Book review : The Art Instinct, by Denis Dutton.
(Editor's note) It would be difficult not to notice the buzz around Denis Dutton's The Art Instinct, (see Nigel Warburton's review, Michael O'Donnell's, and Brian Morton's – and even the Colbert Report). It's always a pleasure to see an accessible and much advertised book written in a naturalistic perspective reach the many. However, it's in the nature of such books to stir controversy. We have asked philosopher Roberto Casati to read and comment Dutton's book, and we look forward to the discussion of his critical review.
Denis Dutton, The art instinct. Bloomsbury Press: 2009.
The importance of Denis Dutton's book lies in its frank endorsement of two very extreme and controversial theses. The theses are, first, that art is an adaptive cultural phenomenon, one that is rooted in an art instinct, and, second, that this rooting has not only, as one may expect, an explanatory import as to how artworks be or look like, but also normative import as to how artworks should be or look like.
The boldness of the two claims is pretty clear. Even if one agrees that the proper explanation of art must use Darwininan resources, one can aling oneself on milder positions, and consider artistic phenomena not the effects of adapatations but by-products or consequences of adaptations; one may even deny that the notion of an art instinct constitute a natural kind. And even if one agrees on Dutton's claim that an art instinct is indeed an adaptation, one can be more cautious in drawing normative consequences therefrom.
Let me confess a general sympathy for an evolutionary approach to culture in general and to art in particular. I take it for likely that if there is a prospect of naturalizing culture, this will be in the framework of a theory of evolution. However, there are many nuances to be discussed and options to be assessed. I think there are various reasons to resist both of the book's claims on the basis of evidence that contrasts with the evidence alleged in Dutton's book, or reinterprets the latter differently.
Dutton's evolutionary hypotheses
At several points in the book, Dutton endorses the strongest possible version of extremely controversial hypotheses, without much arguing.
For example, on p. 147, he endorses Geoffrey Miller's theory of sexual selection for language and creativity, in the following way :
“It is clear that no more than a couple of thousand words at most would have been adequate for communication in the Pleistocene. The excess vocabulary of sixty-thousand-plus words is explained by sexual selection: the evolutionary function of language is not only to be a means of efficient communication byt to be a signal of fitness and general intelligence”.
And later on, pp. 174-175:“We admire clarity, accuracy, and relevance in realistic, descriptive uses of language and regard these qualities as showing that a speaker possesses desirable intellectual qualities (…) Speech performances… are Darwinian fitness indicators”.
Intriguing as it may appear, I cannot see much support for this hypothesis as Dutton states it. The relation with sexual selection appears to be a bit questionable, on the account that as far as the record goes for many centuries the official destinees of many articulated speech performances were males. Elsewhere, the evolutionary theory of art often gains victory by forfeit after a very short fight in which the challengers didn't really stand a chance. This is blatant in the book's account of fiction and stories :
“Story plots are not… unconscious archetypes but structures that inevitably follow, as Aristotle realized and darwinian aesthetics can explain, from an instinctual desire to tell stories about the basic features of the human predicament”.
Here, the victory of the evolutionary explanation is just an artifact of having framed the problem in terms of the alternative between (implausible) Jungian archetypes and evolutionary explanations. But there are other possibilities; for instance, some attractors may command the logical space of plots – in other words, some types of stories may be more likely than others to be told and repeated. Some features of the attractors may indeed be related to adaptations; we do not want stories to be a billion pages long. But I think it is fair to say that we can tolerate a fair lot of distortions, for instance in accepting and enjoying complex temporal structures as those of the movie Memento. But alternatives like this one are not mentioned.
The definition of art
The theory claims to be based on a definition of art. Dutton lists ten criteria :
2.Skill and virtuosity
4.Novelty and creativity
11.Art tradition and institutions
But that is a list of jointly sufficient criteria; hardly a definition (contra Dutton p. 61), that would have to specify necessary conditions as well. Some “background features” are furthermore not listed but should be counted in:
13.Being an artifact
14.Being normally made or performed for an audience
This tolerance for vagueness is a bit surprising given the pretty high standards that are called for elsewhere in the book (For instance, p. 93, Dutton tackles Stephen Jay Gould for his "punch-pulling use of “may”, “most”, and “might” in making his claims (…) he is exasperatingly vague when he turns to patterns of human behavior”) Maybe, as a suggestion, one could ask for at least some hierarchies among the criteria?
