Anthropogenic climate change is the defining challenge of this century. Understanding how societies respond to it can help us anticipate and mitigate its consequences. Most academic literature to date has focused on the short-term effects of weather shocks and natural disasters (see review by Vally Koubi). This approach has generated important insights, but does not speak to the consequences of gradual and irreversible climate change.
In this post, I advocate a more historic and long-run approach to understanding the effects of climate change. Weather anomalies and climactic shifts have been common throughout history. Examining their impact, in all of its complexity, can enrich our understanding of how societies adapt to a changing climate. It can also help us identify tipping points – both environmental and societal – that generate systemic change.
I review recent work on a fascinating period of European history: the Little Ice Age that lasted approximately from the early Renaissance to the Enlightenment. These studies show not only the disastrous consequences of a changing climate, but also the remarkable resilience and ingenuity societies can muster. I also discuss the challenges of attributing political and economic outcomes to the weather.
Did the Little Ice Age Even Happen?
From the fifteenth to the eighteenth centuries, temperatures dropped on average by 1.8-3.6 degrees Celsius. The River Thames froze over nine times in the 17th century alone (see image of a frost fair in London). The weather also became unpredictable, with prolonged cold spells, excessive precipitation, droughts, and frequent storms. The so-called Little Ice Age is generally attributed to a combination of lower solar energy, changing patterns of atmospheric circulation, and heightened volcanic activity, which prevented the sun’s radiation from reaching the earth. Contemporaries blamed it on the configuration of stars and planets, divine punishment, and witchcraft.
Documenting the Little Ice Age is no simple task. Thermometers were not invented until the 17th century. To analyze climate in the more distant past, climatologists have examined “natural archives” — ice cores, tree rings, cave deposits, and lake sediments. Historians have also studied ship logs, letters, diaries, and other written accounts, which help to link broader climactic trends to local weather. These sources offer subjective but vivid images. For example, a messenger approaching Istanbul in March 1640 experienced “such a frost that I caught two frozen birds on the way simply with my own hand.” In England in 1658, the winter was so severe that “the crow’s feet were frozen to their prey; islands of ice enclosed both fish and fowl frozen, and some persons in their boats.” Indirect evidence of the changing climate was also found in the prevalence of winter landscapes and the stunting of human growth.
The precise timing and scope of the Little Ice Age is disputed. Not everyone even agrees that the cooling occurred. In 2013, Kelly and Ó Gráda controversially argued that the Little Ice Age is a product of a statistical artefact called the Slutsky effect: the practice of smoothing data on European weather with a moving average prior to analysis generated an appearance of cooling. As for the freezing of the Thames? It may have been caused by the clumsy arches of the London Bridge, which dammed up the river. (See Kelly and Ó Gráda’s article and their rebuttal by Sam White.)
Since then, other research has argued that while there was cooling, it was punctured by multiple periods of milder weather and confined to specific regions. For instance, recent global palaeoclimate reconstructions by Neukom et al. suggest that only northwestern Europe and southeastern North America experienced colder temperatures in the 17th century.
The Little Ice Age and Crises in Europe
Taking the Little Ice Age as a given, historians were the first to marshal qualitative evidence to study its consequences for states and societies. Many of these early studies focused on the worst-case scenarios and adopted a broad-brush approach. For example, in Global Crisis: War, Climate Change and Catastrophe in the Seventeenth Century, Geoffrey Parker argues that the succession of extreme climate events, famines, and epidemics that started in 1618 killed up to one-third of Europe’s population.
Conclusively establishing causality between climactic anomalies and macro-level political crises is extremely difficult. This is why studies in quantitative social science took a narrower frame, exploring the relationships between weather shocks and smaller episodes of violence. For example, De Juan and Wegenast show that colder temperatures increased the frequency of food riots in England, but only in the 18th century. Problems of interpretation remain even in these narrower studies. For instance, Emily Oster found a link between cooler weather and an increase in witchcraft trials. She argues that lower temperatures produced food shortages and the search for scapegoats. However, Leeson and Russ (2018) attribute the increase in witch trials to the competition of Catholic and Protestant churches for influence in concessionally contested regions rather than the weather.
Conflicts and crises may be easier to spot because they leave more traces in the written record; they also resonate with growing anxiety about our warming atmosphere. However, as Dagomar Degroot and his coauthors argue in this recent Nature article, resilience in the face of an adverse climate has been the rule rather than the exception. Examining a broader range of outcomes, combining multiple methods, and adopting a longer-term view is particularly important for understanding the role of institutions, technology, and culture in human adaptation to a changing environment.
