include a diverse array of mental states. On this account of the reaction-component, the reaction is an aversion, that is, a complex of interrelated affective, motivational, desiderative and cognitive responses to the painful sensation. Thus accepting the composite view pushes us to an aversion theory of pain's intrinsic badness. That is, pain is intrinsically bad because of the aversion it contains.
This theory, I think, comports with the attractiveness and power of a thin conception of dislike. I suspect that part of the attraction to dislike theories lay in the relevant aspect of the experience of pain being much more complex than we can plausibly capture with, for example, my having a desire that the pain cease. Dislike (and unpleasantness) is much more passive, primal and visceral than that. More importantly, we can find all the traditional candidate sources of pain's intrinsic value in the aversion. The reaction-component contains unpleasantness, dislike, motive power and the representation of damage. If I'm right, when we accept the composite view as the answer to Q1, we are led to an irenic answer to Q2 on which a pain's intrinsic badness lies in a conjunction of all of the traditional candidates. The traditional accounts of pain's evil are therefore like the blind men and the elephant. Each held a significant portion of the truth. But elephants are not snakes, and pains are not bad just because they are disliked.
3. PAINS AS USURPERS
I have shown that once we discard the kernel view and adopt the composite view in answer to Q1, we should reject the traditional accounts of why pains are intrinsically bad. That concludes the first part of my project. I shall now argue that on the composite view, for some pains, the aversion theory is not the only correct answer to Q2: one intense pain can have two distinct intrinsic evils. The aversion theory accounts for one; I shall now sketch the other.Footnote 15
On the composite view, intense pains have a distinctive and particular character. These pains are usurpers. In addition to its intrinsic badness, being a usurper gives a pain a further intrinsic evil. Intense pain necessarily makes autonomy and other intrinsic goods impossible. We shall see that it is a mistake to ignore this evil in our normative thought about pain. I suspect that all pains have this character and this additional value. But I shall only press herein for the more moderate thesis that intense pains have it. I shall thus abbreviate ‘intense pain’ with ‘pain’ in this and the next section.
I begin in section 3.1 by arguing that the possibility of autonomy depends on a person having a certain kind of control over parts of her inner life and certain bodily activities. In section 3.2 I shall argue that pains are usurpers in virtue of their undermining this user control. Section 3.3 sketches some of the account's metaphysics and section 3.4 answers some objections. Finally, section 3.5 argues that this heretofore unrecognized intrinsic value is important for our moral reasoning about pain.
3.1. User control
Any creature capable of purposive action has and exercises user control over its thoughts, body and actions. Very roughly,
A person exercises user control over x, where x is some mental or bodily state or process of A, only if A consciously and effectively manipulates or changes x.Footnote 16
User control is a disposition. If a person can exercise user control over x at a time, I shall say that she then has user control over x. All physical and mental acts involve exercises of user control, but one can have user control over x without exercising it. User control pervades every aspect of our mental and bodily lives. Its objects include, inter alia, intentions, emotions, beliefs and bodily movements.
User control comes in degrees. I presently have complete user control over the motion of my left index finger – my willing makes its typing ‘t’ so. But purposive choice is not sufficient for the effective exercise of user control. Despite her heavy concentration and effort in fretting difficult chords, the beginning guitar player still hits sour notes.
To lead an autonomous life, an agent must be free and able to pursue certain of her projects.Footnote 17 User control over some set of the beliefs, desires, intentions, movements, etc., which constitute these pursuits is therefore a necessary condition of autonomy. Of course, while having user control over x is a necessary condition of being autonomous with respect to x, it is not a sufficient condition. Coercion and insanity undermine autonomy, but with both a person normally maintains user control over her acts.
3.2. Pain's evil
Intense pains necessarily affect a sufferer's inner life in two ways:
invasion: Pains invade one's inner life and dominate parts of it
and
passivity: Pains usurp one's user control over parts of herself which she properly has user control over. She is made passive with respect to these parts.
Together invasion and passivity constitute the usurpation of user control. Intense pains are intrinsically bad because they are essentially usurpations. I shall describe how pain is a usurper by saying a bit about each of these aspects.