When it comes to discuss counterexamples, we learn a bit more. A World cup Final satisfies many criteria (1, 2, 5, 7, 9) but is not a work of art. “The reason to resist calling such games works of art has to do with the absence of what must be weighted as one of the most important items on the list: (12) imaginative experience”. So now imaginative experience is a quasi-necessary condition (elsewhere, it is just a criterion among others). I'm afraid we are never given a final word on the real status of criteria.
Dutton is similarly vague when contrasting art with religion (a notion that is the topic of a good deal of work inspired by evolutionary theory). Dutton states, “ Religion by its very nature makes grand claims about morality, God, and the universe. It follows necessarily that  explaining religious beliefs in terms of an evolutionary source attacks religion at its core.  Works of art, however, seldom make overt assertions of fact or instruct people on how they must behave.  Art's world of imagination and make-believe is one where analysis and criticism spoil none of the fun”.
Anthropologists know of many religions that render point (1) false. Even then, (2) does not follow from (1). It is not our world, arguably, but nothing prevents religious beliefs to be adaptive because true – as happens with many other beliefs. And works of art that have normative or political content are not so rare as 3 suggests. Spoiling the fun is not the issue – we have here one of many changes of subject – if the point was to draw a parallel with religion.
Defining, or prescribing art ?
If we leave this issue of fuzziness aside, the most serious problem with Dutton's definition of art is, to put it bluntly, unashamed normativity. The author is never afraid of selling his personal vision of what art should be like, for a definition of what art really is :
“…bringing an understanding of evolution to bear on art can enhance our enjoyment of it. A determination to shock or puzzle has sent much recent art down a wrong path. Darwinian aesthetics can restore the vital place of beauty, skill and pleasure as high artistic values.”
From that statement – and from much of the book – it is clear that the move consists in equating artistic objects with objects that can be the object of an aesthetic experience, where the latter is mostly a type of hedonistic experience. Caution should be obviously exerted at this point, as consensus dictates that there are many unpleasant or anti-aesthetic artistic objects, as well as many beautiful objects that are not artistic. Here is another instance of undisguised normativity in Dutton's definition of art :
“A play in which a man brews a cup of tea, throws it down the drain without tasting it, makes another an throws it out too, and another repeatedly to the end, might be a Dadaist experiment, or an illustration of an on obsessive disorder, but it would be better described as an anti-story rather a story. A character's motivation, as I indicated earlier in this chapter, involves the expression of will, normally towards the fulfillment of a desire, and against resistance and obstruction of some kind”
On top of flagging the trespassing of the line between desciption and prescription, it is to be said that one could have been more generous towards actual artworks. Subtle performances take all their strength from the irresistible force of repetition. (cf. the Jimmie Durham video, "Cousine mutique des Deschiens et de Monsieur Cyclopède"). Likewise, there are more charitable ways of judging works of art based on smell than the author's judgment : “…smell shows no solid signs of becoming the basis for a high art tradition”. The alleged examples in support of this claim are biased by a rather restricted set of imagined possibilities (Beardsley's “scent organ with keys by which perfume or brandy… could be wafted into the air”). It may be that a better proposal would be something on the scale of a garden (a blind friend of mine has indeed created a garden with a large variety of essences disposed partially according to their smells). We could design small paths in such a garden, with dramatic revelations of scents, or preparations to epiphanies built up slowly with small hints, occlusions, false promises. Some imagination is required here, but why condemn smell without even trying?
Art as institution
This normative agenda is what allows Dutton to downplay the institutional aspect of art. This may be seen for example in the way he tells us how to "properly experience" a work of art :
“Properly experienced, a work of art is bracketed off, detached, disengaged – not from close attention but from immediate personal needs, desires, and practical plans”.
