Collapse: How Societies Choose to Fail or Succeed Read online

  Despite the development of these two environmental problems that reduced crop production and virtually eliminated timber supplies within Chaco Canyon itself, or because of the solutions that the Anasazi found to these problems, the canyon’s population continued to increase, particularly during a big spurt of construction that began in A.D. 1029. Such spurts went on especially during wet decades, when more rain meant more food, more people, and more need for buildings. A dense population is attested not only by the famous Great Houses (such as Pueblo Bonito) spaced about a mile apart on the north side of Chaco Canyon, but also by holes drilled into the northern cliff face to support roof beams, indicating a continuous line of residences at the base of the cliffs between the Great Houses, and by the remains of hundreds of small settlements on the south side of the canyon. The size of the canyon’s total population is unknown and much debated. Many archaeologists think that it was less than 5,000, and that those enormous buildings had few permanent occupants except priests and were just visited seasonally by peasants at the time of rituals. Other archaeologists note that Pueblo Bonito, which is just one of the large houses at Chaco Canyon, by itself was a building of 600 rooms, and that all those post holes suggest dwellings for much of the length of the canyon, thus implying a population much greater than 5,000. Such debates about estimated population sizes arise frequently in archaeology, as discussed for Easter Island and the Maya in other chapters of this book.

  Whatever the number, this dense population could no longer support itself but was subsidized by outlying satellite settlements constructed in similar architectural styles and joined to Chaco Canyon by a radiating regional network of hundreds of miles of roads that are still visible today. Those outliers had dams to catch rain, which fell unpredictably and very patchily: a thunderstorm might produce abundant rain in one desert wash and no rain in another wash just a mile away. The dams meant that when a particular wash was fortunate enough to receive a rainstorm, much of the rainwater became stored behind the dam, and people living there could quickly plant crops, irrigate, and grow a huge surplus of food at that wash in that year. The surplus could then feed people living at all the other outliers that didn’t happen to receive rain then.

  Chaco Canyon became a black hole into which goods were imported but from which nothing tangible was exported. Into Chaco Canyon came: those tens of thousands of big trees for construction; pottery (all late-period pottery in Chaco Canyon was imported, probably because exhaustion of local firewood supplies precluded firing pots within the canyon itself); stone of good quality for making stone tools; turquoise for making ornaments, from other areas of New Mexico; and macaws, shell jewelry, and copper bells from the Hohokam and from Mexico, as luxury goods. Even food had to be imported, as shown by a recent study tracing the origins of corncobs excavated from Pueblo Bonito by means of the same strontium isotope method used by Nathan English to trace the origins of Pueblo Bonito’s wooden beams. It turns out that, already in the 9th century, corn was being imported from the Chuska Mountains 50 miles to the west (also one of the two sources of roof beams), while a corncob from the last years of Pueblo Bonito in the 12th century came from the San Juan River system 60 miles to the north.

  Chaco society turned into a mini-empire, divided between a well-fed elite living in luxury and a less well-fed peasantry doing the work and raising the food. The road system and the regional extent of standardized architecture testify to the large size of the area over which the economy and culture of Chaco and its outliers were regionally integrated. Styles of buildings indicate a three-step pecking order: the largest buildings, so-called Great Houses, in Chaco Canyon itself (residences of the governing chiefs?); outlier Great Houses beyond the canyon (“provincial capitals” of junior chiefs?); and small homesteads of just a few rooms (peasants’ houses?).

  Compared to smaller buildings, the Great Houses were distinguished by finer construction with veneer masonry, large structures called Great Kivas used for religious rituals (similar to ones still used today in modern Pueblos), and a higher ratio of storage space to total space. Great Houses far exceeded homesteads in their contents of imported luxury goods, such as the turquoise, macaws, shell jewelry, and copper bells mentioned above, plus imported Mimbres and Hohokam pottery. The highest concentration of luxury items located to date comes from Pueblo Bonito’s room number 33, which held burials of 14 individuals accompanied by 56,000 pieces of turquoise and thousands of shell decorations, including one necklace of 2,000 turquoise beads and a basket covered with a turquoise mosaic and filled with turquoise and shell beads. As for evidence that the chiefs ate better than did the peasants, garbage excavated near Great Houses contained a higher proportion of deer and antelope bones than did garbage from homesteads, with the result that human burials indicate taller, better-nourished, less anemic people and lower infant mortality at Great Houses.

