The ORE of Politics will be available for subscription in late September. Speak to your Oxford representative or contact us to find out more.

Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, POLITICS ( (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy).

date: 20 August 2017

The Global Spread and Contraction of Democracy: A Co-Evolutionary Approach

Summary and Keywords

Much of the literature on international democratic diffusion appeals to mechanisms—competition, learning, emulation or socialization, and coercion—that typically are treated as competing and theoretically separate. All four, however, fit within a co-evolutionary framework, that is, one integrating the concepts of variety, retention, and selection of traits (in this case, regime type). Competition, learning, and emulation are not mutually exclusive and all find support in the large literature on cultural and social evolution. Coercion may seem anti-evolutionary, inasmuch as it implies design and implementation by a powerful rational actor (state, international institution, etc.), but co-evolution can accommodate coercion as well. In co-evolution, agent and environment evolve together: an agent shapes its environment (engages in niche construction), and that reshaped environment alters the fitness of the agents’ traits. A powerful democracy can alter its social and material environment so as to increase the fitness of its own regime. Co-evolution can provide a framework to integrate mechanisms by which democracy and other regime types spread and contract across time and space, and hence can aid empirical research on the effects of global power shifts, including the rise of China, on the fate of democracy in various regions around the world.

Keywords: diffusion, democratization, evolution, learning, hegemony, socialization, empirical international relations theory


What explains the spread and contraction of various regime types across time and space? Why did liberal democracy spread after the 1970s in Europe, Latin America, Asia, and Africa (Huntington, 1991)? A large literature, growing out of some claims of Kant from the late 18th century and the empirical finding that the past two centuries have seen episodic but unmistakable growth in the number of liberal democracies, finds that such states enjoy advantages in international relations. Liberal democracies do not fight one another, allowing them to reap the benefits of peace among a growing number of states (Doyle, 1983; Owen, 1997; Russett, 1992); they win most of the wars they do fight (Lake, 1992; Reiter & Stam, 2002); they trade more with one another; they join and remain in more international institutions (Russett & Oneal, 2001); more generally, they are better at keeping international commitments (Lipson, 2004; Martin, 2000); they enjoy advantages in coercive international bargaining owing to their superior ability to generate audience costs (Fearon, 1994) or their transparency (Schultz, 2001). On the other hand, a growing number of empirical studies claim that hegemons influence over what regime type predominates in their regions. Democracy is more likely to spread across a region that has a democratic hegemon (Cederman & Gleditsch, 2004; Gleditsch & Boix, 2011; Gunitsky, 2014; Gleditsch & Ward, 2006). The same may be true of other regime types (Gunitsky, 2017; Owen, 2010), and analysts have noted worries that somehow the rise of nondemocratic China will weaken democracy in East Asia (Kurlantzick, 2016; Lee, 2015).

The future of democracy is important in a number of regions of the world and, insofar as the Kantian literature is correct, for security and cooperation the world over. But regarding that future, we have here two theses that point in opposite directions. In the case of China and East and Southeast Asia, an argument from hegemony would imply that China’s rise will mean fewer democracies and more authoritarian states. As literature on authoritarian diffusion would put it, China could be a “black knight” (Ambrosio, 2009; Risse & Babayan, 2015; Vanderhill, 2013). On the other hand, the Kantian literature would suggest that any Chinese effect will be small and temporary, and indeed, that China itself will eventually succumb to the selection advantages that accrue to democracies and become one itself.

This article does not try to settle this question. Instead, it aims to make the inquiry more productive by proposing a way to synthesize the theories behind these two theses by means of co-evolutionary logic. The Kantian thesis connects more obviously to evolution: it explains the growth of the number of democracies with the advantages of democracy in today’s international system. More broadly, some historians consider Kant a precursor to Darwin (Lovejoy, 1910). An evolutionary account of the international spread (or contraction) of democracy would stress the spread of ideas and institutions by learning. Agents wanting to preserve and extend their power are boundedly rational and hence do not invent regimes out of whole cloth every time a new problem appears. Instead, agents inherit ideas and regimes and learn from experience—their own and others’—what best helps them reach their goals. Agents who refuse to learn lose power to those who do; thus sometimes evolution produces revolution. Regimes are thereby selected and retained in a set of states according to their fitness. In other words, agents adapt to their environments, and environments have a strong role in selection.

Hegemony, on the other hand, seems anti-evolutionary at first glance, inasmuch as the hegemon by definition has more power and hence more agency. The limiting case would be a hegemon able to design its region by imposing regimes, national borders, and so forth, on its region. Of course, such extreme hegemonic control is impossible even for the most powerful empire. Whatever hegemony China or its leaders exert is instead agency adapted to various constraints. When we see agents adapting to their environment, we should think about evolution. Evolutionary theory has always recognized the agency of units (organisms or otherwise), but in recent years some theorists have stressed how agents use that agency to shape their environment, that is, engage in “niche construction” (Odling-Smee, Laland, & Feldman, 2003). The constructed niche, in turn, may shape agents by selecting for certain traits. Agents and environments co-evolve, each shaping the other over time (Lewontin, 2001).1 Niche construction may be intentional or not; it is a function of power. Either way, inasmuch as the shaped environment still selects for certain traits in agents, niche construction is not an escape from or a thwarting of evolution but a participation in co-evolution. The general implication of co-evolution for international or regional hegemony is that it is constrained but consequential. Even in a world where regime type R appears to enjoy overall selection advantages, rulers of a hegemon of regime type S could suppress those advantages in their own country and region, directly and indirectly, and thereby extend the life of S.

Co-evolution is not a deductive theory that makes point predictions. It is, rather, a paradigm that aids theory and research by identifying agents, constraints, and mechanisms. Co-evolution should not be pitted against the main international relations (IR) theories of realism, liberalism, and constructivism, for co-evolution can accommodate propositions from each of them (e.g., coercion from realism, institutional constraints from liberalism, and learning from constructivism). To explain outcomes, co-evolutionary logic needs to be supplemented with descriptions of the properties and preferences of actors and the properties of environments. Co-evolution is an obvious framework, however, for the analysis of complex systems with significant unintended consequences and feedback effects (Jervis, 1997, pp. 48–50). It has the potential to help explain the uneven spread of democracy across time and space. Social scientists have found evolutionary logic fruitful in understanding complex systems of culture, language, and economics, among others.

