A Causal Theory of Enjoyment

Citation preview

A Causal Theory of Enjoyment Wayne A. Davis Mind, New Series, Vol. 91, No. 362. (Apr., 1982), pp. 240-256. Stable URL: http://links.jstor.org/sici?sici=0026-4423%28198204%292%3A91%3A362%3C240%3AACTOE%3E2.0.CO%3B2-3 Mind is currently published by Oxford University Press.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/journals/oup.html. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is an independent not-for-profit organization dedicated to and preserving a digital archive of scholarly journals. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Sun Jun 17 05:00:56 2007

Mind (1982) Vol. X

~ I 240-256 ,

A Causal Theory of Enjoyment WAYNE A. D A V I S

I shall present a theory of enjoyment, in the form of a definition. T h e basic idea is that an object of enjoyment causes the subject to experience pleasure by causing occurrent beliefs which satisfy desires concerning the experience itself. Pleasure is identified with occurrent happiness, which can be defined in terms of belief, desire, and thought. I also define degree of enjoyment, as that part of the subject's pleasure attributable to the object of enjoyment. Degree of belief and desire are the fundamental quantities. Finally, I briefly treat negative enjoyhent.'

There is a distinction between dispositional and occurrent enjoyment, between 'Jack enjoys the Pathetique Sonata' and 'Jack is enjoying the Pathetique Sonata'. 'Jack enjoyed the Antarctic Symphony' is ambiguous. T h e two are related as follows: A enjoys something dispositionally provided he normally enjoys it occurrently. Our goal then is to define occurrent enjoyment. Let X stand for anything A can be said to enjoy other than himself. We can begin by noting that ( I ) A is enjoying X only if A is experiencing pleasure.

By itself, this tautology should inspire no disagreement. Controversy is possible only concerning what it is to experience pleasure. I believe pleasure here can be identified with occurrent happiness.2 A person experiences pleasure iff he feels happy, or experiences happiness. T h e happier a person is at the moment, the more pleasure he experiences. Pleasure cannot be identified with dispositional happiness, in which sense 'A is happy' means A is predominantly happy in the occurrent sense. A man can enjoy something even though he is normally unhappy. And 'pleasure' I

2

I would like to thank Robert Audi for his extremely constructive and useful criticism of an earlier draft. I discuss this identification at length in 'Pleasure and Happiness'.

-

24 1

A CAUSAL THEORY O F ENJOYMENT

here does not mean pleasure-sensations, which are caused by a variety of stimuli such as massaging a tense muscle, scratching an itch, stepping into a hot shower on a cold day, and stimulating the erogenous zones. We enjoy many things without any such sensations. Pleasure in the sense we are concerned with is not a bodily sensation or complex of sensations. It is not located in any part of the body, or all over the body. I t does not even depend on somatic sensations in the way feeling sick and feeling fatigued do. Note the parallel distinction between 'physical' pain (an all too familiar type of bodily sensation) and 'psychological' pain (suffering, misery, grief, heartache: occurrent unhappiness). Given my identification of pleasure and happiness, ( I ) entails that John must feel happy if he is enjoying the party. If John is enjoying the party, then he is obviously enjoying himself; and it seems evident that if someone is enjoying himself he is happy. One way to spoil someone's enjoyment is to make him unhappy. John's enjoyment of the party may cease when he is told that his dog has been run over. Note that John may not have become happy until half-way through the party, and may remain happy long after the party is over. Experiencing pleasure is obviously not a sufficient condition of enjoying X. Suppose John is enjoying the party despite the smoke biting his eyes. T h e n John is experiencing the smoke while experiencing pleasure, yet he is not enjoying the smoke. T h e reason is that John is experiencing pleasure not even in part because of the smoke. Many factors may explain why John is happy (beautiful women, good Scotch, music), but the smoke is not one of them. So ( I ) should be strengthened to

'

(2)

A is enjoying X only if A is experiencing pleasure at least in part because of X.

John's happiness or pleasure must be one of the consequences of whatever he enjoys.' What we enjoy gives us pleasure and makes us happy. 'Pleasure', heretofore used as an abstract singular term, also occurs as a general term, as in 'His pleasures are few', 'It is a pleasure to meet you', and 'the pleasure of dancing'. A pleasure (general term) is a source of pleasure (singular term). Ryle's claim I

2

Cf. Alston (1967, p. 342). Contrast Ryle: 'To say that a person has been enjoying digging is not to say that he has been both digging and doing or experiencing something else as a concomitant or effect of the digging' (1949, p. 108).

