Global climate change, war, and populationdecline in recent human historyDavid D. Zhang*†, Peter Brecke‡, Harry F. Lee*, Yuan-Qing He§, and Jane Zhang¶

*Department of Geography, University of Hong Kong, Pokfulam Road, Hong Kong; ‡Sam Nunn School of International Affairs, Georgia Institute ofTechnology, Atlanta, GA 30332-0610; §Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Science,Lanzhou 730000, Gansu, China; and ¶Department of Anthropology, University College London, London WC1E 6BT, United Kingdom

Edited by Paul R. Ehrlich, Stanford University, Stanford, CA, and approved October 23, 2007 (received for review April 4, 2007)

Although scientists have warned of possible social perils resultingfrom climate change, the impacts of long-term climate change onsocial unrest and population collapse have not been quantitativelyinvestigated. In this study, high-resolution paleo-climatic datahave been used to explore at a macroscale the effects of climatechange on the outbreak of war and population decline in thepreindustrial era. We show that long-term fluctuations of warfrequency and population changes followed the cycles of temper-ature change. Further analyses show that cooling impeded agri-cultural production, which brought about a series of serious socialproblems, including price inflation, then successively war out-break, famine, and population decline successively. The findingssuggest that worldwide and synchronistic war–peace, population,and price cycles in recent centuries have been driven mainly bylong-term climate change. The findings also imply that socialmechanisms that might mitigate the impact of climate change werenot significantly effective during the study period. Climate changemay thus have played a more important role and imposed a widerranging effect on human civilization than has so far been sug-gested. Findings of this research may lend an additional dimensionto the classic concepts of Malthusianism and Darwinism.

agricultural production � population collapse � price � war–peace cycles �preindustrial era

Scientists have noted that social activities heavily depend onclimate. They have also pointed out that temperature prob-

ably influences our lives more than any other climatic factor andhuman society is especially vulnerable to large, long-term tem-perature changes (1). However, scientific research on the socialeffects of climate change has tended to focus on the economiccosts of current and future climate change and has neglected thestudy of how societies have historically reacted to long-termclimate change. This neglect is unfortunate because a betterunderstanding of how past climatic changes have influencedhuman society may help us better understand our futureprospects.

Recently, important attempts have been made to use high-resolution, reconstructed paleo-climatic data to elucidate indi-vidual cases of prehistoric cultural/population collapses causedby agricultural failure in the Middle East, United States, andChina (2–4). Webster (5) pointed out that warfare was anadaptive ecological choice in prehistoric societies with limitedresources and growing populations, although he was not able touse systematic, scientific data to support his conclusion. Theconcept of environmental conflict has been suggested by severalresearchers, but they focus only on conflicts caused by short-term climate variations and meteorological events (6–9). Gal-loway (10) found that long-term climate change controlledpopulation size in middle-latitude areas. However, his findinglacked quantitative precision because of the absence of high-resolution climate records at the time. We studied a long spanof Chinese history and found that the number of war outbreaksand population collapses in China is significantly correlated withNorthern Hemisphere (NH) temperature variations and that all

of the periods of nationwide unrest, population collapse, anddynastic change occurred in the cold phases of this period(11–13). As a result of recent scientific breakthroughs in estab-lishing more precise paleo-climatic records [see supportinginformation (SI) Text], we extend the earlier study to the globaland continental levels between A.D. 1400 and A.D. 1900, duringthe Little Ice Age (LIA; see SI Text).

The hypothesis we propose posits that long-term climatechange has significant direct effects on land-carrying capacity (asmeasured by agricultural production). Fluctuation of the carry-ing capacity in turn affects the food supply per capita. A shortageof food resources in populated areas increases the likelihood ofarmed conflicts, famines, and epidemics, events that thus reducepopulation size. As a feedback mechanism, population declinehas a dominant tendency to increase the food supply per capita(seen in decreasing food prices), which results in relative peaceand fast population growth. The interactions among thesecomponents in a social system create an important rhythm ofmacrohistory in agricultural societies. The simplified pathwaysof the above chain reactions and feedback loops are representedin SI Fig. 3.

With respect to the character of the causal pathways, therelation between climate and agricultural production has beendemonstrated by many empirical studies (10, 14). Under eco-logical stress, adaptive choices for animal species are the reduc-tion of population size, migration, and dietary change. Depop-ulation typically takes place through starvation and cannibalism.Humans have more pathways, social mechanisms, to adapt toclimate change and mitigate ecological stress. Besides migration,they include warfare, economic change, innovation, trade, andpeaceful resource redistribution. We believe that in late agrariansociety established political boundaries in populated areas lim-ited mass migration; the result of such mass migration, when itoccurred, often was war. Economic change was a costly and slowprocess that involved changing cultures, technologies, and habits.When the speed of human innovation and its transfer were notfast enough to keep pace with rapid ecological change, famineand disease became difficult to avoid. Trade and redistributionunder the condition of shrinking resources would not help muchbecause the ecological stress was at a global or very largeregional scale. Finally, human social development in the form ofinternational and national institutions was not strong enough tobuffer the tensions caused by food resource scarcity. Therefore,war and population decline became common consequences of

Author contributions: D.D.Z. designed research; D.D.Z., P.B., H.F.L., Y.-Q.H., and J.Z. per-formed research; D.D.Z., P.B., H.F.L., Y.-Q.H., and J.Z. analyzed data; and D.D.Z. and H.F.L.wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

Freely available online through the PNAS open access option.

†To whom correspondence should be addressed. E-mail: zhangd@hkucc.hku.hk.

This article contains supporting information online at www.pnas.org/cgi/content/full/0703073104/DC1.

© 2007 by The National Academy of Sciences of the USA

19214–19219 � PNAS � December 4, 2007 � vol. 104 � no. 49 www.pnas.org�cgi�doi�10.1073�pnas.0703073104

climate-induced ecological stress in the late preindustrial era.Recent developments in resource and environmental studies(e.g., refs. 8, 9, 15, and 16) suggest that limited resources andenvironmental degradation would have caused armed conflictsin human history. However, these perceived climate–war–population decline sequences have never been substantiatedwith scientific evidence consisting of long-term time series. Inthe following sections, we verify our hypothesis and evaluate therole of climate change on war outbreak and population declinewith empirical data at global and continental scales.

Results and DiscussionWe adopted a quantitative and macrohistorical approach toexplore the climate–agriculture–war–population relationshipsdepicted in SI Fig. 3 in the preindustrial era. Statistical methodswere used to examine the correlations among the variouscomponents in the pathways. The various relationships betweendifferent factors between the pathways were more thoroughlyanalyzed with the two regions for which there are more data,Europe and China.

Cyclic Patterns of Wars and Temperature Changes. In the samemanner as the NH temperature variations, the incidence ofwarfare in the NH, Europe, Asia, and the arid areas of the NH(i.e., the arid zone from Eurasia to North Africa; see SI Fig. 4)in A.D. 1400–1900 tends to follow a cyclic pattern with aturbulent period followed by a relatively tranquil one (Fig. 1 Aand B). Two periods plagued by unrest and warfare and threerelatively peaceful periods follow the temperature undulationinversely. This pattern not only appears at a continental scale,but also in three war databases with different violence thresholdsat a global scale (Fig. 1C), except for the late 19th century whenthe temperature of the Southern Hemisphere (SH) was its coldestin the last millennium (17) and a great number of wars occurred inthe southern part of Africa.� In short, synchronous periods ofrelative peace and turbulence during those 500 years were a globalphenomenon seemingly linked to temperature change.

To look at the severity of these wars and their impacts onhuman population, a NH fatality index for A.D. 1400–1900 wascalculated from the wars with a known fatality record aggregatedinto 50-year intervals (see Fig. 1D and SI Text). The two peaksof the fatality index appear in the cold 17th and the early 19thcenturies. Two of the greatest population declines since A.D.1400 coincided with the two peaks.

We note that the turning points of war/peace cycles in the NHat the ends of the 17th and 18th centuries are different from theturning points of the two longest mild/cold periods in thecenturies. The time gaps are 20–30 years in all cases, which iscaused by the population feedback, which will be verified in alater section.

We also examined Chinese warfare history back to A.D. 1000(see SI Text and SI Table 2), which shows several war/peacecycles. It can be seen that all war peaks (�25 wars per decade)occurred in a cold climate, and multiple war–peace cycles werefollowed closely the NH’s temperature variations (see SI Fig. 5).The temperature–war correlation is also statistically significant,and population declines followed every high war peak (12, 13).Krus et al. (19) also examined frequency of Chinese wars andfound that the periods that non-Confucian ethical systemsdominated China had more wars. However, the periods oftenwere cold and dry, and such climate drove the northern andwestern tribal states to enter central China, thus changingoriginal Confucian ethical systems.

At the century timescale, war frequencies and the ratio of wars

per year in different centuries have been calculated at global andcontinental geographic scales. Those calculations demonstratethat the number and ratio of wars in the 18th century are thelowest compared with other centuries. The results indicate thatthe worldwide war ratio during cold centuries was 1.93 timesgreater than that of the mild 18th century, and 1.77, 1.91, 1.50and 2.24 times greater for the NH, Asia, the arid areas of the NH,and Europe, respectively. At the country scale, as the incidenceof warfare has been too small to permit correlation analysis tobe conducted, the number of wars in each century was calculatedand the ratios of wars per year were compared. These calcula-tions show that only 27 of 170 countries and areas around the

�Brecke P, Annual Meeting of the Peace Science Society (International), October 8–10, 1999,Ann Arbor, MI.

Fig. 1. Paleo-temperature variation, war frequency, and population growthrate, A.D. 1400–1900. (A) Temperature anomaly (°C) in the NH that issmoothed by 40-year Butterworth low pass filter (17). (B) Number of wars inthe NH (bright green), Asia (pink), Europe (turquoise), and the arid areas in theNH (orange).� (C) Number of wars worldwide as recorded by Wright (39)(turquoise), Luard (38) (orange), and Brecke� (bright green). (D) Twenty-yearpopulation growth rate in Europe (turquoise), Asia (pink), and the NH (blue)�

and the NH 50-year fatality index (bright green). Cold phases are shaded asgray stripes. All war time series are in 10-year units, and the data are listed inSI Table 1. The bright green curves correspond to the right y axis.

