America's Most Wanted was an audacious painting, even by the standards of the contemporary art scene. In 1993, Vitaly Komar and Alexander Melamid, expatriate Soviet artists then in New York, received money from the Nation Institute to study the artistic preferences of people in ten countries, including Denmark, Kenya, Turkey, China, the US and Ukraine. Were the people of each land drawn to abstract or to realist art? What were their favourite colours? What were the subjects they most liked to see depicted?
Komar and Melamid based their witty "Most Wanted" series of paintings on the results of the poll. America's Most Wanted combined the typical American preferences for historical figures, children and wild animals by placing George Washington in a grassy area beside an attractive river or lake. Near him walk three clean-cut youngsters, looking like holidaymakers at Disneyland; to their right, two deer cavort, while in the water behind Washington a hippopotamus bellows. The "Most Wanted" paintings for other countries looked equally absurd.
Joking aside, the project produced some fascinating results. That people the world over tend to prefer realistic art was perhaps predictable. That their favourite colour is blue was curious, though not astonishing. The art theory shocker was this: not only were landscapes the preferred subject for paintings worldwide, but tastes across the globe all gravitated toward a specific kind of landscape - a bluish scene with trees and open areas, water, human figures and animals. This was the case whether the respondents lived in the desert, the city or a rural area - in other words, their landscape preferences were not conditioned by their actual experience.
Arthur Danto, a Columbia University philosopher and key figure in American art theory, weighed in with the suggestion that the results were a product of the worldwide calendar industry. This explanation seemed far-fetched, but it was consistent with the prevailing idea that has been the basis for academic art and aesthetic theory for the past half-century: our tastes in art are a product of enculturation - our aesthetic pleasures derive from our society rather than anything innate.
The Komar and Melamid experiment seemed to refute this idea. Alexander Melamid himself wondered, "Maybe the blue landscape is genetically imprinted in us, that it's the paradise within, that we came from the blue landscape and we want it . . . We now completed polls in many countries - China, Kenya, Iceland, and so on - and the results are strikingly similar. Can you believe it? Kenya and Iceland - what can be more different in the whole fucking world - and both want blue landscapes . . . The blue landscape is what is really universal, maybe to all mankind."
Human and animal life in general is full of interests, inclinations, emotions and sentiments that are not merely learned from experience, though they may be elicited and shaped by experience and learning. An example: the window of my office at the University of Canterbury in New Zealand is high, and its ledge offers a convenient perch and nesting place for pigeons. My solution is a rubber snake on the ledge. The birds land on the ledge, see the snake, and immediately depart, never to return. The odd thing is that, although European pigeons have been in New Zealand for a couple of hundred pigeon generations, there are no snakes in New Zealand and never have been. The phobic reaction of the birds is learned neither from exposure to snakes nor from images of snakes. It is a perfect instance of a natural atavism, an innate fear response that is passed, unnecessarily in this case, from generation to generation.
Human responses to landscapes also show atavisms, and the Komar and Melamid experiments are a fascinating, if inadvertent, demonstration of this. We could argue that the lush blue landscape type is an innate, evolved preference, present in human nature as an inheritance from the Pleistocene, those 1.6 million years during which modern human beings evolved. The calendar industry has not conspired to influence taste but rather caters to prehistoric, pre-calendrical human preferences.
Unknown to most art historians, there exists a body of psychological scholarship that is much more potent in addressing cross-cultural tastes in landscape than hypotheses about enculturation. For example, the biologist Gordon H Orians has described the ideal landscape that human beings would find intrinsically pleasurable. In his formulation, this landscape has much in common with the savannahs and woodlands where hominids split off from chimpanzee lineages and where much of early human evolution was played out; hence, it is called "the Savannah Hypothesis". In brief, this landscape type includes these elements:
- open spaces of low (or mown) grasses interspersed with thickets of bushes and groupings of trees;
- presence of water directly in view, or evidence of water nearby or in the distance;
- an opening-up in at least one direction to an unimpeded vantage on the horizon;
- evidence of animal and bird life; and
- diversity of greenery, including flowering and fruiting plants.
These innate preferences turn out to be more than just vague, generalised attractions towards generic scenes: they are notably specific. African savannahs are not only the probable scene of a significant portion of human evolution, they are to an extent the habitat meat-eating hominids evolved for - savannahs contain more protein per square kilometre than any other landscape type. Moreover, savannahs offer food at or close to ground level, unlike rainforests, which are more easily navigable by tree-dwelling apes.
Human beings are less attracted to open, flat grasslands and more attracted towards a moderate degree of hilly undulation, suggesting a desire to attain vantage points for orientation. Verdant savannahs are preferred experimentally to savannahs in the dry season. The type of savannah that is ideal appears to be the very savannah imitated not only in paintings and calendars but in many great public parks, such as portions of New York's Central Park. Modernly designed golf courses can make stunning use of such savannah motifs.
High-quality savannahs are characterised by Acacia tortilis, a spreading tree that branches close to the ground. Research shows that there is a cross-cultural preference for trees with moderately dense canopies which fork near the ground (a common tree type in 17th-century Dutch landscape painting). Trees with either skimpy or very dense canopies were less preferred, as were trees whose first branches were well out of human reach. A climbable tree was a device to escape predators in the Pleistocene epoch. This life-and-death fact is revealed today in our aesthetic sense for trees (and in children's spontaneous love for climbing them).
