©Ronald de Sousa

This paper is now published in Being Humans, ed.  Neil Roughley. New York & Berlin: De Gruyter (2000)  pp. 287-307, with Reply to Wolfgang Friedlmeier,"  pp. 313-316.  This is a penultimate draft.

1 Why there is no human Nature.

Let me start by proclaiming outright: there is no Human Nature. But the non-existence of their subject has never deterred philosophers from talking about it: we have this much in common, at least, with theologians. Let us then start with the perplexities raised by the phrase itself.

What could 'Nature', in the phrase 'Human Nature', possibly refer to? You might think that Nature is natural if anything is; but Human Nature has been thought partly or wholly divine, or spiritual, and 'divine' and 'spiritual' are among the terms with which 'natural' contrasts. Even if all such talk is interpreted as poetic metaphor, entirely free of any implications about anything super-natural, unnatural, or counter-natural, perplexities still abound about how to understand the term 'Nature' itself.

The cause of these perplexities is that 'nature', which appears to contrast with 'norm', is itself burdened with two sorts of normative implications. The first relates to happiness and the choices conducive to it: one is often told that the secret of happiness is to live according to human nature. The second is concerned with distinguishing what is essential from what is accidental among the characteristics of living human beings. The first is normative in the sense that it implies counsels of conduct, comparisons of value between ways of behaving. The second is normative in a different way, nicely implicit in the slide of meaning undergone by the term 'essential' from belonging to the essence, to of great importance. But while the two enterprises seem different in stressing, respectively, the practical and the epistemological point of view, we shall find that the two can scarcely be disentangled.

2. The problem of Active Passivity.

To make this vivid, let me stay with theology for a moment. In his recent book about Pantheism, Michael Levine rejects Schopenhauer's rather plausible complaint that Pantheism is a doctrine without content, since it simply amounts to renaming the universe "God". Citing Spinoza and the Daoism of Lao Zi as paradigms of Pantheism, Levine objects that both regard nature as a divine and active unity. To think of the universe as a whole as divine, then, amounts to more than simply relabeling it 'God'. While Spinoza's natura naturans and natura naturata are facets of the same substance, the former contrasts with the latter precisely in that it has ethical implications, or at least implications about how one should live. What we must do is be natural, which implies the possibility of being not natural.

But how are we to understand this alternative possibility? We might call this the problem of active passivity. As Mill put it, "While human action cannot help conforming to nature in one meaning of the term, the very aim and object of action is to alter and improve nature in the other meaning". The problem is to say just what belongs in these two senses of nature, and to explain how to assign events to one or the other class. What can this mean in practice? How are we to set about being natural? The very question already suggests that we are in for a long apprenticeship of which the principal difficulty consists, we might say, in learning not to fight our nature. Learning to be natural doesn't come naturally.

Levine suggests that the appropriate practical maxim of the pantheist who aspires to follow Nature is something like the Daoist wu wei. This is usually rendered: Do nothing, but it seems to me that Don't act focuses better both on what the point of the slogan might be and on what is paradoxical about it. The point is this: action, we say, alters the course of nature. (Hence we "can't alter the past" -- which wrongly suggests that we can alter the future.) But in fact that is precisely what is paradoxical, since there's nothing more natural to a living thing than action. The course of nature is the sequence of all events that actually occur, including acts performed by natural creatures. Whatever I do is part of nature (in Mill's first sense) and cannot alter it. Now you may say: "Well yes, you do alter it, in the sense that if you hadn't done that, the course of nature would have been different." But the same goes for any counterfactual: had it not rained today, the course of nature would also have gone differently. Yet the rain was not anyone's action.

In one of his provocative little essays, Lewis Thomas advances the view that most of the ills of complex system, whether they be cities or societies, arise from "meddling", that is, from interventions designed to improve them. "It makes a much simpler kind of puzzle. Instead of trying to move in and change things around, try to reach in gingerly and simply extract the intervener." This seems an appealing thought, until we realize that everything we do can be construed as meddling, and that doing nothing would be fatally unnatural.

Still, one can do better at understanding wu wei. The Daoist precept is best contrasted, I suggest, with the Kantian-Existentialist insistence on the asymmetry between the first and the third person. Sartre proclaimed the doctrine that we cannot escape from freedom. I cannot choose not to choose, for if I did, then that too would be a choice. This striking thought is a variant of a doctrine familiar from Kant: there seems to be nothing incoherent about considering others to be determined by causal forces -- except perhaps, as Strawson has pointed out, in the heat of the moment when we are reacting to what they do. Yet it seems absurd to view myself as so determined at the moment of decision, though I can, of course, take this view of myself in the past. However much I believe that I am determined by causal forces, I can't just not act, letting those forces sweep over me, for if those forces are at work, what they will do is precisely cause me to act.

Of course there is a sense in which one can avoid action: it is the sense expressed in the phrase: 'letting oneself go'. Equivalent phrases such as 'going to seed', or 'going to pot', describe a process of letting nature take its course; they are metaphors derived from plants. Only in plants do we envisage the course of nature as being passive, or at least as not presenting the usual contrast between the active and the passive. Wu wei, then, might be more or less the prescription: be ye like the plants.

