Collapse

by Diamond, Jared

  They survived in ways that strike me as a mixture of ingenious, desperate, and pathetic. For the raw material of adzes, in place of stone, they turned to shells of giant clams. For awls to punch holes, they fell back on bird bones. For oven stones, they turned to limestone or coral or giant clamshell, all of which are inferior to basalt because they retain heat for less time, tend to crack after heating, and cannot be reused as often. They now made their fishhooks out of purse shell, which is much smaller than black-lipped pearl oyster shell, so that it yields only one hook per shell (instead of a dozen hooks from an oyster shell) and restricts the types of hooks that can be fashioned.

  Radiocarbon dates suggest that, struggling on in this way, Henderson’s population of originally a few dozen people survived for several generations, possibly a century or more, after all contact with Mangareva and Pitcairn was cut. But by A.D. 1606, the year of Henderson’s “discovery” by Europeans, when a boat from a passing Spanish ship landed on the island and saw no one, Henderson’s population had ceased to exist. Pitcairn’s own population had disappeared at least by 1790 (the year when the Bounty mutineers arrived to find the island uninhabited), and probably disappeared much earlier.

  Why did Henderson’s contact with the outside world come to a halt? That outcome stemmed from disastrous environmental changes on Mangareva and Pitcairn. All over Polynesia, human settlement on islands that had developed for millions of years in the absence of humans led to habitat damage and mass extinctions of plants and animals. Mangareva was especially susceptible to deforestation for most of the reasons that I identified for Easter Island in the preceding chapter: high latitude, low ash and dust fallout, and so on. Habitat damage was extreme in Mangareva’s hilly interior, most of which the islanders proceeded to deforest in order to plant their gardens. As a result, rain carried topsoil down the steep slopes, and the forest became replaced by a savannah of ferns, which were among the few plants able to grow on the now-denuded ground. That soil erosion in the hills removed much of the area formerly available on Mangareva for gardening and tree crops. Deforestation indirectly reduced yields from fishing as well, because no trees large enough to build canoes remained: when Europeans “discovered” Mangareva in 1797, the islanders had no canoes, only rafts.

  With too many people and too little food, Mangareva society slid into a nightmare of civil war and chronic hunger, whose consequences are recalled in detail by modern islanders. For protein, people turned to cannibalism, in the form not only of eating freshly dead people but also of digging up and eating buried corpses. Chronic fighting broke out over the precious remaining cultivable land; the winning side redistributed the land of the losers. Instead of an orderly political system based on hereditary chiefs, non-hereditary warriors took over. The thought of Lilliputian military dictatorships on eastern and western Mangareva, battling for control of an island only five miles long, could seem funny if it were not so tragic. All that political chaos alone would have made it difficult to muster the manpower and supplies necessary for oceangoing canoe travel, and to go off for a month and leave one’s garden undefended, even if trees for canoes themselves had not become unavailable. With the collapse of Mangareva at its hub, the whole East Polynesia trade network that had joined Mangareva to the Marquesas, Societies, Tuamotus, Pitcairn, and Henderson disintegrated, as documented by Weisler’s sourcing studies of basalt adzes.

  While much less is known about environmental changes on Pitcairn, limited archaeological excavations there by Weisler indicate massive deforestation and soil erosion on that island as well. Henderson itself also suffered environmental damage that reduced its human carrying capacity. Five out of its nine species of land birds (including all three large pigeons), and colonies of about six of its species of breeding seabirds, were exterminated. Those extinctions probably resulted from a combination of hunting for food, habitat destruction due to parts of the island being burned for gardens, and depredations of rats that arrived as stowaways in Polynesian canoes. Today, those rats continue to prey on chicks and adults of the remaining species of seabirds, which are unable to defend themselves because they evolved in the absence of rats. Archaeological evidence for gardening appears on Henderson only after those bird disappearances, suggesting that people were being forced into reliance on gardens by the dwindling of their original food sources. The disappearance of edible horn shells and decline in turban shells in later layers of archaeological sites on Henderson’s northeast coast also suggest the possibility of overexploitation of shellfish.