242

WAYNE A. DAVIS:

(1949, p. 108) that 'His digging was his pleasure, and not a vehicle of his pleasure' is therefore a false dichotomy. Kenny's complaint that 'on such a view . . . [it] would be quite a contingent matter that the pleasure of drinking did not occur while eating . . .' (1963, p. I 33) is similarly misguided. We can agree that 'a pleasure' never means 'a pleasure-sensation'. In contrast, 'a pain' almost always means 'a pain-sensation', seldom 'a source of pain'. 'Happiness' is never a general term. (2) makes possible a certain type of introspective error. '. . . a man might say that he enjoyed the scenery on his honeymoon, when it was really the company of his bride which he had enjoyed; and he might realize this through revisiting the locality alone on a later date' (Kenny, 1963, p. 129). A man can be mistaken as easily about what he enjoys as about what he sees, though perhaps he cannot be mistaken about whether he is happy. A person is often happy for several reasons, as when enjoying many different things at once, such as the music, the girl, and the drink. For this reason, (2) is qualified by at least in part. Given the possibility of overdetermination, we cannot say universally that if A enjoyed X , then A would not have experienced pleasure if X had not occurred. If that song had not been played, John would still have experienced pleasure, because of the girl.1 While A's enjoyment of X depends logically on X , A's pleasure or happiness (due in fact to X ) does not. John could not have enjoyed the song if there was no song, but he could still have experienced pleasure.2 We still do not have a sufficient condition of enjoying X. At the Horowitz concert, I experienced pleasure because of the practising Horowitz did in the decades before the concert, but I did not enjoy his practising. Penelhum (1957, p. 490) and Perry (1967, p. 215) concluded that what is enjoyed must be something occurring at the time of enjoyment. Similar problems still arise, though. During the concert I experienced pleasure because of the actions of m y inner ear; but I did not enjoy them. One problem here is that I was not aware of those actions. (3)

A is enjoying X at t only if A is (nonepistemically) aware of X at t.

T h e temporal references in (3) explain why I did not enjoy, during the concert, the practising Horowitz did prior to the concert: during I

2

Cf. Loeb (1974).

Contrast Quinn (1968, pp. 582 ff.).

-A C A U S A L T H E O R Y O F E N J O Y M E N T

243

the concert, I was obviously not perceiving anything done before the concert. Penelhum and Perry" conclusion is actually incorrect, I might enjoy the explosion of a star even though it took place millions of years ago. I must, however, be aware of the explosion when I enjoy it. It is well known that there is a time gap between awareness and object of awareness. I t is the act of awareness that must be simultaneous with the act of enjoyment. Awareness must include introspection as well as sense-perception. A man can enjoy working out maths problems in his head, and watching a football game. H e can be aware of these mental activities, but they are not objects of sense-perception. We need to be more specific about the kind of awareness required. For during the concert I was well aware of the fact-I knew and could perceive-that Horowitz had practised. We must distinguish nonepistemic from epistemic awareness. This is the distinction between being aware of Horowitz playing the piano and being aware (of the fact) that Horowitz is playing the piano. T h e latter entails knowing and believing that Horowitz is playing the piano. T h e former does not entail knowing or believing anything in particular. Facts or true propositions are the objects of epistemic awareness. Concrete objects, events, or states of affairs are the objects of nonepistemic awareness. If John enjoyed Horowitz playing the piano, he must have been aware of Horowitz playing the piano. It would not be sufficient for John to have been aware that Horowitz was playing, which might have been the case if John were in a sound-proof windowless monitoring booth surrounded by dials and meters. Nor would it be necessary. For John might not know the difference between a piano and a harpsichord, and might not even know who Horowitz is, in which case he could not be aware that Horowitz was playing the piano; he nevertheless could still be aware of, and enjoy, Horowitz playing the piano. There are two different ways of getting pleasure from something. Alan may get pleasure from seducing sixteen-year-old girls, or he may get pleasure from the fact that he seduces sixteen-year-old girls.2 The first way of getting pleasure is the same as enjoyment, and involves nonepistemic awareness. T h e second way differs from enjoyment, and involves epistemic awareness. Alan could not get pleasure from the fact that he seduces sixteen-year-old girls without I

z

This distinction is carefully drawn for visual awareness by Dretske (1969, ch. 11). A similar distinction was noticed by Gordon (1969, pp. 410 ff.).