In his discussion of wu wei, Levine strives to read the phrase sympathetically, yet grants that if rightly understood, it "may mean more 'non-action' than makes sense...." But perhaps our prescription for being natural could stay as close to this as possible, while allowing for the fact that unlike plants I face the problem of endorsing my nature. In order to do this, I must know what my nature is; thus the epistemological form of the problem becomes closely tied to its practical form.

Suppose, for a moment, that some omniscient angel revealed to me my essential human nature. And imagine that I vowed, in gratitude for this knowledge, never again to do anything not specifically enjoined by my nature. It is obvious that my project would be a hopeless one. It may be inscribed in the list of my essential properties that dry food is good for me, or that I am to eat greens, but it won't tell me which to choose between dried spinach and dried broccoli. It may even tell me that I should strive to fulfil my nature, develop my talents, find myself, achieve something in life; but it won't decide what that consists in for me. For what the angel revealed was Human Nature, and even if there were such a thing, it would have to be realizable in countless different ways: my Human essence won't tell me which option is right for this variant of it, me.

3 The reality of possibility.

Underlying these puzzles, are the ambiguous relations between the possible and the actual.

Ab esse ad posse valet consequentia, said the Medievals: If anything is real, it is also possible. The way things are is just one of the many ways they could be. Since Possible-World Semantics became fashionable, this is often expressed by saying that the actual world is just one of many possible worlds. In David Lewis's extreme version, in fact, the actual is just where we happen to be. The word 'actual', in 'the actual world', is a kind of token-reflexive, functioning like 'this' in 'this country' or 'present' in 'the present moment'. We can express this by saying that the actual is contained in the possible. But that way of talking obscures the fact that, from another point of view, the possible is contained in the actual.

The most obvious way in which the possible is contained in the actual is that many properties are dispositional. The actual microstructure of glass is such as to make it possible for light to pass through it; on the arrangement of molecules in water above freezing, supervenes the property of liquidity. And so on. But there are also possibilities that cannot be reduced to the actual dispositions of any substance. These are flagged by a grammatical test: if there is no x such that we can say "it is possible for x to p", then "it is possible that p" expresses such a pure possibility. 'It is possible that it will rain tomorrow' doesn't say that the world has any particular propensity to that effect, but merely that the properties of the world do not exclude it. And while merely possible facts can, as such, play no causal role in the natural world, yet even facts about pure possibility play a determining role in the life of humans. In the words of Goodman again, "a thing is full of threats and promises"; in the literal sense, however, it is typically humans, or at least biological entities, that issue threats and promises, harbor intentions, regrets, resentments, or desires -- all of which states are by their very nature penetrated with non-actual possibility.

In short, the boundaries of the natural stretch well into the non-actual. So how should one set about distinguishing between the natural and the unnatural among nonexistent facts and things? If we are to delineate a Human Nature, that is what we shall need to do.

Aristotle had a way of doing this. Biological entities are potentially what they are before they attain actuality, and while Aristotle holds that some actuality is always prior to any potentiality, any particular actuality is preceded by potentiality. Every oak was first an acorn. If possibility is interpreted in the sense of Aristotelian potentiality, the natural course from possible to actual defines the nature of each biological entity, whether or not this particular one attains its natural end. The Aristotelian program, in fact, is the only one that makes sense of the dual project of discovering our essence and using the result as a guide to how to live.

Unfortunately, Aristotle's is no longer a believable program. Whatever other problems may be posed by the notion of intrinsic teleology, the Darwinian revolution has made it impossible to take seriously either the idea of a human essence or its application to ethical practice. The process of Darwinian evolution is inconsistent not only with the fixity of species, but with the idea that species membership might be defined in terms of necessary and sufficient conditions on shared properties. An individual organism is a member of a given species not in virtue of its properties, but in virtue of its relation to other members of a species, and particularly its lineage. We have the word of biologists that the lineage of humans carries with it no interesting set of defining human characteristics. No set of genes, no set of phenotypic characteristics, is necessary and sufficient to define a human; and if one were to take as defining human nature the impossibly long disjunct of all living human genomes, the next human birth would bring a counterexample. "If evolutionary theory has anything to teach us, it is that variability is at the core of our being."