  Why would outlying settlements have supported the Chaco center, dutifully delivering timber, pottery, stone, turquoise, and food without receiving anything material in return? The answer is probably the same as the reason why outlying areas of Italy and Britain today support our cities such as Rome and London, which also produce no timber or food but serve as political and religious centers. Like the modern Italians and British, Chacoans were now irreversibly committed to living in a complex, interdependent society. They could no longer revert to their original condition of self-supporting mobile little groups, because the trees in the canyon were gone, the arroyos were cut below field levels, and the growing population had filled up the region and left no unoccupied suitable areas to which to move. When the pinyon and juniper trees were cut down, the nutrients in the litter underneath the trees were flushed out. Today, more than 800 years later, there is still no pinyon/juniper woodland growing anywhere near the packrat middens containing twigs of the woodland that had grown there before A.D. 1000. Food remains in rubbish at archaeological sites attest to the growing problems of the canyon’s inhabitants in nourishing themselves: deer declined in their diets, to be replaced by smaller game, especially rabbits and mice. Remains of complete headless mice in human coprolites (preserved dry feces) suggest that people were catching mice in the fields, beheading them, and popping them in whole.

  The last identified construction at Pueblo Bonito, dating from the decade after 1110, was from a wall of rooms enclosing the south side of the plaza, which had formerly been open to the outside. That suggests strife: people were evidently now visiting Pueblo Bonito not just to participate in its religious ceremonies and to receive orders, but also to make trouble. The last tree-ring-dated roof beam at Pueblo Bonito and at the nearby Great House of Chetro Ketl was cut in A.D. 1117, and the last beam anywhere in Chaco Canyon in A.D. 1170. Other Anasazi sites show more abundant evidence of strife, including signs of cannibalism, plus Kayenta Anasazi settlements at the tops of steep cliffs far from fields and water and understandable only as easily defended locations. At those southwestern sites that outlasted Chaco and survived until after A.D. 1250, warfare evidently became intense, as reflected in a proliferation of defensive walls and moats and towers, clustering of scattered small hamlets into larger hilltop fortresses, apparently deliberately burned villages containing unburied bodies, skulls with cut marks caused by scalping, and skeletons with arrowheads inside the body cavity. That explosion of environmental and population problems in the form of civil unrest and warfare is a frequent theme in this book, both for past societies (the Easter Islanders, Mangarevans, Maya, and Tikopians) and for modern societies (Rwanda, Haiti, and others).

  The signs of warfare-related cannibalism among the Anasazi are an interesting story in themselves. While everyone acknowledges that cannibalism may be practiced in emergencies by desperate people, such as the Donner Party trapped by snow at Donner Pass en route to California in the winter of 1846-47, or by starving Russians during the siege of Leningrad during World War II, the existence of non-emergency cannibalism is controversial. In fact, it was reported in hundreds of non-Europea
n societies at the times when they were first contacted by Europeans within recent centuries. The practice took two forms: eating either the bodies of enemies killed in war, or else eating one’s own relatives who had died of natural causes. New Guineans with whom I have worked over the past 40 years have matter-of-factly described their cannibalistic practices, have expressed disgust at our own Western burial customs of burying relatives without doing them the honor of eating them, and one of my best New Guinean workers quit his job with me in 1965 in order to partake in the consumption of his recently deceased prospective son-in-law. There have also been many archaeological finds of ancient human bones in contexts suggestive of cannibalism.