244

WAYNE A. DAVIS:

knowing that he does so, but he could still get pleasure from seducing them (he might not know the age of the girls he seduces, or he might think they seduce him). A father could get pleasure from the fact that his daughter plays the piano beautifully without ever having heard her play, but he could not then get pleasure from her beautiful playing. Epistemic and nonepistemic pleasure-getting are independent. Alan may get pleasure from seducing young girls even though he does not get pleasure from the fact that he does; indeed, it may distress him severely that he seduces young girls, which may send him to a psychiatrist. On the other hand, Alan may get pleasure from the fact that he seduces young girls (he takes it as a sign of youthfulness and sex appeal), even though he does not get pleasure from the act of seducing young girls (due to some physical disorder); indeed, it may hurt him to seduce young girls, and that may send him to a doctor. I said that a pleasure is a source of pleasure. We must get pleasure from it nonepistemically. Playing the piano is one of my pleasures, not the fact that I play the piano.' The requirement of nonepistemic awareness explains Taylor's observation (1963, p. 8) that we cannot be said to enjoy knowing something. While we can be aware that we know things, knowing is not something we are aware of nonepistemically. We similarly cannot enjoy the radio waves around us (even if we are aware, from the music we are enjoying on the radio, that they exist). And Alan cannot enjoy the fact that he seduces young girls (even though he does enjoy the act of seducing them). The requirement also explains Taylor's observation (1963, p. 6 ) that 'enjoy' cannot take as direct object a past participle: Kathy cannot enjoy having seasoned the sauce properly (though she may have enjoyed doing it). While Kathy can certainly perceive that she has seasoned it properly, she cannot perceive herself having done something. I t follows, finally, that we cannot now enjoy future events. Both (2) and (3) entail that we can enjoy only existent things. X must exist (occur, obtain, etc.). John enjoyed the party only if there was a party. Enjoyment is not intentional. This entails that 'where a person enjoys an object or activity in the false belief about its identity or nature, the person can and must be said to enjoy the actual object or activity rather than the object or activity he takes it to be. A person who enjoys an object which is a crude drawing in the delusion that it is a painting by Rubens, clearly enjoys just the crude I

I define epistemic pleasure-getting in 'Pleasure and Happiness'.

A" C A U S A L T H E O R Y O F E N J O Y M E N T

245

drawing at which he gazes with admiration' (Perry, 1967, p. I 14). Perry also points out that when we might be inclined to say that a person is enjoying something that he is daydreaming or hallucinating, we should say that what he enjoys is daydreaming or hallucinating (1967, pp. I 13-1 IS). Enjoyment is also extensional: if A enjoys X and X = Y, then A enjoys Y. If Mary enjoyed the Pathetique Sonata, then she enjoyed Beethoven's eighth piano sonata, since the Pathetique is his eighth. I t makes no difference whether or not Mary knows that what she is enjoying is Beethoven's sonata No. 8. Since enjoyment is extensional, the awareness involved must be nonepistemic, for epistemic perception is intensional.' Ryle (1954) held that attention is a necessary condition of enjoyment.' Attention is stronger than awareness. If I am attending to something, then I am nonepistemically aware of it. But I can be aware of something without attending to it. I can hear one conversation while attending to another. I n many cases, attention is a necessary condition of enjoyment. I t is difficult to enjoy a conversation without attending to it. But it is not a logically necessary condition. I can enjoy a bath while m y attention is riveted to the book I am reading.3 And when I enjoy watching the girls go by, m y attention is focused on the girls, not on m y watching. Epistemic awareness is knowledge. Nonepistemic awareness is experience. T o be aware of the fact that Horowitz is playing the piano is to know that he is playing. T o be aware of Horowitz playing is to experience his playing. Having the experience of being kissed entails more than just being kissed. Someone asleep, unconscious, or dead could be kissed, but could not have the experience of being kissed. Missing is the awareness of being kissed. We cannot, however, watch a game without having the experience of watching it. This is because we cannot unconsciously watch something. It is tempting to say that what we enjoy is always an experience, that experiences are the proper objects of enjoyment. But matters are not so neat. We can enjoy a football game, the music, someone's company, being kissed, playing the piano, and even ourselves. These are things we experience, but are not themselves experiences. I

2

3

Contrast Kenny (1963, p. 130) and Thalberg (1962, p. 71). Dretske (1969, ch. 11) has a detailed discussion of the extensionality of nonepistemic perception. See also Gallie (1954) and Penelhum (1957, p. 497). Cf. Manser (1960-1, pp. 234 ff.), Taylor (1963, p. 4), Perry (1967, p. 95 fn 29 and p. 198), and Gosling (1969, pp. 58 ff.).