4 Emotions as nature's pivot

All that being said, it must be admitted that the temptation to continue talking as if there were such a thing as human nature is hard to resist. I myself must plead guilty, for I've argued in the past that there is a specifically human sort of emotion which gets built up on the basis of innate reactive dispositions. These dispositions may be common to all mammals, or to all primates, or on the contrary they may be idiosyncratic in each human being. But what makes human emotions specific, I argued, was that animals without language are unable to fit their emotional reactions into the framework of a story or "paradigm scenario." The essentialism involved in this theory is rather weak: we shall see why in section 7 below; but recently there have been other more stringent attempts to define the essence of emotions. Both Paul Griffiths and Mohan Matthen claim to downplay analyses based on functional analogies, preferring to anchor their views in the biological notion of homology. In that sense, they might be seen as claiming that certain human emotions can be strictly identified in terms of necessary and sufficient conditions. But even if their theories are viewed as strongly essentialist about emotions, they give no comfort to essentialism about the human beings that have them. For true homologies posit common ancestors for the homologous structures and their bearers. It follows that homologies are no respecters of species boundaries, since according to one of the dogmas of modern biology any two individuals, of the same or different species, have some ancestor in common, however distant. A search for the essence of a homologous trait, such as a particular emotion, is therefore not bound to discover any species essence. Essences, where they can be discovered, are of interest because they make a crucial contribution to the explanation of the behaviour of objects belonging to a natural kind. But once one has cast off the classical view of species as natural kinds, and replaced it with something like Ernst Mayr's "biological concept of species" based on lineage and reproductive isolation, the contribution of species to explanation gets drowned by the contributions of other levels of explanation, from demes to genes. Any explanatory power that belongs intrinsically to species fades into relative insignificance.

Nevertheless, the emotions remain, in my view, the nub to focus on if we want to salvage something useful about that non-existent Human Nature. To see why, let us return to the contrast between the activity of humans and the passivity of plants. I said a few moments ago that insofar as it recommended the passivity of plants, the aspiration to wu wei seemed impossible to achieve. Yet the one phenomenon in human life which may be perfectly suited, by its ambiguous position between the active and the passive, for just that impossible task, is precisely human emotion. Emotions are experienced as undergone rather than as chosen, and yet, at the same time, few if any or our actions fail to be influenced by them. Emotions are, moreover, widely held to be natural.

To follow nature, then, might mean to act only as dictated by emotion. The passivity implied in wu wei would then reside in the fact that emotions are experienced as passive, but we wouldn't fall into the absurdity of identifying wu wei with inertness. My passions are frequently endured as alien forces wresting control from my own self. Yet they just as often guide me in what is felt as the exercise of my greatest freedom, the freedom to be myself as opposed to following rules and conventions laid down for me by a tradition or by the expectations of a community. This ambivalence inherent in the perception of reality was beautifully illustrated in Joanne Greenberg's novel about schizophrenia, I Never Promised You a Rose Garden. The protagonist lives in the elaborate imaginary world of her psychosis, and her psychiatrist promises that while for her to "get well" will mean leaving that alternative world for the "real" world, no one can or will force her to abandon her world. She can leave it only of her own free will. When that transition comes, what is striking about it is that she experiences it both as inevitable and as completely free: as if inevitability were not the opposite but the very condition of freedom.

In that ambivalence, at least some passions are akin to perceptions of reality, a reality that reflects at least my individual nature, and at best something shared with others. If perception, then, is a species of knowledge, and if the passions are akin to perception, then one could expect that they too would be passive without being experienced as forced. The passivity of emotion is analogous to the passivity of perception rather than to the passivity of something endured. We don't choose what to see or hear -- except in self-deception. Yet in perceiving we are doing what is natural, and it is not perception but the deprivation of perception that is experienced as imposed on us against our will.

There is more to say than can be said here about the hypothesis that some emotions constitute perceptions of objective reality. Suffice it to note, for present purposes, that if there is a part of reality that is thus apprehended by emotions, this will comprise facts about values. I will adopt without further argument, as a working hypothesis, this view of emotions as at least sometimes apprehensions of axiological facts. It promises an alternative way to deal with the problem that Aristotle's notion of intrinsic teleology would have solved, if we could only accept it: the problem of delineating those merely possible states which can be classified as natural, and therefore worthy (for whoever wishes to lead a natural life) of being pursued. That problem, as we saw, merges into the practical problem of how to practice active passivity.

First, though, I need to say a word more about about that protean term, emotion. The characterization alluded to above implied that typical human emotions have a certain dependency on language and on the stories that language can be used to tell; but pleasure and pain represent, as it were, the degenerate case of stories, and for present purposes they must count as emotions. As I use it, then, the term emotion does not exclude pleasure and pain; pleasure and pain are limiting cases of emotion.

Learning as an expansion of the scope of happiness.

"Human beings desire above all to know," said Aristotle, "and particularly to see." More broadly, we derive pleasure from the exercise of a competence: we might think of all learning, then, as extending our scope for pleasure and happiness. Spinoza too defined joy as the conscious aspect of an increase in power. From the evolutionary point of view, this is just what we would expect, since our survival depends at least in part on the variety of tasks we are equipped to perform. So it is not surprising that every child (at least until such time as school trains her to turn away in disgust from everything called "learning") displays a passion for learning to do and to know. This fact illustrates some ways that the passions of pleasure and pain act as our guides to "what nature intended." But there are two obvious problems. One is the problem of perversion; the other, which will be the topic of the next section, is the complexity of emotional determination of behaviour.