Zhang et al. PNAS � December 4, 2007 � vol. 104 � no. 49 � 19215

AN

THRO

POLO

GY

world had higher war ratios in the 18th century than in the15–17th and 19th centuries. Most of the ‘‘exceptional cases’’ arelocated near the Equator, where there was no obvious LIA, andthus cooling that would have a significant effect on agriculture.

To refine the analysis, a series of correlation analyses werecarried out at the much lower scale of an annual level timeframe.Although the SH had no obvious LIA, the temperature anomalyfor the SH was used for comparative purposes in the correlationanalysis by examining the temperature and war frequency rela-tionship. The results show that the numbers of wars per year, inall war classifications, were significantly negatively correlatedwith the annual changes of the temperature anomalies at theglobal level (see SI Table 3). The number of wars in the NH andSH also correlated significantly with their temperature varia-tions. The reliability of the correlation analysis was verified by across-examination method that deliberately correlates the warfrequency of a particular hemisphere or region with the otherhemisphere’s temperature anomaly. The results show that allcorrelation coefficients of these ‘‘mismatched’’ combinationswere either insignificant or very low positive. For instance, thecorrelation coefficients between the SH temperature anomalyand NH war frequency were �0.06 for A.D. 1400–1900 and 0.137for A.D. 1500–1900, respectively. This finding suggests that thestrikingly high coefficients of our correlation analysis are by nomeans accidental, and the number of wars in each hemisphereis related only to its own temperature variation. Further analysisof the frequency of wars in Europe, Asia, and the arid areas ofthe NH also shows significant correlations with the NH temper-ature anomaly for their geographical patterns (see SI Table 3).

The values of the coefficients at the annual scale also reflectenvironmental vulnerability and population density, which couldfurther verify our hypothesis. As arid regions are the mostvulnerable to climate change, the highest coefficient should anddoes appear in the war frequencies for the arid areas of the NH.Values are lower for Asia, probably because much of thecontinent for which there is conflict data is subject to a wettropical or subtropical environment, where a fall in temperaturewould have a smaller effect than elsewhere on agriculturalproduction and would not reduce food resources to the sameextent because there is more alternative food. The Americancontinents had a very low population density and a large amountof fertile land during this period, and its values are thereforeeither insignificant or of very low significance. Europe andAfrica have either large cold areas or desert areas relativelyvulnerable to cooling, and we find that warfare there is highlycorrelated with the temperature anomalies. At the global scale,the temperature anomaly is simply an arithmetic mean of the NHand SH temperature anomalies (17). The values of the coeffi-cient for wars at the global scale are thus lower than those of theNH and SH alone.

Population Growth. The ecological stress triggered by climatechange induces population shrinkage of most species, includinghuman beings. Johnson and Gould (20) demonstrated that theagricultural production of Mesopotamia (present-day Iraq)closely followed its changing climate, which led to periodiccollapses of its population by famine and wars (20). Using thebest available world population estimates extending back to A.D.1000, we compared the NH temperature series with the NHpopulation growth rate. We also included the data series defin-ing the spatial extent of cooling in the NH for reference (21) (seeSI Fig. 6). Variations in the population growth rate can bedivided into three periods. In A.D. 1000–1400, when agriculturaltechnology and trade flows were at a low level, the populationgrowth rate was also low, and the world population basicallystagnated. Because of the gradual temperature drop and theincrease in size of the cold area from the ‘‘Medieval WarmPeriod’’ to the LIA, the growth rate also decreased concomi-

tantly. Every sudden temperature drop would induce a ‘‘demo-graphic shock,’’ and the correlation coefficient between the NHtemperature and the population growth rate (in 20-year units) ishighly significant in this period (r � 0.84, P � 0.001, n � 21) asagriculture largely depended on climate. The growth ratereached its lowest level in the 13–14th centuries, primarilybecause of epidemics, wars, and famines. In China, the invasionby the Mongols in the 13–14th centuries was related to theecological stress caused by cooling, which reduced China’s totalpopulation nearly by half (�55 million decline) (22). In Europe,Black Death spread out during the time, which was accompaniedby massive social unrest and economic collapse (23, 24), whichwiped out a quarter to one-third of the population in A.D.1347–1353, the coldest period in the last several hundred years(24). At first glance, a cold climate did nothing to foster plague.However, cooling, according to our model, would cause wars,forced migration, or even famine, which would help to spread outthe ‘‘vectors.’’ In China, historical statistics also reveal that a coldclimate has been accompanied by more epidemics (25).

The second period is A.D. 1400–1700. The first part of thisperiod coincided with a warm phase and the beginning of the‘‘Early Modern Times,’’ and the NH population growth rate roserapidly. The population growth rate stayed at an elevated levelfrom A.D. 1400 to 1600, despite a short cooling beginning in themiddle of the 15th century. This phenomenon can probably beexplained by three factors: (i) a large portion of the northhemispheric land area was not affected by the cooling (see SI Fig.6C); (ii) improvements in trade in the Renaissance Era; and (iii)fewer large magnitude wars during this period (Fig. 1C). How-ever, when the coldest and also lengthiest epoch of the LIA tookplace (17th century), more wars of great magnitude and theassociated population declines in Europe and Asia (especiallyChina) followed (Fig. 1D). The European population was dev-astated by possibly the worst war in its history in terms of theshare of the population killed in A.D. 1618–1648 (24), starva-tion, and epidemics. In China, the population plummeted 43%(�70 million) because of wars, starvation and epidemics in A.D.1620–1650 (22). The rapid ‘‘postshock’’ bounce of the popula-tion growth rate in the late 17th century was related to thedramatically reduced population size and the coupled increase offood supply per capita. A more efficient resource distributioncaused by technological and social advance in Europe might havealso played a role.

A dramatic increase in human population occurred in the lasttwo centuries of the LIA because of the Industrial and Agricul-tural revolutions, especially in Europe. However, a coolingperiod that covered a relatively larger area in the mid-19thcentury still caused a worldwide demographic shock, in part be-cause of the prevalence of warfare in different continents (Fig. 1).

Annual-scale analysis for the period A.D. 1400–1900, whichhas higher-resolution population data, could help verify theabove hypothesis. Correlation analysis showed a very significantrelationship between the NH temperature swings and popula-tion growth rates for the world, NH, Asia, North America, andEurope, as well as China (see SI Table 4). As expected, thecorrelation coefficients for Europe and North America are lowerthan the others. The reason is that Europeans and Americanslived in an ecologically open system since the Industrial Revo-lution in which European population pressure could be relievedby cross-continental colonization and migration. When the 19thcentury (the age of Great Migration in Europe) population datawere not included in the analysis, the correlations for Europe andNorth America were almost as high as those for Asia and thewhole NH. Similarly, the large-scale trans-Atlantic migrationfrom Europe to North America in A.D. 1700–1900, which issupposed to be driven by cooling, can also be substantiated by thesignificant negative correlation between the NH temperature

19216 � www.pnas.org�cgi�doi�10.1073�pnas.0703073104 Zhang et al.

anomaly and North America’s population growth (r � �0.341,P � 0.001, n � 201).

European and Chinese Cases. The above results have shown sig-nificant correlations and coincident timing between temperaturechange, war frequency, and population growth rate in differenttemporal and spatial scales. The analyses also indicate thatcooling may have contributed to widely separated human disas-ters during the cold periods. The paths to those disastersoperated through a reduction in agricultural production. Thereduction is mainly because the cooling brought about a short-ening of the growing season and a reduction of available land forcultivation. Shortages of food resources (reflected in price rises)instigated conflicts and demographic declines. To illustrate therelationship between temperature change and other factors inour pathways, we present figures that show successive fluctua-tions of those factors in Europe and China in A.D. 1500–1800(Fig. 2). Only those two regions have relatively long-spanhistorical records of the factors, and both shared �60% of thetotal world population during that time. China here refers tocurrent territories of the People’s Republic of China, includingso-called China proper, part of Outer China, and some tribalstates of China. Regarding the time span, Europe before A.D.1500 did not have any effective market-pricing mechanisms (26),whereas the Industrial Revolution in Europe spread out afterA.D. 1800. To see the real influence of nature, the upward trendsin agricultural production, population size, cereal prices, and warfrequency, which were caused by technological developmentduring the period, have been detrended in Fig. 2.

The European grain yield ratio followed temperature changevery closely (Fig. 2 A and B). As this data series is only in 50-yearintervals, we estimated the annual agricultural production indexby using cereal prices and population sizes. The estimated valuesmatched closely with the temperature and real grain yieldtrajectories, which imply that temperature change definitelycontrolled agricultural production during that period. The mostsurprising finding is that the fluctuations of all of the compo-nents are the same in terms of macrotrends, turning points, andoscillation magnitude for both Europe and China at a time whenboth regions were detached, economically, politically, and geo-graphically (Fig. 2 B–E). The most important aspect of thisfinding is that the fluctuations of all of the factors were in asuccessive order and corresponded to the temperature change.When agricultural production went down, wheat prices went upin both Europe and China. When prices reached a certain level,more wars erupted. Population growth rates were influenced byboth war frequency and food supply per capita (reflected incereal prices) and dramatically dropped to negative values whenagricultural production was at its lowest levels, cereal pricesreached their highest level, and peaks in war frequency occurred.Such a human ecological and social disaster reduced populationsizes and even caused catastrophic population collapses in bothEurope and China in the 17th century. The period is then calledthe General Crisis of the 17th Century. Research on biologicalliving standards of ancient Europeans also support our data andanalysis; the average height of Europeans in the 17th century isthe shortest during the last two millennia and was 2 cm shorterthan those in 16th and 18th centuries (27). After the slump inpopulation size, food supply per capita increased (as reflected indecreased prices), and the incidence of war in the late 17thcentury declined even as the climate was cold and agricultureproduction was still low. Such a feedback effect can also be seenin Europe in the late 18th century. Although the climate was mildand agricultural production was high, the rapid increase inpopulation dampened the food supply per capita (prices in-creased), and then the number of wars started to go up. Thefeedback mechanism explains why there were time gaps betweenthe turning points of temperature variation and the war–peace

cycles in various NH continents in the late 17th and late 18thcenturies (Fig. 1B). The correlation coefficients between thefactors listed in Fig. 2 have been calculated, and all of thecoefficients for both regions are above the 0.001 significancelevel (see SI Table 5). The level of average correlation is shownby the thickness of the lines in SI Fig. 3. Population size is

Fig. 2. Paleo-temperature variation and its impact on Europe and China,A.D. 1500–1800. (A) Temperature anomaly (°C) in the NH (17). (B–E) Europeandata are in green and Chinese data are in red. The data are in normalized unitsthat indicate the relative amplitude of change only. (B) Detrended estimatedvalues of agricultural production (solid lines) and detrended grain yield ratioin Europe (green-dotted curve) (18). The estimated values of agriculturalproduction are calculated by dividing population size by food price. (C)Detrended population size (solid lines) and population growth rate (dottedcurves) (22, 24). (D) Detrended wheat price index in Europe and detrended riceprice in China (41, 42). (E) Detrended total war frequency (� and ref. 40). Alldata are smoothed by 40-year Butterworth low pass filter. All of the dottedcurves correspond to the right y axis.