246

WAYNE A. DAVIS:

In contrast, watching the game is both an experience and something we experience. However, to enjoy the game is to enjoy experiencing thegame. Of course, there are many different ways of experiencing a game: participating in it, attending it as a spectator, watching it on T V , listening to it on the radio, and so on. Similarly, to enjoy being kissed is to enjoy the experience of being kissed. And to enjoy ourselves is to enjoy w h a t we are experiencing, whether that be a game or the experience of being kissed. In general, let N P stand for any nounphrase other than a reflexive pronoun, and let V P i n g be the gerund formed from any verb-phrase VP, then:

R I A enjoys N P iff A enjoys experiencing N P .

R2 A enjoys V P i n g iff A enjoys the experience of V P i n g .

R3 A enjoys himself iff A enjoys w h a t he is experiencing.

I refer to these tautologies as redundancy rules. They enable us to introduce or eliminate redundancy as we wish. Note that 'experiencing the game' is itself a gerund; Rz can therefore be applied, giving us 'A enjoys the experience of experiencing the game'. 'The experience of experiencing the game' is a noun phrase, so R I yields 'A enjoys experiencing the experience of experiencing the game'. Weeding out all the redundancy leaves 'A enjoys the game'.

Given the redundancy rules, we can confine our attention to defining 'A is enjoying E', where E stands for an experience. T h u s far, we have a necessary condition: A is experiencingpleasure a t least in part because A is experiencing E. This condition is not sufficient, however. Bob's favourite piece is on the radio. Unfortunately, the power button won't stay in. T o keep the radio on, he has to hold his thumb on the button. Bob is enjoying the music, so he is experiencing pleasure. Bob is aware (tactually) of the power button, otherwise he would not be able to keep it pushed in. So Bob is experiencing pleasure in part because he is aware of the button, i.e., because he is having the experience of holding the button in. Bob is not enjoying the button, however. Holding the button in is causing Bob to experience pleasure, but in the wrong way. We have here the familiar problem of w a y w a r d causation. Suppose I am enjoying a dancer. How does my awareness of her cause me to experience pleasure? I t is well known that perception

A CAUSAL T H E O R Y O F E N J O Y M E N T

247

produces countless occurrent beliefs. I t is also known that satisfied desires contribute to happiness or pleasure. By 'satisfied' here I mean subjectively, not objectively, satisfied. M y desire to win the lottery would be subjectively satisfied by my believing that I will win the lottery. It would be objectively satisfied by m y actually winning the lottery. Only subjectively satisfied desires have a direct effect on pleasure or happiness.' M y suggestion is that being aware of the dancer causes m e to have a number of occurrent beliefs about her; these beliefs subjectively satisfy a number of desires; and these satisfied desires cause m e to experience pleasure or happiness. Because I see the dancer, I have numerous occurrent beliefs about her, such as that she has arms, is wearing a white gown, is blonde, is beautiful, is moving gracefully, and is being watched by me. These are not all matters of indifference to me: I want her to be beautiful and to move gracefully, and I want to watch her. I consequently have a satisfied desire that she be beautiful, that she move gracefully, and so on. These satisfied desires contribute to my pleasure. I am happy in part because I am happy that she is beautiful, moving gracefully, and being watched by me. T o generalize, A is enjoying X only if: A has a number of occurrent beliefs about X because A is aware of X , and A is experiencing pleasure (happiness) in part because A wants X to be the way he believes X to be. Problem. M y work is enjoyable at the moment. I am aware of the lamp over m y desk, so I believe that it is on. I want it to be on. Consequently, I am happy i n p a r t because I am happy that the lamp is on. But I am not enjoying the lamp. I believe the solution here is to require that A be happy in large or signi5cant part because A wants X to be the way he believes it to be. Another problem. Variations on the case of wayward causation suggest that there can be no intervening links between A's subjectively satisfied desires and his happiness. Suppose George has a belief and desire detector. H e secretly decides to cut the power to Bob's radio unless Bob believes and desires that the button is in. T h e n Bob is experiencing pleasure in large part because he believes and desires that the button is in. But Bob is still not enjoying the button. While Bob's subjectively satisfied desire is a significant part of the cause of his happiness, it is not a significant part of his happiness. All of this requires considerable elucidation. I