Perversion

Aristotle famously described happiness in terms of virtuous activity in accordance with nature. To exercise your essential nature, then, is what you should do if you are interested in happiness. But it is anyway part of our nature that we are all interested in happiness. Or at least we all normally should be. We shall see in a moment to what extent these Aristotelian thoughts may be translated into the language of modern evolutionary biology. The hypothesis would then be that evolution, rather than some metaphysical principle, has established that convenient pre-established harmony between our passions and what is most likely to fulfil our natural essence. Indeed, now that we need no longer aspire to conformity with any species essence, each of us can understand our own essence as pertaining only to our individual nature. That pre-established harmony however, can be disrupted. Our character can be distorted to the extent that what we pursue no longer conforms to our nature. (Alternatively, we may even lose the natural disposition to pursue happiness in any form. But I'll ignore that problem here.)

If our character is distorted, our pleasures are perverted. This can matter even if there is no God to offend. Many moralists have insisted that perverted pleasures give no lasting satisfaction and don't really make you happy. But unless we have a theory of human nature that is independent of our subjective feelings, that hypothesis is quite vacuous. For the only available test of the natural or perverted character of a given pleasure is whether it gives us lasting satisfaction. And by most reports the pleasures that some regard as most perverted sometimes provide others with their most lasting satisfactions.

That doesn't mean the definition is useless in practice. On the contrary, "what gives me satisfaction" can usefully replace all pretense of talking of what is natural. But to be driven to that is once again to give up on any sort of objective and general characterization of human nature.

Does this matter? We may begin by noting one political advantage. (Although to call it that is to admit that not everyone will regard it as an advantage of any kind). If there is no human nature, then no argument of the form: 'it's bad, and you ought not to do it, because it's unnatural' is acceptable. A fertile source of human oppression is undermined. But does it leave us with any guidance at all? Does it mean, for example, that we no longer have any good argument against the practice of footbinding or clitoridectomy?

Not at all. For those practices hurt, which is prima facie argument enough. The doctrine under consideration has agreed to rely on the verdict of emotion, including pleasure and pain, replacing, rather than providing a criterion for, the concept of 'conforming to nature'.

But here is the rub. A prima facie argument does not end the matter. Education hurts too. (Indeed, the advocates of footbinding or clitoridectomy are likely to think of those practices as improvements on a par with other requirements of a responsible upbringing.) Learning to read involves many painful hours. Yet few loving parents would agree, on that ground, to exempt their child from the toil of learning to read. Much the same can be said for the trouble of adapting to social life itself -- a skill that is hardly likely to be managed without any unpleasant moments. Even if we reject Aristotle's claim that sociality is a necessary part of Human Nature we are unlikely to regard that as undesirable or even optional.

5 Behaviour's Determination by Emotion.

Though the exercise of skills and knowledge typically provides pleasure, skills and knowledge must be learned. In the early stages of mammalian life, the process of learning appears to be wired in: infants' play is not very different from scientific inquiry. But later the games of learning must compete with easier pastimes. Where does the guidance and motive power for such efforts come from? I suggest that it lies in emotion. In the rest of this essay, I shall first sketch some simple examples, and then work up to speculate about some more comprehensive and complicated ways that emotions determine our attempts to learn how to be natural. Emotions, I shall argue, are indispensable guides in our quest for appropriate values in the changing context of our lives.

Some of the ways that emotion governs learning seem fairly straightforward. In the simplest cases of all, emotions facilitate learning without motivating any particular effort. They work in a purely causal mode. In these cases, the effect is produced rather by the intensity of the emotion than by its nature or its content. Any strong emotion has a tendency to fix concomitant events in memory. (All Americans of my generation are said to remember the precise moment when they heard about the assassination of President Kennedy.) Such anecdotal evidence is confirmed by solid research: without activation of the limbic system, which is typically implicated in emotional states, there is no possibility of the kind of conscious experience that can leave a trace in memory. This link between emotional experience and memory retention may well serve some adaptive purpose, but it has nothing to do with the instrumental intentionality of the individual: it can't properly be called motivation of any sort.

Yet emotions are typically thought to motivate. Curiosity, emulation, and some less admirable emotions such as envy, pride, humiliation and fear have the power to motivate the often painful effort required to learn new skills. In these cases, the emotions motivate effort in an instrumental or intentional mode: to say that the threat of humiliation induces me to study for my exams, for example, is to say that I calculate that studying for my exams is a necessary means to avoid humiliation. In this type of case emotions certainly involve desires, and sometimes amount to little more.

In fact, however, to speak of emotions as motivating effort is misleading, in suggesting that there is no significant distinction between emotions and desires. It is desire that typically motivates (some, though not desires all desires, can also be motivated.) And some have indeed thought that emotions were nothing more than a species of desire. But the category of emotions is much broader than desire, and the role emotions play in our choices is different and more fundamental.

To see why, consider first a kind of case that cannot easily be accounted for either in terms of the purely causal model or in terms of an assimilation of emotion to desire.

The Marshmallow Test

Shoda, Mischel and Peake offered a group of four-year-old children a marshmallow with the following choice: You may eat this right away; but if you wait for me to come back, in a few minutes, you can have a second one. A dozen years later, those same children were tested for their social skills, emotional adjustment, and scholastic ability. One might have expected the temptation resisters to have become more efficient, better able to master frustration, and generally better adapted to social life. That did indeed prove to be the case; but more surprising was the fact that the group of resisters scored over 200 points higher on their Scholastic Aptitude tests. The marshmallow test proved twice as reliable for the prediction of scholastic performance at the end of high school as IQ at the same age.

