DESIRE AND SERENDIPITY

Draft of a paper now published in Midwestern Studies in Philosophy vol XXII (1999) pp. 120-134

NB: FOOTNOTES NOT INCLUDED HERE

ABSTRACT:

Serendipity is the finding of something that was not looked for, perhaps in the context of a search for something else, or even for just that. In this paper, I explore the place of serendipity in the sources, in the modes of operation, as well as the a ttainment of the objects of desire. Luck pervades them all, in a diversity of ways. An important source of serendipity is the particularity of the events that fulfil our desires: since that particularity is inexhaustible, it cannot have been fully specified by any statement of our desired end. If so, then in every case involving a particular thing or event, even getting what you wanted involves getting something that you didn't know you wanted. Knowledge and happiness provide broad classes of examples: in the search for knowledge, as in the search for happiness, serendipity is "built in". There would be no point in discovering something that you already fully knew, and none of the delights of true happiness are compatible with the full predictability of their content.

Before there was desire, there were tropisms. Until the selectionist paradigm became available to explain them, tropisms seemed to afford evidence of some hidden level of intention or design. The selectionist explanation is completely satisfactory, but it does imply that the relation between a teleological mechanism and its goal state, while admitting of a general explanation over a population of (individuals or types of cases), is in every particular case contingent. Nothing in the intrinsic present mechanism of the tropism contains more than a clue in context as to its goal.

Desires are intentional states, and so would seem to contain in themselves descriptions of their aim. This would make them entirely different from tropisms in that they would be logically linked to their own conditions of satisfaction. Against this Bertrand Russell once argued for the view that desires have no intrinsic objects defining their conditions of satisfaction. (Russell, 1921, p. 72). Instead, what we call the object of a desire is just whatever condition brings it to an end. The standard objection to this is that all desires could then be characterized as having as their objects the receiving of a sharp blow to the head. And certainly it now seems both rather odd and philosophically unnecessary to solve the problem of intentionality by such heavy-handed means. Nevertheless, this view is perhaps still worth thinking about.

First, by placing desires, at least pre-analytically, on a level with primitive forms of teleology, it challenges us to find the differentiae of desire. Second, what causes a desire to end might sometimes be reinterpreted, retroactively, as the desire' s object: might, in other words, be said to be its serendipitous object. In this way serendipity can introduce a link between tropisms and desires: for while desire contains a specification of its object, it can only be a specification, not an identification, of the object. Even if the event meets the specification contained in the desire, it cannot exclude an infinite variation in the qualities of the particular that instantiates that specification. In accordance with those qualities, in turn, my desire may recur or tend to be abated: in the extreme case, that of the "monkey's paw phenomenon," I may learn "never to desire that again." In other cases, my desire may be blunted in the future, and the overall structure of my desires, my tendencies and inclinations, my character, in short, will change. It will change not by design, but on the model of natural selection. This might hold the grain of truth in Russell's view, implying as it does that all cases of satisfied desire are a kind of lucky accident. And this view, as we shall see, has more to recommend it than one might think at first blush.

In this paper, I want to begin exploring the extent to which our desires are riddled with chance both in their origins and in their mode of satisfaction. I want to pursue the following train of thought: that, in a sense, we never get what we want, or g et it only serendipitously if we do. Further, I shall argue that not getting just what we want is part of the point of some of our deepest desires. What we want, when we want what we most want, is to get what we did not want. That is serendipity.

1. Some ways of not getting what we want

"Serendipity" is a word of rather boring etymology. On the authority of the OED2, it comes from Serendip, a former name for Sri Lanka + -ity. A word coined by Horace Walpole, who says (Let. to Mann, 28 Jan. 1754) that he had formed it upon the title of the fairy-tale `The Three Princes of Serendip', the heroes of which `were always making discoveries, by accidents and sagacity, of things they were not in quest of'. Thus the notion of serendipity appears to presuppose, with common sense, that you can sometimes find what you want, and sometimes fail to; and of those latter cases some turn out to be cases of serendipity, where you find something, but not what you were looking for. Similarly, common sense holds that whether you get what you want is to some degree or other a matter of luck, less so if you planned things carefully, more so if you left things, as we say, too much up to chance. But actually there are several sorts of consideration to suggest that we never get what we want. In this section I want to go over some different senses in which this is true.