For a fuller discussion of this point, see 'A Theory of Happiness'

248

WAYNE A. DAVIS:

Earlier, I identified pleasure with occurrent happiness. I believe occurrent happiness can be defined in terms of belief, desire, and thought.' Take every proposition or thought a person is thinking at the moment, multiply the degree to which it is believed by the degree to which it is desired, add up all the products, and the sum is his degree of happiness. Let p,, p,, . . . ,pn be an enumeration of all n

A's thoughts; then A's degree of happiness h =

b(pi)d(pi).On i-1

the belief scale I use, positive numbers represent belief, negative numbers represent disbelief, o represents the absence of both belief and disbelief, I represents complete certainty, and - I represents complete disbelief. On the desire scale, positive numbers represent desire, negative n u m b e ~ srepresent aversion, and o represents the absence of both desire and aversion. Given such scales, a person is happy if his degree of happiness is greater than zero, unhappy if less than zero. 'Happy' without a complement expresses a nonrelational state. 'Happy that' expresses a propositional attitude. Someone is happy that p only if he believes and desires that p. Someone is unhappy that p only if he believes that p and is averse to it being the case that p. How happy someone is that p is determined by how much he desires that p. Assuming the thought that p is occurring to him, it follows that a person is happier if he is happy that p than if he is unhappy that p; it also follows that the happier a person is that p, the happier he is. Let us say that a thought p adds to happiness when the product b(p)d(p) is greater than zero. A thought subtracts from happiness when the product is negative. An occurrent satisfied desire necessarily adds to happiness. A collection of thoughts p i , . . . ,p, adds to happiness provided the sum of products m

b(pj)d(pj)is greater than zero. T h e collection adds signijicantly j =1

to happiness provided the sum is significantly greater than zero. Precise significance levels cannot be specified. Finally, A occurrently believes (desires) that p iff the thought that p is occurring to A and A believes (desires) that p. Let the class of beliefs concerning experience E include the following: ( I ) Beliefs about E. T h e belief that the experience of kissing Mary is exciting is a belief about the experience. T h e belief that you are having the experience of watching the game is a belief I

See 'A Theory of Happiness'

A CAUSAL THEORY O F ENJOYhIENT

249

about that experience. (2) Beliefs about t h e content of E. T h e content of the experience of kissing Mary is the act of kissing Mary; so the belief that the act of kissing Mary is daring is a belief about the content of the experience. T h e content of the experience of watching the game is the game; so the belief that Dallas is winning the game is a belief about the content of the experience. Some experiences, such as aches and pains, have no content. ( 3 ) Beliefs about parts of E. T h e experience of feeling Mary is part of the experience of kissing her. T h e experience of watching the quarterback is part of the experience of watching the game. Seeing a part of the painting is part of the experience of seeing the painting. Recalling your first date may be part of the experience of reminiscing about high school. (4) Beliefs about t h e contents of parts of E. T h e belief that the quarterback is throwing a pass is a belief about the content of a part of the experience of watching the game. My definition of enjoyment can now be presented.

DI. A is enjoying E iff E is causing A t o h a v e a number of occurrent beliefs concerning E, which collectively a d d significantly t o t h e pleasure (happiness) A is experiencing. T h e idea is this. An experience will normally cause someone to have a stream of occurrent beliefs. Consider only those that 'concern' E. Watching television may remind you that you have a dental appointment tomorrow; but that belief does not concern the experience of watching television, so ignore it. Let p,, . . . ,p, enumerate all the occurrent beliefs E causes concerning itself, and m

let h(E)

=

1 b(pj)d(pj). Then h(E) must be significantly greater j=1

than zero. This entails that at least some of p l , . . . , p, are objects of desire as well as of belief. Some may be objects of aversion, but only if the objects of desire more than compensate for them. D I should be understood as implying that A is experiencing happiness. If p,, . . . ,pn (n >= m) enumerates all thoughts occurring to A, includn

ing those not concerning E, then h =

b(pi)d(pi) must also be i = l

positive. This implies that any thoughts not concerning E must not be so negative as to subtract more than the addition to happiness made by those that do concern E. As I said above, John's enjoyment of a party may be spoiled when he learns that his dog died. Someone may not enjoy being kissed because of excruciating abdominal pain.