Co-evolution implies that rulers and aspiring rulers adapt to their environments by learning from observation in their own state and other states. They attend to what appear “best practices” because their rationality is bounded; they have limited cognitive and material resources and must rely on rules of thumb, authorities, and other shortcuts (Simon, 1982). In particular, elites tend to try to copy states that win important wars, sustain economic growth, and maintain domestic stability over time. Elites copy not only the policies of these successful states but also their basic institutions or regimes. In some periods of history one regime appears to elites nearly the world over as best; in the 1930s it was fascism, in the 1980s and 1990s liberal democracy. But because change is costly and rationality is bounded, agents sometimes resist or cannot identify the optimal type of regime, and copy instead what appears the best regime under the circumstances.

This article has two main sections. The first presents the argument that co-evolutionary logic provides a helpful framework for understanding the progress and regress of various regime types over time and space. In particular, it can incorporate both broad selection advantages of liberal democracy and a role for hegemons in countering those advantages, including coercion, demonstration effects, and externalities. The second argues that co-evolution implies that states, or their rulers, feel pressure to adopt the regime that appears to have fitness advantages in the current international system. But that those rulers may attempt to construct a niche that allows their own regime to continue to flourish, which, in turn, can render the rulers’ regime more fit in ongoing competition.

Why Evolution?

The unit of analysis used here is regime type, not state or international system. To be explained is variation in the relative frequency of particular regime types across a population of countries over time. Regimes are not policies: a policy is a set of rules to solve a specific problem, such as privatization, whereas a regime is a complex of meta-rules—rules about rules—intended to solve the general problem of public order, such as capitalism. Regimes are ideal types and include liberal democracy, absolute monarchy, Marxism-Leninism, and Islamist theocracy. In practice, they may not be mutually exclusive; for example, a social democracy might combine elements of liberalism and Marxism. The article treats states as vehicles of regimes, analogous to organisms in biological evolution; and states’ rulers and other elites, as encoders, analogous to genetic material. With the majority of social scientists who use evolutionary logic, the article is concerned with cultural rather than biological evolution—that is, it makes no claims about how human genes affect or are affected by political regimes or struggles.2

Why use an evolutionary framework to help explain changes in the distribution across space and time of various regime types? Evolutionary accounts are most often contrasted to accounts that posit rational design and that bracket inheritance, copying, and so on. Evolution as applied to culture can incorporate intentional agents and their ideas, so it would be an extreme rational-design hypothesis that was free of evolutionary taint. An evolution-free hypothesis concerning changes in the relative frequency of regimes in a region would claim that a rational agent designs and implements the regimes within each country. This agent would be said to decide at every moment to leave the institutions alone or modify them without selection pressure from its material or social environment, including bounded rationality and learning from others’ examples, and without pressure to retain the institutions already in place. The agent would have perfect information and calculate flawlessly. It would be completely able to implement its decisions, deciding moment by moment whether democracy, fascism, absolute monarchy, or some other option was best for its interests.

This caricature may seem to offer a low bar to evolutionary accounts,3 and that is the point: once we understand what evolutionary theory implies about how political institutions arise, spread, are maintained, and die, then even casual observation of the distribution of regime types makes an evolutionary framework hard to resist. Many regime types have existed in history and the relative predominance of various types has varied—that is, the distribution of regime types is uneven across time and space. Some regimes once dominant in one or more regions, such as absolute monarchy or fascism, have virtually disappeared. New regimes, such as liberal democracy, have appeared and gradually gained predominance in certain regions at particular times. It is not the case that agents have engineered the entire process by which domestic regimes wax and wane internationally. Instead, it is possible to discern the three hallmarks of evolution (Hodgson & Knudsen, 2006): variety (of regimes), selection (some regimes become predominant, while most languish or die off), and retention (regime types can remain after the specific circumstances in which they arose have changed).

Indeed, the strongest reason to use an explicitly evolutionary framework is that the majority of scholarship on the international spread, persistence, and contraction of democracy in world politics appeals to mechanisms that are consistent with Darwinian logic. Evolutionary thinking unites these mechanisms in a coherent framework. It is standard in the literature on policy diffusion to distinguish four mechanisms by which practices spread across states: competition, learning, emulation (socialization), and coercion (Graham, Shipan, & Volden, 2013; Simmons, Dobbin, & Garrett, 2006). Each of these is analytically separate, but the four may be folded into a unified evolutionary logic. Clearly, competition, learning, and imitation are all part of what we mean when we say that domestic regimes evolve. States competing for scarce resources have incentives to become more efficient at securing those resources. States also have incentives to learn from or imitate the successes and failures they observe in other states (Weyland, 2005). The mechanism of coercion may seem outside of evolutionary logic, but it may be folded in under the label of niche construction. The point of using an explicitly evolutionary framework is not to obscure the distinctions among these four mechanisms, but rather to place them within a wider explanatory context that shows how they relate to one another.

Second, and related, it is clear that state rulers and aspiring rulers seek out ideas and information about how to have a more successful state, where success is measured by some combination of stability, security, wealth, and prestige. North Korea is distinctive today precisely because it is anomalous; to one extent or another all other states monitor, learn from, and imitate other states’ successes and failures, including of their regimes. In evolutionary terms, this means that ideas and practices, including those concerning regime type, “reproduce” and “survive” based on how well agents believe they fit current circumstances. For example, the First World War was won by a coalition dominated by liberal democracies (Great Britain, France, and the United States), and in the years following the war liberal democracies multiplied in Europe—Germany, Austria, Hungary, Poland, Czechoslovakia, Romania, Finland, Yugoslavia, Estonia, Latvia, and Lithuania. This flowering of democracy was not a result, at least directly, of coercion by the victorious allies. Rather, the years following the war saw a “vogue for democracy” in Europe because the great Western democracies had won the war (Thomson, 1962, pp. 588–600).