How Europeans Adapted to the Changing Climate
If we look closely, the Little Ice Age also supplies many examples of ingenuity and resilience. Societies learned to diversify their crops, build storage facilities, manage flood risk, or trade with others. Necessity was the mother of invention. Using data from over 5,000 documented innovations and discoveries over 500 years of European history, De Dreu and van Dijk find that cold spells increased rates of innovation.
Thanks to their innovative spirit and geography, some societies grew wealthier and more developed despite the adverse climatic conditions. In The Frigid Golden Age: Climate Change, the Little Ice Age, and the Dutch Republic, 1560–1720, Dagomar Degroot argues that the Dutch prospered by developing new technologies and ways of living, including more sophisticated ship building and the construction of canals for transportation. Importantly, the Dutch economy did not depend on agriculture and the republic imported much of its food. Colder and stormier seas wound up (literally) adding wind in the sails of the Dutch fleet. Degroot also offers a fascinating account of how the shifts between westerly and easterly winds influenced the outcomes of the Anglo-Dutch battles in 1652–1688, though he is careful to emphasize that “no wars were ever won or lost solely because of climate change” (p. 247).
Economic pressures and resource scarcity also increased international trade. Maria Waldinger analyzes panel data for 2,120 European cities, combining temperature reconstructions with data population size, agricultural yields and prices, and trade from the Registers of the Sound Toll. As expected, she finds that cooler temperatures shortened growing seasons and thus reduced agricultural yields and increased grain prices. Reduced agricultural productivity, increased mortality, and outmigration all lowered urban populations across Europe. At the same time, cities responded to climate change by increasing their volume of trade, as measured by the number of incoming ships and total taxes paid to the Danish government on goods arriving in each port (yet another way in which the Dutch benefited during the Little Ice Age). Waldinger also finds that regions less affected by temperature changes adjusted their land use to take advantage of their comparative advantage in agriculture.
This strategy was pursued by early modern Poland as well, where the amount of land under cultivation increased as forests were cleared to cultivate rye for export. As discussed in the PNAS article by John Haldon and co-authors, the Polish-Lithuanian Commonwealth experienced an economic and cultural revival as a result. In the mid-seventeenth century, however, the destructive wars with Sweden and Russia reduced the population of the Commonwealth by a third, shrank the amount of cultivated land, and eventually crippled the economy. The economic and cultural decline of the Poland seems to have been caused not by the cooling temperatures, but by the convergence of economic and political crises. It is up for debate whether environmental challenges were the last straw.
Challenges and Opportunities in Studying Historic Climate Change
Statistically significant relationships between climate shocks and historical outcomes are not sufficient to prove causality. The use of the weather as an exogenous source of variation is also problematic. Frequently used as instruments for income shocks, grain prices, institutional weakness, or migration patterns, weather shocks may be bad instruments (see Ali’s recent post here). The exclusion restriction is likely to be violated given the many outcomes affected by the weather. Relatedly, our weather measurements may themselves be a product of development or state capacity (see this article by Schultz and Mankin). Paleoclimate data are immune to this, but such data are less precise and require expertise in natural sciences to interpret. Ideally, we would combine data from many sources to develop more accurate and high-resolution measures of weather patterns and consider multiple channels through which weather affects human livelihoods.
Most importantly, economic and political crises have multiple causes, so explaining their occurrence requires integrating endogenous factors other than the weather. While the link between climate and the upsurge in conflict in 17th century Europe sounds fascinating, for example, it is incomplete without considering the Protestant revolution, the rise of the modern state, changing agricultural and military technology, and other endogenous variables.
Given that the current challenges of climate change are global in nature, scholarship on historical, long-run effects would also benefit from a wider geographic scope. Our understanding of the consequences of the Little Ice Age, for example, is largely confined to western and northern Europe. There is fertile ground to study how the Little Ice Age impacted societies in central and eastern Europe, which exported grain to western and northern Europe during that time.
 The term was coined by geologist F.E. Matthes in 1939, though it originally referred to the entire 4,000-year period of mountain-glacier expansion and retreat.
 Reconstructing past climate is a complex and imprecise process, and estimates vary by study and period.
 Quoted in Geoffrey Parker, Global Crisis, p. 5.
 John Evelyn quoted in Parker, Global Crisis, p. 6.
 The Toll was levied by the King of Denmark on shipping through the strait between Denmark and Sweden.