250

W A Y N E A. D A V I S :

Let me apply D I to our original case of wayward causation. T h e experience of listening to the music is not part of the experience of holding the button in (or vice versa). T h e occurrent beliefs in virtue of which Bob is experiencing pleasure concern the music, not the button. Bob does have some occurrent beliefs about the button, such as that it is in and that it has to be held in. T h e first belief adds to Bob's happiness, since he wants the button to be in (so that he can listen to the music). But Bob is averse to the fact that he has to hold the button in. In general, the button is largely a matter of indifference to Bob and he does not give much thought to it. A similar case should be contrasted. Like many people, I often look at snapshots in order to have the pleasure of reminiscing about the past. Reminiscing is not part of the looking. Yet here I enjoy not only reminiscing, but also looking at the snapshots. Why? I have many beliefs about the picture as I look at it: I am looking at the picture; the picture is o f . . .; it is a good picture; the picture reminds me o f . . .; and so on. These are all things I desire. So there are plenty of occurrent beliefs about the picture adding to my happiness. Suppose John is playing the slot machines, and is winning between I 5 and IOO dollars every few times he puts a dollar in. It is clear what satisfied desires account for John's elation. I t is fairly obvious why pleasure-sensations contribute to enjoyment: we are happy that they occur, and hence both believe that they occur and want them to occur. Other cases are far from clear. What satisfied desires account for Mary's enjoyment of a piece of music? None, it is commonly held.' We can discover many, however, by answering the question W h a t did she like about it? Mary may have liked the melody, the sound of the strings at a certain point, a particular modulation, the driving rhythm, a certain chord, the way a particular note was played, the tinkle of the triangle there, and so on. These are all features of the music Mary was happy with. They seemed good, right, nice, or pretty. Now to be happy with something is to be happy that i t is the way i t is. And if Mary is happy that p, then she desires and believes that p. Consequently, corresponding to everything M a r y liked about the piece of music, there is a satisfied desire. There is no need to require that Mary be able to tell us what she liked about the music, or be able to describe 'the way it is'. For one I

See Sharp (1928, p. ~ o z )Beebe-Center , (1932, p. ~ o I )Bedford , (1959, p. 86), Thalberg (1962, p. 73), Taylor (1963, p. ~ g )and , Perry (1967, pp. 59 ff.).

. A C A U S A L T H E O R Y OF E N J O Y M E N T

251

thing, given the complexity of a piece of music, and its fast changing nature, she will have forgotten most of the things she liked by the time the piece is over. For another, Mary's command of the language may be limited. Children, and most adults for that matter, simply lack the words to describe music adequately. Furthermore, none of the relevant desires had to exist prior to acquiring the occurrent beliefs about the music.' Mary may have had no idea the triangle was going to tinkle. But afterward she may have been glad that it did. Perception may be followed immediately by evaluation. According to Perry, we enjoy something only if we like some of its 'intrinsic' features. 'We could not understand a worker who said, "I enjoy my job, but the only thing I like about it is the Christmas bonus" ' (1967, p. 209). On my analysis, such a worker would have to be thinking of his Christmas bonus most of the time he does his job, a near impossibility. Only intrinsic features of the job will be continually before the worker's mind. Besides, while the thought that he will get the bonus should add significantly to his happiness, the thought that he has to work in order to get it could easily subtract a comparable amount. Ryle (1949, p. 108) held that desiring something is a necessary condition of enjoying it.' His example was digging, for which the thesis is quite plausible. It is hard to imagine a person enjoying digging who did not want to dig. Then again, it is hard to imagine someone simply digging without wanting to dig (though easy to imagine him not 'really wanting' to dig). Ryle's thesis has several problems. First, if I enjoy Horowitz, do I have to want him? At most, I want to experience him. T h e redundancy rules would be helpful here. Second, a person can enjoy something he does not recognize as such and therefore could not d e ~ i r eDick, . ~ completely ignorant musically, could enjoy Horowitz playing the Waldstein Sonata without knowing who Horowitz is, what piece is being played, or even what instrument it is being played on, in which case Dick could not want Horowitz to play the Waldstein Sonata. Dick could at most want him to do that. T o avoid these objections, Ryle's thesis should be reformulated as follows: A enjoys E only if A I

2

3

Contrast Sharp (1928, p. 402), Ta~lor(1963,p. 19), and Perry (1967, pp. 59 ff.). SOhave Gallie (1954, p. I ~ I )Kenny , (1963, p. I Z ~ Armstrong(1968, ), p. 176), and Warner (1980, pp. 508 ff.). Stout (1899, p. 232), Woodworth (1921, pp. 177 ff.), Edwards (1975, p. 280), Tatarkiewicz (1976, p. 37), and Brandt (1979, pp. 38-42) hold that a person who enjoys X must want X to continue. Cf. Alston (1967, p. 345).

