On a recent trip to Eastern Australia, I was fortunate to see some of the spectacular ancient "art" in Northern Queensland. "Art" is is quotes because the original meaning and function of these ancient images is unknown. What we do know derives mostly from contemporary aboriginal interpreters, who have grown up with these paintings and wall markings. On their telling, many of the depicted figures are "Quinkans," or spirits. Visiting the Quinkans offers an opportunity to reflect on rock paintings more broadly. Prehistoric imagery raises deep questions about the origins of art. The Queenland sites are near a town called Laura on the Cape York Peninsula. It's about a four hour drive from the closest airport (in Cairns), followed by a harrowing 45 minute drive into sandy bush roads (above). Rachel did the driving, fearlessly facing dust, ditches, roots, and rocks in a rental 4-wheel drive. We were guided by Roy (we never learned his last name), a local guide who works for the Quinkan and Regional Cultural Center. If we'd called in advance, we could have travelled in a guide's car. We were offered a small discount on the the costly tour for the effort of taking our own vehicle. Roy took us to an area called the Quinkan Galleries, one of several sites that allow visitors, which is known for having particular vivid paintings. After the long winding ride, we came to a clearing (below), and from their Roy escorted us along with a small handful of other visitors to a group of rocky overhangs, decorated with overlapping images of local animals, people, and spirits. The term galleries seemed apt, and evidently there are many other sites nearby that are off-limits, and thousands of others elsewhere in Australia. Most people associate rock painting with the caves of Western Europe, but Australia hosts the most expansive complexes of ancient art galleries in the world. Roy told us that he learned about these from his grandfather, but also confided that much knowledge has been lost, and that more might be lost in his children's generation. It's not surprising that knowledge has been lost. The images on these rocks are thousands of years old. Dates range considerably, because these cites were used by countless generations. The oldest extant images are probably around 25,000 year old. For perspective, recall that the world's most famous prehistoric paintings--the caves of Lascaux--are about 15 thousand years old. There are some older paintings in Europe and elsewhere in Australia, but the Laura sites certainly correspond in both content and antiquity to some of the world's earliest images. This art began squarely in the Upper Paleolithic (50-10 thousand years ago), the period in which painting is widely believed to have begun. The images we saw may include some that are much more recent. The oldest graphic elements are abstract carvings (I include an example above from a nearby site called Split Rock, which is open to the public for a nominal fee). The paintings are in several different styles, and often overlap, with older, more faded images in the background. The newest styles may be as young as 6,000 years old, which is still a millennium older than the old kingdom of Egypt. Paintings in these newer styles may have been added and embellished up until the last century, but many of the extant images are well over 10,000 years old. It is unclear whether the Quinkan galleries were in continuous use since Paleolithic times. Given their remote location, it is possible that they fell out of use for periods of time. It is possible, however that there is continuity between these ancient artists and more modern residence of the area. The local Kuku Yalanji people are descended from groups that date back 50 thousand years, when this part of Australia was first inhabited. Locals have inherited a rich lore corresponding to the images here. Their interpretations came into the consciousness of white Australians largely through the efforts of Percy Tresize (left), a pilot, painter, and explorer who began to explore and document the caves around Laura in the 1960s. Tresize consulted local informants and began to reconstruct their stories for a broader audience. His 1969 book, Quinkan Country, remains the most thorough introduction. Tresize collaborated with an aboriginal artist named Dick Roughsey (pictured with Trezise, right). Roughsey was from Mornington Island, not Laura, but he worked with Tresize to collect information. The two also developed complementary painting styles and collaborated on children's books, including one called The Quinkins, which was adapted into a marvelous animated short. It can be viewed on YouTube. I included a still below, showing two spirits -- I will return to them below. The film is narrated by actor and activist, Brian Syron. Another Roughsey film is narrated by David Gulpilil, Australia's most famous indinous actor. Roughsey painted until his death in the 1980s and his work remains popular in Australia. He signed his work using his tribal name, Goobalathaldin. I include two examples below. The first, from 1970, shows Roughsey and Trezise looking at rock art, and the second, from 1981, shows the arrival of the first missionaries. In celebrating ancient aboriginal art, we mustn't forget that one of the world's oldest cultural groups fell victim to genocidal colonization with the arrival of Europeans. According to the current interpretations there are two kinds of Quinkans, or spirits, represented along side the people and animals on these stone walls. The first are the timaras, who are tall and thin and live in the cracks in the mountainsides (see central figure below). The tamers are playful trickers, but they are benevolent and protect children. The other spirits, called imjins, are malevolent. The males bounce around on their