I am not sure this is an accurate description of “proper experience”? I pride myself in possessing some works of art. It is part and parcel of my attitude to them that I possess them. And it is part and parcel of my attitude towards some masterpieces hanging in museums that I desire them. Possession is a necessary condition of pride (it may be loose possession, as when your country's museum acquires a masterpiece), which in turn is a source of pleasure. But maybe this is a reason Dutton needs to use the detachment requirement? The pleasures of possessing an artwork are better explained in an institutional framework.
Roughly speaking, an institutional theory of art states that a work may be genuinely artistic merely by being included in a network of social practices and institutions such as the art market, the european institution of museums, etc. Such theories of art are one of the main target of Dutton's book (full disclosure : I have defended what I take to be a substantive version of the theory elsewhere).
For instance, according to institutionalists, Dada artworks are interesting not much because they incorporate the essence of art, but because they reveal an aspect of artworks that was always present but not visible because of many other functions artworks happened to play. Second, Dada artworks – as many a Duchamp's statement makes clear – are not (or I should say: ought not) supposed to introduce a new standard of taste. You may never really end up liking or enjoying Webern's melodies or Duchamp's Fountain or Joyce's Finnegan's Wake, no matter the amount of exposure to those and similar works of art you are prepared to endure. This fact does not deprive them of the “artwork” label. Failing to explain this amounts to a change of subject in aesthetic discussion.
That's what makes Dada the real hard test case for Dutton's definition. Duchamp's readymades only satisfy his criterion (11); which is of course a necessary condition for the Institutional Theory of art. Dada is discussed extensively later on: “…any artifact that has all, or nearly all, of the other twelve features on the list does not need to have this one to be a work of art; such an object could not fail to be a work of art in the absence of only this feature” (p. 200). This amounts to saying that the institutional feature is a non-necessary feature. In my view, this amounts to a change of subject.
One of Dutton's main argument against institutional theories of art is a charge of naïve social constructivism, that he leads against XXth century art theory. Art criticism thrives on a ridiculously shallow empiricist/associationist psychology, according to which matters of taste and of cultural acceptance are a matter of exposure to paradigms. He gives a nice illustration when mentioning the work of Kumar and Melamid. They famously found important similarities in people's aesthetic preferences, which they polled and then graphically presented in paintings reflecting those preferences in terms of surface allocated to the canvas. Dutton wittingly quotes Arthur Danto's worried remarks on discovering their performance : Danto appears to be so puzzled by the polls that he proposes to explain the data in terms of exposure to reproductions that massively confer an advantage to landscapes among artworks – wich does not seem to put culturalist accounts of universal art preferences on a sound basis.
However, we should not be distracted by Danto's reaction and attempt to explain preferences in terms of cultural imprint. Nor should we be distracted by Anton Webern's claim that “the postman on his rounds might someday be overheard whistling an atonal tune” (p. 205). We may simply concede that no amount of “time and familiarity” could suffice to make atonal music or Dada enjoyable – incidentally, Duchamp accepted this premise. The proper claim is not that by exposing people to calendars or atonal melodies from their infancy they will end up liking those things. The proper claim is that liking or disliking certain artifacts is immaterial to their being or not being artworks.
Dutton is so busy showing that art is much less constrained by cultural institutions than is usually assumed by art theorists, that he sometimes lets mistakes slip into his demonstrations. On the relations of artists to science, he states that “the use of mixed colors in the history of painting has always been intuitive and singular, and never especially dependend on observing the rainbow or on Newton's demonstration of the spectrum”. This appears to be inaccurate. A number of artists were keen on color science, although more on psychological than physical discoveries. Many Impressionists read Chevreuils' work on the effects adjacent colors had on each other – and used it. In older ages, the use of complementary colors in chiaroscuro was systematic.
