2. Three concepts of virtue

We can talk of the virtues of things (hammers or poems, for instance) and the virtues of people. Restricting ourselves to the latter, a virtue is a kind of excellence of a person, where purely physical excellences such as being strong or fit, are excluded. There may be other conditions on this broad notion, such as being acquired by habituation, being deep and enduring, and being integrated with other such states in a certain way. For instance, in one of the more influential contemporary formulations, Zagzebski defines a virtue as ‘a deep and enduring acquired excellence of a person, involving a characteristic motivation to produce a certain desired end and reliable success in bringing about that end’.Footnote ³ For our purpose, which concerns an examination of the relation of virtue to motivation, we can be largely silent about these further conditions.

In a broad sense of the term, a virtue is an excellent state of a person (where excellence is understood as a valuable property rather than necessarily a supremely valuable one); and states come in a wide range, comprising not just traits (dispositions) but also capacities of one sort or another, including skills and abilities, both practical and intellectual.Footnote ⁴ One can have a capacity without having a motivation to exercise it: someone can have the ability to cycle, even though she has not cycled for years. Though to acquire these abilities in real life requires practice, a person could in principle have acquired them without any practice at all (perhaps the cycling fairy endowed her with the ability with a touch of her magic wand). So one could in principle have some excellences, those that are capacities, without having ever exercised them.

Excellences are in this way distinct from what I will call dispositional excellences or excellent traits. These are states that have to be exercised when appropriate conditions present themselves if a person is to possess them. Consider traits such as being reliable, hardworking and nice. These are all valuable states, but for a person to possess them it is not sufficient that she have the capacity to be reliable, hardworking or nice: unreliable, lazy and nasty people presumably have those capacities too. Rather, the requirement is that on suitable occasions a person acts in a reliable, hardworking or nice way. A disposition or a trait of a person requires the exercise of the appropriate capacity, so its possessor must be appropriately motivated.

The nature of that motivation can vary. Someone may for instance be nice because she wants to be nice for its own sake. I will call this kind of motivation an intrinsic one, in a sense that I will clarify shortly. Alternatively, someone may be motivated to be nice out of fear of social sanctions (being disliked, ostracised, etc., if she is not nice) and so have a non-intrinsic, that is, an extrinsic motivation. One may be reliable because one takes being reliable as a fundamental end, or alternatively because one believes that it is the best way to advance one's career. One may be industrious because one thinks that hard work is a good thing, or because it is a mere means to other things one values, such as a good salary and a job. So dispositional excellences can be actualised by either intrinsic or extrinsic motivations.Footnote ⁵

The third notion of virtue is what I will call a ‘fully-fledged’ virtue. This is a dispositional excellence that requires an intrinsic motivation, rather than requiring either an intrinsic or an extrinsic one. This notion applies to paradigm moral virtues: for instance, a kind person must be motivated to perform a kind action because it is kind, and not because it is an action that will advance her own interests. This is a standard requirement on the moral virtues in ordinary thought, and many philosophers from Aristotle onwards have endorsed it. In what follows, I will often for the sake of brevity call fully-fledged virtues simply ‘virtues’.

How should we understand the notion of an intrinsic motivation invoked in the notion of a virtue? A kind person does not do simply what kindness requires: if someone gives money to a street person merely in order to impress his socially conscious girlfriend or to get rid of the loose change weighing in his pocket, he does what a kind person would do, but he does not do it from kindness, and so he is not really kind. I will say that the actions of such a person conform to kindness or are in accordance with kindness; but that he does not act from kindness, he does not act for the sake of kindness.Footnote ⁶ Saying that someone acts from kindness requires that the person act with the aim of being kind. But having such an aim does not suffice for acting from kindness. If a person acted merely because he thought that kindness in general advanced his interests, that ‘kindness is the best policy’, he would still not be genuinely kind. As Archbishop Richard Whately famously remarked of another virtue, ‘Honesty is the best policy, [but] a man who acts on that motive is not an honest man.’Footnote ⁷ The policy motivation is ruled out if not only a person acts with the aim of being kind, but also he does not have this aim merely because he believes that it is the best means to some further end. I will say that kindness is his ultimate end, that is, it is an end that is adopted not merely because it is the means to some further end that he has.