  Thus, environmental damage, leading to social and political chaos and to loss of timber for canoes, ended Southeast Polynesia’s interisland trade. That end of trade would have exacerbated problems for Mangarevans, now cut off from Pitcairn, Marquesas, and Societies sources of high-quality stone for making tools. For the inhabitants of Pitcairn and Henderson, the results were even worse: eventually, no one was left alive on those islands.

  Those disappearances of Pitcairn’s and Henderson’s populations must have resulted somehow from the severing of the Mangarevan umbilical cord. Life on Henderson, always difficult, would have become more so with the loss of all imported volcanic stone. Did everyone die simultaneously in a mass calamity, or did the populations gradually dwindle down to a single survivor, who lived on alone with his or her memories for many years? That actually happened to the Indian population of San Nicolas Island off Los Angeles, reduced finally to one woman who survived in complete isolation for 18 years. Did the last Henderson Islanders spend much time on the beaches, for generation after generation, staring out to sea in the hopes of sighting the canoes that had stopped coming, until even the memory of what a canoe looked like grew dim?

  While the details of how human life flickered out on Pitcairn and Henderson remain unknown, I can’t tear myself free of the mysterious drama. In my head, I run through alternative endings of the movie, guiding my speculation by what I know actually did happen to some other isolated societies. When people are trapped together with no possibility of emigration, enemies can no longer resolve tensions merely by moving apart. Those tensions may have exploded in mass murder, which later nearly did destroy the colony of Bounty mutineers on Pitcairn itself. Murder could also have been driven by food shortage and cannibalism, as happened to the Mangarevans, Easter Islanders, and—closer to home for Americans—the Donner Party in California. Perhaps people grown desperate turned to mass suicide, which was recently the choice of 39 members of the Heaven’s Gate cult near San Diego, California. Desperation might instead have led to insanity, the fate of some members of the Belgian Antarctic Expedition, whose ship was trapped by ice for over a year in 1898-1899. Still another catastrophic ending could have been starvation, the fate of Japan’s garrison stranded on Wake Island during World War II, and perhaps exacerbated by a drought, typhoon, tsunami, or other environmental disaster.

  Then my mind turns to gentler possible endings of the movie. After a few generations of isolation on Pitcairn or Henderson, everyone in their microsociety of a hundred or a few dozen people would have been everyone else’s cousin, and it would have become impossible to contract a marriage not in violation of incest taboos. Hence people may just have grown old together and stopped having children, as happened to California’s last surviving Yahi Indians, the famous Ishi and his three companions. If the small population did ignore incest taboos, the resulting inbreeding may have caused congenital physical anomalies to proliferate, as exemplified by deafness on Martha’s Vineyard Island off Massachusetts or on the remote Atlantic island of Tristan da Cunha.

  We may never know which way the movies of Pitcairn and Henderson actually ended. Regardless of the final details, though, the main outline of the story is clear. The populations of Mangareva, Pitcairn, and Henderson all inflicted heavy damage on their environments and destroyed many of the resources necessary for their own lives. Mangareva Islanders were numerous enough to survive, albeit under chronically terrifying conditions and with a drastically reduced standa
rd of living. But from the very beginning, even before the accumulation of environmental damage, the inhabitants of Pitcairn and Henderson had remained dependent on imports of agricultural products, technology, stone, oyster shell, and people from their mother population on Mangareva. With Mangareva’s decline and its inability to sustain exports, not even the most heroic efforts to adapt could save the last people alive on Pitcairn and Henderson. Lest those islands still seem to you too remote in space and time to be relevant to our modern societies, just think about the risks (as well as the benefits) of our increasing globalization and increasing worldwide economic interdependence. Many economically important but ecologically fragile areas (think of oil) already affect the rest of us, just as Mangareva affected Pitcairn and Henderson.

