Critical perspectives on historical collapse

Karl W. Butzera,1 and Georgina H. Endfieldb aDepartment of Geography and the Environment, University of Texas, Austin, TX 78712; and bSchool of Geography, University of Nottingham, Nottingham NG7 2RD, United Kingdom

Historical collapse of ancient states or has raised new awareness about its possible relevance to current issues of sustainability, in the context of global change. This Special Feature examines 12 case studies of under stress, of which seven suffered severe transformation. Outcomes were complex and unpredictable. Five others overcame breakdown through environmental, political, or socio-cultural resilience, which deserves as much attention as the identification of stressors. Response to environmental crises of the last millennium varied greatly according to place and time but drew from traditional knowledge to evaluate new information or experiment with increasing flexibility, even if modernization or intensification were decentralized and protracted. Longer-term diachronic experience offers insight into how societies have dealt with acute stress, a more instructive perspective for the future than is offered by apocalyptic scenarios. | social-ecological resilience | multicausality

Current Visions of Collapse interaction. Together, such programs have tific advances in simulating the physical he breakdown of historical states fostered a series of research themes and outcomes of climatic change seem to be or societies (1) has drawn atten- concepts, relevant for and, in many cases more advanced than the modeling of social tion since the 18th century, with addressing, . response to that change (18, 19). This may T fl shifting explanations that reflected These themes and concepts include re ect the profound of social prevailing intellectual climates (2). An complexity and network theory, historicity analysis (20, 21) and the lack of unanimity early, macrohistorical mode gave way to and legacies, tipping elements and points, among the physical and social sciences or biological analogs or the overriding signif- and associated path-dependent relation- humanities in regard to concepts, assump- icance of technological progress. However, ships of coupled systems. Through paleo- tions, legacies, validity, and contingencies. two world wars of unspeakable brutality environmental and historical research, These issues pose difficult problems for reawakened millennialist anxieties, more various international science programs model designs to simulate coupled systems. recently compounded by growing fears of seek to identify the dynamics that stimulate exponential population growth, environ- societal adaptations to such elements, so as Social Science Agenda for Collapse mental deterioration, and global climatic to provide insight into the resilience of Without downplaying the importance of change. In this spirit of crisis, some authors past societies that arguably may be of use hard science and modeling in studies of have searched for alarming, futuristic sce- to contemporary and future societies. collapse, it can be argued that socio-cul- narios in historical digests. However, an The challenge remains to develop an tural and political processes need greater increasing number of scientists has begun outline for comprehensive, integrated attention. We therefore suggest a broader, to counter this sometimes chaotic dis- models that convincingly explicate past integrative definition: Societal collapse course with sophistication, to suggest more socio-ecological interactions, or that cap- represents transformation at a large social measured estimates for change or to steer ture the broader, dynamic principles cross- or spatial scale, with long-term impact on attention to the desirability of remedial cutting human–environmental systems combinations of interdependent variables: action. Diverse efforts are underway to while also accounting for the finer- (i) environmental change and resilience; model social response and offer simulated resolution evidence and complex outcomes (ii) demography or settlement; (iii) socio- predictions of short- or long-range envi- that they entail. Unfortunately there are economic patterns; (iv) political or socie- ronmental change. insufficient empirical, rather than simu- tal structures; and (v) ideology or cul- In particular, concern with climatic lated, data on the nature of societal re- tural memory. change, global environmental change, and sponse to cross-disciplinary inputs, triggers, These are intrinsic properties rather sustainability has stimulated the formation or tipping points. The public is confronted than a shopping list. They encompass en- of formal or informal programs and by metanarratives of global change or ergetic structures and flows, but equally so working groups to address environment– by semipopular works that suggest over- have cultural and psychological dimensions relationships and their implications simplified causal correlations. Such hy- that are grounded in human perceptions, for sustainability. These include Inter- potheses can readily be misunderstood as values, and solidarity. Temporal parame- national Council of Science-sponsored facts. Historical examples should instead ters are relative, because the interactive efforts such as Past Global Environmental be carefully selected to study the societal cluster of processes waxes and wanes in Change, the Integrated Land Ecosystem– implications of predicted, future environ- shifting constellations as a transformation Atmosphere Processes Study, and a newly mental scenarios (13). begins, climaxes, and concludes, with or minted Program on Ecosystem Change Other difficulties arise when normative without obvious discontinuities. Such a and Society. environmental systems are coupled with definition of collapse calls for a genuinely Related groups and programs focus on human systems that emphasize information, interdisciplinary field of analysis, one that land-use change (3–5), resilience (6–8), technology, and social organization. Human pays explicit attention to social and vulnerability (9, 10), and sustainability groups and individuals introduce cognition, (11). Apart from the underlying need for information, and communication as pow- institutional funding, such organizational erful variables in dealing with values, atti- Author contributions: K.W.B. and G.H.E. designed research; strategies serve to draw attention to a new tudes, and decision making (14–17), so that K.W.B. performed research; and K.W.B. wrote the paper. generation of directed, collaborative re- , perception, and behavior condition The authors declare no conflict of interest. search (12) and help sketch alternative how societies will interact with their envi- 1To whom correspondence should be addressed. E-mail: thematic agendas for cross-disciplinary ronments or define their priorities. Scien- [email protected].