252

W A Y N E A. D A V I S :

wanted to have that experience. A must have noticed the experience, but he need not have identified it as E. A third problem seems intractible. A person can enjoy something despite an aversion to it. ' Imagine a woman strongly attracted sexually to her fiancCe but morally opposed to premarital sex. Imagine further that in the heat of an embrace, he overpowers her; and she-unable to help responding-enjoys being raped, much to her subsequent shame. It is implausible and unnecessary to assume that at the moment of ecstacy her conscience was over-ruled. My definition entails that had she wanted him to do it, she would have enjoyed it more; and further, that had her aversion been stronger, it would have prevented her from enjoying it. D I also implies that she must have wanted something concerning the experience. But it does not follow that she wanted to have the experience itself as a whole. Ryle's companion thesis that desiring something is a sufficient condition of enjoying it is also incorrect. Alan has always had a burning desire to kiss a Playboy bunny. Suppose one day he kissed Monique, who in fact is a Playboy bunny. It does not follow that Alan enjoyed the experience. He may not have knozvn that Monique was a bunny; and even if he did, he may have temporarily forgotten the fact (it somehow 'slipped his mind'). That is, it may not have occurred to Alan that he was kissing a Playboy bunny. T o promote enjoyment, a desire must be subjectively satisfied and must be occurrent. But even if it occurred to someone that he did what he wanted to do, it still would not follow that he enjoyed doing it. People often want to do their duty without enjoying it. Jack may want for various reasons (to study its structure, to see what depths modern music has sunk to) to listen, and finish listening, to an avant-garde duet for garden hose and bicycle wheel, without enjoying a single note of it. Imagine finally a terminal cancer patient in constant, excruciating pain. His daughter kisses him and he wanted her to. But he is too . ~o enjoy E, A must experience miserable to enjoy a n ~ t h i n g T pleasure or happiness; his desires concerning that experience must add significantly to his pleasure; and all of his desires concerning the experience must be taken into account. Satisfied desires do play a role in enjoying something. But the connection is not as simple and direct as Ryle envisaged. I

z

Cf. Penelhum (1964, p. 86), Perry (1967, pp. 106, 129, 152, and 203)~and Cowan (1968, p. 31). This case also shows that the definition of enjoyment recently developed by Warner (1980) is too weak.

253

A CAUSAL THEORY O F ENJOYMENT

Warner (1980, pp. 511, 518) holds that the desires in virtue of which something is enjoyed must be intrinsic desires. This is an attractive suggestion, for it easily explains why I do not enjoy the lamp being on (to recall an earlier example) and might explain why Jack does not enjoy the avant-garde duet. Warner's thesis is no help with cases of duty, however, for the desire to do one's duty is an intrinsic desire (most of the time at least). I may keep a promise for the sake of doing my duty. But I do my duty for its own sake. His thesis also seems too strong. Suppose Joe is playing the slot machines. Money suddenly streams out. Joe enjoys winning the money. He very much wanted to have the experience of winning it. There do not, however, seem to be any intrinsic desires here. T h e desire for money is obviously not intrinsic. T h e desire to win is not intrinsic either. Even if Joe is playing 'just for the fun of it', he desires to win not for the sake of winning, but for the pleasure of winning. T o desire something intrinsically is to desire it for its own sake, not for the sake of anything else, not even for the sake of pleasure.

We can enjoy things a lot, or a little. We enjoy some things more than others. In short, enjoyment varies quantitatively. T h e conditions under which E is enjoyed to some extent are given by D I . H o w much E is enjoyed remains to be defined. Let e(E) designate the degree to which A is enjoying E. We want to define e(E) in such a way that e(E) > o when A is enjoying E. A natural suggestion is to identify degree of enjoyment with that part of A's pleasure or happiness attributable to E. Total pleasure is n

measured by h =

b(pi)d(pi),where p,,

. . . ,pn enumerates all

i=l

thoughts occurring to A. A ready candidate for that part of A's m

pleasure attributable to E would seem to be h(E) =

1 b(pj)d(pj), j=1

where p,, . . . ,p, (m < n) enumerates all occurrent beliefs E causes concerning itself. Both h(E) and h are greater than zero when A is enjoying E. Unfortunately, h(E) could properly be said to represent a p a r t of A's happiness only if h(E) < h. In this case, A has other net sources of pleasure besides E. There is no serious problem though if h(E) = h, for h(E) then represents all of A's happiness, an improper part. But what if h(E) > h ? Suppose all of A's thoughts concern