3.2.1. Invasion
Being in pain is being invaded by an alien. Pains are not merely unwelcome – as are embarrassment and shame – but are experienced as entities that are not part of the sufferer. The depth and scope of this invasion increases with their severity.Footnote 18invasion has two parts:
Alien Nature: The constituent elements of a pain are experienced as alien presences. A sufferer does not identify with the elements of a pain.
and
Disassociation: A sufferer is disassociated from a pain's constituent elements. She retains the dim awareness that the elements belong to her.Footnote 19
alien nature and disassociation are probably logically distinct. For example, a schizophrenic who does not identify with the voice of god in her head need not feel that the voice is somehow hers. But their logical inseparability would not affect my view. I shall treat them as distinct.Footnote 20
The key feature of alien nature is the lack of identification with the constituent elements of a pain. The sensations and desires it imposes upon her are alien in this way, as are her contortions and groans. We normally identify with parts of ourselves to different degrees. Spasms and interloping thoughts are to some degree alien; so are the unfamiliar movements involved in learning a musical instrument or sport. The elements of a pain can similarly be more and less alien. The less a sufferer identifies with the elements, the worse the pain will tend to be. Of course, nothing of normative significance follows immediately from something satisfying alien nature. The badness of pain depends on both invasion and passivity. While the notion of identifying with parts of oneself is complicated and contentious, this rough characterization will suffice here.
‘Disassociation’ can refer to sundry phenomena, but only those which satisfy disassociation are directly part of invasion.Footnote 21 A case of disassociation satisfies disassociation only if sufferer S being disassociated from x entails that S retains a dim awareness that x belongs to her. While she does not identify with the desires a pain imposes, a sufferer is aware that they belong to her. She does not identify with her contortions but remains aware that it is her body which moves.Footnote 22
This dim awareness tempers the lack of identification that alien nature captures. Though a person does not identify with her pain, she still feels it as hers. This is not simply because the pain happens to be occurring in her. She feels the pain as hers because she is aware that it is her body and mind which have been turned against her. By contrast, a cancer patient's tumor occurs in her body but she needn't think of it as belonging to her. Scarry brings this out nicely:
Regardless of the setting in which he suffers . . . and regardless of the cause of his suffering . . . the person in great pain experiences his own body as the agent of his agony. The ceaseless, self-announcing signal of the body in pain . . . contains not only the feeling ‘my body hurts’ but the feeling ‘my body hurts me.’ This part of the pain . . . sometimes becomes visible [to an observer] when a young child or an animal in the first moments of acute distress takes maddening flight, fleeing from its own body as though it were a part of the environment that could be left behind. If self-hatred, self-alienation, and self-betrayal . . . were translated out of the psychological realm where it has content and is accessible to language into the unspeakable and contentless realm of physical sensation it would be intense pain.Footnote 23
Being in pain involves feeling that part of oneself (here: her body) has been turned against her. That involves thinking of her body as something independent of herself, but at the same time remaining aware that it is hers. I claim that this is also true of the beliefs, desires, emotions and other components of a pain. disassociation attempts to capture this quasi-schizophrenic aspect of pain.
3.2.2. Passivity
Being passive in the sense that satisfies passivity is not mere inertness. In this sense, passivity involves the loss of user control, and feeling helpless and controlled. Consider:
Your Trial by Ordeal: Your hand is placed in a pot of water which is slowly brought to a boil. If you remove your hand before it reaches a boil, your child will be killed. If you succeed, she will be safe.
Your pain involves the urge to withdraw your hand. The urge takes several forms: You want to remove your hand, and it feels as though your hand is being involuntarily pulled toward the water's surface. Your pain also involves cognitive elements: you find yourself rationalizing removing your hand, trying to make excuses for giving in and allowing your child to die.
At the same time, you try to keep user control over your hand and your thoughts. You strain to keep your hand in the water. You remind yourself of the stakes and your love for your child; and you actively resist the lure of the rationalizations. Insofar as you are successful, you retain user control over your hand, desires and thoughts. Insofar as you fail, you lose user control over these aspects; you become passive with respect to them.