Third, evolution’s assumptions are relatively weak. Selection mechanisms in evolution do not require the strong axioms of rationality—perfect information and perfect ability to calculate—that have called into question standard rational choice and game-theoretic approaches. Hence evolution does not imply that units divine or practice the universally optimal strategy, whatever that may be. The claim is more modest: whatever traits become predominant are the best among available alternatives at mitigating some local problem, and they may in the long run bring other, even worse problems (Hodgson & Knudsen, 2006, p. 5). Agents need only be “boundedly rational,” with limited resources and thus incentives to use heuristics and other shortcuts to making decisions. As concerns a state’s domestic regime, elites cannot achieve absolute certainty about the superiority of one regime type; they can only observe what works and what does not within their purview. As the failures of European democracy in the 1930s attest, the winning regime type may prove less competitive at time t+1 than agents thought at time t.

Fourth, for many years, disciplines outside biology have profited by importing evolutionary logic. One need not accept the ambitious claims of “Universal Darwinism”—that all complex systems are adequately explained by evolution—to acknowledge that the mechanisms of variation, selection, and inheritance help explain outcomes in many domains, including language (Berwick & Chomsky, 2016), the practices of organizations (Aldrich, 1999; Nelson & Winter, 1982), and technology (Dafoe, 2015). As in most social science using evolutionary logic, genetics here is bracketed and analysis is conducted at a higher level of analysis, namely regime type. Thus my argument falls under the label of cultural or social evolution (Boyd & Richerson, 1985). Scholars have used evolutionary logic to explain the emergence of international rules and practices (Barnett, 2009; Florini, 1996; Modelski, 1990; Wendt, 1999), of globalization (Modelski, Devezas, & Thompson, 2007), of sovereign states (Spruyt, 1994), of war (Levy & Thompson, 2011), of the global political economy (Schwartz, 2012), and of world politics itself (Kahler, 1999; Tang, 2016). Other scholars, cited in the following section, have used evolutionary logic to explain the global spread of democracy and peace.

Some object to the application of evolution to social phenomena because the latter involve agents with intentions, whereas Darwinian evolution involves genes (which cannot think), disciplined by a purposeless environment. The introduction of rationality in agents and purpose into the environment would then mean a completely different and separate logic. But this objection does not appreciate the capaciousness of agency. Lewontin notes that organisms, conscious and intentional or not, act upon their environment, and sometimes thereby affect their own evolution, “The environments of organisms are made by the organisms themselves as a consequence of their own life activities” (Lewontin, 1983, p. 280). As Hodgson and Knudsen (2006, pp. 11–12) write, the line between animal and human intentionality is difficult to draw. Another objection would be that “evolution” might connote claims about progress (including moral progress) and Aristotelian final causes and, in world politics, an inexorable movement of the human race toward universal liberal democracy (Fukuyama, 1992). But evolution need not rely on teleology and, as argued in a later section, can easily accommodate both movements away from liberal democracy and countries and entire regions that resist democracy.

Evolutionary logic, of course, brings some hazards. The 19th-century social Darwinism of Herbert Spencer, which divided humanity into allegedly natural races analogous to species and posited a survival of the fittest, still casts a long shadow (Offer, 2000). As Spencer’s critics during his own time understood, some aspects of his appropriation of Darwinism did not necessarily follow from the theory. Spencer and others thought that evolution necessarily implied progress (with baleful normative consequences), a view that Darwin himself was to abandon (Toulmin, 2000, pp. 92–93). Further, handled carelessly, evolutionary logic can produce unfalsifiable just-so stories. A phenotype survives, we say, because it was most fit; but if we identify fitness by survival, we argue in a circle. More broadly, there is a reason why structural-functional explanations were routed from the field years ago by rational choice: the former took structures to be agents and could not provide convincing microfoundations for their claims. However, evolutionary accounts can both link the micro to the macro and yield empirically testable propositions concerning information about fitness, learning, pressure to conform, and niche construction.

Evolution and Regime Types

As Huntley (1996) argues, an evolutionary logic is implied in the famous argument of Kant about perpetual peace (1795). IR scholars today know Kant’s essay mainly for its claim that republics will form a league of peace and international law. But Huntley notes that for Kant, the trend toward republicanism and peace is generated by fear of violence in the state of nature—that is, by the dangerous social environment in which states interact. As war becomes more violent and counterproductive, more and more people will demand that their states set up institutions that guarantee that the national interest (rather than the interests of the elites) are served—namely, republican institutions. Evolutionary theory distinguishes more from less severe selection pressures, and expects more homogeneity among units as selection pressure grows. A Hobbesian state of nature, recognized by Kant, selects for a growing league of republics.

More recently, a number of scholars have built on this basic insight concerning liberal democracies. Mitchell, Gates, and Hegre (1999) have modeled interactions between domestic regimes and international outcomes; they found that (as Kant would expect) war tends to be followed by democratization and that more democracies decreases the number of wars. Cederman and Gleditsch (2004) noted that democratic transitions tend to cluster in time and space, and argued for the co-evolution of democracy and peace in regions. In their model, as long as democracies practice collective security (defending one another when attacked), democracy spreads in a geographically proximate set of states. Other literature, meanwhile, attends closely to the mechanisms through which democratizations cluster. Many of these works argue for “diffusion,” or the spread of democracy from state to state (Brinks & Coppedge, 2006; Starr, 1991). Although evolution could help explain the clustering of regime transitions across states without diffusion—selection by the states’ environment still would be operating—it would be difficult to account for the clustering of transitions to the same regime type. Why would the environment select for liberal democracy specifically, in state after state, unless the people in those states were imitating other states? Soviet elites visited the West and saw the consumer goods there; East Germans could watch West German (and hence a great deal of American) television (Markham, 1984).

When we say that evolution helps cause a given international distribution of regime types—democracy, fascism, monarchy, and on on—what do we mean? In keeping with the three hallmarks of evolution (variety, selection, retention), we mean that:

  1. 1. A variety of regime types exists across an international system (regional or global).