Zhang et al. PNAS � December 4, 2007 � vol. 104 � no. 49 � 19217

AN

THRO

POLO

GY

correlated with cereal prices, and the population growth rate iscorrelated with war frequency and agricultural production.

Theoretical and Social ImplicationsThe very significant correlations between temperature change,war frequencies, and population declines at the global andcontinental scales have been verified by different quantitativemethods. The coincidence of multiple war–peace cycles andpopulation decreases worldwide at various temporal scales isunlikely to be accidental. It has been shown that elevated levelsof war outbreak and population decline in populated areasoccurred during a cold climate in the past millennium. Europeand China are strong cases to illustrate that the links andfeedback effects between temperature change and social events,long-term cycles of war outbreak and population decline, orig-inate in large part from a fundamental driver, climate change. Allof the above analytical results indicate that war and depopulationhave been the important adaptive choices in preindustrial soci-ety. The results also have implications for the following out-standing issues.

Other Pathways for Human Adaptation. Our quantitative resultsalso show the importance of the roles of some other humanadaptive choices in mitigating war outbreak and populationdecline. At the country and local scales, these social mechanismsmight have lessened, postponed, or even eliminated on occasionwar outbreak and population declines during bouts of mild orshort cooling. For instance, before A.D. 1400, war outbreaks andpopulation decreases in China immediately followed every de-cline of temperature (see SI Fig. 5). After then, the short coolingin the 15th century did not generate a period of war outbreak andpopulation decline, and the times of war outbreak and popula-tion decrease in response to great temperature declines werepostponed by �30 years in the 17th and 19th centuries, possiblybecause of the work of institutions and technological develop-ment. After A.D. 1700, the base level of war frequency in Chinadecreased because the Qing emperors had united all trouble-some tribal states in the western and northern marginal areas(see SI Fig. 5B).

In the long run and at a global scale, technological and socialdevelopment raised the population growth rate by 30% for A.D.1400–1700 and 310% during A.D. 1700–1900, compared withthe A.D. 1000–1400 period (Fig. 1D). It also reduced climatedependence of growth rate of population (after A.D. 1400),postponed the time of population decrease, and acceleratedsubsequent population recovery. For instance, the demographicshocks in the NH before A.D. 1400 have no time delays betweenthe population growth rate drops and the temperature drops, butafterward the shocks are delayed by 20 and 40 years, respectively,in the 17th and 19th centuries (see SI Fig. 6). The gradualincrease in time delays for NH population declines as we movedinto the modern era may reflect that at least some socialmechanisms may becoming more effective over time at themacroscale. We cannot evaluate quantitatively the relative im-portance of these mechanisms because measurement of theseparameters is not feasible with current data sources. However,these adaptive choices that are positive to humanity have not letthe human race escape from social calamities such as populationcollapse caused by severe cooling at both the global and conti-nental scales as shown in the history of the past millennium. Forarmed conflict, the positive social mechanisms could neitherreduce the number of wars nor indefinitely postpone the timesof war outbreak in any cooling periods. However, the change ofeconomy after A.D. 1900 (the change of basic resources inindustrial society) had made that the great conflicts and demo-graphic shocks were to only a limited extent affected by agri-cultural production at a global scale.

Cycle Theories. Many social scientists have noted cycles of socialorder and disorder in history, in which downturn periods werecharacterized by inflation, a fall in real wages, political crisis,war, and state breakdown (SI Text and refs. 28 and 29). There aremany theories and hypotheses that attempt to explain this typeof crisis and the dynamics behind the cyclical pattern of history(24). They have told us much about the dynamics behind thecycles. However, those models cannot explain parallel occur-rence of these crises in widely separated regions that were indifferent stages of civilization, culture, and resource endowment,nor predict the timing of such crises. The agrarian and environ-mental models are similar to our model. But because there wereno high-resolution paleo-temperature reconstructions at thetime these models were proposed, and the explanations in thesemodels lack population feedback, the models are not entirelyconvincing.

Demographic Theories. Malthus, as well as Darwin and many otherecologists, who state that ‘‘positive checks’’ occur when popu-lation growth overshoots the level of livelihood resources basedon the assumption that the level of resources needed for basiclivelihood is essentially constant or possibly monotonically in-creasing, is only partly correct (see SI Text). However, our resultsshow that the level of those livelihood resources (food) will, atbest, increase in an oscillating, not always positive, mannerbecause the impact of climate change, and population growthwas not. An anti-Malthusian, Boserup (30), claimed that popu-lation pressure would lead to the growth of agricultural produc-tion. She is correct with respect to limited scale regions andshort-term timeframes if the land still has potential for intensivecultivation or for a very long-term trend concerning humanhistory. However, such intensive cultivation could not keep pacewith the needs of a growing population during periods when theclimate cooled in Europe and China, as can be seen in variousEuropean economic records and the evidence from this study.

War Theories. War is, no doubt, an extremely complicated socialphenomenon, and many scholars have investigated the problemof the causes of war since the days of Thucydides (SI Text andref. 31). Sadly, scholars have made at best modest progress on theproblem even though members of all of the social sciences haveaddressed it (32). Some of these theories may explain certainwars, and some of the theories may even explain sizable classesof wars, but none of them can explain the temporal and spatialpatterns of warfare presented above. Our approach has not beento examine causes of individual wars, but to investigate thefrequency of all wars during a specific period of human historyand find out when, where, and why the war–peace cyclesoccurred. Our results not only answered when and where mostof wars occurred, but also imply that relative food scarcity wasa fundamental cause of war outbreaks according to our quan-titative analysis on various physical and social components andin different geographical regions. Such a resource scarcitymanifested itself in two causal pathways: a direct cause, in whichresource-oriented wars erupted as most of the world’s popula-tion still struggled to satisfy the lower levels of Maslow’sHierarchy of Needs (33), and an indirect cause, as constrainedfood resources and economic difficulties stemming from thatintensified different social contradictions, that increased thelikelihood of war outbreaks.

Implications for the Future. The findings from our study aregermane to current situations even though they pertain to theconsequences of climate cooling. We are now in the warmestclimatic phase of the past two millennia, another climaticextreme like the 17th century in LIA, which was the coldestperiod of the past two millennia (17). Both higher and lowertemperatures would lead to change of the human ecosystem. The

19218 � www.pnas.org�cgi�doi�10.1073�pnas.0703073104 Zhang et al.

important direct impact of the change from the climate coolingwas on agricultural production that dominated the economy ofagricultural societies. Global warming could also create animpact on agriculture. Although rising of temperature will, atleast for a while, increase total bio-productivity according tobiological principles, recent research suggests that the negativeeffects of global warming on agriculture seems greater than thepositive effects (34, 35). Significantly, most of the world’spopulation will continue to rely on small-scale agriculture, whichremains as vulnerable to climatic fluctuations as it was in thehistorical period covered in this study. Other direct impacts ofglobal warming, such as sea level rising, spread of tropicaldiseases, increase of extreme weather events and glacial retreat,would also add costs on the current economy that is supportedby cheap energy. In severe cases, the economic burden wouldcause conflict for resources and intensify social contradictionsand unrest as we have seen in the past. However, we believe thatthe greater threat from global warming comes from the uncer-tainty of the ecosystem change, because the current high globalaverage temperature (which has never been experienced in thelast two millennia) is continuing to rise at an accelerated speed.Perhaps we are reaching the point at which it might break thebalance of a human ecosystem that has been long established ata lower temperature, and in addition, many secondary andtertiary effects of global warming cannot be predicted based oncurrent knowledge. A change of one key component in theecosystem under global warming would likely cause disastrousresults in human societies dependent on the existing humanecosystem. Is the changed ecosystem sufficiently adaptable orare the adaptation choices affordable for all of us?

Although we have more robust social institutions at bothinternational and national levels, and much more advancedsocial and technological developments at present, a much largerpopulation size, higher standard of living, and more strictlycontrolled political boundaries will limit some adaptive choicesto climate change. We hope that positive social mechanisms thatare conducive to human adaptability will play an ever moreeffective role in meeting the challenges of the future.

MethodsPaleo-Climate Data. Research in the past 10 years has led tosignificant improvements in high-resolution paleo-climate re-

constructions (see SI Text). Our temperature data come from themost comprehensive and latest published records of the NHannual temperature trajectory, supplemented by the SH andglobal temperature trajectories (17). Besides, it has been con-firmed that a cold period (LIA) existed globally during the lastmillennium (36, 37) (see SI Text).

Historical Socioeconomic and Demographic Data. Historical socio-economic and demographic data for the early preindustrial eraare scarce. It was not until the Late Middle Ages (�A.D. 1400)that some of the desired records started to be kept. For war data,three of the most representative worldwide warfare series (�, 38,39) were used to measure war frequency. Besides, a compre-hensive China warfare dataset (40) that documents 1,664 armed-conflicts fought in A.D. 1000–1900 was also used in this research(see SI Text). For population data, they were extracted from theAtlas of World Population History (24) and Recent History ofChinese Population (22) (see SI Text). For food supply per capita,it is represented by historical cereal prices because it is deter-mined by supply (agricultural production) and demand (popu-lation size). Nevertheless, only the cereal price data for Europe(41) and China (42) are long enough time series for our dataanalysis (see SI Text). For agricultural production, it is a per-ceived wisdom that a cold climate would lead to harvest failure,especially in the past. Although the data on agricultural pro-duction for many parts of the world are unavailable in our study’stime span, we could use two parameters to represent agriculturalproduction in our quantitative analysis. The first one is the grainyield ratio in relation to seed for Western and Central Europe(18), but the data are in 50-year intervals. The second one is theestimated annual values of agricultural production for Europeand China, which are calculated by population size over cerealprice (see SI Text).