(i) Gorgiastic desire and Platonic advertising.

Note, first, that if it could be established that (in some or all cases) getting what you want would be a logical impossibility, this shouldn't be assumed to be entirely unfortunate. It would have a bad side, to be sure, but also a good one. The bad side was much stressed by Plato, who had two sorts of reasons for adopting this view. First, desire and pleasure seemed (at least sometimes, for Plato is always large enough to contain contradictions) a kind of laughably self-defeating pursuit: desire as leaky vessel, desire as sieve, an orgiastic or Gorgiastic cycle of itching and scratching, replenishment and loss. Pleasure is essentially conditioned by pain. Desire is a void, but not an inert void. For a void has nothing to do with desire unless it is perceived in relation to the possibility of being filled. Indeed, what Plato finds wrong with desire is precisely this dynamic dimension -- if only because it implies temporality, as if the fact that pleasure and desire can't have any existence outside of time is what debases them. Insofar as the desire for pleasure is also a desire for the absence of pain, therefore, it turns out that for the relevant class of pleasures true satisfaction is logically impossible.

The second reason for condemning desire to perpetual dissatisfaction, on Plato's view, depends not on the mechanics of pleasure but on the metaphysics of desire. It is that the proper explanation for our desire is always the same, and almost always not the one we believe it to be. We seem to desire this and that; but what explains our desire is the Form of the Good. From which Plato fallaciously infers that the Form of the Good is what we really desire. Excepting the case -- rare to the vanishin g point -- where you can bring yourself to desire the Form directly, and -- what is even more rare -- where you can actually somehow attain it, this means that our desires are systematically mistaken and therefore systematically frustrated.

That is the bad side. The good is nicely illustrated by the more mundane case of the use of sex in advertising. A wonderful device, when you come to think about it, in that it turns desire into a sort of commercial perpetual motion machine. If you get people to buy a product by appealing to desires for something else, then you indefinitely postpone the saturation point. Since customers will never be getting what they really want, their desire will not wane and they will keep coming back for more. Thoug h much reviled, this device is actually just an implementation of the Platonic theory of Forms, where the ultimate erotic object takes the place of the Form of the Good. If that is the true object of all our desires, then all our apparent desires are mist aken. "What we really want is never quite what we get": that is the ultimate engine of economic progress and prosperity.

(ii)Two senses of satisfaction.

The Platonic example requires us to distinguish between "real wants" and "apparent wants". While this distinction is usually exemplified by cases which result in disappointment, there is a parallel class of cases in which serendipity might arise: for i f I don't know what I am really looking for, I might indeed find it by accident. Happiness was in my own back yard all along, we say. This thought requires us to accept that one might be mistaken about the object of one's own desire.

Adam Phillips, in an essay called "Looking at Obstacles" [reference] cites (or invents) a Chinese Proverb: "Wherever you look there's something in the way." Phillips is interested in the way that we choose obstacles to put in our own way: these define not just the conscious content of our desire to overcome them, but the unconscious desire not to. A desire for one thing obscures other things that we might desire: so desire itself, regardless of the status of its satisfaction conditions, is an obstacle as well as a goal, just as anything we might look at obtrudes our view of something else.

This situation no doubt requires us to concede that many of our desires are unconscious: as would be, for most people, the desire for the Good that Plato places in the position of being the aim of all desire. Unconscious desire complicates the picture, but also simplifies excessively the explanation for the simultaneous satisfaction and frustration of desiring persons: for the mix of satisfaction and frustration can too conveniently be understood in relation to corresponding mixes of conscious and unco nscious desires. So I would rather, for my present purposes, conduct the discussion without appeal to unconscious desire. A more direct and descriptive approach is available, and the situation becomes less paradoxical if we allow ourselves to make a disti nction between two kinds of satisfaction. I'll call these semantic satisfaction and emotional satisfaction. The sufficient condition for semantic satisfaction is that the proposition formulated in the characterization of the desire be true. If I desire that I eat an apple, then my eating an apple constitutes satisfaction in the semantic sense. But this is not sufficient to settle the question of emotional satisfaction: on eating the apple, I may discover that I am not satisfied , or (what may be something different again) that I feel no satisfaction. In such cases I will say that my desire fails of real satisfaction.