We must infer that the capacity to resist temptation is related to the capacity for learning. It might be objected, however, that the role of emotion in the marshmallow test is not distinct from the modes already described. Or, on the contrary, it might be claimed that emotions aren't actually involved at all.

If the first objection is correct, the case must be classed either with the instrumental-intentional or with the merely causal. But whatever emotions were involved, their effect can't have been purely causal, since it was manifestly the content of a thought -- the expectation of a reward -- rather than its affective intensity alone that was relevant to the result.

It is less obvious that the fortitude that allowed the resisters to hold out couldn't have acted in the same way as in the instrumental-intentional case. There would then be two alternative accounts of the case. We could either say that the two groups managed the ends-means calculation differently, or else that they differed in their actual desires.

The first suggestion, however, seems gratuitous: the calculation is hardly one which could have escaped either group, particularly when we recall that their difference was uncorrelated with IQ. The children's problem was not to calculate a suitable means to a pre-established goal, but rather to hold by the result of such a calculation. What they required, in other words, was something like a capacity to avoid "weakness of will" or acrasia. In cases of acrasia, as Aristotle pointed out, the reasons are already in: adding more reasons won't help: "When water chokes, what will you wash it down with?"

But this still leaves the possibility that the resisters were simply subject to a stronger desire for the second marshmallow. It is a plausible principle of folk psychology that only wants can fight wants. Despite Kant's insistence that acts motivated by "inclination" can claim no moral worth, even the dictates of the pure Rational Will must be transmuted into wants if they are to mount an effective resistance to temptation. The application of that principle here, it might be claimed, is that my desire for the immediate consumption of one marshmallow is curbed only by the greater desire to get another. But this principle too runs into difficulties before the ancient problem of acrasia. If acrasia exists, the principle must be false, insofar as acrasia consists in acting on the weaker desire. If, on the other hand, the principle is true analytically, it rules out the existence of acrasia. No one ever acts on the weaker desire, and so the two groups were simply distinguished by the different strengths of their desires: the first group had a stronger desire to consume the first marshmallow at once, the second a stronger desire to acquire a second. But is it really plausible to suppose that future scholastic ability is correlated with the strength of a desire for two marshmallows rather than one?

The capacity to resist acrasia is obviously important for the accomplishment of many an entreprise. It is a major component of what used to be called character, and it seems to have been lacking in Phineas Gage and "Elliott," the patients described by Damasio. After a frontal lobe lesion, these patients seemed unchanged in their intellectual abilities and even in their personality as measured by written tests such as the MMPI, yet they became systematically acratic. Their lives fell apart, as if they had lost the adult equivalent of the capacity to pass the marshmallow test. And what is most striking in the clinical descriptions that have been given of these characters, is that their capacity for instrumental-intentional calculation appears to be unimpaired. We might say they have not lost their ability to reason practically in theory, they are just incapable of applying this skill in practice.

But if that is the case, then why not say that perhaps emotion has nothing to do with the matter at all? That was the second objection. What distinguished the resisters, the objection goes, isn't any special emotion, but on the contrary some altogether different disposition, amounting to the capacity to overcome emotion.

A knock-down argument is not to be expected against this second objection. But a broader view of the role of emotion in our mental and behavioral economy may help to see why the objection is misguided. As the case of acrasia suggests, emotions do not typically act on behaviour merely in the manner of desires. To get a more comprehensive sense of the role of emotions in solving our problem of how to learn to be natural, we must first return to the question of natural thriving, and face a problem I have so far barely hinted at.

6 Emotions and Axiological Norms

I have so far spoken as if thriving in accordance with one's nature were something I can do, as it were, all by myself. I have left out the social and the moral. Yet among the skills one is expected to learn is the skill of living among other people. That too must be learned, but it must be conceded that it is not a skill the exercise of which invariably leads to an increase in the range of one's pleasures. This regrettable fact commonly gives rise to the claim that morality is opposed to nature: "Nature," says Katherine Hepburn to Humphrey Bogart in African Queen, "is what we are put in the world to rise above." Yet we do generally manage to learn some social skills and moral norms, and that we do not necessarily regard the process whereby we do so in the same light as footbinding or clitoral excision. Those distasteful examples, however, serve to remind us that even in the most sceptical mind, some measure of conviction remains that the wickedness of some practices relates to their being in some sense unnatural.

Can we, after all, make that good? Certain recent arguments about the notion of function might nourish some hope. If we can give sound biological reasons for thinking that certain dispositions have natural functions, then perhaps those dispositions can be ranked more highly, in our striving for the natural life, than those that reflect no such utility.