Thus we have the following condition on virtuous motivation, generalising from kindness to other virtues: a virtuous agent must act for the sake of virtue, where acting for the sake of virtue is understood in terms of the agent's aiming at being virtuous and this aim being her ultimate end:

An agent is intrinsically motivated by virtue just in case she fulfills those two conditions. My claim is that a dispositional excellence is a (fully-fledged) virtue only if the actions that actualise it must be intrinsically motivated in this sense. If the actions that realise the disposition need not be intrinsically motivated, the disposition is not a virtue.Footnote ⁸

Hence we can take the satisfaction of the V-schema as a necessary motivational condition on a state's being a (fully-fledged) virtue. For instance, substituting ‘kind’ for ‘virtuous’ in it, makes it come out true: necessarily a kind agent aims at kindness and she does not pursue kindness merely as a means to some further end. But niceness is not a (fully-fledged) virtue: one can be nice even though one performs nice actions out of fear of social sanctions, thus violating the first condition, or merely because one believes that nice people tend to get ahead, thus violating the second condition.

The virtuous agent may, though need not, have an explicit belief that she is acting virtuously and desire to so act, if she is acting for the sake of virtue. She may, for instance, desire to perform kind actions for their own sake and believe that the action she is currently performing is kind. I will call this sort of virtuous agent highly reflective. But one can act for the sake of virtue without having any explicit belief or desire concerning virtue; that is, agents other than highly reflective ones can be virtuous. Someone may simply be moved by the plight of a street person to hand him some money, without conceptualising her action as kind: she may just believe that he needs help and feel moved by that belief to help him; or she may simply feel a surge of sympathy towards him; and so on. Beliefs, desires and feelings may be realisers of the state of acting for the sake of virtue without containing explicit reference to the concept of virtue.

We can visualise the relationships between these three concepts of virtue in terms of a ‘bull's eye’ structure, with a fully-fledged virtue sitting at the centre, dispositional excellences surrounding and including them, and these in turn being surrounded by and included in excellences. If a state is a fully-fledged virtue it follows that it is a dispositional excellence, since the former requires an intrinsic motivation, whereas the latter is satisfied by either an intrinsic or an extrinsic motivation. And if a state is a dispositional excellence, it is also an excellence, since the former requires some motivation, but the latter may be motivated but need not be.

3. Creativity and concepts of virtue

In which, if any, of these senses is creativity a virtue? We are setting aside putative extra conditions such as being a state acquired by habituation. So to establish that creativity is an excellence, we merely have to establish that it is a valuable psychological quality. That is widely acknowledged: we praise and admire people for their creativity, we seek to foster it in various ways through educational and cultural projects, and so on. We also do not value creativity merely as a means, i.e. for its effects, but also value it as an end: that is, we think that it has final or intrinsic value.

One might challenge the excellence claim: we do not invariably value creativity: the creativity of a torturer and creative testimony in a law court are bad things. There are several points to be made about such examples, but for our purposes we can note that to say that something is a good, even a final good, is not to say that it is an unconditional good, i.e., good under all circumstances. Being intelligent is a final good, but if one is trapped in a boring, stultifying job one might reasonably wish that one were less intelligent, for one would suffer less. Likewise, happiness is a final good, but it may not be good under conditions in which it is at the expense of achievement. So noting that creativity is not an unconditional good does not undermine the excellence claim.

Is creativity a dispositional excellence? Some hold that creativity is a mere capacity or ability. For instance, Margaret Boden defines creativity as ‘the ability to come up with ideas or artefacts that are new, surprising and valuable.’Footnote ⁹ Thinking of creativity as an ability would make it an excellence but not a dispositional excellence. However, there are reasons to think of it as a trait, i.e., something that must be exercised under appropriate conditions. The poet Arthur Rimbaud gave up poetry by the age of twenty-one, having decided to pursue a life of adventure. He presumably still possessed the ability to produce creative poetry, but he was no longer a creative poet. To be that, he would have had to actualise the ability on appropriate occasions: and he was no longer a poet, let alone a creative one. Or consider someone who possessed the ability to write creative poetry but who had never exercised it at all. We would not say that this person was a creative poet, but that he merely possessed the ability to become one. ‘Being creative’ is a success-term: one must have actually done something creative in order to qualify and not merely have the ability to do something.Footnote ¹⁰ In this it is like traits such as kindness, niceness, reliability and so on.