Before describing the various ways in which this can happen, I want to differentiate the distinction just made from a closely related one, between semantic satisfaction and success. The latter distinction is particularly important when one seeks to und erstand the difference between desire and belief: for belief contrasts with desire as well as other intentional states in being the only state for which the criteria of success and of semantic satisfaction are identical. Semantic satisfaction, for any propositional attitude, is simply the truth of the associated proposition. Success is the achievement of the aim or formal object of the state, which is determined by its formal object. Thus, since the aim or formal object of belief is truth, semantic satisfaction of a belief is a sufficient criterion of success. For wants and desires, on the other hand, the criterion of success (the formal object) is desirability: to desire something undesirable is, in terms of this terminology, a kind of failure that parallels that of a belief that fails to be true. Now in the standard case (leaving aside desire that have no propositional object) the satisfaction of a desire is the coming to be of its propositional object. The notion of success applied to the desire allows for the possibility that the propositional object was not the right thing to have wanted in the first place. So the notions of success and emotional satisfaction are not the same, but are related in this way: it may at least s ometimes be that emotional satisfaction is not achieved precisely because the object of desire was actually undesirable.

The suggestion that there is an actual or objective matter of fact about what is or is not desirable may well meet with the same sort of scepticism as Plato's suggestion, in the Philebus, that certain pleasures are false. And I don't want to pre ss the point here. I want merely to suggest that there may be something about the object of desire, whether merely relatively to the desirer of in some more objective way, that explains the resulting failure of emotional satisfaction. But for the moment I leave it aside and turn to the task of further describing some of the occasions that may give rise to emotional non-satisfaction.

2 Source, Mechanism, and Object

There are three places, or phases of desire at which luck may intervene. The most obvious is in the object, but it can also occur at the very source or our desire, or in its mode of operation. I briefly consider each of these in turn, and then expand on an aspect of the first which deserves special attention, namely the opportunities for serendipity afforded by the fact that the objects of desires may involve particulars.

(i) Serendipity in Objects

Here are some things -- not necessarily mutually exclusive -- that may prove to have gone wrong as I bite into my apple, thus securing the semantic satisfaction of my original desire.

a) It may be that the apple disappoints: I found the apple mealy and soft, tasteless or not tart enough. In this case I might have told you, or myself, had I bothered more elaborately to detail my desiderata: what I wanted was to bite into a crisp, juicy, sweet, flavorful, and tart apple. This examples makes it appear that the vagueness lies all in the specification, and is remediable with an effort of precision. But this is illusory: for there is no reason to believe that the specification could ever end, proof against all misunderstanding by the genie of destiny. Will it be sufficient to add that the apple must be non-poisonous, unadulterated or only normally adulterated? Only if we can count on normal attending circumstances: provided, that is, that we need not specify such additional conditions as that the apple must not come from the garden of a wicked tyrant who promises death to whoever steals or eats his apples. This condition may be harder to fulfil than it seems, since it amounts, in e ffect, to our having a solved the notorious frame problem in Artificial Intelligence: the problem of know what concomitant circumstances are liable to change and what to stay the same, given any given change.

b) In a second type of case, there was nothing lacking in my specification of the target or my desire. Yet still I am not satisfied, because the very event that constitutes the semantic satisfaction of that desire acts, at the same time, as the cause of a change of heart. The cases of Platonic dissatisfaction already mentioned might be classed under this general heading: the fulfilment of a desire may be sufficient for us to realize that this was not what we really wanted. I find I no longer lik e this kind of apple, even though it does indeed taste exactly as I remembered or imagined it. Or, perhaps, though my tastes have not changed, I find it doesn't taste as it used to; and while I couldn't have specified how it did taste, I would have recogn ized it. The second of these options seems to bring the case back into the class of those where the problem lies merely in the vagueness of the prior characterization; but Daniel Dennett has argued convincingly that neither phenomenology nor physiological research could conclusively establish any real distinction between the two cases. So these, and all intermediate cases in which we apparently can't tell for sure whether or to what extent our tastes have changed, or actual expectations been disappointed, should all be classed together as cases where our powers of anticipation are not able to encompass the full dimensionality of what the event might bring.

c) Actually there is, in the two preceding cases, a disconnection between the specification of desire and the fulfillment of its satisfaction more radical than I have so far suggested. For while we can expect some correlation between semantic and emotional satisfaction, the correlation rests on the presence of the right physiological conditions -- on the right neurotransmitters finding the right receptors, in the right quantity, and so forth. The religious expression of the vicissitudes of moods exemplified by a poet such as George Herbert, who in the Garden describes the gratuitous quality of what Christians call grace:

Grief melts away
Like snow in may
As if there were no such cold thing.