  CHAPTER 4

  The Ancient Ones: The Anasazi and Their Neighbors

  Desert farmers ■ Tree rings ■ Agricultural strategies ■ Chaco’s problems and packrats ■ Regional integration ■ Chaco’s decline and end ■ Chaco’s message ■

  Of the sites of societal collapses considered in this book, the most remote are Pitcairn and Henderson Islands discussed in the last chapter. At the opposite extreme, the ones closest to home for Americans are the Anasazi sites of Chaco Culture National Historical Park (Plates 9, 10) and Mesa Verde National Park, lying in the U.S. Southwest on New Mexico state highway 57 and near U.S. highway 666, respectively, less than 600 miles from my home in Los Angeles. Like the Maya cities that will be the subject of the next chapter, they and other ancient Native American ruins are popular tourist attractions that thousands of modern First World citizens visit each year. One of those former southwestern cultures, Mimbres, is also a favorite of art collectors because of its beautiful pottery decorated with geometrical patterns and realistic figures: a unique tradition created by a society numbering barely 4,000 people, and sustained at its peak for just a few generations before abruptly disappearing.

  I concede that U.S. southwestern societies operated on a much smaller scale than did Maya cities, with populations of thousands rather than millions. As a result, Maya cities are far more extensive in area, have more lavish monuments and art, were products of more steeply stratified societies headed by kings, and possessed writing. But the Anasazi did manage to construct in stone the largest and tallest buildings erected in North America until the Chicago steel girder skyscrapers of the 1880s. Even though the Anasazi lacked a writing system such as the one that allows us to date Maya inscriptions to the exact day, we shall see that many U.S. southwestern structures can still be dated to within a year, thereby enabling archaeologists to understand the societies’ history with much finer time resolution than is possible for Easter, Pitcairn, and Henderson Islands.

  In the U.S. Southwest we are dealing with not just a single culture and collapse, but with a whole series of them (map, p. 142). Southwestern cultures that underwent regional collapses, drastic reorganizations, or abandonments at different locations and different times include Mimbres around A.D. 1130; Chaco Canyon, North Black Mesa, and the Virgin Anasazi in the middle or late 12th century; around 1300, Mesa Verde and the Kayenta Anasazi; Mogollon around 1400; and possibly as late as the 15th century, Hohokam, well known for its elaborate system of irrigation agriculture. While all of those sharp transitions occurred before Columbus’s arrival in the New World in 1492, the Anasazi did not vanish as people: other southwestern Native American societies incorporating some of their descendants persist to this day, such as the Hopi and Zuni pueblos. What accounts for all those declines or abrupt changes in so many neighboring societies?

  Favorite single-factor explanations invoke environmental damage, drought, or warfare and cannibalism. Actually, the field of U.S. southwestern prehistory is a graveyard for single-factor explanations. Multiple factors have operated, but they all go back to the fundamental problem that the U.S. Southwest is a fragile and marginal environment for agriculture—as is also much of the world today. It has low and unpredictable rainfall, quickly exhausted soils, and very low rates of forest regrowth. Environmental problems, especially major droughts and episodes of streambed erosion, tend to recur at intervals much longer than a human lifetime or oral memory span. Given those severe difficulties, it’s impressive that Native Americans in the Southwest developed such complex farming societies as they did. Testimony to their success is that most of this area today supports a much sparser population growing their own food than it did in Anasazi times. It was a moving and unforgettable experience for me, while I was driving through areas of desert dotted with the remains of former Anasazi stone houses, dams, and irrigation systems, to see a now virtually empty landscape with just the occasional occupied house. The Anasazi collapse and other southwestern collapses offer us not only a gripping story but also an instructive one for the purposes of this book, illustrating well our themes of human environmental impact and climate change intersecting, environmental and population problems spilling over into warfare, the strengths but also the dangers of complex non-self-sufficient societies dependent on imports and exports, and societies collapsing swiftly after attaining peak population numbers and power.

  Our understanding of southwestern prehistory is detailed because of two advantages that archaeologists in this area enjoy. One is the packrat midden method that I’ll discuss below, which provides us with a virtual time capsule of the plants growing within a few dozen yards of a midden within a few decades of a calculated date. That advantage has allowed paleobotanists to reconstruct changes in local vegetation. The other advantage allows archaeologists to date building sites to the nearest year by the tree rings of the site’s wood construction beams, instead of having to rely on the radiocarbon method used by archaeologists elsewhere, with its inevitable errors of 50 to 100 years.