3628–3631 | PNAS | March 6, 2012 | vol. 109 | no. 10 www.pnas.org/cgi/doi/10.1073/pnas.1114772109 PCA ETR:INTRODUCTION FEATURE: SPECIAL

humanistic issues, which should be more disasters, or feedbacks reflecting adjust- with multiple lines of convergent evidence than an afterthought. ments to land access and land use. and social science sophistication, a possi- In pursuing such a line of research, an- Such shifts or discontinuities in the ble diagnosis of protohistoric “collapse” alytical case studies are particularly pro- prehistoric record may well have been must rely on a fragmentary data set. This ductive. If based first and foremost in traumatic or even catastrophic at some explains the significance of the insider empirically grounded data, from specific scale, but the archaeological record often perspective for later societies that do have areas, case studies can be examined in- lacks the necessary sensitivity to detect this, a complex written record (37), as was the ductively and deductively. For example, the even when regional depopulation is ap- case in Egypt, Colonial Mexico, or Cyprus. 1930s Dust Bowl phenomenon on the parent. That indeterminate picture im- Great Plains has been reexamined many proves when the archaeological record Dating, Timing, and Correlation times (22–26), with an explicit historical acquires a much larger palette of criteria, Reliable time frames are as much a prob- model by Robert Kates, to arrive at in- such as socially differentiated architectural lem in identifying collapse as are records creasingly closer approximations of per- sites, suggestive of more complex and that include direct narratives (38, 39). In ceptions, issues, and processes. Renewed stratified communities, especially if there is a specific case study, the criteria to date attention to good heuristic exemplars is evidence of concentrated craft production, key information on environmental vari- integral to scientific investigation, and in exploitation of mining resources, hints or ability, depopulation, political simplifica- the case of historical collapse also provides better of exchange networks, and nested tion, or social resilience may simply be occasion to examine how societies dealt hierarchies of larger and smaller sites with inadequate to integrate or to correlate with crises to avoid breakdown. possible economic differentiation. Enter with other sites or outlying regions. — In addressing the issues and nature of the concept of the city highlighted by The collapse of the in historical collapse, the articles of this special economic, political, or religious or of Old Kingdom Egypt Special Feature rely heavily on case studies prestige or control (27). illustrates the problem. Accurate calibra- that specifically draw from archaeology, For urban societies there may also be tion of Accelerator Mass Spectrometry , economics, and geography, documentary sources that help identify (AMS) dates for the mid to late third as well as the geological and biological collapse of a group of towns or an archaic millennium B.C.E. is tenuous, given the sciences. However, the cross-disciplinary state. Urban centers in the Near East reach irregularities of the calibration curve (40). fields of interaction are much broader and back into prehistoric times in the case of Attempted calibrations may have a range proto-literate, high or “civili- directly intersect with overarching concepts ” of some 300 y, with multiple intercepts. On such as environmental , land-use zations. Investigation can move beyond the other hand, chronological recon- change, institutional structures, resilience, archaeological survey to unravel the suc- structions from regnal years are beset by cessive horizons of settlement mounds, and sustainability. A consistent component problems of incomplete, incorrect, or il- with the preferred assistance of geo- is scale, in both space and time. legible records or elite accounts for in- archaeological research (32). House dividual reigns, in addition to periods with floors, former streets, and