It cannot be
That I am he
On whom thy tempests fell all night....

Taking God to be a metaphor for Biology, we could say that there would be no emotional satisfaction but for the gift of chemical grace.

(ii) Serendipity in the Source of desire

Chemical grace, in this rather metonymic sense, is involved not only in the end point of desire, but in its very sources. A very general sort of luck, good or bad, is involved in my having the desires that I have, as opposed to some others. That observation may seem to be nothing but an excuse for generalized irresponsibility, but it's demonstrable by a simple a priori argument concerning what we might call, with a nod to Borges, The Great Lottery of Desire.

1 The Lottery of Basic Desires

A number of philosophers have drawn attention to an ambiguity in the word `desire'. In Wayne Davis's terminology, it is ambiguous between appetitive and motivational desire: the former is the sense of desire in which we have a desire; its variants are craving, yearning, and it can can be more or less intense. Moreover appetitive desire can give us reasons for other desires, but can't itself arise for reasons. Motivational desire is a more general sort of wanting; it i s the sense in which we can desire to do something even though we experience no feeling of desire for it. It arises from reasons, including the existence of appetitive desire.

In those terms, then, the a priori argument involves the relation between appetitive and motivational desires. Consider the set of all my present desires, both appetitive and motivational, AD. Of these, suppose, first, that the subset of AD that consists in all my actual motivational desires are traceable back to a basic set of desires BD, such that AD are either caused by or instrumentally related to BD. Then BD, perhaps augmented by the appetitive desires that form part of AD, are the desires that are "given" to me: and I can't be responsible for those. There is, then, a kind of moral luck involved in my being born with the desires that I was born with, where `born' is a metonymy for all the influences of genetics and environment which I could in no way have caused or generated. This kind of brute moral luck seems to me to remain, even if some of my very appetites turn out to be influenced by my more elaborate desires. If I hadn't wanted to develop sophisticated tastes, I would have never exposed myself to the kind of experiences that resulted in this craving for foie gras, say. But somehow, at some time, my desire to develop sophisticated tastes arose out of some more basic appetitive desire.

Another feature of appetitive desire that has been several times noted is that they are "self-referential" in the sense that the object "does not `transcend' the existence of that desire: on contemplating a possible world in which he does not have the desire concerned, the agent should see no desirability [in its realisation]." (Platts 1980) But most real appetites, such as thirst are actually counter-examples: when we are thirsty, we need water; if we didn't feel it that would still be true(G.F. Schueler). This brings in a rather odd kind of luck "at the source": while there is nothing mysterious from the biological point of view about the fact that generally we feel thirst when we need water, that is also, from the point of view of any particular episode of thirst, a lucky accident. For evolution will never guarantee that some general truth holds in a particular case: here too, each particular case involves chemical grace.

2. Utopian Desire

A second sort of serendipity "at the source" opens up when we consider the possibility of desire opening out (as wishes more properly than desires can be said to do) onto its very own context, as it does in utopian desire. A utopian desir e is one that not only targets a point in the context of a set of possible futures, but actually designs the set of possible futures itself. A utopian desire, in this sense, can be assessed only according to the most minimal standards of rationality, beca use there can be no exhaustive list of possible worlds to which it can be compared. (Leibniz might, for all we know, have been right). Utopian desire finds itself lost in some random place in design space, and must define a whole world from scratch. So in the case of utopian desire, serendipity lodges at the heart of the point of origin of desire, as well as its destination. There is no frame-free framing of utopia.3 The succession of desires.

More modestly, there is a role for luck in the way that our desires succeed one another when they are not arrived at as means. Annette Baier, discussing Hobbes's notion of desire, has pointed out that

the instability of the Hobbesian desirer stems neither from the impossible intentional object of his desire nor from his failure to appreciate desired goods once they are possessed, but rather from the regeneration of his desires from their ancestors' satisfaction. (Baier 45) Where desires engender other motivational or instrumental desires, there is a certain predictable logic to their succession; but where desires follow one another not instrumentally but contingently, the further desire to which an original desire gives rise is a matter of luck. So this is one more opportunity for serendipity at the source of desire.