oversized penises (called "tails" in the children's story) and the females bounce on their pendulous breasts. Imjins kidnap children and eat people. Here are examples from Split Rock. Roy also told us that one of the figures on a Quinkan Gallery wall represented a female evil spirit who disguised herself as a beautiful young woman to lure men. She is the dark figure with delicate locks of hair and large breasts in the center, below. This evil succubus spirit is standing beside two men, a large bird, and some other animals. A great serpent forms a framing motif over her head. In stark contrast to this busy array, just around the bend on a stretch of the same wall, there is a solitary image of a person lying horizontally, perhaps one of her victims lying dead (below). Just after the section with the succubus figure, there were two rock walls that converged forming a triangular passageway, flanked by more paintings (below). This area, Roy told us, was used for the initiation of young men. Consistent with that story, one wall is covered by juvenile handprints (also below). These are not the only juvenile handprints to be found in ancient sites. It has even been suggested that much rock art was produced by adolescents. Others have used finger-length ratios to suggest that much rock art was produced by women. Such questions of authorship will forever allude us. Though named after the spirits, the Quinkan Galleries are equally impressive for their many depictions of animals. Creatures include kangaroos and wallabies (top), dingos (see the first imjin picture above), scrub turkeys, emus, ibises, echidnas, bats (below), baramundi, snakes, turtles, and crocodiles (below). These animals would have likely been hunted by inhabitants of the region, though there is also a long tradition of individuals adopting a "totem animal," which they cannot eat. The significance of these animals is hard to determine, since the images do not typically show interactions between animals and people. Indeed, the images are all curiously independent of each other. The walls often have a kind of compositional coherence, but the individual elements are rarely interacting, suggesting that these are not images of scenes or events. Some of the images were presumably painted together, such as a group of bats, but many others could have been painted at different times, perhaps years apart. Trezise observes in his book that the Quinkan Galleries contain at least ten different styles corresponding to multiple stages of development, from simple geometric petroglyphs to naturalistic polychrome representations. His list includes: geometric engravings, figurative engravings, stencils, silhouettes, outlined silhouettes, large polychrome figures with light bodies, large polychrome figures with red bodies, small figures in all colors, and large bichromes of men with interior decorations. He also notes that there are small monochrome figures that are poorly drawn, which he identifies as more recent additions. Trezise's sequence is surmised based on patterns of fading and overlapping, rather than chemical dating techniques (For a shorter list, see the image from the Quinkan Cultural Center, right). This account of artistic development, from simple geometry to naturalistic representation, is a standard refrain in studies of prehistoric art. It is sometimes said that the history of painting recapitulated ontogeny: we see children progress through stages that are analogous to how painting developed overtime. The geometric forms in ancient art are often compared to children's early markings, or to absent-minded doodles make by adults. But various facts complicate this simple story. One complication is that, in Europe at least, highly naturalistic three-dimensional sculptures appear before there is evidence of naturalistic painting. The Venus of Hohle Fehls from the Rein valley (right) is 35-40 thousand years old. At that time, painting in Europe was limited to geometric forms. This may make sense, on reflection, however, since three-dimensional forms are, in some sense, more like the 3D objects they depict. Depicting 3D objects on a flat rock wall may have been a greater breakthrough. Another complication is that some of the oldest paintings in Australia--those of Kimberley in the Northwest--are highly naturalistic (see example below). They are quite different from the paintings near Laura: elegant, lyrical, and intricate, where as Laura works are more monolithic, weighty and whimsical. Some have suggested that the Kimberley works were not made by aboriginal artists at all, but rather by immigrants from Africa, who came some 70,000 years ago. I attended a fascinating and provocative lecture in Brisbane by biologist, J.D. Pettigrew, who has compiled evidence suggesting that Kimberley art dates back to these earlier times, and is African in origin. Some would scoff at this theory, and complain that it doesn't give adequate credit to the ancestors of aboriginal Australians, who descend from populations in Central and South Asia. There is also scant evidence of human life in Australia dating back 70,000 years, nor is there any art in Africa that has been dated that far back. As dating methods improve, we can hope to get issues of chronology and authorship straightened out. What is clear, however, is that early paintings were created by a number of different cultural groups, not just in different parts of Australia, but in different parts of the world. Paleolithic rock art has been found in Africa, Asia, Australia, and Europe. The oldest reliable dates are from El Castillo in Spain. There is a red dot in those Spanish caves that has been dated as about 40,000 years old. The paintings at Chauvet begin to appear few thousand years after that. Kimberley painting in Australia could as old, and the rock