So creativity is not just an excellence, it is a dispositional excellence. Is it also a (fully-fledged) virtue? We have seen that a necessary condition for this is that creativity satisfies the V-schema. It would do so if the following were true:

Does creativity satisfy the first condition? Creativity requires the production of original and valuable items (how the items are produced matters as well). So the creative person must aim at producing original items if the schema is to be satisfied. But there are many examples of people being creative who do not aim at producing anything original. For instance, the anthropological evidence shows that not all societies value originality. Members of the Abelam, a New Guinea tribe, were insistent that their carvings were not original but in the ancestral style, even though they were discernibly very different from those produced a hundred years earlier, and they persisted in this belief even when this was pointed out to them by the anthropologist Anthony Forge. They did so because they were aiming to produce very traditional carvings, not original ones. Serbo-Croatian epic poets claimed that they performed their songs identically each time, even though tape-recording by the folklorist Albert Lord showed that the length of songs could vary by several thousand lines at different times even when performed by the same singer.Footnote ¹¹ Members of these cultures value stability of tradition and not originality: when confronted by the claim that they have produced something original, they deny that they have done so, since they do not want to produce anything original. So it is not part of their aim to produce original things and therefore not part of their aim to be creative, even though they do in fact produce creative things. The point here is not that they do not conceptualise their activities as original: the point is that their reactions, when it is pointed out to them that they have produced something original, show that they are not aiming at being original; indeed, their reactions show that they are aiming at being not original.

These kinds of cases are not uncommon. Many non-Western and early societies regard their art as a form of ritual or magic, so it is important to make art exactly according to the correct, established method. If that is the function of art in their societies, it is important not to be original in any respect that would affect a work's ritual or magical efficacy. But, given the individuality of human beings and because it is very hard to maintain exact historical knowledge in such societies, almost inevitably their artists end up doing something original. And given their aims it is entirely understandable that they deny that their work is original when this is pointed out to them. There are many other examples: for instance, Ancient Egyptian tomb art was a way of making sure that the deceased navigated his way successfully through the afterlife. One did not want a dead pharaoh stranded due to some artist's having done something original; nevertheless, some of these artists ended up doing just that. And there are even some creative artists in Western cultures who disavow the pursuit of originality, but instead regard their work as preserving or reviving a tradition: this may be true of the composer John Tavener, for instance. More generally, an artist can produce work that is unintentionally creative: he comes across some original device in the course of working without having aimed at it. These cases show that it is not a necessary condition on being creative that one aim at being creative.

There is another reason why a creative person does not always have being creative as her aim. For the pursuit of originality can sometimes be self-defeating as a strategy for being creative, since one strives too hard for something new and thereby produces something worthless. Think of the plethora of modernist and contemporary works that strive for novelty in some salient way, but lack any significant value. Kant notes the existence of ‘original nonsense’, and calls the phenomenon in art ‘Mannerism’, which he describes as a ‘sort of aping, namely that of mere individuality (originality) in general, in order to distance oneself as far as possible from imitators, yet without having the talent thereby to be exemplary at the same time.'Footnote ¹² The Mannerist, intent on originality, runs the risk of producing worse work than he otherwise could, and thus of defeating his aims. So there is good reason why creative people do not always aim to be creative.

So creativity fails the first motivational condition on being a fully-fledged virtue. Is the second condition always fulfilled, in those cases where a person does aim at being creative? That is, does a creative person with this aim invariably adopt creativity as her ultimate aim, or is it adopted merely as a means to some further end? We say that a person is not really kind if she aims at kindness merely because she thinks that kindness has instrumental benefits. Would we say the same about someone who aims at being creative but who is motivated by purely instrumental considerations – that is, would we say that she is not really creative?

Consider George, a successful, creative graphic designer. When he was younger he used to design for the love of it, and was motivated by the thought that designing creative products is worthwhile in itself. But as middle age has crept over him, he has grown disillusioned, but he still wants the income, lifestyle and fame that his design success wins him. So now he still aims at and produces creative designs, but does so only because they are a means to securing the material and social goods that he wants. He has in short lost his intrinsic motivation; now his motivation is merely instrumental, and so extrinsic. He continues, nevertheless, to be a creative designer. So creativity does not require an intrinsic motivation and hence is not a fully-fledged virtue; rather, it is a dispositional excellence.