We thank Profs C. Y. Jim, G. C. S. Lin, T. McGee, and Dr. D. J.Wilmshurst, the editor, and two anonymous reviewers for stimulatingdiscussion and valuable comments on this manuscript. This research wassupported by the Research Grants Council of the Hong Kong SpecialAdministrative Region Government for Project HKU7243/04H and theLouis Cha Fund.

1. National Research Council (1998) Decade-to-Century-Scale Climate Variabilityand Change: A Science Strategy (Natl Acad Press, Washington, DC).

2. Polyak VJ, Asmerom Y (2001) Science 294:148–151.3. Weiss H, Bradley RS (2001) Science 291:609–610.4. Wu W, Liu T (2001) Q Sci 21:443–451.5. Webster D (1975) Am Antiquity 40:464–470.6. Cowie J (1998) Climate and Human Change: Disaster or Opportunity? (Parthe-

non, New York).7. Ferguson RB (1984) Warfare, Culture, and Environment (Academic, Orlando, FL).8. Homer-Dixon TF (1994) Int Security 19:5–40.9. Suhrke A (1997) in Conflict and the Environment, ed Gleditsch NP (Kluwer,

Dordrecht, The Netherlands), pp 255–272.10. Galloway PR (1986) Population Dev Rev 12:1–24.11. Lee HF, Fok L, Zhang DD (2007) Clim Change, in press.12. Zhang DD, Jim CY, Lin GCS, He YQ, Wang JJ, Lee HF (2006) Clim Change

76:459–477.13. Zhang DD, Zhang J, Lee HF, He YQ (2007) Hum Ecol 35:403–414.14. Lamb HH (1995) Climate, History, and the Modern World (Routledge, London).15. Stranks RT (1997) in Conflict and the Environment, ed Gleditsch NP (Kluwer,

Dordrecht, The Netherlands), pp 157–175.16. Westing AH (1988) Cultural Norms, War, and the Environment (Oxford Univ

Press, Oxford).17. Mann ME, Jones PD (2003) Geophys Res Lett 30:1820.18. Slicher van Bath BH (1967) Acta Historiae Neerlandica 2:26–106.19. Krus DJ, Nelsen EA, Webb JM (1998) Psychol Rep 83:138–143.20. Johnson DL, Gould L (1984) in Climate and Development, ed Biswas AK

(Tycooly, Dublin), pp 117–138.21. Osborn T, Briffa KR (2006) Science 311:841–844.22. Jiang T (1993) Recent History of Chinese Population (Hangzhou Univ Press,

Hangzhou, China).

23. Fischer DH (1996) The Great Wave: Price Revolutions and the Rhythm of History(Oxford Univ Press, New York).

24. McEvedy C, Jones R (1978) Atlas of World Population History (Allen Lane,London).

25. Gong S (2003) Acta Geogr Sinica 58:870–878.26. Beaud M (1984) A History of Capitalism, 1500-1980 (Macmillan, London).27. Koepke N, Baten J (2005) Eur Rev Econ History 9:61–95.28. Deway ER, Franden DE (1947) Cycles: The Science of Prediction (Holt, New

York).29. Goldstone JA (1991) Revolution and Rebellion in the Early Modern World (Univ

California Press, Berkeley).30. Boserup E (1965) The Conditions of Agricultural Growth: The Economics of

Agrarian Change under Population Pressure (Allen & Unwin, London).31. Howard ME (1983) The Causes of Wars and Other Essays (Temple Smith,

London).32. Van Evera S (1999) Causes of War: Power and the Roots of Conflict (Cornell

Univ Press, Ithaca, NY).33. Maslow AH (1970) Motivation and Personality (Harper & Row, New York).34. Furuya J, Koyama O (2005) Japan Agric Res Q 39:121–134.35. Lobell DB, Field CB (2007) Environ Res Lett 2:014002.36. Esper J, Cook ER, Schweingruber FH (2002) Science 295:2250–2253.37. Hughes MK, Diaz HF (1994) Clim Change 26:109–142.38. Luard E (1986) War in International Society: A Study in International Sociology

(Tauris, London).39. Wright Q (1942) Study of War (Univ Chicago Press, Chicago).40. Editorial Committee of Chinese Military History (1985) Tabulation of Wars in

Ancient China (People’s Liberation Army Press, Beijing).41. Beveridge WH (1921) Econ J 31:429–452.42. Peng X (1965) History of Chinese Currency (Shanghai People’s Press, Shanghai).

Zhang et al. PNAS � December 4, 2007 � vol. 104 � no. 49 � 19219

AN

THRO

POLO

GY

Environmental Pollution 83 (1994) 37-43

CLIMATE CHANGE AND HUMAN HISTORY. SOME INDICATIONS FROM EUROPE, AD400-1400

Neville Brown

Environmental Change Unit, University of Oxford, la MansfieM Road, Oxford, UK, OX1 3TB

Abstract The influence of climate change on history is discussed in such a way as to take account of recent research while seeking to herald, rather than foreclose, the much more Jbcussed and definitive debates that will become possible by the turn of the century. By then, much more elabo- rated time series of global climatic change over the last several millennia should be available.

The period chosen for examination is the 1000 years or so following the collapse of Roman Europe. During this time, civilisation seems to wane, wax and wane again, very much in phase with the climate deteriorating, improving and then deteriorating once more. What this overview high- lights is the great vulnerability to climatic perturbation of societies that are marginally poised for other reasons. In several major respects, the whole of world society will be marginal in the 21st century.

Keywords." climate change, Europe, medieval, impact.

INTRODUCTION

In the course of the last 20 years or so, the diverse evi- dence about climate change within the span of recorded history has been subject to searching analysis by scholars such as Jean M. Grove, Hubert Lamb and Emmanuel Le Roy Ladurie. Their work has followed in the wake of that by such pioneers early this century as C. E. P. Brooks and Ellsworth Huntington. At the same time, analysis is becoming more accurate and complete on the palaeoclimatic scale, by which is primarily meant that extending back through the start of our Holocene or 'post-glacial' era and on through the successive glaciations of the Pleistocene. Duly, those concerned with climate prediction in connection with the current debate on 'global warming' have started to look to this work to provide the kind of extreme scenarios against which they can best test their General Circulation Models (GCM). No matter for this purpose that the present tendency to global warming is being forced by pollution of the natural atmosphere by mankind's economic activity whereas palaeoclimatic fluctuations owed little or nothing to this human factor.

By the same token, one might look to recorded history to see how societies and political regimes adapt

Environ. Pollut. 0269-7491/93/$06.00 © 1993 Elsevier Science Publishers Ltd, England. Printed in Great Britain

37

to climate change. At first sight, of course, there is a contradiction here. One looks to the future with appre- hension of global warming whereas, as regards the past, warming is seen in the main as beneficial. It extends growing seasons and allows more intensive crop growth. It reduces the spread of ice and of permafrost subsoil. Hopefully, any rises in sea level it induces through ice melt will not be too abrupt.

However, this contradiction is by no means as acute as may appear. For one thing, there is a counter-current of scientific opinion (especially associated with the State Hydrological Institute at St Petersburg) that stresses that even the current prospect of global warm- ing holds out some hope of real advantage, not least through a shrinking (Budyko & Izrael, concern of analysts

of the areas of precipitation deficit 1991). For another, the prime

today may be less with the direc- tion of temperature change than with the pace of it. After all, this pace has seemed set to be several times faster than that of any comparable secular trend in recorded historical experience. Nor will a comparison couched in these broad terms be affected so very much if the targets set at Rio for curbing the emission levels of greenhouse gases are met in full.

What the historical sweep well confirms is the tendency for climate zones to move, however erratically and irregularly, towards the equator during times of cooling but polewards if general warming is under way. Of special importance in this connection are the respec- tive subtropical belts of high atmospheric pressure: the zones characterised by anticyclonic winds, descending air, blue skies and steppe or desert landscapes. These belts can be traced back in the geological sequence across hundreds of millions of years. Their key role in offsetting the huge convection currents of ascending air around the equator was confirmed mathematically quite early in the history of modem meteorology. In short, they are among the strongest features in the general circulation of the atmosphere.

Yet while they may be very enduring, they can also be mobile, with markedly differential implications for the water balance at regional level. Hubert Lamb has deduced that, 4000 years ago in the European sector in winter, the median axis of highest barometric pressure was 16 ° of latitude further north than is the case today. Yet European annual mean temperatures seem not to have risen then more than 2 ° above present levels

38 N. Brown

(Lamb, 1982a): an increase very liable to be exceeded within the next century, even on 'Rio' assumptions.

Most of those who have written sometime this century about the impact of climate change have been geographers. Historians have tended thus far to be either blankly indifferent or positively scornful (Grove, 1988). Exceptions are to be found among the community of French historians associated with Annales, a journal founded in 1929 and very committed to forging links with other subjects, especially geography. Ferdinand Braudel was a doyen in this respect. Appropriately, it is Le Roy Ladurie, 'by common consent the most brilliant of Braudel's pupils' (Burke, 1990a) who has produced quite the most widely quoted of all the climate-and- history studies. None the less, in Times of Feast, Times of Famine (a work mainly devoted to the influence of climate change on Europe across this last millennium), he stresses how complex is the challenge of assessing the effect on crop yield of secular changes of mean tem- perature that may well not exceed 1 ° Celsius. Among the complications he draws attention to is the reality that rainfall trends are much more regionalised or even localised than temperature ones. Accordingly, they 'show long, interregional and a fortiori world tendencies much less easily' (Le Roy Ladurie, 1971a). He also perceives an inclination to want things both ways on the question of folk migrations: 'The Teutons of the first millennium before Christ are supposed to have left their countries of origin because of the cold. The Scandinavians of the period before AD 1000 are supposed to have done the same thing for exactly the opposite reason--the mildness of the climate, stimulat- ing agriculture and thus also population growth, is said to have led to the departure of surplus male warriors' (Le Roy Ladurie, 1971b).

It may, of course, be possible to square this circle by distinguishing between those migrations, born of economic stress, that have taken the form of the whole- sale displacement of the people in question; and those, rooted in enrichment, that have involved a people's expanding out of what then still remains their estab- lished homeland area. Here one link can be that such improvement eases poverty and so reduces what may otherwise be a horrendous rate (a half of all live births within the first year?) of infantile mortality. Hence more warriors and then more farmers. All the same, it is possible to take the more fundamental point that Professor Le Roy Ladurie goes on to make. This is that judgement on a lot of presumed correlations between climate change and the human historical process ought to be suspended until more thorough climate research has been done.