The art instinct
For what I could find in the book, I fail to see that Dutton has made a case for an art instinct: a specific, dedicated module or system modeled on the language module or instinct, that would react automatically and mandatorily in front of artworks in the way, say, in which the language system reacts when presented with linguistic stimuli. I simply suspect that we are not in the presence of a natural kind here. One can hold a weaker thesis, according to which interpreting artworks is the business not of Darwinian module, but of a system has neurological dignity and gets consolidated in an intensive training process – such as the system for reading (Cf. Stanislas Dehaene's work on reading). However, this would be a completely different, nonadaptive hypothesis, and would require a close inspection of a vast domain of relevant evidence.
This alternative account is seldom given due attention. For example, speaking of Stephen Jay Gould's idea that “high-order cultural activities [are] spin-offs of the oversized human brain”, Dutton claims it is a “false picture” ignoring “the fact that arts, like language, emerge spontaneously and universally in similar forms across cultures, employing imaginative and intellectual capacities that had clear survival value in prehistory”. We find many patterns of argument like this in the book. Use is made of an alleged fact – the universality of art – and of a hypothetical explanation of the fact, in order to criticize an adverse hypothesis. In the case in point, the explanation as it is does not refute Gould's hypothesis, as it is perfectly consistent with it. Thus I cannot see the relevance of statements such as “Art may seem largely cultural, but the art instinct that conditions it is not”. There is no fact to be explained that an art instinct would explain here.
Towards metaaesthetics ?
The most promising line of enquiry in Denis Dutton's book is his exploration of agreement and disagreement about aesthetic matters.
“If we look at [work in aesthetics]… we find that the paradoxes on which aesthetics as a discipline depends – the conflicts that generate incisive analysis instead of bland description – are manifestations of varied and conflicting feelings about art that lie deep in the psyche. The logical analysis requirement… is unable… to explain where these competing feelings and intuitions come from. For this level of explanation, we need to turn to evolution”
If what the author is interested in is a metaaesthetical study of aesthetic agreement and disagreement – a very stimulating idea – and not a characterization of art or a normative claim about what is art, I am indeed prepared to concede much of what the book says! I would really love to see developments along those lines in psychology or sociology. Data are also needed on the converse topic of aesthetic agreement. Dutton writes :
“From Lascaux to Bollywood, artists, writers, and musicians often have little trouble in achieving cross-cultural aesthetic understanding. The natural center on which such understanding exists is where theory must begin”.
The use of 'often' renders the claim less urgent, of course. But what is the psychological evidence for cross-cultural understanding? I would expect a whole area of experimental aesthetics to fluorish here; but so far, the results of this forthcoming research programme are a desideratum, not a fact.
Before concluding, let me make a couple of remarks about the style and attitude of the book : to my taste, it tries to impress readers in a slightly too conspicuous way. The author's habit of intimidating name-dropping occasionally gets tiresome – "the Iliad, the cathedral at Chartres, Leonardo's Lady with an hermine, Breughel's Hunters in the snow, Hokusai's Thirty-Six views of Mount Fuji" etc. (p.7, the list goes on and on). I would like to object here to the rhetoric. If I want to learn something about the arts, I need to know what is it that makes Schubert's Winterreise a masterpiece, and it is not by enlisting it along other masterpieces and adding that “their nobility and grandeur … flow from their ability to addres deep human instincts” that we'll make progress in understanding. In the midst of this Victorian grandeur, Dutton occasionally cracks a few untasty jokes that are all the more surprising in a book so keen on the value of beauty (notice the the pages about Manzoni's cans).
Overall, in my view, Dutton's account of art leaves out too many aspects of art. For instance, like it or not, art has today a remarkable political function (among many others), one that is recognised as such in many contexts (incidentally, p. 232 appears to run together the political function of art with a politically based criticsm of the arts.) And an educational function as well. Many parents have remarked the intriguing preference children have for contemporary art over old master's museums, and exploited it to bring childrens in contact with the artworld. These functions – and the many others that art has, had, and will have – do not request any of the alleged titillations of ancient idyosyncratic preferences. I, for one, am quite happy with the idea of a mature, unconstrained art.