The leading attempts to define in a naturalistic spirit a normative notion of function have advocated an etiological account of the normativity of biological function. The common core of these proposals is that while the causal powers of functions might be explained in terms of proximate causes leading to the elaboration of structures that underlie dispositional properties, there is nothing in any particular case, considered in itself, that differentiates those effects of an organ or device which constitute its function, from others which are mere effects. A device may have many effects; but only those that it is meant to have count as its function -- including, not infrequently, effects that it fails to bring about. The classic example compares two effects of the heart's functioning: the circulation of the blood, and the production of beating sounds. Both can be seen as playing a part in the orchestration of some larger system (The Symphony of the Human Body, perhaps, in the case of the heart sounds), so no analysis that rests solely on their present causal powers to contribute to some larger system, can account for our intuition that the first, but not the second, is the function of the heart. The phylogeny of the heart, by contrast, can justify our calling the former, but not the latter, a (or the) function of the heart. For hearts exist in virtue of the advantage conferred, on the ancestors of the organisms they now inhabit, by other hearts circulating the blood.

Could we then not extend this account to emotions, asking what their respective functions are, and thus gain relatively direct access to the value of their axiological verdicts? The function of fear is to protect us from danger, that we may survive and procreate; the function of love is that we may increase the likelihood of procreation; the function of curiosity is to enhance our learning curves. And so on.

If this view can be made out, we can jettison metaphysical teleology, and still rehabilitate the Aristotelian program. The norms of our lives are built into our emotional makeup by the process of evolution, and our emotions do indeed give us direct access to objective values. These values are not, of course, absolute in the sense of existing independently of the contingent facts of human nature. On the contrary, they derive from facts of human nature, they are axiological facts for humans.

Unfortunately, it won't work quite so readily, for two very different reasons.

First, while one may classify a particular emotion as of a certain type -- a case of grief, a case of jealousy, a case of fear -- and while one may think that grief, jealousy or fear in general may have a biological function, it seems entirely incredible that just any particular token of that type should have, in every case, just that function. Similarly, one might say, one might have some vague notion as to the adaptive function of language -- to conceal one's thoughts, was a good guess attributed to Talleyrand -- yet it would stretch credulity to suggest that every occasion of speech served just that purpose. Every occasion of speech is different, and has its own purposes to serve. So it may be with emotions: each emotion is different, and if it serves any purposes at all, those are determined uniquely by the context at hand.

Second, the legacy of evolution includes all too obvious examples of emotional dispositions whose functional credentials leave little doubt, but which most moralities find repugnant. Murderous jealousy, ethnic hatred, unbridled lust and greed, are surely solidly in place because their effects favoured the propagation of our ancestors over those might-have-been-our-ancestors whose dispositions were tolerant, humane, gentle and modest. Yet few would infer that they should invite the first rather than the second set of emotions to dictate their axiology.

Some have criticized the approach of Millikan et al. because of its focus on historical facts, and suggested a forward looking variant of the same naturalizing strategy. The forward looking variant stresses, again very roughly, the contribution of some feature of an organ or mechanism to the propensity (the intrinsic, one-time only probability) of the organism to propagate, in relation to the average propensity for other members of the population. We can be pretty sure that the heart's sounds, unlike its capacity to circulate the blood, leave this propensity unaffected. This variant therefore appears to account for the fact that circulating the blood, but not producing sounds, is a function of the heart, without requiring that we look to the past for it.

But the problem won't be solved by turning away from the past to the future. For what is objectionable about jealousy, ethnic hatred, lust and greed is not that they are no longer likely to lead to reproductive success. They may or may not; but we're not willing to let their status rest on the answer. The point is not that they don't now serve our reproductive ends, but that they no longer serve our moral ends. But if emotions are, in general, the last court of appeal for morality, what can it mean to condemn certain emotions on moral grounds? Are we not threatened with a descent into the bear-pit of emotivism, where somehow our different emotions must just have it out by brute force?

7 A proposed resolution

Two moves can come to our aid here. One is a crucial refinement introduced by Millikan in the etiological view of functions, which enables us to meet the first objection. The other I'll call axiological holism, which serves to protect us from the second. When taken together, these considerations will support the conclusion that some sense can, after all, be made of the concept of living naturally under the guidance of our emotions.

Particular emotions as derived functions.

I have argued elsewhere that human emotions typically involve more or less elaborate stories, the plots of which are traceable to paradigm scenarios adapted from the earliest dramas lived in our childhood. On this view, mere desires are, like pleasures and pains, degenerate or zero-level cases of emotions. (even here there can be two basic plots that will affect the quality of the emotion: want and expect to get, and want and expect frustration.) But the implication is that in the general case the role of emotions in motivating behaviour will be more complex than that of simple desire in this way: when motivated by an emotion, we have not just some specific goal, but a goal that fits into a larger pattern to be emulated: the total scenario that identifies the particular emotion.