Is this counterexample successful? First, one might object that, as we noted, creativity has final or intrinsic value, i.e. it is valuable for its own sake, and since this is so, in order for an agent to be creative she must be motivated to act because of the intrinsic value of creativity and hence she cannot regard it as being of merely instrumental value. George's case is therefore not a coherent possibility: he must be motivated by the belief that creativity is intrinsically valuable and hence take it as an ultimate aim.

However, that argument fails for two reasons. To say that a state is intrinsically valuable does not entail that an agent believes that it is intrinsically valuable. Also, it is possible to believe that some state is intrinsically valuable but not to be motivated to actualise it for that reason. So George may either not believe that creativity is intrinsically valuable or if he believes it, it may play no role in his motivations to be creative. It is important not to confuse the claim that a state is intrinsically valuable with the distinct claim that an agent is intrinsically motivated to produce it.

Second, one might object that George is not really creative. To be creative it does not suffice that one produce original and valuable items; it also matters how one does this. For instance, producing original and valuable items merely by luck or by mechanical search procedures devoid of all understanding would not count as exercises of creativity.Footnote ¹³ One of the constraints on how items are produced is that the creative agent must exhibit a certain kind of reasons-sensitivity: he takes the actual and potential values of the product he is working on as reasons to fashion it in one way rather than another.Footnote ¹⁴ Since George is interested only in money and fame, he does not exhibit this kind of reasons-sensitivity.

However, George does exhibit this kind of rationality: he takes the fact that making a design in a certain way would make its design values more original as a reason to make it in that way. Of course, he takes it only as an instrumental reason: design values serve his ends of securing fame and fortune, and he does not take them as intrinsic reasons to produce something. But he does take them as reasons because of their beneficial effects, so he does exhibit sensitivity to reasons directed at the value of the product. And this reasons-sensitivity allows him to take credit for his designs, as would not be the case if they were the product of mere luck or of mechanical search techniques.

Third, it may be objected that George is not fully or properly creative because his activities lack the degree of reliability that we require in a creative person's activities. It is just a contingent fact outside his control that his design activities are socially rewarded; were this to alter, he would give up designing. Kieran, for instance, appeals to considerations of reliability in holding that extrinsically motivated agents do not exhibit what he calls ‘exemplary’ creativity.Footnote ¹⁵ Recall that Zagzebski holds that a virtue must be ‘deep and enduring’ and Aristotle that it is ‘a firm and unchanging state'. Sarah Broadie has interpreted Aristotle here to mean, in part, that whereas one could give up a skill voluntarily without its counting against the claim that one possessed that skill, giving up a state voluntarily is incompatible with that state's being a virtue.Footnote ¹⁶

However, Aristotle offers the ‘firm and unchanging’ test as one for fully-fledged virtues such as fairness and courage, so this objection misfires, since it merely strengthens the claim that creativity is not this kind of virtue. George is creative but his state is not deep and enduring in this sense, for he would voluntarily give it up if social incentives changed.

Moreover, the fact that he might voluntarily give up the state does not impugn its status as one of full or proper creativity. Rimbaud, as noted, voluntarily abandoned poetry by the age of twenty one to pursue a life of active adventure, but that he did so in no way impugns his claim to have been a highly creative poet before then. One can be creative at a spectacularly high level without the state's being deep and enduring. And recall Tynnichus, the poet mentioned in Plato's Ion, who produced only one praise-song of any merit: he was creative once, but evidently not reliably so.Footnote ¹⁷

It is also important not to confuse reliability with degree of creativity. Creative actions are, by ordinary standards of reliability, often highly unreliable: they involve going beyond established outcomes, procedures or techniques, so are more likely to fail than routine actions. Indeed, Zagzebski thinks of creativity as an unusual kind of virtue in part because it does not involve reliable success in the same sense that holds for normal kinds of virtue.Footnote ¹⁸ But even when we set reliability at the less demanding success-rate appropriate to creativity, it does not follow that the more creative someone is the more reliable they are in their creativity. An amateur watercolourist may far more reliably produce modestly creative paintings than an extremely creative but very experimental painter, most of whose paintings are failures. Indeed, there is a general reason to think that the more original work will have a higher failure rate, since it goes beyond established outcomes or procedures more than does less original work.