community or Importance of Direct, “Insider” an uncertain number of short-term rulers. monumental structures can be studied at Information No generally accepted chronology has the microlevel to identify different kinds Over the span of the Holocene, telling been established for Mesopotamia of sediments that accumulate during set- evidence of social–ecological change is before ≈2100 B.C.E. For example, within tlement growth or decline, until the time commonly provided by archeological sur- the Syrian site of Tell Brak a very large of abandonment or deliberate destruction vey and geoarcheological evidence (27) for — (14, 33, 34). On a wider scale, transport of body of isotopic dates for three distinct settlement discontinuities. The number, archaeological levels that represent suc- urban sediments to adjacent streams may — density, or size of settlements can decline allow testing of relationships between cessive ruling administrations are in- or shift to new concentrations. Archeo- occupation and regional degra- distinguishable (41). In Egypt, the Old logical inventories may change, or whole Kingdom dynasties seem to be reasonably – dation or environmental change. Ex- landscapes may be abandoned (28 30). planations for discontinuities or collapse well dated, but the subsequent First In- When resettlement takes place, new ar- become more tangible. termediate Period is not (42). As a result it chaeological components often dominate, Although some written sources become is uncertain which Egyptian dynasties or and few older loci may be reoccupied. available for parts of the Near East toward rulers correlate with which Akkadian Settlement shifts or discontinuities are 3000 B.C.E., they initially are very frag- kings, let alone how either region in- common enough in the Neolithic, Bronze mentary or thematically limited. In con- terdigitated with events in the Aegean Age, or Iron Age records of Europe (31) junction with later recording of oral world or Indus Valley. The ancient capital and typically take place over centennial traditions, they offer a skeletal framework of Akkad has not yet been found and is time scales. Change may represent partial of changing dynasties but, until late in probably dispersed deep in alluvial de- or complete archeological succession, with the third millennium, little by way of ex- posits of the Euphrates River, whereas some shifts rapid or even catastrophic planation for or perceptions of change events referred to in the chronicles can be and others marked by a measure of evo- (35, 36). For example, the demise of the difficult to relate to actual places, let alone lutionary continuity. Human impact may Akkadian Empire is incompletely docu- scientifically excavated sites. have effected environmental modification, mented by coeval sources. Instead, It is a stretch to correlate the reputed or biotic evidence may record a shift to the changes have to be inferred from younger collapse of several Near Eastern civi- exploitation of new habitats or resources, texts or chronicles that were ideologically lizations (supposedly around 2200 B.C.E.) accompanied by changing adaptations or tailored with respect to an idealized role of across western Asia and beyond when in more complex patterns of resilience, in the “good ruler.” Although the conven- fact major events are imprecisely dated line with new, inferred vulnerabilities. tional canons of Near East historiography within 2 or more centuries. Such temporal External parameters can also shift, as are complemented by phenomenally rich uncertainty also does not warrant the use a result of climatic anomalies, invasion, or archaeological and artistic records, initial of proxy climate records from distant epidemic disease—beyond the ebb and reconstructions of administrative struc- oceans or continents to explain socio- flow of endogenic change, and driven by tures and social changes depend to some historical processes proceeding apace in new opportunities or priorities, famine degree on anthropological models. Even Mesopotamia or Egypt.