Fortunately, the development of the requisite tech- niques for the identification and dating of climate indicators may soon generate data time series suffi- ciently definitive and reliable to engage the interest of political, economic and social historians. Thus in Britain one of the streams of research initiated by the Past Global Changes (PAGES) core project within the International Geosphere-Biosphere Programme is to

concentrate o n the past 2000 years. The aim is to reconstruct a detailed sequence of climatic and environ- mental change for the entire globe over this span. The temporal resolution that is being sought is at least decadal or, ideally, annual or seasonal. One thing that can very confidently be predicted is that local climatic anomalies will be more evident on such a scale.

The provision of such information will confront historians of every hue, from neo-Marxists to Popperian conservatives, with stark choices as to how far climate shifts are to be seen as factors in historical change and, of course, how far these shifts may result from human activities. In the meantime, it may be useful to review ex- isting knowledge in order to identify apparent correla- tions that will merit more stringent examination when the time is ripe. Looking at this on the millennial time- scale, one relationship that most strikingly suggests itself is that between the civilisation of Europe and its climate between the fall of the Roman Empire and the end of the Middle Ages. As civilisation seems to wane, wax and wane again so, too, does climate seem to deteriorate, im- prove and then deteriorate once more. In both spheres the thirteenth century stands out as an optimum.

Here attention will be paid mainly to the effect of climate on history. But were one looking more parti- cularly at how climate change may be forced anthro- pogenically, one would concentrate, so far as this era is concerned, on the removal of temperate forest, especially in Europe itself and in China. Not only does deforestation release carbon dioxide into the global atmosphere. In addition, it may modify the water cycle and alter the albedo of the land surface, the proportion of any incident radiation immediately lost through reflection. The average albedo for short-wave solar radiation is 0.15-0.18 for deciduous forest, 0.09-0.15 for coniferous, but 0.18-0.25 for grass or cereal cover (Barry and Chorley, 1990).

In Europe, deforestation went ahead quite steadily within the bounds of the Roman Empire. Then it slowed down abruptly during the fifth and sixth centuries: the depths of what have traditionally been known as the 'Dark Ages'. In the seventh, it resumed, as witness the extensive clearance of the Ardennes (Latouche, 1981). However, the great phase of forest clearance in Europe was between 1050 and 1250, this against the background of steep population growth. Italy may have acted as something of a pacemaker on clearance (Fichtenau, 1991). German colonisation eastwards certainly played a part (Goudie, 1990).

In China, the population fluctuated erratically around the 60 million mark between the Han dynasty (206 Bc to AD 220) and the year 1600. However, the percentage living in South China (the Yangtse valley and southwards) rose from 8 in the middle Han to an estimated 49 around the year 1450. This shift betokened a very vigorous and sustained colonisation of the South with the indigenous peoples of the fertile valleys being partially assimilated but widely displaced (Kolb, 1971). Associated with this aggrandisement was exten- sive deforestation. All in all, it will be surprising if more

Climate change and human history 39

detailed climatic profiles do not afford confirmation that anthropogenic factors were contributing to global climate change.

THE ROMAN WORLD

As one turns to consider instead the impact of climate change, the first question that presents itself is one that has intrigued scholars for over 200 years. How far did climate change affect the struggle of the later Roman Empire against continual onslaught by barbarian tribes? Forays from that quarter had been something the ancient empires had had to live with for centuries. Witness the Great Wall of China begun in the third century Bc to ward off the Hsiung-nu nomads and the Huns. In the third century AD, however, the incursions into the Roman domain became massive, resulting inter alia in the devastation of much of Gaul. In the fourth, the situation stabilised awhile. In the fifth, the Roman Empire in the West was finally overwhelmed.

Gleaning evidence from writers such as Pliny the Elder and Strabo about such indicators as the distri- bution of vineyards, Hubert Lamb has concluded that the regional climate started to warm sometime before 100 BC. He has further inferred that, subject to some cold intermissions, this trend was sustained until AD 350 or thereabouts (Lamb, 1977a). This perspective stands rather diametrically in opposition to that adopted by the great classical scholars, French and English, of the eighteenth century. Their consensus was that, through- out the era of Rome on the wane, central Europe was appreciably colder than even in their own chilly times.

In his magisterial Decline and Fall of the Roman Empire, published in 1776-88, Edward Gibbon laid much stress on the progressive worsening of that vast polity's intrinsic defects, most of which would derive from climate fluctuation either very indirectly or not at all. Nevertheless, he did note how frequent and thick winter freeze-ups of the Rhine and the Danube had often afforded the barbarians 'a vast and solid bridge of ice' across which to deploy large armies. He also noted that reindeer flourished in the forests of Germany and Poland. He saw Europe then as akin to the Canada of his own day, presumptively that much colder because of more forest cover (Gibbon, 1970). Two generations later, Dr Thomas Arnold (famous not only as the headmaster who shaped at Rugby the English public school system but also as a classical historian) described the Alps of the Hannibal era as far colder and more snowladen than in Amold's own day (Arnold, undated). Judgement on the specific point is made harder by our still being uncertain through which alpine pass Hannibal's army entered Italy in the autumn of 218, thereby taking the Romans utterly by surprise. However, the more general proposition that the later Roman era was a cold one around the Mediterranean basin has lately been supported by a UNESCO sponsored scientific study. More particularly, this identifies as centuries of notably severe cold the first AD and, albeit less acutely, the fifth (Issar, 1992a).

If this inference of general cooling is valid, it carries the usual corollary that the climatic zones were dis- placed, however erratically and irregularly, towards the equator, and one thing this connotes is that the anti- cyclones characteristic of Scandinavia in winter could become stronger and more persistent, thereby drawing dry easterlies across much of the continent more regu- larly. Recent evidence of pronounced glacial recession across Scandinavia around AD 400 may corroborate this interpretation (Issar, 1992b).

Also, the notion of displacement fits well enough the received image of Roman North Africa as an exotic, and productive, 'granary' region for the Empire at large. That in its turn completes the picture of the Roman Empire as having been, as from the final conquest of Carthage in 146 ac, an essentially Mediterranean one. True, we are warned that the customary elements in the said image are either spurious or unproven (Shaw, 1981). But this may mean that it is extravagant, not that it is baseless.

Moreover, a prevalence of dry continental winters could be related to a notion entertained by at least two of the earlier students of climate change, Huntington and Brooks. It was that the closing stages of Imperial Rome witnessed more arid conditions across a broad swathe of territory extending through Germany and across much of Russia. Thus Ellsworth Huntington observed a continual shrinkage of the Caspian from 500 BC to AD 500, with an acceleration from the first century of the Christian era (Huntington, 1907a). Then in the fourth and fifth centuries, said C. E. P. Brooks, 'many German settlements were established on low ground, now swampy' (Brooks, 1926). But although Brooks was a bold and determined researcher, his chronologies were not always accurate in relation to modern practice. More generally, one has to say that the evidence about aridity in that era has remained incomplete so far as West Central Europe is concerned. For Italy there are few indications of very wet years between AD 200 and 550 (Lamb, 1982b). However, Italy's situation is ambiguous in relation to the above analyses. In any case, the evidence from these stands in contrast to the finding that Lake Van in Turkey tended to be 'abnormally' high from 250 to 750 (Issar, 1992c). Still, perhaps the most pertinent analysis is one done of peat bogs by the Soviet Academy of Sciences. This showed that, in a broad zone from Sweden through the Moscow region levels of annual precipitation in the fifth century tended to be 50-100 mm below the longer term norms (Chernavskaya, 1990).

So overall the signs are of aridity and cold sufficient to stimulate the v6lkerwanderung in inner Eurasia that bore so hard on the Roman and other settled realms. A well-recognised driving force was the migration of Hunnish tribes out of East and then central Asia. In 372, they marched across the lower Volga, in continua- tion of their long-term westerly shift. In 406, they impelled German tribes into a major surge into Gaul across a frozen Rhine; and only 4 years later Visigoths, also out to avoid them, sacked Rome itself.

40 N. B r o w n

TOWARDS AN OPTIMUM

Soon, however, one has started to move towards the 'Little Climatic Optimum' of the Middle Ages. A particularly pronounced rise in mean temperature was to occur between AD 900 and 1250, at least across much of the Northern Hemisphere. On the glacial margins of Greenland it may have been as much as 5 ° . Drift ice became rare around Iceland and the Denmark Strait after 800; and seems to have been virtually unknown there between 1020 and 1200 (Lamb, 1966). Freeze-ups of the Nile recorded in 829 and 1010 may confirm rather than negate this analysis. This is because they indicate the Azores anticyclone extending regularly into central Europe even in winter, thereby drawing north-easterlies across Egypt.

The usual tendency for a warming of the Northern Hemisphere to be especially pronounced in high latitudes would have displaced the polar front northwards and, at the same time, weakened the thermal gradient across it. Perhaps the initial onset of the more salubrious climate that resulted would have assisted the sixth century Irish monks whose primitive open boats made it to Iceland, probably to Greenland and maybe well down the American coast (Ashe et al. , 1971). More remarkable, however, is the subsequent expansion of the Vikings, moving by sea and land and doing so as warriors, traders and colonists. In 789, the first Viking raid burst upon what had become a very tranquil Anglo- Saxon England. Around the turn of that century, the Vikings reached Iceland; and the frontier society they established there was to lead the early medieval Nordic world as a model of parliamentary democracy (the Athling being formed in 930), of the rule of law and, in due course, of inspirational literature. In 985 or there- abouts, they started what became a colony of up to 10 000 people in what they saw as 'Greenland'. Shortly afterwards, they founded a small settlement in New- foundland. The artefacts they eyentually traded with the indigenous peoples have been discovered as deep into the Canadian Arctic as Ellesmere Island. Mean- while, one branch of the Eastern Vikings, the Rus, had founded Novgorod as their forward capital in 862. Other Rus were soon to found Kiev, a stepping stone to Constantinople.

Presumably assisted by stronger penetrations north- eastwards of the Gulf Stream Drift, Viking navigators also struck north. They discovered Jan Mayen and Spitzbergen and entered the Barents Sea. There is even a tradition in the Japanese archipelago that suggests a Viking visitation from whatever quarter. Yet the Vikings did not actually settle in the territory of Finn- mark (the furthermost province of modern Norway) until the thirteenth century (Stagg, 1952a). Nor did the Norwegian Vikings urbanise very much within their own country. Only eight settlements in medieval Norway were accorded the legal status of townships; and the most northerly of them was Trondheim (Lunde, 1985). Something eventually to consider in this connection might be the implications for the far North in winter of

any realignment towards the Pole of the Eurasian anti- cyclones.