A fourth objection to the George scenario notes that his creativity is less admirable than that of someone, such as his earlier self, whose motivations are intrinsic. We admire people who create out of love of their activity or subject in a way that we do not admire those who are motivated by factors such as fame and money. But someone's creativity is deficient if we have reasons not to admire it: so George's creativity is not full or true creativity. Kieran for instance appeals to reasons for admiration in his defense of creativity as a virtue.Footnote ¹⁹

However, we do admire George – he is after all a great designer. And if we admire him less than his younger self, it is not because his motivation is now extrinsic, but rather because of the values he is now pursuing. Consider a creative thief, who finds ways to rob people in original ways. One kind of creative thief robs out of the sheer love of robbery, a second robs the rich to give to the poor; but in so far as we admire them, we admire the first, intrinsically motivated thief less than the second, instrumentally motivated one: a kleptomaniac is less admirable than Robin Hood. An investor who is intrinsically motivated invests for the sheer love of accumulating money, but an extrinsically motivated investor may invest in order to generate a large sum of money to give away to charity. Here too, we admire the extrinsically motivated agent more than the intrinsically motivated one. Likewise, we may find more admirable a father who is motivated to produce a creative toy to cheer up his sick son than someone who produces the toy simply for the love of being creative. So the explanation of our diminished admiration for the later George is because of the values (fame and money, rather than design values) that ultimately motivate him, not because he now has an extrinsic motivation. Having an intrinsic motivation per se does not explain admiration: the nature of the values that ultimately motivate one does. So we should not hold that George's creativity is somehow deficient because of its extrinsic motivation.

Finally, one may object that the George scenario merely shows that there are two types of creativity: intrinsically motivated creativity is a fully-fledged virtue but extrinsically motivated creativity is not. Hence there can be virtuous and non-virtuous creativity, and that is all that the virtue theorist of creativity needs to claim.

But this objection concedes the point to the dispositional excellence view. If there are some exercises of creativity that are instances of fully-fledged virtue and some that are not, then creativity per se cannot be a fully-fledged virtue. There cannot be, for instance, virtuous and non-virtuous exercises of kindness, because kindness is a virtue, so one cannot exercise it non-virtuously. But dispositional excellences can have fully-fledged virtuous exercises (someone is nice because she takes being nice as an ultimate end) and exercises that are not instances of fully-fledged virtues (she is nice out of timidity or from a desire to gain advancement). So creativity is a dispositional excellence, not a fully-fledged virtue.

So we should admit that George is someone who is genuinely creative but has purely extrinsic motivations, and thus does not exhibit a fully-fledged virtue. A variant case would give him both a purely intrinsic motivation and a purely extrinsic one. He would have mixed motivations. Clearly he would be creative in this scenario too. But note the difference to a fully-fledged virtue: if we discovered that someone were fair partly because she thought that it was right to be fair and partly because she thought cultivating a reputation for fairness would help her advance her career as a politician, then in so far as her actions were motivated by the extrinsic motivation, we would say that they were not fair. This is not true of creativity.

Thought-experiments prove conceptual possibility, but one might wonder how psychologically plausible such scenarios are. Though one can of course always dispute what the motivations are in actual cases, the presence of extrinsic motivations in cases of even landmark creativity is plausibly common. James Watson and Francis Crick discovered the structure of the DNA molecule in 1953. According to Watson's account in his book, The Double Helix, they were extremely competitive and motivated by the desire for scientific glory, reputation, and to win a Nobel Prize. Their main rival in the race was Linus Pauling, a celebrated biologist who developed some of the methods they used. Watson writes that their goal was ‘to imitate Linus Pauling and beat him at his own game.'Footnote ²⁰ When they discovered that Pauling had developed his own solution, the triple helix, they were distraught, and when shortly thereafter they discovered he was mistaken, they were jubilant. There is no reason to deny that they also had an intrinsic motivation to produce valuable, original science. Their motivations were mixed, but they were certainly creative and their extrinsic motivations added to the incentives for them to come up with their breakthrough discovery.