Butzer and Endfield PNAS | March 6, 2012 | vol. 109 | no. 10 | 3629 Implementing a Social Science Agenda the abrupt, cold-climate oscillation at the ers from Iceland, by Streeter et al. (49), The research articles of this Special Fea- end of the Pleistocene (the Younger shows that plague-related depopulation on ture constitute an alternative framework to Dryas, ≈12,900–11,600 cal. B.P.) delayed Iceland was followed by reduced pastoral a metanarrative-driven understanding of the transition to food production in the impact on the soil balance, which eventu- historical human–environment relation- open woodlands of the southern Levant. ally may have improved environmental ships and collapse. These empirically In response, resilient foraging groups re- resilience in the face of the LIA. The grounded articles represent four con- organized to implement earlier, successful North Atlantic articles illustrate the ef- tinents and come from many distinct en- strategies, exploiting a broader spectrum fectiveness of comparative analyses of fi vironments, addressing a wide range of of resources (identi ed by plant and large datasets from an expanded range of global variability and resilience. They span animal residues), in combination with variables. They go well beyond the singular different time ranges, so as to offer greater mobility. Later in the cycle, emphasis of abrupt climatic change, by a spectrum of contextual factors and a greater investment in prime, high-yield teasing out the role of systemic resilience issues, with cumulative contributions to foods narrowed the resource spectrum and and niche specialization. a repertoire of ideas. decreased mobility. However, hunting Since 1878 British colonial and Cypriot The opening Perspective [including pressures eventually reduced the avail- postcolonial enacted envi- supporting information (SI), available ability of large game, favoring a belated ronmental policies based on the premise of online] by Butzer (2) argues that a plausi- shift to food production. Multidisciplinary ongoing and degradation of ble cause for sociopolitical breakdown archaeology can indeed develop useful a fragile Mediterranean ecosystem (i.e., models for prehistoric times. coupled system degradation). According to should be grounded in direct, insider in- ≈ – formation, drawn from internal histories. The Maya Collapse ( 750 900 C.E.) is Harris (50), this narrative is based on in- Case studies from Old and New Kingdom controversial, with explanations ranging accurate assumptions and interpretations Egypt record how economic decline, cor- from long-distance teleconnections to ar- that oversimplify the goals of , guments centered on different environ- ruption, insecurity, or war precondition fail to recognize the difference between mental responses in various regional, a state for concatenations of declining degradation and change, and scapegoat ecological contexts, originally including productivity, recession, and possible cli- the pastoral lifestyle and its presumed tropical forests and woodland. As noted by matic perturbations. With low resilience, environmental impact. Archival and field Luzzadder-Beach et al. (46), this process cascading devolutionary feedbacks can research instead underscore the environ- spanned time and distance. Complex then destabilize a system through famine, mental and societal resilience of “de- proxy data indicate several drier periods, internal conflict, and political simplifica- graded” areas and the effects of changing linked to major transitions in human ad- tion, eventually leading to subsistence cri- economic, political, and social contexts. aptation or response. However, during the This case study of natural resource man- ses, breakdown of the social order, and Terminal Classic, collapse was pervasive, agement and environmental resilience civil wars. However, with high societal re- even in perennial wetlands that should (see ref. 43) illustrates that what the co- silience, buffering feedbacks may instead have been less affected by drought. Dun- lonial government viewed as unsustainable lead to the emergence of military leaders, ning et al. (47) address the different rates fi was, in all probability, sustainable. support by new elites, a reaf rmation of of Maya population recovery from such “ ” The Special Feature concludes with the cosmic order, and ideological shifts recurrent droughts and environmental Endfield’s (51) archivally based study of that allow reconstitution of the state. In degradation. In some areas long-term en- resilience and adaptive response in co- effect, multiple interactive thresholds led vironmental changes required develop- “ ” lonial Mexico. She marshals archival doc- to different regime shifts (see ref. 43) in ment of new adaptations, whereas Old vs. New Kingdom Egypt. elsewhere cultural factors may have de- umentation to argue that although much Expansion of these case studies to in- layed repopulation, including the re- of human history can be viewed through ’ fl clude Islamic Mesopotamia, Egypt s