Needless to say, climate is an aspect of change the established historians of the Viking era direct little attention to. On the other hand, they do seem agreed that land hunger played a part, especially in Norway. It did so by stimulating much strife between petty king- ships. That led to the desperate struggle of Harold Fairhair (c. 850-933) to consolidate central govern- ment; and this in its turn to many going into exile, not least to Iceland. So the underlying factor could be seen as rising population, due to climate change, in the valleys of the South. What does seem clear is that, unlike the Western European age of discovery from the fifteenth century, things were not triggered off by the advent of new marine technology (Kirkby, 1977). On the contrary, it was only after a desire to sea-rove had gripped them that the Vikings utilised and improved that which had long been round the North Sea (Foote & Wilson, 1980).

Yet dramatic and lasting though the Viking impact on so much of Europe was, it ought not to divert one too completely from processes of recovery intrinsic to what had been the Roman domain. Of especial interest in this connection is a debate generated by the out- standing economic and social historian of the medieval period, Henri Pirenne. Some measure of his standing is that, although himself Belgian, he was repeatedly in- vited to become the first editor of Annales (Burke, 1990b).

The relevant Pirenne thesis is that, all through the Dark Ages, the Mediterranean remained, even for the Germanic tribes, 'the very centre of Europe', especially as a m a r e n o s t r u m across which commerce was multi- laterally preserved and from which it regenerated. Then came the great expansion of the ArabS after the death of Mohammed in 632. In 711, they brought Islam to Spain; and were only to be checked at the battle of Tours in southern France in 732. After that, the argu- ment goes, the Mediterranean became 'a Moslem lake' and the centre of gravity of Christian Europe switched abruptly northwards: 'Without Islam the Frankish Empire would never have existed and Charlemagne, without Mohammed, would be inconceivable' (Pirenne, 1976).

Against that a number of writers have persuasively argued that the relative decline of the Mediterranean well preceded the Arab expansion. It has further been claimed that the ascendancy of the North was even- tually to be confirmed by a multiple breakthrough in agricultural practice: 'By the early ninth century all the major interlocking elements of this revolution had been developed: the heavy plough, the open fields, the modern harness, the triennial rotation--everything except the nailed horseshoe which appears a hundred years later . . . . The agricultural revolution of the early Middle Ages was limited to the Northern plains where the heavy plough was appropriate to the high deep soils, where the summer rains permitted a large spring planting and where the oats of the summer crop sup- ported the horses to pull the heavy plough. It was on

Climate change and human history 41

those plains that the distinctive feature both of the late medieval and the modern world developed' (White, 1976). Meanwhile, in the Eastern Roman Empire, still surviving under rule from Constantinople, the sixth century had ushered in a syndrome of catastrophe: drought, locusts, plague, foreign attack, urban unrest ... (Mango, 1980). In the light of what was said earlier about Mediterranean rainfall, none of this is easy to relate immediately to climate fluctuation around that basin then. But it is fair to surmise that early medieval warming helped to stimulate the rise of the North. Meanwhile, there seems to be a consensus that the Muslim explosion out of Arabia was fired by drought rather than plenitude (Lamb, 1982c).

More or less throughout Europe, medieval Christen- dom attained its zenith between 1100 and 1300. The tree line across northern areas was typically 80 m higher by then than it is today (Grace, 1989). Correspondingly, the frontiers of agriculture widely ascended 60 m. Great monastic orders were founded. In 1115, St Bernard established in Champagne the Abbey of Clairvaux, which was to be the headquarters of the Cistercians, an order dedicated to pushing the margins of farming outwards and upwards. Hundreds of towns were established during these centuries. By the year 1300, fourteen Christian universities had been established in Italy or Iberia; and another six in France or England (Matthew, 1985). In 1241 an axis was formed between Hamburg and Lubeck which became the basis of the Hanseatic League, the most obvious manifestation of how closely the chilly Baltic was rivalling the Mediter- ranean as a commercial thoroughfare. Granted, no Baltic cities even began to match the burgeoning during this time and beyond of Genoa and Venice but that owed much to a near elimination of Constantinople in the wake of the Fourth Crusade (1204).

Longer growing seasons encouraged the incipient Russian statehood to shift its centre of gravity, cultural and commercial, northwards to Novgorod (Milner- Gaillard & Dejesky, 1989). That shift may well have been crucial to the survival, during an infancy spent under the shadow of the Tatar branch of the Mongol expansion, of what was to become in our generation the Eurasian superpower. Moscow was founded in 1156.

A COOLING TREND

During the climax of this phase, rainfall tended to rise. Not least did it in inner Asia, around the Caspian and Lop Nor-- the very fluctuating lake near the main watershed in Sinkiang (Huntington, 1907b). Here again, falling infantile mortality as food became more plenti- ful can be presumed. So can a like trend with young foals. Together these effects will have given the Mongols the extra manpower and horsepower that allowed of the short but unparalleled orgy of military adventurism initiated by Genghis Khan (1162-1227).

If so, it was a product of climate change that was very comprehensively destructive at the time and largely catastrophic in its long-term connotations; and that is

even apart from the strong probability that marauding Mongol horsemen casually brought the main Black Death bacillus out of obscurity in the eastern Himalayas and gave it on the Asian steppes a base from which to radiate (McNeill, 1983).

One great tragedy was that the Mongols inflicted on a hitherto splendid 'Arab civilisation a blow from which it has never fully recovered'. They did so by the devastation of Mesopotamia and by leaving the way open for what proved to be five dreary centuries of Ottoman domination of the Arab peoples (Guillaume, 1954). One consequence must have been to retard the development of science and leave it too Eurocentric. Another certainly was that, whether through expansion or decline, the Ottomans were continually to be a source of international instability.

Arguably, too, another tragedy was that, although Russia did survive, it was not with its free spirit unimpaired. Indeed, various writers from Karl Marx onwards have inferred that this spirit was irrevocably subdued by Mongolian notions of blind submission to martial authority (Szamuely, 1974). It is fair to say, however, that many theories have been in contention about the roots of Russian/Soviet political culture (Bell, 1958), not a few of them influenced unduly by 'Cold War' stereotyping. It is a subject to address less tendentiously as the new historical dialogue develops between West and East.

While this moister phase was making itself manifest in inner Eurasia, mean temperatures were beginning what were to be five or six centuries of unsteady but insistent descent virtually worldwide, the era that became what is now meant by the 'Little Ice Age'. Increased vulcanism will have been a trigger. Between 1025 and 1250 there were remarkably few volcanic eruptions whereas the subsequent record became more like the longer term average together with some surges of high activity, notably from 1390 to 1490. Around Iceland, however, colder conditions seem to have set in during the 1180s. Except for one decade of intermission, 1212- 1223, abnormally cold seasons return sporadically through to the last two decades of the thirteenth century. Then many severe winters in quick succession were to cause repeated famine (Grove, 1988b). By early in the thirteenth century, the crossings to Iceland and Greenland were getting rougher. At the same time, the incidence of severe storms and sea floods around the North Sea basin increased markedly. In 1212, for instance, as many as 306000 were reckoned to have drowned in North Holland (Lamb, 1977b).

By the fourteenth century, the climate had become more characterised by extremes of all kinds (Lamb, 1987). In 1315, after summer rains so incessant that they were compared with the biblical flood, crops failed all over Europe, ushering in widespread famine. Hence- forward famine became more recurrent generally, while both tree lines and arable margins began to recede and villages to be abandoned. At the same time Europe lost the remarkable freedom from rampant diseases that it had enjoyed for nearly 600 years. Pulmonary tuber-

42 N. Brown

culosis, not quite an epidemic disease but a highly contagious one, markedly increased. Between 1348 and 1351, perhaps a 'third of the world' the Europeans knew died in the Black Death. Leprosy sharply declined but largely because more efficient killers were in ascent.

Other factors had accentuated the vulnerability of society to worsening climate. The population of Europe trebled between 950 and 1250 to reach about 80 million, perhaps the highest total for 1000 years. A wheat-based monoculture accentuated soil exhaustion (Gottfried, 1983). Transportation improved enough to spread disease but not sufficiently for the correction of harvest imbalances. As these strains aggravated the contradic- tions already deepening within the feudal system, the resultant social tensions found expression in the febrile utopian visions of such movements as the pastoureaux in France and the flagellants in and around the Rhine valley. As part and parcel of this negativism, anti- semitic outbursts also became more commonplace and vicious (Tuchman, 1978). On the constructive side, the development of a chronic labour shortage strengthened the economic position of the peasantry within the feudal class war. This explains the longer term success of uprisings like the Peasants' Revolt in England in 1381. Immediately, they might end in defeat and repression (Green, 1955). Eventually, they helped eliminate not just particular instruments of feudal control (e.g. poll taxes or ordinances for wage-fixing) but the whole idea of villeinage. However, these interactions are among those that closer climate analysis should do much to elucidate.

In Iceland the climate crisis came early and so did the political. Vicious feudalistic wars in the early 13th century favoured Norwegian intervention. From the 1260s, Norway established a careless yet oppressive suzerainty which gave way in 1380 to an even more constrictive Danish regime.

The Viking community in Greenland had died out by 1480, isolated from Europe since 1408 and lacking the adaptive resilience that had characterised its early years. Its members had locked themselves obdurately into pre-established patterns 'elaborating their churches rather than their hunting skills' (McGovern, 1981). In this they contrasted markedly with the Norwegians in Finnmark whose farming and fishing positively thrived on climatic adversity until the late eighteenth century (Stagg, 1952b). An explanation for this contrast may lie in the proposition that a people responds positively to environmental change provided that change is not impossibly acute.