  2. 2. States retain regimes as power seekers and their supporters inherit the ideology and regime and pass it on.

  3. 3. States, constrained by scarce resources, select the particular regime that seems most successful at addressing the challenges that their domestic and/or international environment imposes upon them.

What follows is a simple evolutionary story about how a particular domestic regime comes to predominate. More finely grained attention to agents and power via the concept of niche construction then complicates the story.


Evolution may explain the initial emergence of particular regime types, for example, how Russia became communist (rather than something else) in 1918, which would require analysis of how Marxism-Leninism came into being and triumphed over its anarchist and socialist competitors in the early 20th century. No doubt selection and retention are heavily involved. I leave off the initial emergence of ideologies and regimes, however, and begin with the fact of regime and ideological diversity across states. Globally, we see a wide variety of regimes. Today, most regions are dominated by one regime type, for example, Europe and the Americas by liberal democracy, the Arabian Peninsula by monarchies. East Asia’s heterogeneity makes it especially interesting: it comprises liberal democracies (Japan, South Korea, Taiwan, Philippines, Indonesia), a semi-democracy (Malaysia), a democratizing state (Burma), a military dictatorship (Thailand), market-Leninist states (China, Vietnam, Laos, Cambodia), and an opaque, but evidently absolute, dictatorship (North Korea).


Regimes do change. But change is less common than stasis, and evolution has a well-developed account of the retention of traits. In biology, genetic material is passed on and inherited when the phenotype it produces enjoys a reproductive advantage. Cheetahs are fast (phenotype) because they inherit the genetic material (genotype) that enables speed. In cultural evolution, ideas, practices, and institutions are reproduced across a population over time and thereby retained. In social phenomena, however, the concept of “reproduction” is not quite the same as in biology. People who live under an advantageous institution are not necessarily going to have more robust offspring, nor will those offspring have a preference for that institution imprinted into their genes. Institutions instead are reproduced over time and across people through habit, example, education, and path dependency, or the collective investment by groups in them (Pierson, 2000). Cultural “traits,” such as institutions, are inherited through culture. Cultural evolution, then, employs a kind of Lamarckian logic in which acquired traits (such as institutions) are passed on. Some skeptics of cultural evolution have disqualified it from consideration for this reason. Hodgson and Knudsen (2006, pp. 14–15) have argued persuasively, however, that Lamarckism is a red herring: Darwin himself admitted the possibility that acquired traits could be genetically passed on. In any case, the concern here is not to trace the retention of political regimes to human alleles.

The important point is that a political regime is not perpetually being invented by the people who live under it. Thomas Jefferson believed that every generation has the right to a new set of laws; no generation can bind its successor to its constitution.4 But Jefferson, a consummate rationalist, was unrealistic. A better depiction of why a country has the particular regime it has during any normal time comes from his contemporary Edmund Burke, for whom a society’s laws were an accumulation of wisdom over ages.5 Even in the United States, with its attachment to novelty, the Constitution, laws, and judicial interpretations are inherited and passed on. Regimes normally change only gradually.


But regimes do change, and the question is how they do so. Evolutionary theory points to the mechanism of selection under competitive pressure. Cheetahs inherit speed because antelope are scarce, and over time, slow cheetahs were less able to catch antelopes and so had fewer offspring. Slowness was selected out of the cheetah’s genetic material; speed was selected for. Likewise, states live in a world of scarcity: material goods, time, and prestige all are limited. The regimes that survive are those best able to help states address their local scarcity; those selected out are less able to meet their rulers’ needs. Thus China began to jettison Maoist state socialism in the late 1970s because it was failing to build a strong, independent China. At a maximum, states that do not select the best available regime may die. States do die (Fazal, 2007), and state death is dangerous for at least some state elites. It is difficult to deny that the Soviet Union disappeared owing at least in part to its inability to compete with the United States, and that the Soviets’ communist regime, which by the 1970s was failing to innovate and carry on the impressive economic and technological growth of the 1950s, was partly responsible (Brooks & Wohlforth, 2000; Owen & Poznansky, 2014). Gorbachev’s attempt to adopt a version of democratic capitalism (roughly, European-style social democracy) came too late, but what matters here is his attempt at imitation.

To say that evolutionary logic rejects the strong assumptions of rational choice theory does not mean that it posits agents that are not purposive. The claim instead is that the scarcity of resources sets bounds on agents’ rationality (Simon, 1982). States cannot have perfect information about which regimes work best in their particular circumstances. Thus they have incentives to seek out information about various regime types (actual and potential), interpret the resulting information, and follow what appear to be virtuous examples. Bounded rationality also implies that they will not necessarily select the best option. States adapt to their environment in what they take to be the optimal way (i.e., what will best serve their interests as they conceive of them), but their choices are heavily conditioned by that same environment and may even harm them. They must halt their search for the best option at some point, and they cannot know with certainty that the point is optimal (Callebaut, 2007); the decision as to when to halt the search may partly be a function of the agent’s environment (e.g., an ideology pressed on him by a group to which he belongs).6

Use of evolutionary logic to explain changes in the distribution of regimes across states points us to the importance of information in selection as well as retention. China and the Soviet Union alike gathered information about how state socialism was working vis-à-vis its alternatives, chiefly capitalism or a market-based economy. They compared their own statistics with those of the United States and other Western countries and (in China’s case) with Singapore, which managed to combine single-party rule with capitalism and growth. New regimes do appear abruptly, as the appearance of the Soviet Union itself in 1917 shows. Normally, however, bounded rationality requires states to select from an existing menu of regimes.

Two Complications

Thus far, the story would seem to be that agents inherit state regimes and select a new regime for their state when new information about which regimes make for success and which for failure becomes available. This story is too simple, of course. The literature on regime change, particularly on democratization, tells us two things. First, when a regime changes, elites seldom all agree that the change is desirable; those who have power under the old regime want to keep it and often fight to do so. Just as evolutionary biology looks to the genetic level of analysis, a finely grained evolutionary treatment needs to go to a lower level of analysis, that of agents. Agents—individuals and groups—vary in utility function and capability, and a state’s selection and retention of a regime type will involve their struggles, sometimes to the point of violence and revolution.