These scenarios, like the occasions of our learning words, with their resulting associations and the value paradigms which they embody, are indefinitely varied. The result of this process is that no two persons' emotional repertoires are identical, any more than any two persons share an identical idiolect. Our individual natures are the outcome of unique genomes meeting unique environments, and the different emotional repertoires thus generated determine the options we see as live ones in our lives. The useful concept of a live option comes from William James, who points out that a nineteenth century New England gentleman felt no need to make a serious comparison between, say, Protestant Christianity and Islam. Whatever its objective virtues, Islam was not a live option for him. In the different context of the twenty-first century, altogether different options may be experienced as live, not only among religions but in very aspect of our social lives and individual choices. Moreover, two contemporaries may differ in the options they apprehend as live. This is enough to account for the fact that our emotions frequently lead us to different conclusions in the face of apparently identical "facts": We cast ourselves as actors in those dramas derived from our idiosyncratic paradigm scenarios, and expect others to join us in roles of our devising. (So no wonder disappointment is all too frequent.) But if I'm right in my argument so far, a world of axiological unanimity is neither more likely to be achieved nor more intrinsically desirable than a world of universal factual agreement.

Paradigm scenarios were built on innate capacities for affect and reaction, but they are (with the possible exception of a few more or less hard-wired "basic affect programs") as far from being simple applications of stereotyped dispositions as our sophisticated utterances are far from the innate articulatory capacities on which language learning rests. They are shaped, both in their origins and in their re-enactments through our lives, by the increasingly complex contexts in which they occur.

Millikan's notion of "adapted proper functions" or "categorial functions" is designed to apply to thought and language rather than emotions. But it is easily adapted to our purposes. We can explain, in terms of it, how we can think of psychological items such as emotions as biological despite the fact that individual tokens can be entirely novel. "Heredity," she notes, "does not directly dictate traits but rather patterns of interaction with the environment, thus controlling development"; thus "to have biological functions an item need neither have the same categorial properties, e.g. the same absolute structure, as items that participated in the life cycles of ancestors, nor need its functions, when categorially described, be functions performed by any of its ancestors." When applied to emotions, this means that a token of an emotion, like a token of an utterance, can be utterly novel without weakening the claim that it has a biological function. As an instance of the exercise of dispositions the capacity for which is was shaped by natural selection, it exhibits a Millikian "proper function". But in the present context, the possibly unique function that it is "meant" to serve in its particular context is enabled by a proper function which evolved precisely to allow such appropriate responses in novel environments.

Axiological Holism.

Even if that is right, it doesn't establish that each emotion as currently experienced embodies an axiologically correct apprehension. It leaves intact the second of the problems just raised, of the manifest nastiness of many particular emotions.

Two considerations can help to ease this worry. The first consists in granting the point. To say that vision has the function of providing distal information about our environment is not to say that visual illusions never occur. Similarly, to say emotions in general constitute apprehensions of axiological reality is not to say that every emotion is equally to be trusted. Emotional mistakes don't show that there is nothing emotions are supposed to be getting right. So how do we tell which is right and which is wrong? Much in the same way as we test the veracity of perceptual information: by appealing to other perceptions or, mutatis mutandis, to other emotions: that is the second consideration. Something like the method of reflective equilibrium is commonplace in science as well as in ethics; what is less often noticed is that in the case of ethics, the items that need to come to equilibrium are typically not ordinary empirical facts, but emotional responses. Without such responses, even the most persuasive argument in favour of painful effort in aid of learning and change will remain powerless to move us. On the other hand, no single response in isolation can claim to deliver a definitive verdict. The analogue of a belief that is reasonable but false is an emotion that is reasonable in the light of the resemblance between the sitation that elicits it and the paradigm scenario that defines the axiological property of which it constitutes the perception, and unreasonable in the light of other, more inclusive scenarios of which the present one can be conceived as a mere fragment. Among available theories of truth, the pragmatic theory may in the end be the most plausible, in its appeal to an objectivity rooted in the settling of opinions at the end of inquiry. Similarly, we might envisage an (equally utopian) summation of all emotions, in the light of all relevant apprehensions, not excluding all the comparative judgments, rational calculations, and thought experiments, that rationality can provide. Against that utopian sum all more partial emotional states will be judged, just as all opinions will be judged against the state of knowledge at the end of inquiry. Meanwhile, the best we can do is mess about looking for reflective equilibria in the local environment of thought, knowledge, reason and feeling.

Even a novel emotion experienced in a novel context may provide axiological guidance. Once we come fully to take stock of this possibility, we are on the way to solving the most puzzling of the problems we've encountered: how to make sense of active passivity in a world that is constantly changing in consequence of our own intervention.

Earlier in this essay I dismissed Lewis Thomas's suggestion that we could refrain from "meddling": the choice of not meddling does not exist. All that people do, as complex emotional and psychological systems, and all that societies "do", as higher-order complexes of people, is nothing but meddling with the inextricable result of infinite earlier meddlings. Despite soothing clichés about how nothing is new under the sun, the fact is that the chances are vanishingly small that even a system of relatively low complexity, such as a cell, should ever find itself twice in precisely the same state. No two animals, even clones, will ever be identical. Both technological change and social change present us with worlds that are strictly new. It is all perfectly meddled with, and all perfectly natural. How, then, comes the question again, can we expect an emotional equipment built for other worlds ever to guide us in this one?