art of Laura makes its debut around 25,000 years ago. In Africa, representational paintings date back at least 23,000 years (the date of the Namibian image below), and abstract designs are over three times that age. This chronology shows that representational paintings first appear at different times in different places, but mostly within the Upper Paleolithic. It seems unlikely that works that spread across the globe have a common cultural ancestry, so we must ask why different groups came up with the idea to paint, and why this happened during the same period of human prehistory. The puzzle is compounded by a further fact: the first paintings appear many thousands of years after the initial appearance of our species. Modern humans have been around for about 200,000 years (see the 195,00 year old homo sapiens femur below). But representational painting is only about 50,000 years old. If that's right, we didn't paint depictions for three quarters of our history as a species. Why did we start? And why did different groups start independent of each other? One answer is that there was a biological mutation that changed the way humans think (e.g., here). On this view, the brain continued to evolve after the body and some change in our ancestors' brains gave rise to art. I don't buy this story. First, as just noted, representational painting seems to appear in different places in a population that is already widely distributed geographically. This makes it unlikely that there was a single mutation before this spread. Second, there are some authors who think neanderthals were making representational paintings. If so, the practice has less to do with our biology than some conditions of life that were affecting biologically distinct groups. Third, there are many examples of major changes in human behavior that take place without biological mutations. The most obvious and important are the changes that took place during the neolithic revolution. These two (including the development of farming) were found across geographically isolated groups. It seems more likely that the rise of representational painting relates to changes in how our ancestors lived. The most obvious factor is demography (see this, this, this, and this). Representational painting may reflect increased population sizes. Population increases are known to have occurred in the Paleolithic due to changes in climate and resultant patterns of migration. Around 70 thousand years ago, the world temperature plummeted, causing population bottlenecks, and then, when things began to warm up, expansion began to warm up, 60 thousand years ago, new patterns of expansion and migration were possible (see temperature chart above, from here). Migration would have also increased the probability of different groups coming into contact with each other. Such factors would have had two important effects: group sizes would have grown and intergroup conflicts would have increased. Both effects require new practices of cultural cohesion. Large groups can be unified by sharing a common culture, and they they can differentiate themselves from competitors this way. These pressures may have lead to a proliferation of new behaviors, including systems of religion, characteristic dress, music, and visual symbolism. Population growth may have also made life more sedentary. It is hard for large groups to move around, and risky because they are likely to encounter other groups. Setting up permanent dwellings, meeting places, and sacred sites would naturally follow. These are good conditions for the emergence of rock art. The transition from portable arts, like small sculptures and instruments, to petroglyphs and painting may reflect this change in group mobility. This demographic story also helps explain another mystery. In addition to asking why painting began, we must ask why it remains unchanged for so long. At many sites (like Chauvet above), we see essentially the same form recurring over thousands, even tens of thousands, of years. This makes sense if rock art is not an expression of creativity, but rather of group identity. Expressions of group identity (flags, emblems, religious iconography, beloved characters, and so on) tend to remain constant. Egyptian art remained relatively constant for three millennia. Once a population arrives at a stable way of life, there is little pressure to innovate, and much pressure to retain a constant identity. I noted above that sculpture became representational before painting. Humans had been making complex hunting tools for millennia, and making these non-functional items may have begun with the emergence of religion, trade, and other social practices associated with large populations. The revolutionary idea of representational painting may have been a kind of inevitable accident. As groups became less nomadic and began marking walls with abstractions, they may have eventually noticed that some forms resemble things seen in life. Representational art reflects features of the human visual system: people and animals are represented in "canonical orientations" that allow for easy object recognition. Many of the earliest representations are solid, rather than outlined, because the objects we see are do not have contours that differ in color from their surfaces (see the example from the Quinkan galleries above). Perhaps all groups that make abstract marks on walls eventually see one that accidentally resembles the silhouette of a person or animal, which then gives rise (perhaps over thousands of years) to the eventual experiments with representational forms. Inevitable accidents could also explain the prevalence of handprints in rock art (the example below is from Borneo, and others can be found around the world). When working with pigments, it is inevitable that such prints