Second, power is unevenly distributed among states, and big countries or hegemons often interfere with the regime choices of elites in smaller states. A hegemon intervenes through various means, from overt military intervention to covert action to economic sanctions to rhetoric, to see that smaller states have the regime it wants. Hegemons and their environments co-evolve.

Agents and Regime Change

It is agents who retain and select regimes for states, agents who pass on and inherit the practices of which regimes consist, agents who gather information about which regimes work and which do not. Agents must be part of an adequate evolutionary account because agents are not all the same with respect to preferences or capabilities. In a given country, most people would assert that they favor what is best for their country, but the fact that they often disagree on which regime is best means that we must look at their struggles to retain and change regimes. We also must attend to their different capacities to affect their state’s regime type.

Owing to different origins and tastes, some people want much more power and public influence than others. Some are politically ambitious—either those with what Machiavelli called virtù, who want to rule, or those who care deeply about public order and justice and believe they can have a salutary effect on those things—while others care more about goods such as art, family life, sensual pleasure, devotion to God, or other things. Roughly, this means the standard social-scientific distinction between political elites (power seekers) and the public. Elites—actual and aspiring rulers and their advisers—care much more about ruling, and hence about regimes, and so devote many more resources to gathering and interpreting information about the relative performance of regimes. To preserve the regime that enables their rule, they monitor the successes and failures of other states and implement changes accordingly. The public, by comparison, is far less engaged in political questions during normal times.

Elites, of course, want a scarce good—power—and so compete with one another. Thus they inevitably disagree over policy and sometimes the political regime itself. They compete for power in part by vying for the loyalty of constituencies, such as the military, police, business owners, clergy, bureaucracy, media, and general public. One way elites demonstrate their legitimacy to their constituents is through ideological consistency, or loyalty to the same regime regardless of circumstances. If a monarchist changes her mind and becomes a republican, her credibility will suffer, as monarchists will regard her as a traitor, and republicans will question her sincerity; she will bear audience costs (Fearon, 1994). Thus, although elites may change their minds about what regime is best, they will normally be consistent. Elites in power under an extant regime thus will have a stake in the retention of that regime.

Hence regime change does not require that all or even most elites change their minds about what regime is best. Rather, it can happen when the balance of power among the elites in a country shifts (Gleditsch & Ward, 2006, p. 912; Risse & Babayan, 2015). In the modal case a monarchy will become a republic not when the king becomes a republican, but when republicans gain power at the expense of monarchists. Republicans can gain power when new information gives fence-sitting elites and those less committed to monarchy per se higher confidence in the efficacy of republicanism and less confidence in the efficacy of monarchy. Those energized elites then try to rally the public to the new ideology. Even when elites change their minds, as when democrats become authoritarians, they will normally do so when they judge the balance of power within the country has shifted in favor of the challenger regime.

To return to the logics of retention and selection: the balance of ideological power within a country can shift when new information becomes available about the relative success of various regime types. When a republic unexpectedly wins an important war over a monarchy, or when the republic’s economy develops rapidly over a sustained period while the economies of monarchies stagnate, republicanism gains credibility across states and monarchy loses credibility (Gleditsch & Ward, 2006; Miller, 2016). Elites in these neighboring states who favor republicanism will have new moral and material capability and may attempt to rally the public to their cause and force reforms or even a regime change. The transnational rise of republicanism and weakening of monarchism amounts to a change in the social environment of the states in the region—a change to which elites of all types have incentives to adapt.

Niche Construction and Co-evolution

Recent studies find that hegemonic states can have significant influence over the regime types of their smaller neighbors (Boix, 2011; Gunitsky, 2017; Owen, 2010). It might seem that hegemony’s effects on the distribution of domestic regimes across states are outside of evolution, inasmuch as the hegemon is overriding the retention and selection mechanisms. But in the natural world, power, understood as the ability to influence outcomes, is unevenly distributed as well, and organisms routinely shape their environments in their favor. Thus the concept of “niche construction.” To say that agents build niches is to say that they shape their environment, intentionally or not, so that it will select for and retain traits, behaviors, or institutions. In that sense, environment evolves along with agent. Often, niches benefit their constructors. Earthworms consume and excrete the soil they burrow in, thereby altering its properties to favor more plant growth, which in turn benefits the earthworms by giving them more plant litter. In enriching soil over generations, earthworms cause the soil to select for a particular epidermis structure in earthworms themselves. Niches may also harm their constructors. The larger point is that feedback loops connect environmental changes, phenotypes, and genes (Odling-Smee et al., 2003, pp. 11–12). For original Darwinism the environment was exogenous to evolution, inasmuch as the properties it selected for were not affected by living things. (Sociobiology and evolutionary psychology as a rule retain this assumption of exogeneity.) More recent thinking on evolution has endogenized the environment, seeing it as co-evolving with agents (Lewontin, 2001; Odling-Smee et al., 2003, pp. 242–244).

As biologists have endogenized environment with agency, so should those who study political regimes in world politics. Human agents attempt to alter their environments so as to solve problems they face, and the altered environment can feed back into the traits of the agents, causing certain properties to be selected for (Jervis, 1997, pp. 48–50). The process happens domestically: when republicans find themselves with more power and effect regime change, the new republican regime empowers republicans in the future, who in turn use their power to reinforce the republican regime.