consecration of abandoned lands (kax) an ontogenetic lens of emergence, ores- Fayum Oasis, and Axum (Ethiopia) (ref. 2 before reoccupation as cultivated space cence, and decline, intersecting with ex- and its SI) adds experience on irrigation treme events or natural disasters, to do so (kol). As in the Levant, cyclic growth and fi failure or demographic decline in the wake decline involved both environmental and would obscure the speci c socio-eco- of ethnic change, climatic anomalies, or cultural resilience, but given its diversified nomic, cultural, political, and background environmental degradation. Although the criteria, the Maya record is more complex. environmental contexts that helped shape many interlinked inputs, triggers, and Two other articles offer insights on re- them. The interactions between the fl feedbacks can precondition or lead to silience and social transformations in the environment and society in uence how breakdown, environmental change was far North Atlantic realm, as based on regional livelihoods may be vulnerable to subordinate to internally driven processes, survey, excavation, and a battery of ana- disruption. However, her detailed case rather than the primary cause of de- lytical techniques, applied to multiscale analyses also inform on the degree to volution. Cognition, values, and priorities comparative analyses of examples from which different societies and groups can are critical, so that the conceptual model Medieval Iceland and . Dug- develop institutions and cultural coping proposed (figure 1 in ref. 2) not only em- more et al. (48) assess differences of social strategies to deal with environmental phasizes environmental resilience but also transformation, sustainable practice, envi- changes at different scales, to show that the role of leadership, elites, and ideology ronmental change, isolation, mobility, vulnerability to change can lead to an im- for either breakdown or reconstitution. and choices about subsistence and social proved understanding of risk (10). Envi- These historical examples of cyclical col- organization. Under what circumstances ronmental crises could therefore challenge lapse were not “abrupt” but mainly played can population levels not be maintained? but also improve societal resilience, in- out over centennial rather than decennial How did divergent adaptations benefit creasing opportunities for learning and time scales. Iceland and the Greenland colonies with innovation, to broaden the repertoire of In a strictly archaeological setting, the onset of the Little (LIA)? This adaptive responses (17, 52). Collapse is Rosen and Rivera-Collazo (44), with the informs on the consequences of locational not an inevitable result of transformations, help of good paleoecological data, show choices for settlements, with respect to although the transformations themselves the utility of adaptive cycles from resil- subsistence and external connectivity. offer opportunities to examine the com- ience theory (45). The authors suggest that High-resolution correlation of tephra lay- plex structure of social interactions.

3630 | www.pnas.org/cgi/doi/10.1073/pnas.1114772109 Butzer and Endfield Overall, the case studies present six through a web of complex social responses. how societies have avoided collapse by examples of collapse in response to mainly Rather than assemble an anthology of revitalizing a common will to overcome societal, interacting factors (cascading historical collapse, the Special Feature adversity, drawing from both old experi- feedbacks), in which climatic or ecological highlights multicausality, resilience, un- ence and new information to revise or change was no more than a coagency, with predictability, and how societies cope develop collective strategies for survival. a single case of significant degradation with crises. fi Voluntary transformation can be painful, (Axum). Some ve examples without col- Societal collapse raises productive but it does offer hope for reconstitution lapse illustrate different facets of envi- questions about diachronic coupled sys- and recovery. The case studies developed ronmental or societal resilience. This set of tems. Such issues would include the fol- supporting articles cuts across ecological lowing: why collapse is important, how in this Special Feature suggest that optimal variability and time to articulate comple- common it is in the historical record, solutions ultimately are cognitive and col- mentary insights about multicausality, whether it is inevitably linked to environ- laborative. However, solutions to acute rather than single-factor environmental mental disasters, or whether it offers crises of sustainability cannot be devised or explanations. The environment is of course precedents to correct contemporary out- implemented if remedial response is critically important for sustainability, comes or devise solutions for the future. modeled with stereotypic assumptions but such interrelationships are filtered Perhaps the most trenchant would be about human behavior.