This kind of competiveness and focus on reputational and material benefits is certainly not confined to science. Pablo Picasso was famously competitive and this shaped his artistic creativity at many points. The art historian Jack Flam has argued that the professional rivalry between Picasso and Henri Matisse was a driving force in both their careers, leading to the creation of some of their most radical works: Les Demoiselles d'Avignon (1907) was in part a response to Matisse's Blue Nude (1907).Footnote ²¹

Inventions are often motivated by extrinsic factors. Paul McCready won the Kremer Prize in 1977 for the first human-powered glider to complete a mile-long flight in a figure of eight. In an interview, McCready explained that he had accumulated a lot of debt at the time and that ‘I felt that I didn't have the time to mess with such things, but I had this strong economic motivation to take an interest in man-powered flight, so I charged around trying to figure out a way to solve it.'Footnote ²² And according to Simon Singh, the major early work on decoding the Germans' Enigma Code was done by Poles, drawing on French intelligence, whereas the French and American intelligence communities gave up on it, since it was so hard. Singh notes that this was because the Poles were more terrified of a German invasion than were the French or Germans.Footnote ²³ Fear of death is a more powerful motivator than the intrinsic satisfactions of code breaking.

One could multiply examples, but let me give a final pertinent one. The Research Excellence Framework exercise and its predecessor the Research Assessment Exercise have been a major influence on shaping academic resourcing and careers in the UK for more than twenty years. Its criteria for the quality of academic outputs are that they should show rigour, originality and significance; so they should be creative. A good deal of overproduction of run-of-the-mill academic work has resulted from this major extrinsic motivation, but it has also produced a good deal of very creative work. And of course academics are also familiar with the extrinsic incentives provided by promotion decisions, including the rigours of the US tenure decision procedure. It is hard to believe that academics would be anywhere nearly as productive, including productive of creative work, without such incentives. Simple reflection on the everyday conditions of their existence ought to convince academics that creativity can be extrinsically motivated.

4. The psychological evidence

I have argued that creativity is a disposition, as well as an excellence, so it requires some motivation to be present, but this motivation need not be an intrinsic one; hence creativity is not what I have called a fully-fledged virtue. Not only can we coherently imagine cases of creativity that lack intrinsic motivations or that combine extrinsic and intrinsic motivations, but there are also many real examples where the motivations are plausibly mixed or even purely extrinsic. However, the claim that creativity is not a fully-fledged virtue seems to stand in tension with a major research programme in psychology, which argues that intrinsic motivation plays a central role in creativity.

One of the earliest studies in this programme was the charming 1973 ‘magic marker’ study, which gave three to five year old nursery children magic markers (felt pens) with which to draw. The fifty-one children who showed an interest in drawing with the markers were divided into three groups. The group who were told that they were drawing in order to obtain a ‘Good Player Award’ showed significantly lower interest in the task (as witnessed by their shorter periods of playing with the markers in subsequent free play sessions) and lower quality of drawings than the groups who were offered no reward or who were given an unexpected one.Footnote ²⁴ So offering an extrinsic reward reduced children's intrinsic interest in the task and lowered their creative expression. This and numerous other studies were the basis for Teresa Amabile's formulation of the Intrinsic Motivation Hypothesis (IMH), which in its 1983 version runs: ‘the intrinsically motivated state is conducive to creativity, whereas the extrinsically motivated state is detrimental.'Footnote ²⁵