When you fail, you feel helpless. You may, for example, feel like a spectator watching in horror as your hand pulls from the water. These feelings of helplessness are part of passivity. In Scarry's nice turn of phrase: ‘In physical pain . . . suicide and murder converge, for one feels acted upon, annihilated, by inside and outside alike.’Footnote 24 This connects with the dim awareness that disassociation involves. On disassociation, one is aware that the usurped aspect is properly her own. passivity adds a sense of helplessness that is tied to the awareness that the aspect is something she ought to control.
3.3. Overview of metaphysics
I've now claimed that intense pains are usurpers of user control. As such they necessarily undermine the possibility of autonomy. If the undermining of the intrinsically good is intrinsically bad, then the usurpation of user control is intrinsically bad. The metaphysics of this account are complex. I shall give only a quick outline here.
Certain forms of user control are necessary conditions of autonomy. Hence the usurpation of user control is itself intrinsically bad.Footnote 25 However, the possession of user control is not itself intrinsically good. Rather, a usurpation is bad because undermining certain forms of user control is undermining that which is valuable. The usurpation of these kinds of user control does not cause autonomy to be undermined. It is the undermining of autonomy. Therefore, the usurpation of user control is intrinsically bad in virtue of its metaphysical relationship to that which is good in a person's life.Footnote 26
If this is correct, intense pain qua usurper is intrinsically bad in virtue of its relationship to intrinsic goods like autonomy. By setting aside some issues about the categories of value, we can say that: intense pains have two distinct intrinsic evils.
3.4. Two objections
Before moving on to the significance of intense pain having two evils, let me quickly sketch answers to two natural objections.
3.4.1. Overinclusion
I've argued that the usurpation of user control is a necessary condition of pain's evil. Since this is only a necessary condition, there is room for other phenomena to involve the usurpation of user control and be similarly intrinsically bad. This may seem problematic. There are many cases in which we lose user control, but in which we find no evil. Indeed, with the experience of immersion in music, wallowing in the sun, orgasm and other pleasures, the loss of control seems to be a large part of what makes the experience good.
Of course, the concern does not arise for just any loss of user control. Usurpations involve a specific kind of experience – one which satisfies invasion and passivity. But we identify with the pleasure of wallowing in the sun, and do not find the sensation invasive in the sense described by invasion.
More importantly, the badness of losing a particular kind of user control depends on what intrinsic value it is a necessary condition of. This depends on substantive accounts of the intrinsic goods. An account of pleasure's intrinsic value will entail that certain forms of user control are its necessary conditions. Thus if it turns out that losing user control over some aspect x is part of what makes pleasure good, then the loss of user control over x cannot be bad.
But there are some cases in which a pleasure or other innocuous sensation does satisfy passivity and invasion, and therefore involves the usurpation of user control. Here are two examples.
Sentry: Sarah and her Army squad are deep in enemy territory. They are all exhausted. Sarah remains awake as sentry while the others sleep. If she falls asleep they will all die. Yet as each wave of fatigue washes over her, she finds herself, to her disgust, wanting nothing more than to fall asleep.
Ascetic**: Frances's tormentors inject her with a combination of heroin and MDMA which causes overwhelming sensations of pleasure and ecstasy. Despite the fact that her deepest convictions demand that she eschew pleasure, she finds herself wanting more.
Both cases involve the experience of the usurpation of user control. The sensations are invasive and make each passive with respect to the changes in her beliefs and desires. Since the usurpation of user control is intrinsically bad, it follows that these feelings of sleepiness and pleasure are intrinsically bad. I believe this is the correct result.
The sensation of pleasure is invasive and usurping for Frances precisely because of the deeply held beliefs that undergird her autonomous life. She has shaped herself around a life eschewing pleasure. Thus while she, at the moment, wants the pleasures that force themselves on her in Ascetic**, her life is built around wanting not to want them. This is in stark contrast with the rest of us. Many of us would welcome the temporary ecstasy just as we normally welcome the sensation of sinking into the pillow.Footnote 27
More importantly, Frances's pleasure forcibly draws her to it – it forces her desires away from her deep commitments. By twisting what she wants, the pleasure undermines her autonomy. It not only forces her to act contrary to what she most wants; it also warps what she wants away from what she is deeply committed to. Mutatis mutandis in Sentry.