would arise spontaneously when, for example, dry ochre blows over the hand of painter, leaving a hand-shaped impression. Anyone who has followed recent work on rock art will wonder why I haven't mentioned a topic that has become extremely trendy in that literature: shamanism. David Lewis-Williams has spearheaded a movement in recent decades, which interprets much rock art as related to shamanistic rites. He points out, for example, that many rock paintings include geometric forms characteristic of the visual effects that occur when one is entering a hallucinogenic induced trance. For example, he notes that many rock paintings include dots, which are akin to visual spots that appear in early stages of trance states (the image below comes from Peche-Merle, France). Lewis-Williams also notes that rock art often occurs in places that are far removed from sites of habitation, and that animals depicted are often ones that would have been revered, but were not regularly hunted. These, he surmised, has spiritual significance, and representational art was meant to resemble the hallucinations that arise when spiritual leaders enter into states of altered consciousness. The shamanism hypothesis has been enormously influential and is now well known outside cave-art circles. But, within cave art scholarship, it remains highly controversial. Rock art expert Paul Bahn has been a leading critic. He collaborated with neuroscientitist, Patricia Helvenston, to show that Lewis-Williams exaggerates the parallels between cave imagery and trance states. He also argues that the painters of these caves did not have hallucinogens, and that contemporary shamans are not known to produce rock art (for another critic, see Alice Kehoe). More concessive critics admit that some rock art may relate to shamanism but insist that this should not generalize. Robert Layton, for example, has distinguished shamanism from totemism in depictions of animals, and he distinguishes both of these from secularism. Then, using the distribution of animals in various regions where rock art has been found, he argues that different rock art serves different functions in different places (Layton's scheme appears above). He argues that secular uses of imagery are characterized by use of a wide range of creatures, in equal numbers, painted in any place suitable for habitation. The paintings of Lascaux are deep inside a cave and depict a small range of animals, suggesting a religious function. But other sites show the secular pattern. This brings us back to the Quinkan galleries. The pattern here is secular. As a visitor, I was expecting the art to be located in places that had overtly special significance. Places that were usual, hidden when foraging in the area, and perhaps hard to get to by foot. On the contrary, they appear on any wall that seems suitably flat and sheltered. The image above shows the setting, with one of the major galleries just visible on the far left. These walls also include images of many different kinds of animals. No creature is singled out as sacred. The Quinkans themselves are believed to be supernatural beings, but for these artists they may have seemed as real as kangaroos: they were believed to be residents of the forest, whose voices could be heard regularly, and whose mischief left regularly signs and introduced daily dangers. The geometrical patterns appear in more or less random locations. They do not look like hallucinogenic auras surrounding the creatures. There is little reason to believe that these images relate to trance experiences. This leaves us with an important reminder. Rock art is found all over the world, but each place in different. Each has its own style and its own function. One cannot use Lascaux (above) as a model for the world. Even within Australia, each region has its own character, and within a single site, there may be a history of different groups making work with unique significance. I mentioned stylistic variation in the Quinkan galleries, and the contrast with the graceful Kimberley figures. Kimberley also exhibits a second style, called Wandjina, that is quite different from both (below), and there is also the famous x-ray images in Kakadu National Park in Northern Australia (below). Such variation rock art provides a spectacular demonstration of the fact that human pictorial practices cannot be given a unified analysis. That is why it is risky to use the term "art" in this context. The functions that we associate with "art" may be extremely different from the functions of these paleolithic paintings. Some may have been used for rituals, some for instruction, and some may be ancient doodles. Were any just made for aesthetic admiration, like contemporary galleries? We will never know. Given this diversity, the mystery of why humans began painting becomes both harder and easier. It is easier, because painting is not a single cultural practice, but many. There is no single thing to be explained--no human universal. The puzzle becomes harder, however, because there are some striking similarities in rock art. If these works lack a common function, why do they arise in so many places and why do they share features in common, such as depictions of animals painted in profile. I've tried to sketch some answers, having to do with population growth and visual object recognition, but many puzzles remain. It is crucial that we don't let our search for the origin of painting lead us to ignore variation. Paleolithic rock art shows that humans faced similar challenges across the globe, and that our visual systems work in the same way. But we mustn't forget particularity. The figure from Split Rock (above), with outstretched arms and a gridded cloak, reminds us that each example of ancient art has its own character, its own meaning, its own charm.