Subsequent research complicated the picture. In one study, a group of children were given intrinsic-motivation training, using ‘immunisation’ techniques. They were taught how to maintain their intrinsic interest in a task, despite the offer of rewards, by being shown videos of children talking about how to maintain intrinsic interest, and then by discussing this in groups with a teacher. If these children were then offered a reward, the creativity of their stories was actually enhanced relative to those who had the same training but who were offered no reward, and also relative to those who were offered a reward but no training.Footnote ²⁶ So contra the IMH, extrinsic motivation can actually enhance creativity. This and similar studies led Amabile to reformulate her hypothesis as the Intrinsic Motivation Principle (IMP): ‘Intrinsic motivation is conducive to creativity; controlling extrinsic motivation is detrimental to creativity, but informational or enabling extrinsic motivation can be conducive, particularly if initial levels of intrinsic motivation are high.'Footnote ²⁷ So the claim is that some (‘synergistic') extrinsic motivations enhance creativity, since they support intrinsic motivations (for instance, they enhance the subject's sense of being self-determining, or they give information about what is working and what is not) and other (‘non-synergistic') extrinsic motivations undermine creativity, particularly where the offer of a reward tends to suggest control over the subject's behaviour. One of IMP's adherents claims that it is now so well empirically supported that it is ‘an undisputed principle’ in social psychology.Footnote ²⁸

Does this evidence show that creativity is a (fully-fledged) virtue? The first problem is that the claim concerns only probabilistic causal influence: intrinsic motivation is ‘conducive’ to, i.e., tends to promote, creativity, whereas non-synergistic extrinsic motivation is ‘detrimental’ to, i.e., tends to reduce, creativity. But for a trait to be a virtue it is necessary that the agent have the appropriate intrinsic motivation. Being motivated to perform an act for the sake of kindness does not tend to make the act kind; rather, having that aim is necessary for the act to be kind. And being motivated by self-interest in performing a kind act is not just detrimental to the act's being kind: rather, in so far as that is one's motivation, one is not kind, and if it is one's sole motivation, one is not kind at all. So the empirical, causal, claims are too weak to support the necessary connections required for the existence of a virtue.

However, this way of construing IMP assumes that synergistic extrinsic motivations are capable of independently motivating creative actions, as well as motivating them by enhancing intrinsic motivations. But if one holds that synergistic extrinsic motivations work only by enhancing intrinsic ones, then it follows that intrinsic motivations are necessary for creativity: since creativity is a disposition, a trait, it requires some motivation; synergistic extrinsic motivations on their own cannot motivate it, since they work only by enhancing intrinsic motivations; hence an intrinsic motivation must be present for creativity to exist. And this enhancement construal of IMP seems to be how Amabile, at least at some points, understands IMP. For instance, she refers to ‘the intrinsic motivation that is essential for creativity.'Footnote ²⁹ That requires the enhancement, rather than the independence, construal of IMP. Amabile also states that ‘the first (problem identification) and third (response generation) stages of the creative process, where the novelty of the outcome is importantly determined, may require intrinsic motivation that is unencumbered by any significant extrinsic motivation.'Footnote ³⁰ Since there is no creativity without the generation of novel responses, it follows that intrinsic motivation may indeed be required for creativity. And that would support the claim of creativity to be a fully-fledged virtue.

So construed, though, IMP goes considerably beyond what is warranted by the evidence that Amabile marshalls. What her evidence shows is that sometimes extrinsic motivations reduce creativity (the magic marker case) and sometimes enhance it (the immunisation case). Neither kind of case shows that intrinsic motivations are necessary to creativity. The case required to prove that would be one where people who are completely lacking in intrinsic motivation for some task, when offered rewards of some kind, were incapable of producing anything creative. Not only is no evidence of this kind offered, but it is also highly implausible as a psychological claim. For instance, it would mean that George is a psychological impossibility, even though we have no problem in making sense of him. It would also mean that a student who took a required course for credit, but who lacked all intrinsic interest in the subject, would be incapable of producing creative work in the course, even though getting the high grades that he desired depended partly on his producing creative work.

There is even some experimental evidence against IMP understood in its enhancement version. Robert Eisenberger and Linda Rhoades gave 115 students taking an introductory psychology module a class assignment of producing five creative titles for a short story about popcorn being cooked in a pan. One half of the class was also offered a financial reward for coming up with creative titles. The students offered the financial reward produced more creative titles than those who were not offered a reward.Footnote ³¹ Some of these titles were rather good: ‘The Little Kernel That Could’, ‘The Golden Years’, ‘Coming Out’ and ‘Growing Pains'. But all these students were acting from extrinsic motivations (this was true even of the ones who were not promised a reward: this was a class assignment) and there is no reason to think that they had any intrinsic interest in producing short story titles: they were, after all, taking a psychology class, not a creative writing one. Yet they produced something creative. Of course, the supporter of the enhancement construal could maintain that they must have had some intrinsic interest in writing titles, but unsupported psychological speculation does not warrant any claim, let alone one about what is essential.