Axiological holism, together with the sort of two-stage functionality of emotions that I've borrowed from Millikan, holds some promise for an answer. We are forever launched into a dialectical dance: our emotional dispositions respond to, but also create, entirely new contexts. We obviously can't judge new situations with anything other than our old dispositions; but if we apply ourselves to take as comprehensive a view of the new context as possible, the totality of our emotional reactions will generate new paradigm scenarios, from which emerge fresh axiological judgments.

To make this claim a little more concrete, let me cite three examples.

The first bears on our capacity to countenance certain philosophical possibilities: it lies in the reflections of this very essay. The possibility of viewing the philosophical problem of the sources of value in the present light is one that did not exist, say, before Nietzsche and Darwin. This is not, I suggest, because such thoughts would have been unthinkable: it is rather because Nietzsche and Darwin have changed the emotional dispositions that set the framework of our philosophical vision, in such a way as to make such thoughts into live options. It is easy to imagine cultural contexts in which they would seem blasphemous, depressing, or mad, and arouse corresponding emotional reactions. But to me, and doubtless to any reader who shares the feeling that there is nothing original about them, these reflections seem simply nothing less than natural. For now, for me, to live in accordance with Nature is to live in accordance with these thoughts.

Perhaps the example of a philosophical vision seems to far removed from concrete emotional life. So here are two other examples, derived from the fact that Western countries have seen, in the century that is ending, the emergence of radically new norms governing the nature of families and family life.

One consequence of that process is that for educated members of liberal societies, it is difficult even to imagine what could justify or even explain the hostility traditionally evoked by what used to be called "unnatural relations". Those who still find repulsive the idea of homosexual marriages, for example, can only do so at the cost of wearing, as the common metaphor aptly puts it, "blinkers" which prevent them from bringing together all the values that form the separate fragments of their own axiology. For I surmise that most of those who live in the context of liberal and democratic assumptions would find that, when probed separately, their own direct emotional reactions would place individual love, freedom, tolerance, and certain forms of equality high on their axiological scale. But they may be unable to bring these together, and apply them to a situation that they have learned to code in terms of an unquestioned paradigm scenario. The supreme moral value of literature and art, particularly movies, stems from their power to bring such emotions to bear on situations previously judged in the light of rigid stereotypes.

Here is one last example. Nothing seems more "natural", I venture to surmise, than the assumption that children are best raised by their own natural parents. Yet our century has seen several experiments, such as those of the Kibbutzim, in which alternative arrangements have been tried. These were -- naturally enough -- frequently attacked as unnatural. But whatever their measure of success, the present perspective raises the possibility that such alternative arrangements cannot be judged solely in terms of their conformity with the past of our species. Their naturalness can only be judged in the light of the total context in which they arise, a context which is, of course, itself a historical product. Thus the changing relations between the sexes, the demands of modern economic life, and changing assumption about the relation between the various ages of life may engender a situation where very different patterns of child raising come to seem not only normal, but natural. Perhaps that will bring us, in the twenty-first century, the flourishing of a pattern once familiar in a very different context of historical China, where the actual upbringing of any given child is entrusted not to her parents, but to her grandparents. For consider the naturalness of the following facts, in the context of modern Western life: First, women have the highest chance of giving birth to healthy babies in their twenties. Second, both men and women are most likely to be at their intellectual peak between their third and sixth decades. Third, both are less likely, when young, to be emotionally stable and mature, and therefore less fit to raise children. Fourth, men and women in their fifties and sixties are still vigorous enough, as well as likely to be more patient and wise. Does it not seem natural, then, that women should give birth to one generation of children, but that men and women together should bring up not their own children but those of the following generation? It seems to me at least conceivable that the verdict of our emotions considered globally should one day be that such arrangements would indeed be the most natural.

8 Conclusion

I have argued that we must abandon the idea that our emotions, including but conceived far more broadly than pleasure and pain, are mere indicators of some independently existing naturalness. The sum of our emotional reactions to the worlds we create is all there is to "Human Nature." These global dispositions to emotional response constitute, if only by default, that Human Nature, and if they are brought into intelligent and comprehensive dialectical conversation with one another and with the changing contexts of our lives, they constitute all there is to know of moral value.

But emotions can provide reliable guidance only in the context of the totality of their axiological verdicts. That we should ever attain such comprehensive emotional vision is, of course, utopian; but no more so, after all, than the traditional phantasms of moral theory, the Kingdom of Ends, the Original Position, the Greatest Happiness. Such a conception is, I suggest, as close as we can come to the object of our quest: a solution to the problem of active passivity, and a way of forming rational judgments abou(t what is relevantly natural, which make some sense of a notion of living in accordance with nature, even while that "nature" never stands still. Emotions both determine the live options before us, and shape our responses to those options. They do this, in part, in terms of the stories we are able to tell ourselves. As history changes, so do those stories. One of them, which flourished in the Enlightenment and will, I hope, survive the twenty-first century, is a story that prizes coherence and comprehensiveness of vision. This ideal of comprehensiveness, applied to the dialectic of emotions, yields a distant but perhaps still recognizable picture of life lived naturally, even as we create ourselves as the universe's most perfectly functionless artefacts.

NOTES
[not available in this version]

TOP