We should therefore agree that Sentry and Ascetic** involve normatively significant usurpations of user control, and thus involve ordinarily innocuous sensations that are intrinsically bad in the same way as pains qua usurpers. This is not a problem for my view.
3.4.2. The role of phenomenology in usurpation
I have described the usurpation of user control in roughly phenomenological terms. This meets an important objection. Upon hearing my view, many remark that it's impossible for a pain's intrinsic value to lie in its relation to autonomy. Any intrinsic value of pain, they believe, must lie in the way it feels. I agree. But once we've rejected the kernel view, we can no longer assume that ‘how pain feels’ refers solely to its sensory-component. On the composite view, ‘how pain feels’ can be extremely rich and complex. If I'm right, fear can be an intrinsic property of pain, and it is part of how some pains feel. Far from removing pain's phenomenology from the center of its intrinsic badness, the complex phenomenology of usurpation gives a much richer and more powerful account of what being in pain feels like than was possible with the kernel view.
3.5. Significance of the two evils
The fact that intense pains have two distinct evils is not a mere curiosity. I believe that in some cases each value commends different alternatives. I shall argue that it can be a mistake to ignore the usurpation's intrinsic badness.
Consider an acute pain:
Fiona's Fall: Crossing 6th Avenue on a cold New York night, Fiona slips on some ice and lands hard on her left arm breaking the ulna. She lies in the street, clutching her forearm. Traffic is approaching. Passersby urge her to get up, but she moans that she can't. She sees the cars and wants to rise, but finds she can't move.
We should take her at her word when she claims that she cannot get up. While it is physiologically possible for her to do so – nothing is wrong with her legs and she has one good arm – the pain has paralyzed her. Every time she moves, the pain pushes her back down. In trying to get up and believing that she should, she fights the pain. In being unable to rise, she loses. She feels taken over by it. The pain is an oppressor which she is helpless to resist. The helplessness and paralysis that she feels are part of the usurpation of user control. Lying in the street, Fiona is subject to the two distinct intrinsic bads of the aversion and the usurpation.
Fiona's pain is such that the aversion is bad50 and the usurpation is bad50 (written badaversion/usurpation, it is bada=50/u=50). Now suppose that we offer Fiona a choice:
Drug A: Diminishes the aversion's badness to bad45; but leaves the usurpation unchanged. Thus the pain with Drug A is bada=45/u=50.
Drug B: Diminishes the usurpation's badness to bad20; but leaves the aversion unchanged. Thus the pain with Drug B is bada=50/u=20.
Surely it is rationally permissible for Fiona to choose Drug B's much greater reduction in the pain's overall badness. Whatever the relationship between the two values, it's implausible that the aversion is lexically prior to the usurpation. Indeed, we don't think this about instrumental value. I'd be irrational to accept the ruin of my career to salve a paper cut. I think it is also likely permissible for her to choose B when the effects are equivalent for both drugs; where A yields bada=20/u=50. However, the absence of lexical priority is enough to show that we should not ignore the intrinsic badness of usurpation.
But why should we believe that it is possible for the aversion and usurpation's values to vary independently of each other? Drug B operates by diminishing her feeling of helplessness. But, like fear, the feeling of helplessness is part of the aversion. Thus the diminution of the helplessness seems to be ipso facto a diminution of the aversion. The decrease in the usurpation's badness seems to entail a commensurate decrease in the aversion's intrinsic badness.
We can avoid this problem by stipulating that Drug B's diminution of the helplessness has the side-effect of increasing the contribution of another component, such as the pain's meaning. Imagine that Drug B must be administered with an instrument shaped like a baseball bat. That makes the fact that she will not be playing second base this season weigh more strongly in the aversion. This increased contribution of the meaning is exactly equal to the helplessness's decreased contribution. Thus the feeling of helplessness present changes, but the degree of aversion does not. Therefore, the usurpation decreases without a commensurate decrease in the aversion.
If the two values can vary independently, then it seems that Fiona can rationally choose Drug B. Thus, considering only the intrinsic properties of her pain, what she ought to do is determined by the usurpation's badness, and not just by the badness of the aversion. Therefore, we should not ignore the usurpation's intrinsic badness. Pain has two evils.Footnote 28