  The tree ring method depends on the fact that rainfall and temperature vary seasonally in the Southwest, so that tree growth rates also vary seasonally, as true at other sites in the temperate zones as well. Hence temperate zone trees lay down new wood in annual growth rings, unlike tropical rainforest trees whose growth is more nearly continuous. But the Southwest is better for tree ring studies than most other temperate zone sites, because the dry climate results in excellent preservation of wooden beams from trees felled over a thousand years ago.

  Here’s how tree ring dating, known to scientists as dendrochronology (from the Greek roots dendron = tree, and chronos = time), works. If you cut down a tree today, it’s straightforward to count the rings inwards, starting from the tree’s outside (corresponding to this year’s growth ring), and thereby to state that the 177th ring from the outermost one towards the center was laid down in the year 2005 minus 177, or 1828. But it’s less straightforward to attach a date to a particular ring in an ancient Anasazi wooden beam, because at first you don’t know in what year the beam was cut. However, the widths of tree growth rings vary from year to year, depending on rain or drought conditions in each year. Hence the sequence of rings in a tree cross-section is like a message in the Morse code formerly used for sending telegraph messages; dot-dot-dash-dot-dash in the Morse code, wide-wide-narrow-wide-narrow in a tree ring sequence. Actually, the ring sequence is even more diagnostic and richer in information than the Morse code, because trees actually contain rings spanning many different widths, rather than the Morse code’s choice between only a dot or a dash.

  Tree ring specialists (known as dendrochronologists) proceed by noting the sequence of wider and narrower rings in a tree cut down in a known recent year, and also noting the sequence in beams from trees cut down at various unknown times in the past. They then match up and align ring sequences with the same diagnostic wide/narrow patterns from different beams. For instance, suppose that this year (2005) you cut down a tree that proves to be 400 years old (400 rings), and that has an especially distinctive sequence of five wide rings, two narrow rings, and six wide rings for the 13 years from 1643 back to 1631. If you find that same distinctive sequence starting seven years from the outermost ring in an old
beam of unknown felling date with 332 rings, then you can conclude that the old beam came from a tree cut down in 1650 (seven years after 1643), and that the tree began to grow in the year 1318 (332 years before 1650). You then go on to align that beam, from the tree living between 1318 and 1650, with even older beams, and you similarly try to match up tree ring patterns and find a beam whose pattern shows that it comes from a tree that was cut down after 1318 but began growing before 1318, thereby extending your tree ring record farther back into the past. In that way, dendrochronologists have constructed tree ring records extending back for thousands of years in some parts of the world. Each such record is valid for a geographic area whose extent depends on local weather patterns, because weather and hence tree growth patterns vary with location. For instance, the basic tree ring chronology of the American Southwest applies (with some variation) to the area from northern Mexico to Wyoming.

  A bonus of dendrochronology is that the width and substructure of each ring reflect the amount of rain and the season at which the rain fell during that particular year. Thus, tree ring studies also allow one to reconstruct past climate; e.g., a series of wide rings means a wet period, and a series of narrow rings means a drought. Tree rings thereby provide southwestern archaeologists with uniquely exact dating and uniquely detailed year-to-year environmental information.

  The first humans to reach the Americas, living as hunter-gatherers, arrived in the U.S. Southwest by 11,000 B.C. but possibly earlier, as part of the colonization of the New World from Asia by peoples ancestral to modern Native Americans. Agriculture did not develop indigenously in the U.S. Southwest, because of a paucity of domesticable wild plant and animal species. Instead, it arrived from Mexico, where corn, squash, beans, and many other crops were domesticated—corn arriving by 2000 B.C., squash around 800 B.C., beans somewhat later, and cotton not until A.D. 400. People also kept domestic turkeys, about which there is some debate whether they were first domesticated in Mexico and spread to the Southwest, or vice versa, or whether they were domesticated independently in both areas. Originally, southwestern Native Americans just incorporated some agriculture as part of their hunter-gatherer lifestyle, as did the modern Apache in the 18th and 19th centuries: the Apache settled down to plant and harvest crops during the growing season, then moved around as hunter-gatherers during the rest of the year. By A.D. 1, some southwestern Native Americans had already taken up residence in villages and become primarily dependent on agriculture with ditch irrigation. Thereafter, their populations exploded in numbers and spread over the landscape until the retrenchments beginning around A.D. 1117.