1. Tainter JA (2000) Problem solving: Complexity, history, 19. Wainwright J (2010) , capitalism, and Sasson JM (Simon & Schuster Macmillan, New York), II, sustainability. Popul Environ 22:3–41. the challenges of transdisciplinarity. Ann Assoc Am pp 831–841. 2. Butzer KW (2012) Collapse, environment, and society. Geogr 100:983–991. 37. Butzer KW (2005) Environmental history in the Proc Natl Acad Sci USA 109:3632–3639. 20. Crooks A, Castle C, Batty M (2008) Key challenges in Mediterranean world: Cross- disciplinary investigation 3. Clark WC, Dickson NM (2003) Sustainability science: agent-based modeling for geo-spatial simulation. of cause-and-effect for degradation and soil erosion. J The emerging research program. Proc Natl Acad Sci Comput Environ Urban Syst 32:417–430. Archaeol Sci 23:1773–1800. USA 100:8059–8061. 21. Loule MA, Carley KM (2008) Balancing the criticisms: 38. Lentz DL, Hockaday B (2009) Tikal timbers and 4. Turner BL, 2nd, Lambin EF, Reenberg A (2007) The Validating multi-agent models of social systems. Simul temples: Ancient Maya agroforestry and the end of emergence of land change science for global Model Pract Theory 16:242–258. time. J Archaeol Sci 36:1342–1353. environmental change and sustainability. Proc Natl 22. Miewald RD (1978) Social and political impacts of 39. Hunt TL (2007) Rethinking ’s ecological Acad Sci USA 104:20666–20671. drought. North American Droughts, ed Rosenberg NJ catastrophe. J Archaeol Sci 34:485–502. 5. Dearing JA, Braimoh AK, Reenberg A, Turner BL, II, (Westview, Boulder, CO), pp 79–101. 40. Hornung E, Krauss R, Warburton DA, eds (2006) Van der Leeuw S (2010) Complex land systems: The 23. Worcester D (1979) Dust Bowl: The Southern Plains in Ancient Egyptian chronology. Handbook of Oriental need for long time perspectives in order to assess their the 1930s (Oxford Univ Press, New York). Studies (Brill, Boston). future. Ecol Soc 15:21. 24. Bowden MJ, et al. (1981) The effect of climate 41. Oates D, Oates J (2001) Archeological reconstruction 6. Nelson DR, Adger WN, Brown K (2007) Adaptation to fluctuations on human populations: Two hypotheses. and historical commentary. Excavations at Tell Brak, environmental change: Contribution of a resilience Climate and History: Studies in Past Climates and Their Vol. 2, Nagar in the Third Millennium BC, eds Oates D, framework. Annu Rev Environ Resour 32:395–419. Impacts on Man, eds Wigley TML, Ingram MJ, Farmer G Oates J, McDonald H (McDonald Institute Mono- 7. Redman CL, Kinzig AP (2003) Resilience of past (Cambridge Univ Press, Cambridge, United Kingdom), graphs, Oxbow Books, Oxford), pp 379–396. landscapes: Theory, society and the longue durée. pp 479–513. 42. Bronk Ramsey C, et al. (2010) Radiocarbon-based Conserv Ecol 7:14. 25. Butzer KW (1983) Human response to environmental chronology for Dynastic Egypt. Science 328:1554–1557. 8. Gil-Romera G, et al. (2010) Interpreting resilience change in the perspective of future global climate. 43. Kinzig A, et al. (2006) Resilience and regime shifts: through long-term ecology: Potential insight in Quat Res 19:279–292. Assessing cascading effects. Ecol and Society 11:20. Western Mediterranean landscapes. Open Ecol J 3: 26. Cunfer G (2005) On the Great Plains: Agriculture and 44. Rosen AM, Rivera-Collazo I (2012) Climate change, 43–53. Environment (Texas A&M Univ Press, College Station, TX). adaptive cycles, and the persistence of foraging 9. McLaughlin P, Dietz T (2007) Structure, agency and 27. Butzer KW (2008) Other perspectives on urbanism: economies during the Pleistocene/Holocene tran- environment: Toward an integrated perspective on beyond the disciplinary boundaries. The Ancient City, sition in the Levant. Proc Natl Acad Sci USA 109:3640– vulnerability. Glob Environ Change 18:99–111. eds Marcus J, Sabloff JA (Sackler Symposium, School 3645. 