254

WAYNE A. DAVIS:

experiences E and F , so that h = h(E) 4- h(F); h(F) might well be negative. Imagine that Mary is enjoying the music (E) despite the fact that she is sitting in an uncomfortable chair (F). Mary's enjoyment of the music is being spoiled to some extent, though not completely, by the chair. When h(E) > h, we can still attribute all of A's pleasure to E. No more than h can be attributed to E, however, for E cannot account for more pleasure than A is experiencing. If the spoiling experiences were absent, more pleasure would be attributable to E, so A would enjoy E more. h(E) would not have to change as long as h increased. All of A's pleasure would still be attributable to E, but there would be more of it. Mary would enjoy the music more if she weren't sitting in an uncomfortable chair, even if a more comfortable chair would not increase her attention to or appreciation of the music. Our suggestion can now be worked into a definition. There are two cases to consider: h > h(E) and h(E) h. In the first case, e(E) = h(E); only part of A's pleasure is attributable to E. In the second case, e(E) = h; all of A's pleasure is attributable to E. In general, e(E) = min[h(E), h], the lessor of h(E) and h. Thus:

D2. e(E) = min[h(E), h], provided A is enjoying E. It follows from Dz that if A is enjoying E, then h 2 e(E). This implies that if A is enjoying something a lot, then he is very happy. It also implies that, other things equal, the more A enjoys something the happier he is. None of this would follow if e(E) were defined equal to h(E) in all cases. D2 also makes it possible to enjoy something to an insignificant extent, despite the fact that h(E) must be significantly greater than zero if A is enjoying E. Finally, as the previous paragraph illustrated, thoughts that do not concern E can influence how much E is enjoyed. Other things equal, Mary would enjoy the music more if her attention to the music were interrupted by pleasant thoughts (e.g., that her chair is comfortable, or that she is going to get married soon) than if her attention were interrupted by unpleasant thoughts (e.g., that her chair is uncomfortable, or that her mother is terminally ill). This would not be possible if e(E) always equalled h(E). Why not equate e(E) with h in all cases of enjoyment? Because two things enjoyed at the same time need not be enjoyed to the same extent. Suppose Mary was at a dance concert. She may well have enjoyed the music more than the dancing. She may have found the musicmoving and beautiful, while the dancing struck her as just so-

A CAUSAL THEORY O F ENJOYMENT

255

so. She may simply appreciate music more than dance. Furthermore, h may exceed e(E). Jack is ecstatic, since he has just married the girl of his dreams. H e is enjoying some champagne at the reception. If e(E) always equalled h, it would follow that he is enjoying the champagne immensely, since he is so happy. Indeed, it would follow that he has never enjoyed any other champagne more (or anything else for that matter), since he has never been so happy. Yet Jack might well have found the wine to be good but not great. D 2 defines degree of occurrent enjoyment, how much someone is enjoying a particular experience on a particular occasion. Degree of dispositional enjoyment would be more difficult to define since a type of experience may be enjoyed to different extents on different occasions. It is hard to assign a single definite value to the degree to which I enjoy playing tennis, since on some occasions I enjoy it a lot while on others I enjoy it a little or not at all. Perhaps some sort of average value could be used.

I have been concerned thus far with positive enjoyment. There is also such a thing as negative enjoyment. A piece of music, or a party, may be unenjoyable. T h e word 'enjoyment', however, does not have its own antonym in English. Replacing 'pleasure (happiness)' with its antonym 'pain (unhappiness)' in D I yields a definition of negative enjoyment. D3. A is negatively enjoying E iff E is causing A to have a number of occurrent beliefs concerning E, which collectively add significantly to the pain (unhappiness) A is experiencing. Thoughts concerning E add to unhappiness when h(E) < o. This entails that at least some of those thoughts are objects of aversion as well as belief. A experiences unhappiness when h < o. T h e scale of enjoyment can now be extended so that negative values represent negative enjoyment: Dq. e(E) = max[h(E), h], provided A is negatively enjoying E. Since in this case both h(E) and h are negative, e(E) must be set equal to the greater of the two, which is the least negative. I t follows from D 2 and Dq that e(E) = o when E is neither enjoyable nor unenjoyable. GEORGETOWN UNIVERSITY