Landscape preference is not always for wildness, a sense of virgin territory, which can appear intuitively forbidding. In particular, the attraction of natural savannah-like scenes can be increased by signs of human habitation - control and intervention. Low grasses that appear to have been grazed by domestic stock can add appeal, as do such modern clichés as a cottage with smoke curling up from the chimney. Such features seem to humanise a landscape, rendering it less threatening.
Responses to landscape also depend on possibilities for exploration and orientation: "reading" a terrain. Experimental work by the psychologists Stephen and Rachel Kaplan shows that the most desirable landscapes have a moderate degree of complexity. Extremes of intricacy, such as an impenetrable jungle, or boring simplicity, such as a flat, open plain, are undesirable. Preferred landscapes are characterised by coherence and legibility - terrain that provides orientation and intelligibility invites exploration.
A sense of a natural or man-made path is the most common cue for exploration, along with a surface that is even enough for walking. A path or a riverbank that can be followed into the distance can greatly increase the appeal of a landscape. This feature is found in landscape arts across the world, and is particularly potent if the scene suggests that a fertile valley or cool mountains might be where the path leads.
The Kaplans have also stressed a preference for an element of mystery, which they define as a feeling that "one could acquire new information if one were to travel deeper into the scene" - following the path or looking around the bend. They speculate that a sense of mystery implies a "longer-range, future aspect" of landscape preference. More than any other component of landscape characteristics, mystery stirs the imagination and as such is vitally important to landscape as an art form.
In a well-known experiment that has been replicated, the psychologists J D Balling and J H Falk showed photographs of five natural landscape types to six different age groups, each of which was asked about its preferences to "live in" or to "visit" each. The landscape types were tropical forest, coniferous forest, deciduous forest, East African savannah and desert. None of the photographs included water or animals. The age groups were eight, 11, 15, 18, 35, and 70 and over. From age 15 onwards, preferences were varied, with an equal liking for deciduous forest, savannah and coniferous forest, all three of which outrated tropical forests and desert, the latter being the least preferred by all age groups. The most striking finding was in the youngest group: eight-year-olds preferred savannahs for both living and visiting above all the other age groups. It is hard to explain this result from habituation, as none of the eight-year-olds had ever been in a savannah environment.
Choice of habitat was a crucial, life-and-death matter for people (and proto-people) in the Pleistocene. From our day back to the time of Socrates and Plato is a mere 120 generations. If we go further back from their Athens to the invention of writing, agriculture and the first cities, it is a lot longer: another 380 generations. But the Pleistocene epoch itself - the evolutionary theatre in which we acquired the tastes, intellectual features, emotional dispositions and personality traits that distinguish us from our hominid ancestors and make us what we are - was 80,000 generations long.
Over such a vast period of time, human beings moved out of Africa and into environments very different from the savannahs. Our ancestors walked along coastlines, went inland, learned to survive as Arctic hunters, and managed to sustain life in the deserts of Asia and Australia. They populated rainforests both temperate and tropical, followed the receding glaciers northward through Europe, and found islands off the east coast of Asia. Human evolution occurred not in any single geographical place, but over much of the globe. Unlike many animal species that are adapted to a single physical habitat and will die out if that environment disappears, human beings - clever, social, language-using tool- makers - devised ways to live in almost all physical environments on earth.
Nevertheless, the desirability of the original savannahs is an innate idea that lies deep in the human mind. We remain emotionally attached to them today because having an emotional predisposition toward such landscape types was a survival advantage for our prehistoric ancestors, not unlike a liking for sweet and fat, or sex. Even if an emotional attachment to such landscapes and a longing to go down the roads they offered had only a small survival advantage in the Pleistocene, it would still have become deeply engrained in the emotional life of the species over thousands of generations. Landscapes cannot be eaten or experienced by touch. They can, however, be seen, and their visual beauty is evolution's way of directing us first to the most fruitful and survivable landforms, just as the beauty and charm of a child is evolution's way of ensuring we treasure our offspring.
The emotions felt by our distant ancestors towards advantageous landscapes are of little use to us today, as we are no longer nomadic hunters who survive off the land. Nevertheless, as we still have the genetic legacy of those ancient nomads, these emotions can flood into modern minds with surprising and unexpected intensity. People who have spent their lives in cities can find themselves on a country road. Rounding a bend, they are confronted with a turn-off that leads up a valley. Pastures and copses of oaks dominate the foreground; farther up the valley the road winds and disappears into older forest. A stream lined with lush foliage follows the road for some distance and then is lost from view, though its route is indicated by groves of older trees. Far up the valley, a last bend in the road can be glimpsed. Beyond that, higher hills take on a bluish, hazy cast, blending imperceptibly into distant mountains flanked by great cumulus clouds.
Such scenes can cause people to stop in their tracks, transfixed by an intense sense of longing and beauty, determined to explore that valley, to see where the road leads. We are what we are today because our primordial ancestors followed paths and riverbanks over the horizon. At such moments, we confront remnants of our species' ancient past.
"The Art Instinct" by Denis Dutton is published by Oxford University Press (£16.99)