There is a second reason why the psychological evidence does not support the (fully-fledged) virtue view. This concerns how the notion of an intrinsic motivation is defined in the psychological literature. Amabile in 1983 defined it thus: ‘Persons who engage in an activity for its own sake are intrinsically motivated; persons who engage in an activity to achieve some goal external to task engagement are extrinsically motivated.'Footnote ³² With its reference to engaging in an activity for its own sake this definition is clearly related to the notion of intrinsic motivation as I have characterised it. But it is also puzzling: what would we mean, for instance, if we said that we went walking for its own sake, or did some cooking for its own sake? To make sense of this we have to identify some characteristic of the action that the agent finds desirable.Footnote ³³ So we have to specify the kind of reason the agent has more exactly: for instance, she engages in pleasurable activity for its own sake, i.e., for sake of the pleasure she takes in it. And likewise, for each value characteristic that the agent might pursue for its own sake: e.g., she engages in a challenging activity for its own sake, i.e., for the sake of the challenge it provides; or a cognitively interesting activity for its own sake, i.e., for the sake of its cognitive interest. The definition we developed earlier was used to characterise the kind of intrinsic motivation that a virtuous agent must have: she performs a virtuous action for its own sake, for the sake of virtue: that is, she performs an action because it is virtuous and this is her ultimate end. On Amabile's definition this is only one kind of intrinsic motivation and her definition allows for a wide range of kinds of intrinsic motivation. So even if the psychological evidence showed that intrinsic motivations were necessary for creativity, this would not show that the kind of intrinsic motivation that is necessary for a disposition to be a fully-fledged virtue is required to be creative. Even if one showed, for instance, that enjoying some activity were necessary to be creative in it, this would not prove that creativity is a virtue, since this would require showing that one must act with the ultimate aim of being creative. Suppose that psychological tests of the Abelam showed that they had to enjoy their wood-carving in order to be creative in it; this would not show that creativity is a virtue, since, as we noted, they are creative without aiming to be so.Footnote ³⁴

Amabile later refined her definition:

If we read the list of five ways in which a reaction can be positive as complete, then the problem is that the definition is set up so that the intrinsic motivation of creativity – aiming at creativity as an ultimate end – does not figure at all, so that showing that a motivation satisfies one of these conditions cannot show that creativity is a virtue. And if we read it as a list of open-ended conditions, then as with the earlier definition, showing that intrinsic motivation in some of these senses is necessary to be creative does not entail that the intrinsic motivation is of the right kind to constitute creativity as a virtue.

So the evidence produced by Amabile's research programme does not show that creativity is a virtue. Proponents of the programme also sometimes overstate its claims. Hennessey, for instance, glosses IMP as holding that ‘Intrinsic motivation is conducive to creativity, and extrinsic motivation is almost always detrimental.'Footnote ³⁶

However, that sits ill at ease with the fact, as we noted earlier, that competition and rivalry are common between creative people in various domains, from arts to sciences; that there are institutional structures in many domains, such as academia, that give overt rewards to creative results and moreover appear to produce them; and that there are prizes, including literary, architectural and philosophical ones, which are intended not just to mark creative achievement but also to foster it. Much of the research programme's data comes from work on schoolchildren and undergraduates. But if we turn to the world of high-end creativity associated with professionals, the existence of extrinsic rewards of various kinds is endemic and given how creative some of these people are, the evidence does not suggest that extrinsic motivation is ‘almost always’ detrimental to creativity. Rather, it suggests that it can be very helpful. Given the years of preparation that are required to acquire the expertise needed for creativity in a domain, how long it can take to achieve creative success and how many utterly dispiriting failures someone may have to endure before she does so, it is easy to explain why any motivation, extrinsic or intrinsic, can help someone to get through the hard times. However much one loves one's subject, faced with continual failure, or self-agonising doubts about one's ability to pull off an achievement, or paralysing uncertainty about how to proceed, the thought that one may not just fail in some creative endeavour that one values, but may also lose one's job or forfeit one's career if one does so, is bound, like hanging, to concentrate the mind wonderfully.