(iii) Serendipity in the mode of action of desire.

Donald Davidson has described a striking example to illustrate the point that an action is not just anything that is caused by a want. In addition, the desire must cause the event in the right way. His example is of a man hanging on to a rope at the other end of which his rival is holding on for life. The man realizes that he entertains a murderous desire to let go, and is so disturbed by this desire that he lets go indeed -- but unintentionally. Hence the outcome, though in some way caus ed by a desire,doesn't really count as an action. Here there is indeed something serendipitous about the outcome, about the very fact that the event that occurs is the one that was desired.

Closely related to Davidson's cases are some others, which Jon Elster has famously explored, in which the outcome does not merely happen to be serendipitously related to the desire, but necessarily is so. Such, for example. are going to sleep or falling in love. In these sorts of cases the desire cannot be the cause "in the right way" of the event which is its object, because that event is by definition not an intentional one. These are not strong cases of serendipity in the sense that mo st interests me, however, because the actual aim of the desire is sufficiently specified even if it cannot be attained as a direct exercise of the desire. Some of them, however, such as falling in love (though not, as far as I can see, falling asleep), may involve particular targets; and this is the element which introduces the most striking sort of gap between the specification of the desire's aim and the actuality of its fulfillment.

3 The Infinitude of Particularity

I shall sidle up to the special problems raised by particularity by way of the medieval puzzle concerning a paradox of promising.

Suppose that you have promised someone a horse, and would not prefer not to keep your promise. Here is the method recommended by a scholastic sophist for defending yourself when dunned: show the promisee every actual horse, and ask of it, Have I promis ed you this one? The answer is no. Could we but run through all existing horses, there would be none of which we would have promised it. Therefore, if there is no horse which I have promised to give you, how can you still caim that I have fa iled to keep my promise?

This puzzle arises from a simple ambiguity of scope in the specification of desire. Hence its air of sophistry. But beneath it there lies a deeper problem. For in truth every satisfaction of a desire that involves a particular necessarily brings a measure of serendipity. For the properties of any actual particular are infinite in number, hence it is logically impossible to have specified all of them explicitly.

Is the restriction to actual particulars necessary? It may seem not. For the properties of any possible particular are also infinite: it is particularity, and not actuality, to which the privilege of having infinite properties belongs.

Barring ontological scruples about possibilia, this is true. But there is a crucial epistemological difference. An actual particular can be specified by a simple act of ostensive definition. Its infinite properties are thereby determined, without their needing to be known. But in the general case it is impossible to specify a merely possible particular in such a way that its infinite properties are determined, since this would involve either listing those properties or fully specifying the possible wor ld inhabited by the individual specified, and either task is endless. I say "in the general case", because it might be argued that some possible individuals can be specified parasitically, as in "the possible cup of tea that is identical with the o ne on my desk in every respect except one". I'm sceptical of even this manoeuvre, however, because I don't see how one could control the proliferation of differences. If the cup contains a single molecule of water more than the actual one, for example, then other differences follow (the tea is that much more diluted, for example) and other possibilities for variety open up: where exactly is the additional molecule located? which isotope are involved? What is the past history of that molecule? -- from which again the infinite depth of the particular must fail to be determined.

4 Future particulars

Let us make a fresh start, and look at the way that desire aims at its object. Picture the eye of desire as focused on a point in the future. The space of possible futures stretches before us, like a map on a screen: "that one," says desire: but doesn't necessarily yet move. Note that this picture hints at (`captures' would claim too much) the intentionality of desire, but not what is distinctive about the pro-attitude. It could equally well represent fear, or neutral expectation. This doesn't matter, for the species of luck with which I am concerned can attend all of these. The problem that is raised by this picture is the problem of reference to future particulars. Does the target actually consist in an actual future event, or is it merely a specification in a representational space? When I "look forward to" the future, am I actually referring or merely specifying?