  Nevertheless, many or most European and American anthropologists, brought up to regard cannibalism with horror in their own societies, are also horrified at the thought of it being practiced by peoples that they admire and study, and so they deny its occurrence and consider claims of it as racist slander. They dismiss all the descriptions of cannibalism by non-European peoples themselves or by early European explorers as unreliable hearsay, and they would evidently be convinced only by a videotape taken by a government official or, most convincing of all, by an anthropologist. However, no such tape exists, for the obvious reason that the first Europeans to encounter people reported to be cannibals routinely expressed their disgust at the practice and threatened its practitioners with arrest.

  Such objections have created controversy around the many reports of human remains, with evidence consistent with cannibalism, found at Anasazi sites. The strongest evidence comes from an Anasazi site at which a house and its contents had been smashed, and the scattered bones of seven people were left inside the house, consistent with their having been killed in a war raid rather than properly buried. Some of the bones had been cracked in the same way that bones of animals consumed for food were cracked to extract the marrow. Other bones showed smooth ends, a hallmark of animal bones boiled in pots, but not of ones not boiled in pots. Broken pots themselves from that Anasazi site had residues of the human muscle protein myoglobin on the pots’ inside, consistent with human flesh having been cooked in the pots. But skeptics might still object that boiling human meat in pots, and cracking open human bones, does not prove that other humans actually consumed the meat of the former owners of those bones (though why else would they go to all that trouble of boiling and cracking bones to be left scattered on the floor?). The most direct sign of cannibalism at the site is that dried human feces, found in the house’s hearth and still well preserved after nearly a thousand years in that dry climate, proved to contain human muscle protein, which is absent from normal human feces, even from the feces of people with injured and bleeding intestines. This makes it probable that whoever attacked that site, killed the inhabitants, cracked open their bones, boiled their flesh in pots, scattered the bones, and relieved himself or herself by depositing feces in that hearth had actually consumed the flesh of his or her victims.

  The final blow for Chacoans was a drought that tree rings show to have begun around A.D. 1130. There had been similar droughts previously, around A.D. 1090 and 1040, but the difference this time was that Chaco Canyon now held more people, more dependent on outlying settlements, and with no land left unoccupied. A drought would have caused the groundwater table to drop below the level where it could be tapped by plant roots and could support agriculture; a drought would also make rainfall-supported dryland agriculture and irrigation agriculture impossible. A drought that lasted more than three years would have been fatal, because modern Puebloans can store corn for only two or three years, after which it is too rotten or infested to eat. Probably the outlying settlements that had formerly supplied the Chaco political and religious centers with food lost faith in the Chacoan priests whose prayers for rain remained unanswered, and they refused to make more food deliveries. A model for the end of Anasazi settlement at Chaco Canyon, which Europeans did not observe, is what happened in the Pueblo Indian revolt of 1680 against the Spaniards, a revolt that Europeans did observe. As in Chaco Anasazi centers, the Spaniards had extracted food from local farmers by taxing them, and those food taxes were tolerated until a drought left the farmers themselves short of food, provoking them to revolt.

  Some time between A.D. 1150 and 1200, Chaco Canyon was virtually abandoned and remained largely empty until Navajo sheepherders reoccupied it 600 years later. Because the Navajo did not know who had built the great ruins that they found there, they referred to those vanished former inhabitants as the Anasazi, meaning “the Ancient Ones.” What actually happened to the thousands of Chacoan inhabitants? By analogy with historically witnessed abandonments of other pueblos during a drought in the 1670s, probably many people starved to death, some people killed each other, and the survivors fled to other settled areas in the Southwest. It must have been a planned evacuation, because most rooms at Anasazi sites lack the pottery and other useful objects that people would be expected to take with them in a planned evacuation, in contrast to the pottery still in the rooms of the above-mentioned site whose unfortunate occupants were killed and eaten. The settlements to which Chaco survivors managed to flee include some pueblos in the area of the modern Zuni pueblos, where rooms built in a style similar to Chaco Canyon houses and containing Chaco styles of pottery have been found at dates around the time of Chaco’s abandonment.