But of greater interest is an international hegemon that affects this process—that is, that willy-nilly shifts the balance of power in a smaller state from supporters of one regime to supporters of another.7 State rulers always have an interest in a friendly international environment, in particular in neighbors whose policies are helpful to their purposes. Leaders of a democratic hegemon may prefer democracies, because democracies do not fight one another (Doyle, 1983; Owen, 1997) and are generally more reliable partners (Lipson, 2004). Leaders of a capitalist hegemon tend to want economies open to foreign trade and investment. Rulers of a communist hegemon may want neighbors whose economies they can plan and control. Rulers of any hegemon will want to degrade domestic ideological threats and foreign ideological enemies (Owen, 2010, pp. 45–46). If selection pressures favor their own regime type, they may spread their regime abroad. If pressures favor a competing regime type, they may simply try to block the spread of that regime. A hegemon’s government may mistake the type of selection pressures, as the G. W. Bush administration did when it promoted democracy in Iraq from 2003 and pressed for free elections in the Palestinian territories in 2006. Rulers may use a variety of means (Börzel & Risse, 2012), including military action (invasion, air strikes, special forces), covert action, economic sanctions, capacity building, externalities (e.g., constructing international rules that reward states with the desired regime), and rhetoric.

Clever leaders of a hegemon no doubt will use all available tools that they believe will be efficacious; for example, they will not invade the target if they believe that domestic or international resistance would make an invasion counterproductive. That leads to the first sense in which a regime-promoting hegemon’s leaders participate in evolution: they must adapt their policies to the environment they are trying to shape. The hegemon’s rulers will find that the environment selects for some tools and against some others. Second, in shaping the regional or international environment via regime promotions, the hegemon’s rulers are trying to set up a feedback loop to their own state, one that makes it more likely that its own regime will be retained over time, precisely because regime retention means power retention for them. Agency and structure are endogenous, even with great powers (cf. Braumoeller, 2013).

Hegemonic influence also may be less direct and even unintentional, for example, by inspiring emulation or learning in unanticipated places, or by inadvertently creating or empowering particular agents in smaller states. The victorious powers of the First World War implemented the Wilsonian principle of self-determination to break up the Austrian, German, and Turkish empires, but the message was received in the British and French empires as well. Delcour and Wolczuk (2015) find that Russian attempts to block democratization in Georgia and Ukraine have backfired by mobilizing countering forces in favor of democracy.

Co-evolution as a Framework

Evolutionary logic holds great promise for those who wish to understand and explain the varied fortunes of democracy across the world and through history. To reiterate, a co-evolutionary account cannot make point predictions about the spread or contraction of regime types across regions. It is rather a framework that pulls together a number of extant theses concerning the temporo-spatial diffusion of regimes. Its logic points us to competition among states, and hence their rulers, for scarce resources and also to transnational learning and imitation of evidently successful regime types and the fading away of unsuccessful ones. Adding co-evolution—the endogenous evolution of agents and their environments—allows us to take on board the differences in power across agents, differences that political scientists are bound to attend to. Among the observable implications of a co-evolutionary framework are:

Retention and Selection

  1. 1. Elites across states will search beyond their national borders for information on which regime types are successful and which are not.

  2. 2. Clear evidence that one regime type is superior to another—for example, an unexpected war victory or sustained economic development and growth—will shift the balance of ideological power within other states.

Niche Construction

  1. 1. Leaders of a hegemon whose regime is favored by the selection process in (2) will intervene in smaller states to aid agents who prefer that advantaged regime.

  2. 2. Leaders of a hegemon whose regime is not favored by the selection process in (2) will intervene in smaller states to block or contain agents who prefer the advantaged regime.

A co-evolutionary framework also can guide research on diffusion that seeks to address skeptics who argue that democratization is caused by local (state-specific) conditions rather than by diffusion. Co-evolution makes clear that elites’ local environments will influence their actions: they will try to shape their environment but cannot invent it out of whole cloth. Thus local conditions are causal in a co-evolutionary framework, but so is information that elites gather from other countries. Evolutionary theory also has long recognized that copying errors occur, as when parents pass genetic material on to their offspring. Finally, the evolutionary mechanism of genetic drift has implications for regime diffusion. Genetic drift takes place when random events (having nothing to do with environmental fitness) cause a trait to be passed on or to die off. In a time of global power shifts, especially toward China, with its market-Leninist regime, these propositions are of increasing importance.

Most important for the purposes of this project, co-evolution can pull together hypotheses that seem disconnected, and are sometimes seen as necessarily competing, into a unified theoretical framework. If empirical research ultimately shows that liberal democracy spreads or contracts more by rational learning than by social emulation, or more by coercion than by rational learning, better to interpret that as the triumph of one mechanisms within a theoretical research program rather than as the triumph of one theory over another. What mechanisms predominate under what conditions does matter enormously, but so does our ability to make overall sense of macrotrends, such as the progress and feared regress of democracy in global politics.


Earlier versions of this article were presented at the Swedish Institute for International Affairs in Stockholm, the German Institute of Global and Area Studies in Hamburg, the Peace Research Institute of Frankfurt, the 2016 meeting of the American Political Science Association, and the Program on International Political Economy and Security at the University of Chicago. The author thanks Seva Gunitsky, Stein Tønnesson, Oliver Turner, Kevin Weng, and Kurt Weyland for comments, and the Center for Transnational Relations, Foreign and Security Policy at the Free University of Berlin, and the Global Governance Group at the WZB Berlin Social Science Center for support. The author drafted this article while on a Humboldt Research Award from the Alexander von Humboldt Foundation.


Aldrich, H. (1999). Organizations evolving. Beverly Hills, CA: SAGE.Find this resource:

Ambrosio, T. (2009). Authoritarian backlash: Russian resistance to democratization in the former Soviet Union. Farnham, England: Ashgate.Find this resource:

Barnett, M. (2009). Evolution without progress? Humanitarianism in a world of hurt. International Organization, 63, 621–663.Find this resource:

Berwick, R., & Chomsky, N. (2016). Why only us: Language and evolution. Cambridge, MA: MIT Press.Find this resource:

Boix, C. (2011). Democracy, development, and the international system. American Political Science Review, 105, 809–828.Find this resource:

Börzel, T. A., & Risse, T. (2012). From Europeanisation to diffusion: Introduction. West European Politics, 35, 1–19.Find this resource:

Boyd, R., & Richerson, P. J. (1985). The origin and evolution of cultures. Oxford: Oxford University Press.Find this resource:

Boyd, R., Richerson, P. J., & Henrich, J. (2011). The cultural niche: Why social learning is essential for human adaptation. Proceedings of the National Academy of Sciences, 108 (Suppl. 2), 10918–10925.Find this resource:

Braumoeller, B. F. (2013). The great powers and the international system: Systemic theory in empirical perspective. Cambridge, U.K.: Cambridge University Press.Find this resource:

Brinks, D., & Coppedge, M. (2006). Diffusion is no illusion neighbor emulation in the third wave of democracy. Comparative Political Studies, 39(4), 463–489.Find this resource:

Brooks, S. G., & Wohlforth, W. C. (2000). Power, globalization, and the end of the Cold War: Reevaluating a landmark case for ideas. International Security25(3), 5–53.Find this resource:

Burke, E. (1909). Reflections on the revolution in France (1790). Retrieved from

Callebaut, W. (2007). Herbert Simon’s silent revolution. Biological Theory, 2(1), 76–86.Find this resource:

Cederman, L. E., & Gleditsch, K. S. (2004). Conquest and regime change: An evolutionary model of the spread of democracy and peace. International Studies Quarterly, 48, 603–629.Find this resource:

Dafoe, A. (2015). On technological determinism: A typology, scope conditions, and a mechanism. Science, Technology and Human Values, 40, 1047–1076.Find this resource:

Delcour, L., & Wolczuk, W. (2015). Spoiler or facilitator of democratization? Russia’s role in Georgia and Ukraine. Democratization, 22(3), 459–478.Find this resource:

Doyle, M. W. (1983). Kant, liberal legacies, and foreign affairs, I. Philosophy and Public Affairs12(3), 205–235.Find this resource:

Fazal, T. M. (2007). State death: The politics and geography of conquest, occupation, and annexation. Princeton, NJ: Princeton University Press.Find this resource:

Fearon, J. D. (1994). Domestic political audiences and the escalation of international disputes. American Political Science Review, 88, 577–592.Find this resource:

Florini, A. (1996). The evolution of international norms. International Studies Quarterly, 40, 363–389.Find this resource:

Fukuyama, F. (1992). The end of history and the last man. New York: Simon and Schuster.Find this resource:

Gleditsch, K. S., & Ward, M.D. (2006). Diffusion and the international context of democratization. International Organization, 60, 911–933.Find this resource:

Graham, E. R., Shipan, C. R., & Volden, C. (2013). The diffusion of policy diffusion research in political science. British Journal of Political Science, 43, 673–701.Find this resource:

Gunitsky, S. (2014). From shocks to waves: Hegemonic transitions and democratization in the twentieth century. International Organization, 68, 561–597.Find this resource:

Gunitsky, S. (2017). Aftershocks: Great powers and domestic reforms in the twentieth century. Princeton, NJ: Princeton University Press.Find this resource:

Haas, M. L. (2005). The ideological origins of great power politics, 1789–1989. Ithaca, NY: Cornell University Press.Find this resource:

Hodgson, G. M., & Knudsen, T. (2006). The nature and units of social selection. Journal of Evolutionary Economics, 16, 477–489.Find this resource:

Huntington, S. P. (1991). The third wave: Democratization in the late twentieth century. Norman: University of Oklahoma Press.Find this resource:

Huntley, W. L. (1996). Kant’s third image: Systemic sources of the liberal peace. International Studies Quarterly, 40, 45–76.Find this resource:

Jefferson, T. (n.d.). Letter to James Madison, September 6. 1789. The Founders’ Constitution. Retrieved from this resource:

Jervis, R. (1997). System effects: Complexity in political and social life. Princeton, NJ: Princeton University Press.Find this resource:

Kahler, M. (1999). Evolution, choice, and international change. In D. A. Lake & R. Powell (Eds.), Strategic choice and international relations (pp. 165–198). Princeton, NJ: Princeton University Press.Find this resource:

Koremenos, B., Lipson, C., & Snidal, D. (2001). The rational design of international institutions. International Organization, 55, 761–799.Find this resource:

Kurlantzick, J. (2016). State capitalism: How the return of statism is transforming the world. Oxford: Oxford University Press.Find this resource:

Lake, D. A. (1992). Powerful pacifists: Democratic states and war. American Political Science Review, 86, 24–37.Find this resource:

Laland, K. N., Odling-Smee, J., & Feldman, M. W. (2000). Niche construction earns its keep. Behavioral and Brain Sciences, 23, 164–172.Find this resource:

Lee, J. (2015, November 6). Why does China fear Taiwan? American Interest. Retrieved from this resource:

Levy, J., & Thompson, W. (2011). The arc of war: Origins, escalation, and transformation. Chicago: University of Chicago Press.Find this resource:

Lewontin, R. C. (1983). Gene, organism, and environment. In D. Bendall (Ed.), Evolution from molecules to men (pp. 273–285). Cambridge, U.K.: Cambridge University Press.Find this resource:

Lewontin, R. C. (2001). The triple helix: Gene, organism, and environment. Cambridge, MA: Harvard University Press.Find this resource:

Lipson, C. (2004). Reliable partners: How democracies have made a separate peace. Princeton, NJ: Princeton University Press.Find this resource:

Lovejoy, Arthur O. (1910). Kant and evolution. I. Popular Science Monthly, 77, 538–553.Find this resource:

Markham, James M. (1984, February 13). TV brings Western culture to East Germany. New York Times. Retrieved from this resource:

Martin, L. L. (2000). Democratic commitments: Legislatures and international cooperation. Princeton, NJ: Princeton University Press.Find this resource:

Miller, M. (2016). Democracy by example? Why democracy spreads when the world’s democracies prosper. Comparative Politics, 49, 83–116.Find this resource:

Mitchell, S. M., Gates, S., & Hegre, H. (1999). Evolution in democracy-war dynamics. Journal of Conflict Resolution, 43, 771–792.Find this resource:

Modelski, G. (1990). Is world politics evolutionary learning? International Organization, 44, 1–24.Find this resource:

Modelski, G., Devezas, T., & Thompson, W. R. (Eds.). (2007). Globalization as evolutionary process: Modeling global change. London: Routledge.Find this resource:

Nelson, R., & Winter, S. (1982). An evolutionary theory of economic change. Cambridge, MA: Harvard University Press.Find this resource:

Odling-Smee, F. J., Laland, K. N., & Feldman, M. W. (2003). Niche construction: The neglected process in evolution. Monographs in Population Biology 37. Princeton, NJ: Princeton University Press.Find this resource:

Offer, J. (2000). Herbert Spencer: Critical assessments (Vol. 2). London: Taylor and Francis.Find this resource:

Owen, J. M. (1997). Liberal peace, liberal war: American politics and international security. Ithaca, NY: Cornell University Press.Find this resource:

Owen, J. M. (2010). The clash of ideas in world politics: Transnational networks, states, and regime change, 1510–2010. Princeton, NJ: Princeton University Press.Find this resource:

Owen, J. M., & Poznansky, M. (2014). When does America drop dictators? European Journal of International Relations, 20, 1072–1099.Find this resource:

Pierson, P. (2000). Increasing returns, path dependence, and the study of politics. American Political Science Review, 94, 251–267.Find this resource:

Ratner, E. (2009). Reaping what you sow: Democratic transitions and foreign policy realignment. Journal of Conflict Resolution, 53, 390–418.Find this resource:

Reiter, D., & Stam, A. C. (2002). Democracies at war. Princeton, NJ: Princeton University Press.Find this resource:

Risse, T., & Babayan, N. (2015). Democracy promotion and the challenges of illiberal regional powers: Introduction to the special issue. Democratization, 22(3), 381–399.Find this resource:

Russett, B. (1992). Grasping the democratic peace: Principles for a post–Cold War world. Princeton, NJ: Princeton University Press.Find this resource:

Russett, B., & Oneal, J. (2001). Triangulating peace: Democracy, interdependence, and international organizations. New York: Norton.Find this resource:

Schultz, K. A. (2001). Democracy and coercive diplomacy. Cambridge, U.K.: Cambridge University Press.Find this resource:

Schwartz, H. M. (2012). An evolutionary approach to global political economy. In R. Palan (Ed.), Global political economy: Contemporary theories (pp. 129–139). Abingdon, U.K.: Routledge.Find this resource:

Simmons, B. A., Dobbin, F., & Garrett, G. (2006). Introduction: The international diffusion of liberalism. International Organization, 60, 781–810.Find this resource:

Simon, H. A. (1982). Models of bounded rationality: Empirically grounded economic reason. Cambridge, MA: MIT Press.Find this resource:

Siverson, R. M., & Starr, H. (1994). Regime change and the restructuring of alliances. American Journal of Political Science, 38, 145–161.Find this resource:

Spruyt, H. (1994). The sovereign state and its competitors: An analysis of systems change. Princeton, NJ: Princeton University Press.Find this resource:

Starr, H. (1991). Democratic dominoes diffusion approaches to the spread of democracy in the international system. Journal of Conflict Resolution, 35, 356–381.Find this resource:

Tang, S. (2016). The social evolution of international politics. Oxford: Oxford University Press.Find this resource:

Thayer, B.A. (2000). Bringing in Darwin: Evolutionary theory, realism, and international politics. International Security, 25(2), 124–151.Find this resource:

Thomson, D. (1962). Europe since Napoleon. London: Longmans.Find this resource:

Toulmin, S. (2000). Interlude: Evolution and the human sciences, in J. Offer (Ed.), Herbert Spencer: Critical assessments (Vol. 2, pp. 90–119). London: Taylor and Francis.Find this resource:

Vanderhill, R. (2013). Promoting authoritarianism abroad. Boulder, CO: Lynne Rienner.Find this resource:

Wendt, A. (1999). Social theory of international politics. Cambridge, U.K.: Cambridge University Press.Find this resource:

Weyland, K. (2005). Theories of policy diffusion: Lessons from Latin American pension reform. World Politics, 57, 262–295.Find this resource:

World Bank (n.d.). GDP per capita (current US$). Retrieved from


(1.) In their book on the evolution of war, Levy and Thompson (2011) define co-evolution not as agent-environment evolution but as the interactive evolution of multiple processes, such as political organization and weaponry.

(2.) Ambitious natural and social scientists are working to build links between biological and cultural evolution; see Lewontin (2001), Laland, Odling-Smee, and Feldman (2000), and Boyd, Richerson, and Henrich (2011). For an application of evolutionary psychology to international relations see Thayer (2000). Evolutionary psychology and sociobiology recognize no feedback effects from culture to biology (Odling-Smee et al., 2003), and so my argument is quite different from that of Thayer.

(3.) The research program on the rational design of international institutions is predicated on the notion that institutions evolve, but the emphasis there is on agents’ responses to environmental pressures, not their co-evolution with the environment. Koremenos, Lipson, and Snidal (2001, pp. 766–768).

(4.) Jefferson’s formula was “the earth belongs in usufruct to the living,” meaning the living ought not to harm those yet to be born. Jefferson (n.d.).

(5.) Thus Burke: “As the ends of such a partnership [i.e., a society] cannot be obtained in many generations, it becomes a partnership not only between those who are living, but between those who are living, those who are dead, and those who are to be born” Burke (1909).

(6.) Agents are free, of course, to ignore how other states address their problems and how far they succeed or fail. In the early 21st century, North Korea’s regime comes closest to that kind of willful ignorance. That country’s condition as a self-impoverishing pariah, intermittently extorting aid from the United States and China, is the exception that proves the rule; it supports the proposition that agents do better when they try to learn from other states.

(7.) This leaves aside the question of how a state becomes a hegemon. A co-evolutionary framework would insist that it does so by adapting to and shaping its environment better than other states. An explanation as to why the hegemon does this better than other states would need to recognize co-evolution at the domestic level, perhaps resulting in a complex picture of nested evolution, or evolution within evolution. Thank you to one of the anonymous reviewers for suggesting this point.