It's not obvious how one is to decide this question. A certain kind of Kripkean essentialism might suggest that under certain conditions an individual is actually referred to, not merely specified: if I refer, say, to someone's next child, then the reference might be countenanced even if no prediction is possible about the sex, appearance, personal characteristics, etc of the person in question. But desire is more exigent than belief in its minimal requirements for particular reference. I can desire the birth of a child, but not the birth of this particular as yet non existent child. The reason for this difference between reference and desire, as it seems to me, is an intriguing one: for it is for the very same reason that reference to a future particular is possible yet desire of one is not, namely the impossibility of non-satisfaction. In one case, nothing (except the non event) could arise to undermine the claim to have made a reference to this rather than some other particular: the intention to do so isn't linked, and doesn't need to be linked, to any specification of its properties. On the sort of view I have in mind, its origins determine its identity, but it doesn't determine its qualitative properties.

By contrast, desire needs to specify some properties: no state will count as a desire unless there is a possibility of being disappointed because the event does not live up to the desire, does not meet its specifications. This is certainly true of appetitive desire, though I'm not sure I have a definitive argument for it in the case of motivational desire. An appetitive desire is typically linked to a need, and needs call for kinds of states or things, not particular states or things. They are the heirs of tropisms, and like them are conditioned by particulars long since gone.

The point can be made in term of the two senses of satisfaction previously distinguished. Recall that semantic satisfaction is the mere truth of the sentence specifying the object of the desire, but affective satisfaction requires that one likes what one gets. The first must be subject to the same conditions as reference, since reference is one of the semantic properties of the sentence. But whether the same holds of the second is moot, and one argument against it seems to me to have some force. This is precisely the argument from the rationality of disappointment: if the event is exactly as you specified it, what can there be to complain about? You can't be rationally disappointed that event a was not event b, if there is no conce ivable way that you could have told the difference. The mere fact of having secured reference is therefore sufficient to provide for satisfaction in BOTH senses.

An objection suggests itself. Could not reference be secured, in spite of a failure of matching between the specification and the event? Suppose I desire to see your next child, whom I identify as your "future daughter". Actually your next child is a b oy. Now my specification has been sufficient (under certain well worn conditions) to refer, but may they not allow me to be disappointed when I find out that this future daughter is actually not a daughter? Only the desirer, it seems, can determine whether such a failure of matching will matter or not. But if this is so, then there seems to me to be no grip to the notion of being disappointed because the particular specified in the description is not the particular that figures in the actual event. I may go back on my original description, and say: "well, I desired to see your child regardless of what sex it turned out to be"; but it seems to make no sense, then, to complain that it wasn't actually the particular individual whom you had "in mind".

5. The Desire for Discovery

Among our most human, if not our most insistent of desires, is the desire to know. When Aristotle said: Human beings above all desire to know, he may have been exaggerating about human nature, but he surely spoke truly of some of us. A desire for knowledge could sometimes just be a desire for a larger dose of certainty in what we already believe. And indeed Meno's Paradox enjoins us from looking for anything new, because we couldn't recognize it even if we hit upon it by chance. Yet the more interesting cases are where the desire for knowledge aims precisely at the unknown: it is a desire for some acquisition of new knowledge. This isn't a paradox, despite Plato's efforts to make it look like one, because the specification is in the form of a question, and the answer needn't be known to be recognized as an answer (unless, perhaps, it's a philosophical question....) So we might say that the desire for knowledge is one in which the need for serendipity is built in.

6 Art and Happiness

Meno's original paradox was about knowledge. But the point applies also to every form of art, including writing philosophy essays. (In thinking about this paper I was moved by the hope that given the posing of a problem, something would come up by way of an interesting elaboration upon it, if not actually a solution. If it has, that was serendipity: I couldn't have predicted it, by hypothesis, and yet it was, in a sense, precisely what I wanted.) The autographic plastic arts, indeed, involve the quest for serendipity in both the ways that I have distinguished. First, it is almost a defining property of the process of invention that while one can desire to invent, one cannot desire to invent X under a description that fully specifies the nature of X, since that would mean one would have already have invented it. So one can look for something that will produce a certain effect; or experiment with the effects produced by various devices; but one cannot simply want to produce an effect which one has fully predicted will be produced by means that one has already fully understood. Second, however much one has specified both the means and the effect, one cannot hope to have produced just this particular object: so this also involves the type of serendipity that is due to the infinitude of the particular's properties. When Picasso said, "Je ne cherche pas, je trouve", we do not have to suppose that he had no desire to produce a work of art.