THE RENAISSANCE AND TOMORROW

With the fifteenth century one enters into the Renais- sance. In the great age of optimism that ended with the First World War, the Renaissance was seen as a time of very genuine 'rebirth', a dramatic reassertion of everything positive and noble in the human spirit. Today our responses are more complex. Cultural historians regard it less as a sudden departure than as the accelerated evolution of intellectual and cultural

themes that can be traced back to the High Middle Ages, university development being one of them. Political and social historians stress how violent the two centuries after 1450 were, this at every level from widespread witchburning (Trevor Roper, 1969)~ to continental warfare, religious and geopolitical. Social philosophers attribute the terrible tensions this revealed partly to a widening gap between the cultural elites and the masses (Fromm, 1942; Weil, 1971). Environ- mentalists are conscious of how abruptly and extensively the world was thrown into ecological disequilibrium by the very sudden projection across almost every ocean of European imperialism (Wilkinson, 1973; Ponting, 1992).

Certainly the manifold complexities of this era will offer much scope for the assessment of climate impact, as and when the elaborated climate time series become available. From the Renaissance onwards, however, one has continually to be concerned with the impli- cations for Europeans and everybody else of climate change across the world at large.

Thus early in the fifteenth century, several centuries of technological advance in China culminated in succes- sive maritime expeditions (undertaken by up to 25 000 men) around the Indian Ocean as well as nearer home. Yet on the death in 1431 of Cheng Ho, the general who had led this great endeavour, the links he had forged with some 30 regimes were broken as the Ming emperors turned their attentions inwards leaving the realm of maritime expansion open to Europeans who were, in fact, largely relialat on technologies (notably the compass, printing and gunpowder) transferred from China. To some extent the Ming were concerned to secure their exposed northern borders and also to curb opium imports (Giles, 1964). No doubt, too, they were fearful lest the new social forces technology could set in train were to destabilise their regime internally (McNeil, 1985). But perhaps they were concerned as well about the agricultural crisis then developing across China in association with a much more erratic climate (Jin-Qi Fang, 1992). Weighing that consideration would be important to an understanding of world history (and so, derivatively, European) at the time and for several centuries thereafter.

But for now one can attempt no more than what has been essayed above, a strictly Eurocentric view of climate impact in the 1000 years or so from the decline of Rome. One inference that can readily be drawn is that this impact tends to be most pronounced on societies that are already marginally placed: on the tribal or colonial fringes of settlement, near upland or polar or desert limits, and so on.

It is an inference that has ominous implications for the twenty-first century, no matter how great its technical virtuosity or its global awareness. For against the background of growth of population and rise in material expectations, that century threatens to be desperately marginal in several critical respects. It does in relation to the control of the wherewithal for armed conflict. It does in relation to its whole resource

Climate change and human history 43

base; and to a very diverse spectrum of environmental pressures. It does apropos social cohesion. It will not be a century that can afford to be casual about the added menace of climate change.

REFERENCES

Arnold, T. (undated), Hannibal's Passage of the Alps. In Half Hours With Standard Authors, ed. L. Valentine. The Library Press, London, pp. 133-64.

Ashe, G. et al. (1971). The Quest for America. Praeger, New York, pp. 271-2.

Barry, G. R. & Chorley, R. J. (1990). Atmosphere, Weather and Climate. Routledge, London, Table 1.3.

Bell, D. (April 1958), Ten theories in search of Soviet reality. Worm Politics, 10, 327-65.

Brooks, C. E. P. (1926). Climate Through The Ages. Ernest Benn, London, p. 340.

Budyko, M. I. & Izrael, Yu I. (ed.) (1991). Anthropogenic Climatic Change. The University of Arizona Press, Tuscon, Chapter 10 (translated from the Russian by the authors).

Burke, P. (1990a). The French Historical Revolution. Polity Press, Cambridge, p. 61.

Burke, P. (1990b). The French Historical Revolution. Polity Press, Cambridge, p. 21.

Chernavskaya, M. M. (1990). Moistening changes on the Russian plain during the historical period. In Climatic Change in the Historical and Instrumental Period, ed. R. Bradzil. Masaryk University, Brno, pp. 130-3.

Fichtenau, H. (1991). Living In The Tenth Century, Univer- sity of Chicago, Chicago, p. 338.

Foote, P. C. & Wilson, D. M. (1980). The Viking Achieve- ment, Sidgwick and Jackson, London, pp. 260-1.

Fromm, E. (1942). Escape from Freedom. Kegan Paul, London, p. 40.

Gibbon, E. (1970). The Decline and Fall of the Roman Empire, Vol. 1. J. M. Dent, London, Ch. IX.

Giles, H. A. (1964). A Chinese Biographical Dictionary. Taipei, p. 112.

Gottfried, R. S. (1983). The Black Death. Robert Hale, London, pp. 15-25.

Goudie, A. (1990). The Human Impact on the Natural Environment. Blackwell, Oxford, p. 37.

Grace, J. (1989). Tree lines. In Forests, Weather and Clim ate, ed. P. G. Travis et al. Royal Society, London, pp. 233-45.

Green, V. H. H. (1955). The Later Plantagenets, Edward Arnold, London.

Grove, J. M. (1988a). The Little Ice Age. Routledge, London, p. 3.

Grove, J. M. (1988b). The Little Ice Age. Routledge, London, p. 22.

Guillaume, A. (1954). Islam. Pelican, Harmondsworth. Huntington, E. (1907a). The Pulse of Asia. Houghton and

Mifflin, Boston, Ch. XVII. Huntington, E. (1907b). The Pulse of Asia, Houghton and

Mifflin, Boston, p. 287. Issar, A. S. (1992a). The Impact of Climate Variations on

Water Management Systems. Ben Gurion University of the Negev, Sede Boker, Fig. 1.

Issar, A. D. (1992b). The Impact of Climate Variations on Water Management Systems. Ben Gurion University of the Negev, Sede Boker, Fig. 11.

Issar, A. S. (1992c). Tire Impact of Climate Variations on Water Management Systems. Ben Gurion University of the Negev, Sede Boker, Fig. 5.

Jin-Qi Fang (July August 1992). Establishment of a data

bank from records of climatic disasters and anomalies in ancient Chinese documents. In International Journal of Climatology, 12, 499-519.

Kirkby, M. H. (1977). The Vikings. Phaidon, Oxford, pp. 56-7. Kolb, A. (1971). East Asia. Methuen, London, p. 47. Le Roy Ladurie, E. (1971a). Time of Feast, Times of Famine.

George Allen and Unwin, London, pp. 88-9. Le Roy Ladurie, E. (1971b). Times of Feast, Times of Famine.

George Allen and Unwin, London, p. 293. Lamb, H. H. (1966). The Changing Climate. Methuen, London,

p. 64. Lamb, H .H. (1977a). Climate, Present Past and Future, Vol. 2.

Methuen, London, Ch. 12. Lamb, H. H. (1977b). Climate, Present Past and Future~ Vol. 2.

Methuen, London, Tables 13, 3 and 13, 4. Lamb, H. H. (1982a). Climate, History and the Modern

World. Methuen, London, pp. 118-2 I. Lamb, H. H. (1982b). Climate, History and the Modern

World. Methuen, London, Fig. 59. Lamb, H. H. (1982c). Climate, History and the Modern

World. Methuen, London, p. 160. Lamb, H. H. (May 1987). What can historical records tell

us about the breakdown of the medieval warm climate in Europe in the fourteenth and fifteenth centuries--an experiment. Beitrdge zur Physik der Atmosphgire, 60(2), 131-43.

Latouche, R. (1981). The Birth of the Western Economy. Methuen, London, p. 189.

Lunde, Oivind (1985). Archaeology and the medieval towns of Norway. Medieval Archaeology, XXIX, 120-35.

McGovern, T. H. (1981). The economics of extinction in Norse Greenland. In Climate and History, ed. T. M. L. Wigley et al. Cambridge University Press, Cambridge, pp. 379-403.

McNeil, W. (1983). The Pursuit of Power. Basil Blackwell, Oxford. pp. 48-9, 60.

Mango, C. (1980). Byzantium, The Empire of a New Rome. Charles Scribners, New York, Ch. 3.

1Hatthew, D. (1985). Atlas of Medieval Europe. Equinox, Oxford, p. 165.

Milner-Gaillard, R. & Dejesky, N. (1989). Cultural Atlas of Russia and the Soviet Union. Equinox, Oxford, p. 24.

Pirenne, H. (1976). Medieval cities. In The Pirenne Thesis, ed. Alfred F. Havighurst. D. C. Heath, Lexington, pp. 1-26.

Ponting, C. (1992). A Green History of the WorM. Penguin, Harmondsworth, Ch. 7.

Shaw, B. (1981). Climate, environment and history: the case of Roman North Africa. Climate and History, ed. T. M. L. Wigley et al. Cambridge University Press, Cambridge, pp. 379-403.

Stagg, F. N. (1952a). North Norway. A History. George Allen and Unwin, London, pp. 61-2.

Stagg, F. N. (1952b). North Norway. A History. George Allen and Unwin, London, pp. 75, 104-20.

Szamuely, T. (1974). The Russian Tradition. Seeker and Warburg, London, Ch. 2.

Yrevor Roper, H. R. (1969). The European Witch Craze of the 16th and 17th Centuries. Penguin, Harmondsworth.

Tuchman, B. (1978). A Distant Mirror. The Calamitous Four- teenth Century. Alfred A. Knopf, New York, Ch. 2, 5.

Weil, S. (1971). The Need For Roots. Harper and Row, New York, p. 45.

White, L. (1976). The northward shift of Europe's focus. In The Pirenne Thesis, ed. Alfred F. Havighurst. D. C. Heath, Lexington, pp. 166-8.

Wilkinson, R. G. (1973). Poverty and Progress: An Ecological Model of Economic Equilibrium. Methuen, London.

A brief history of climate change and conflict

BY JAMES R. LEE | 14 AUGUST 2009

Article Highlights

The interaction between climate change and conflict started as early as 35,000 years ago.

The Neanderthals, Vikings, and Mayans all benefited and suffered from a changing climate that affected resources such as water, game, and agriculture.

By analyzing historical case studies of climate and societal collapse, we can identify a set of discernible lessons for today.

In recent years, many foreign affairs experts have attempted to demonstrate the linkages between climate change and the social tensions that can lead to conflict. While critics may believe this is simply a fad in international affairs, history suggests otherwise. Over the last few millennia, climate change has been a factor in conflict and social collapse around the world. The changing climate has influenced how and where people migrate, affected group power relations, and provided new resources to societies while taking away others. Such circumstances cause large-scale alterations in lifestyles and illustrate pathways from climate change to conflict.