  Jeff Dean and his colleagues Rob Axtell, Josh Epstein, George Gumerman, Steve McCarroll, Miles Parker, and Alan Swedlund have carried out an especially detailed reconstruction of what happened to a group of about a thousand Kayenta Anasazi in Long House Valley in northeastern Arizona. They calculated the valley’s actual population at various times from A.D. 800 to 1350, based on numbers of house sites containing pottery that changed in style with time, thereby permitting dating of the house sites. They also calculated the valley’s annual corn harvests as a function of time, from annual tree rings that provide a measure of rainfall, and from soil studies that provide information about the rise and fall of groundwater levels. It turned out that the rises and falls of the actual population after A.D. 800 closely mirrored the rises and falls of calculated annual corn harvests, except that the Anasazi completely abandoned the valley by A.D. 1300, at a time when some reduced corn harvests sufficient to support one-third of the valley’s peak population (400 out of the peak of 1,070 people) could still have been extracted.

  Why did those last 400 Kayenta Anasazi of Long House Valley not remain when most of their relatives were leaving? Perhaps the valley in A.D. 1300 had deteriorated for human occupation in other ways besides its reduced agricultural potential calculated in the authors’ model. For instance, perhaps soil fertility had been exhausted, or else the former forests may have been felled, leaving no nearby timber for buildings and firewood, as we know to have been the case in Chaco Canyon. Alternatively, perhaps the explanation was that complex human societies require a certain minimum population size to maintain institutions that its citizens consider to be essential. How many New Yorkers would choose to remain in New York City if two-thirds of their family and friends had just starved to death there or fled, if the subway trains and taxis were no longer running, and if offices and stores had closed?

  Along with those Chaco Canyon Anasazi and Long House Valley Anasazi whose fates we have followed, I mentioned at the start of this chapter that many other southwestern societies—the Mimbres, Mesa Verdeans, Hohokam, Mogollon, and others—also underwent collapses, reorganizations, or abandonments at various times within the period A.D. 1100-1500. It turns out that quite a few different environmental problems and cultural responses contributed to these collapses and transitions, and that different factors operated in different areas. For example, deforestation was a problem for the Anasazi, who required trees to supply the roof beams of their houses, but it wasn’t as much of a problem for the Hohokam, who did not use beams in their houses. Salinization resulting from irrigation agriculture hurt the Hohokam, who had to irrigate their fields, but not the Mesa Verde
ans, who did not have to irrigate. Cold affected the Mogollon and Mesa Verdeans, living at high altitudes and at temperatures somewhat marginal for agriculture. Other southwestern peoples were done in by dropping water tables (e.g., the Anasazi) or by soil nutrient exhaustion (possibly the Mogollon). Arroyo cutting was a problem for the Chaco Anasazi, but not for the Mesa Verdeans.

  Despite these varying proximate causes of abandonments, all were ultimately due to the same fundamental challenge: people living in fragile and difficult environments, adopting solutions that were brilliantly successful and understandable “in the short run,” but that failed or else created fatal problems in the long run, when people became confronted with external environmental changes or human-caused environmental changes that societies without written histories and without archaeologists could not have anticipated. I put “in the short run” in quotation marks, because the Anasazi did survive in Chaco Canyon for about 600 years, considerably longer than the duration of European occupation anywhere in the New World since Columbus’s arrival in A.D. 1492. During their existence, those various southwestern Native Americans experimented with half-a-dozen alternative types of economies (pp. 140-143). It took many centuries to discover that, among those economies, only the Pueblo economy was sustainable “in the long run,” i.e. for at least a thousand years. That should make us modern Americans hesitate to be too confident yet about the sustainability of our First World economy, especially when we reflect how quickly Chaco society collapsed after its peak in the decade A.D. 1110-1120, and how implausible the risk of collapse would have seemed to Chacoans of that decade.