Finally, a confluence of several of the kinds of serendipity I have discussed might be called happiness. While all Americans are bound by patriotic duty to believe in their right to pursue it, happiness is, in other quarters, notorious as a quality of experience whose relation to its pursuit is sadly random. Think only, for example, of the contribution to happiness that we owe to the sense of sight. Many of the sights that bring the sweetest pleasure -- the grace of a stalking cat, the radiance of a child's face, the rippling of a field of grass in the wind, the curve of the hills against a mottled sky -- in short all that is naturally beautiful illustrate many of the levels of serendipity of which I have spoken:

Insofar as such sights are enjoyed independently of any previous condition of appetitive desire, they belong to that blessed class of pleasures which were exempted from even Plato's denigration. We can do a little to put ourselves in their way, but no more. The crucial elements of happiness are essentially serendipitous.
Whether or not such sights are preceded by desire, there is only a vague biological story we can tell about why it should be these and not others that so lift our hearts. But at the level of individual experience, the fact that those are objects of enjoyment is a given, and the match between desire and experience is a matter of luck at the level of the very source of our desires.
The objects of such desires are never fully specified as part of their content. Indeed, a desire that specifies its content too closely is likely to be an obsessive or pathological one, a case of fetishism. This is presumably what makes the "specialised" sex trade both lucrative and relatively predictable: but the specialized sex trade's contribution to happiness is probably rather a minor one.
Most obviously, happiness is sometimes serendipitously found at the heart of misery. I don't mean in a spirit of masochism, but only in the sense that the converse of the `dust and ashes' phenomenon has been known to occur. Witness Descartes's vivid example, somewhere in the Passions of the Soul, of a man finding unexpected relief amid his tears at the funeral of his wife.
The contemplation of beauty is not knowledge, but what it does share with the pursuit of knowledge is precisely the character of serendipity. For beauty contemplated only remains beauty if it is "ever new", and what keeps it so is precisely the inexha ustible character of the particular scene, even if the particular scene can be described in the same words as a thousand others. The same is true of the pursuit of knowledge. We can want to do it, and even thirst for it, but part of what we want is to dis cover something that we haven't predicted.

7 Conclusion: the Pervasiveness of Luck

I fear I may claim no jewel of serendipity for these plodding lucubrations. But I hope I have at least made a start on a taxonomy of some of the roles played by chance in desires and their satisfaction. What emerges from this perspective is the rather unoriginal observation that our lives are shot through with luck. Our desires are born in luck: we struggle to give meaning to the notion of responsibility, and indeed can barely live without assigning responsibility at least to ourselves; but I can't see how to disagree with Aeschylus, Plato, Lao-zi or Freud, for all of whom the deepest sources of our desires lie far outside the realm of what we can control. These sages differ only on the question of how far these sources can be articulated or comprehended, and here, my Enlightenment instincts pull me back far enough to side with Freud and Plato. Moreover, luck attends not only the birth of our desires but the rough fit between them and their objects which none but the serendipitous attain. Best, then, to espouse our destiny (if we can only manage to desire this) and think of our desires more modestly as hopes or opportunities.

As for the objects of our desires, I have suggested that they present their deepest form of serendipity in two areas where it is conceptually impossible to get what we desire, and where, nevertheless, what we desire is precisely to get what it was beyond us to desire.

One of these areas is the one Aristotle put at the center of our humanity, namely the desire to know. The other is equally central, and relates to our attachment to particulars. What seemed to Plato to be one of the main defects of individuals was their temporality. For us, on the contrary, temporality is the mark of the particular and enhances its value.

I have argued that there is indeed a sense in which we can desire only types, even when these are types of events that involve individuals. But then serendipity turns out to be at the heart of our desire. We can desire only event types, but it be a constituent of an event type that it involve some individual or other, and anyway what we get are always event tokens. Therefore, at least where our desire goes beyond the satisfaction of a simple need (for those aren't for particulars), it is part of our desire that we should be surprised: and it is of the essence of particulars that they should have this power to surprise. This places Aristotle's emphasis on the desire for knowledge in an interesting new light. For the most worthy objects of knowledge according to Aristotle are substances. And substances are particulars, though not just any particulars; in virtue of being particulars, however, their attraction as objects of knowledge lies partly in the fact that they can never be completely known.

What we get is ever both more and less than we can want: less in that particulars never attain the perfection that is compatible only with the thinness of the ideal. More, precisely in that particulars present us with a dimension of thickness th at only actual particulars in space and time afford.

BACK to Ronnie's HOME PAGE