Because climate change can be a contentious subject, it's worth taking a moment to answer some basic questions and put forth a series of assumptions. First, what is climate change? The Intergovernmental Panel on Climate Change, the most authoritative source on the subject, assesses climate change by measuring changing temperatures and precipitation. Since trends in temperature often (but don't always) drive trends in precipitation, scientists consider temperature a more robust and stable measure of climate change. But in simplest terms, climate change is the long-term change in the patterns of these two meteorological characteristics. Second, does climate change affect the world the same everywhere? In fact, climate change is a heterogeneous phenomenon and produces different outcomes in different places. The subsequent case studies demonstrate that a changing climate can have acute regional effects such as near the equator or North Pole, the "hot" and "cold" zones, respectively. In both cases, "hot" and "cold" conflicts demonstrate how rising and falling temperatures have had different impacts on human survival and prosperity.

Finally, a fact: The relationship between climate and conflict isn't as simple as cause and effect. Instead, climate change events--such as temperature shifts of a few degrees or a precipitation change of a few inches--contribute to conflict gradually over the long term. Because the climate has been changing for millennia, it's possible to look at the past for examples. Do we see cases of climate-conflict interaction when rates of climate change have diverged? In fact, it's possible to assess historical events and records in order to construct pictures of how climate affects conflict. Historical case studies, therefore, allow us to identify three paths from climate change to conflict: sustained trends, intervening variables, and the need for conflict triggers.

Consider each of the paths. The first is that conflict has the potential to emerge after a sustained period of divergent climate patterns. While people can survive aberrant, short-term climate change by exploiting existing or stored resources, this strategy has temporal limits. On this particular path the issue isn't one of surviving an especially fierce rain or

harsh winter, but the cumulative effects of many fierce rains and many harsh winters. Next, climate change alone won't cause conflict but, along with other factors, will contribute to and shape it. It's one variable among many others, such as cultural, economic, or demographic factors. Last, unless a society learns to adapt to sustained climate change, its wealth will decline and its social fabric will weaken with each passing year. But even if a society faces these environmental challenges, a trigger--such as an assassination, extreme natural event, or random act of group violence--is usually required to ignite violent conflict.

We can, nevertheless, draw lessons from natural (climate) and human (conflict) interaction that may be transferable to today's global climate challenge by considering three periods in human history: human pre-history (the time before written records), the medieval climate optimum, and the "little ice age."

During human pre-history, probably in today's Middle East region, humans found plentiful game and encountered the Neanderthals. Over time, humans pushed into Europe and forced the Neanderthals further north into the less hospitable parts of the continent, where game wasn't as abundant and temperatures were much colder. Although the Neanderthals had survived several ice ages over the course of hundreds of thousands of years, they couldn't survive both an ice age and the humans, who enjoyed advanced weaponry and social organization. Of course, theories about the end of the Neanderthals are controversial and unresolved. However, there is no question that climate change provoked the interaction of human societies and the Neanderthals and subsequently led to conflict.

The medieval climate optimum lasted from 500 to 1000. It brought about a period of sustained progress in Europe as warmer conditions allowed for longer growing seasons in the largely agricultural societies. However, conditions elsewhere were quite different. Between 700 and 900, rainfall in China was scarce due to weak summer monsoons that failed to develop over the Pacific Ocean. Gerald Haug and other researchers haveconcluded PDF that famines caused by the failed monsoons resulted in peasant revolts and fueled the intrastate conflict that drastically weakened, and then led to the complete collapse of, the Tang Dynasty.

Across the globe in North America, the Mayans had settled in the lowlands around 8000 BC and began practicing large-scale farming as early as 2000 BC. By the beginning of the medieval climate optimum in AD 500, the population was nearly 14 million, making it one of the largest centers of civilization anywhere. But the thriving Mayan cities began to experience diminished long-term rainfall patterns. Dry conditions began in 760 and, after a 50-year wet period, drought again set in about 860. Another drought followed in 910. The boom-and-bust cycles of rainy and dry periods contributed to eras of both growth and decline. Technology, population sizes, and agricultural intensity overwhelmed the land. Yields declined with the dry conditions and these structural incongruities led to ongoing wars between Mayan city-states that eventually contributed to their collapse.

The warming in Central America that was disastrous for the Mayans was, on the other hand, fortunate for the Vikings. Warmer temperatures in the north meant their land was more hospitable to live on. Complex push-and-pull factors allowed the Vikings to expand their settlements from Scandinavia to Iceland, Greenland, and later Newfoundland. It was in

Newfoundland that they encountered Native Americans. The Native Americans, too, perceived the warmer climate as a new opportunity and fought to control the increasingly abundant land. The Vikings and Native Americans would alternately trade and fight with each other throughout the Vikings' time in Newfoundland.

Enter the "little ice age"--a period marked by abnormally cooler temperatures. Scholars differ on the exact duration of this period; some researchers believe it started as early as 1000 in certain northern regions, whereas other historians, such as noted scholar Brian Fagan, believe it lasted from 1300 to 1850. Regardless, when the climate turned cold, the Viking colonies that had flourished in the warmth of the medieval climate optimum collapsed in Newfoundland (which they had abandoned because of ongoing conflict with the Native Americans). The western Greenland colony was the next to collapse, and around 1350, coinciding with the time of the Black Death, the eastern colony also began to decline. It survived only into the early 1500s.

During the same period, the Anasazi, a hunter-gatherer people who over centuries settled into a sedentary lifestyle, lived along the rivers of what is today the Southwest United States. With gradual improvements in technology and a beneficial climate, their population grew. But the little ice age brought a period of long-term drought, and Anasazi population growth exceeded its resource base. Timber, game, and other resources had to be imported from neighboring areas. The Anasazi had survived a long-term drought and many smaller ones in their long history. So why were they unable to cope with the little ice age drought? Like the Mayans, Anasazi city-states came into conflict as resources dwindled.

Because of its recency, scholars are able to theorize more completely about the little ice age. There were sharp extremes in temperature during the period. In fact, it had two temperature low points, one during the late 1400s and early 1500s, and another during the late 1700s and early 1800s. During the latter extreme cold period, a catastrophic geologic event occurred. In April 1815, Mount Tambora, on the Indonesian island of Sumbawa, threw a massive amount of volcanic dust into the atmosphere in one of the largest volcanic explosions in modern times. The volcanic dust travelled worldwide and blocked the sun's rays, lowering temperatures, especially in the northern hemisphere. The period of extreme cold, coupled with this sudden volcanic eruption, produced the "year without summer."

The year without summer illustrates two different stories--one in Europe and another in North America. In Europe, the cold forced people to migrate. But land across the continent was relatively densely settled, so conflict often came with population movement. On top of this structural problem, the Napoleonic wars culminated shortly before, sapping the vitality and stability of the economy. People were already suffering through instability, but the cold compounded social strife. There were no "pressure valves," such as open land, for populations under temperature-induced duress. The result was social upheaval, riots, and disease.

The cold was just as bad in eastern North America and Canada, where most summer crops were lost. But unlike Europe, there was ample land for climate migrants west of the United States. The building of the Erie Canal in 1817 gradually opened up a westward route for easy migration. Thus, where in Europe people lacked the pressure valve necessary to cope with

difficult times, in North America the valve not only existed but its exploitation was encouraged.

These are just a few accounts that allow us to synthesize history into a set of discernible, often regionally specific, lessons on conflict manifestations for today. The impact of climate change is obviously differential in the "hot" and "cold" wars, where trends may reward one part of the world while punishing another--for example, in terms of economic subsistence. In particular, four lessons of conflict emerge from the aforementioned cases that may have application for today's challenges:

The decline or growth in general resources, such as arable land and fresh water, can cause significant societal impacts. During extremely warm periods in the equatorial zone, marginal lands gradually lose resource assets. This is evident in the drift of the Sahara Desert southward into sub-Saharan Africa, which has occurred over many millennia and continues today. At the same time, in the polar zone, lands and their resources can become more abundant. It was because of this that the Vikings were able to survive for 500 years in Greenland before the little ice age.

Conflict can exist between societies or within them. By nature, the "hot" and "cold" war zones show divergent paths. In the polar areas, "cold" conflict emerges between states seeking to exploit the new resources that warming makes available. In the equatorial area, which includes many deserts, conflict erupts over declining resources, especially in warmer periods. Such livelihood conflicts often transcend borders and lead to migration. This is particularly evident for historic peoples such as the Mayans and the Anasazi, but also in today's conflicts in the North African Sahel and Sudan's Darfur region.

There is a full menu of climatic causes of conflict, depending on where the conflict is ("hot" and "cold" war areas), the type of climate change (temperature or precipitation), and the trends in the patterns (increasing or decreasing). In the polar zone, the change in temperature indirectly drives conflict behavior and the impact on inhabitable land is most important. In the equatorial zone, change in precipitation patterns is clearly a major driver, but temperature changes also can influence evaporation rates. Here, water is most important.

The resiliency of conflict is different in the "hot" and "cold" zones. In the polar zone, conflict is episodic; wars come and go with changes in temperature. In the equatorial zone, conflict is more gradual and continuous. "Hot" wars often stretch on as human population growth and changes in habitat tend to exacerbate changing climate conditions. Simply put, they are "cold" wars of opportunity versus "hot" wars of desperation.

Today, we see the manifestations of climate change slowly emerging in melting glaciers and drying fields. We need to imagine how changes in climate will create possibilities for conflict, using these historical lessons as guides. It's important to note that reacting to the challenges of "hot" and "cold" wars will require different strategies. The past is a good guide, but new types and modes of conflict emerging from climate change are to be expected. In that respect, finding historical examples is easier than contemplating how climate change and conflict will create new models of interaction and present new challenges in the future. After all, the idea that climate change causes conflict is not revolutionary, but evolutionary.

AUTHOR BIO

James R. Lee

Lee is the associate director of the Center for Teaching Excellence and a professor in theSchool of International Service at American University. He has extensive experience as an analyst of international trade, environmental policy, and security issues. He is the author ofClimate Change and Armed Conflict: Hot and Cold Wars and Exploring the Gaps: Vital Links between Trade, Environment, and Culture.

RELATED CONTENT = SUBSCRIPTION REQUIRED

Copyright © 2013 Bulletin of the Atomic Scientists. All Rights Reserved. Terms of Use.

Remote Address: 128.114.25.218 • Server: www.thebulletin.org (205.186.130.136) Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.31 (KHTML, like Gecko) Chrome/26.0.1410.64 Safari/537.31

http://www.thebulletin.org/web-edition/features/brief-history-of-climate-change-and-conflict