10. Endfield GH (2008) Climate and Society in Colonial for Advanced Research Press, Santa Fe, NM), pp 77–94. 45. Gunderson LH, Holling CS (2002) Panarchy: Under- Mexico: A Study in Vulnerability (Blackwell, Oxford, 28. Potter TW (1979) The Changing Landscape of South standing Transformations in Systems of Humans and United Kingdom). Etruria (Elek, London). Nature (Island Press, Washington, DC). 11. Turner BL, II (2010) Vulnerability and resilience: 29. Jameson MH, Runnels CH, Van Andel TH (1994) A 46. Luzzadder-Beach S, Beach TP, Dunning NP (2012) Coalescing or paralleling approaches for sustainability Greek Countryside: The Southern Argolid from Wetland fields as mirrors of drought and the Maya science? Glob Environ Change 20:570–576. Prehistory to the Present Day (Stanford Univ Press, abandonment. Proc Natl Acad Sci USA 109:3646– 12. Janssen MA, Schoon ML, Ke W, Börner K (2006) Stanford, CA). 3651. Scholarly networks on resilience, vulnerability and 30. Marcus J (1998) The peaks and valleys of ancient states: 47. Dunning NP, Beach TP, Luzzadder-Beach S (2012) adaptation within the human dimensions of global An extension of the dynamic model. Archaic States, Kax and kol: Collapse and resilience in lowland environmental change. Glob Environ Change 16: eds Feinman GM, Marcus J (School for Advanced Maya . Proc Natl Acad Sci USA 109:3652– 240–252. Research Press, Santa Fe, NM), pp 59–94. 3657. 13. Meyer WB, et al. (1998) Reasoning by analogy. Human 31. Milisauskas S, ed (2002) European Prehistory: A Survey 48. Dugmore AJ, et al. (2012) Cultural adaptation, Choice and Climatic Change, eds Rayner S, Malone EL (Kluwer, New York). compounding vulnerabilities and conjunctures in (Batalle Press, Columbus, OH), Vol 3, pp 217–290. 32. Butzer KW (2008) Challenges for a cross-disciplinary Norse Greenland. Proc Natl Acad Sci USA 109:3658– 14. Butzer KW (1982) Archaeology as Human Ecology: geoarchaeology: The intersection between environ- 3663. Method and Theory for a Contextual Approach mental history and geomorphology. Geomorphology 49. Streeter R, Dugmore AJ, Vesteinsson O (2012) Plague (Cambridge Univ Press, New York). 101:402–411. and landscape resilience in premodern Iceland. Proc 15. Adger WN (2000) Social and ecological resilience: Are 33. Rosen AM (1986) Cities of Clay: The Geoarchaeology of Natl Acad Sci USA 109:3664–3669. they related? Prog Hum Geogr 24:347–364. Tells (Univ of Chicago Press, Chicago). 50. Harris S (2012) Cyprus as a degraded landscape or 16. Liu J, et al. (2007) Complexity of coupled human and 34. Beach TP, Luzzadder-Beach S (2008) Geoarchaeology resilient environment in the wake of colonial intrusion. natural systems. Science 317:1513–1516. and aggradation around Kinet Höyük, an archaeol- Proc Natl Acad Sci USA 109:3670–3675. 17. Folke C (2006) Resilience: The emergence of a per- ogical mound in the Eastern Mediterranean, Turkey. 51. Endfield G (2012) The resilience and adaptive capacity spective for social-ecological systems analyses. Glob Geomorphology 101:416–428. of social–environmental systems in Colonial Mexico. Environ Change 16:253–267. 35. Shaw I (1999) The Oxford History of Proc Natl Acad Sci USA 109:3676–3681. 18. Costanza R, et al. (2007) Sustainability or collapse: (Oxford Univ Press, London). 52. Walker BH, Holling CS, Carpenter SR, Kinzig AP (2004) What can we learn from integrating the history of 36. Franke S (1995) Kings of Akkad: Sargon and Naram- Resilience, adaptability, and transformability. Ecol humans and the rest of nature? Ambio 36:522–527. Sin. Civilizations of the Ancient Near East,ed and Society 9:5.

Butzer and Endfield PNAS | March 6, 2012 | vol. 109 | no. 10 | 3631