Democracy and Climate Change for Realists: An Integrated Empirical and Normative Assessment

Abstract

Debates about the role of domestic political institutions in addressing global climate change often cite an American politician and a British scientist as points of reference. Writing in 1992 as a U.S. senator and presidential candidate, Al Gore claimed that “an essential prerequisite for saving the environment is the spread of democratic government to more nations of the world.”¹ Two decades later, James Lovelock suggested that a “suspension of democracy” may be required “when climate and other changes become as serious and as deadly as a major war.”² Researchers on the politics and policy of climate change routinely cite Gore and Lovelock for their contrasting views.³

Gore's and Lovelock's comments share a broad sense of ecological crisis but differ on political institutions. Gore's stance implies a theory of green democracy: liberal-democratic institutions uniquely and consistently promote climate-change mitigation and other environmental benefits. Lovelock's stance is usually read as a climate-specific version of eco-authoritarianism. Scholars who cite them assume they cannot both be right. But they can both be wrong.

This is the conclusion of the attempt reported here at an integrated empirical and normative assessment of scholarship on democracy and climate change. An integrated assessment is important because all empirical research makes normative assumptions (and vice versa) and has normative implications (and vice versa). Bridging exercises have value, and not only because of their scarcity.

Empirical and normative specialists alike tend to regard Gore as substantially correct and Lovelock as ill-informed or injudicious. Whereas scholars who focus on environmental issues as international collective-action problems typically assume that national regimes are essentially alike in key respects,⁴ scholars of comparative political institutions point to internal features of domestic regimes to rebut their presumed parity. Even as some climate-change specialists have drifted away from viewing the issue in binary terms,⁵ a significant body of academic opinion maintains liberal democracies’ superiority as forces for mitigation: in other words, Gore over Lovelock.

My conclusions are different. When we focus on the particular problem of climate-change mitigation, instead of generically “environmental” issues as a class, Gore's and Lovelock's claims both provide poor guidance. The best theory and evidence, I argue, are consistent with a regime-agnostic or neutral stance on the future of climate-change policy.

Regime-neutrality holds that, as a means to a given end, neither democracy nor autocracy is systematically better or worse than the other. Here, such a position on climate change emerges, first, from close consideration of the theoretic logic behind green (environmentally protective) democracy, as converted to a narrower logic of clean (climate-fixing) democracy; second, from a review of cross-national statistical studies that probe the empirical evidence for democracies’ and autocracies’ climate-change policy-making.

This skeptical or neutral stance does not entail a normative endorsement of climate-fixing autocracy, for which theory and evidence are perhaps even weaker than for climate-fixing democracy. But it can revitalize normative debate in other ways. Exploring difficult questions about normative trade-offs reveals that liberal democracy fares poorly in arguments from intrinsic value, in the context of climate-change policy and therefore must resort to novel arguments from instrumental value. In the absence of more compelling causal logics on their behalf, liberal democracy and illiberal autocracy alike will have to cede the climate-change field to other conceptual markers. The most eligible successors would be a post-liberal, more egalitarian form of democracy and a clean (climate-fixing) hybrid regime combining liberal and authoritarian features.

Regime-neutrality can be considered a key ingredient in a broader climatic realism. Whole regimes cannot be redesigned overnight by Rousseau's fictive Legislator in some architectonic effort to fight climate change. Yet massive amounts of money, energy, and time have been and continue to be expended on influencing the development of domestic political institutions within states around the world. Climate change is steadily and deeply transforming the social and economic frameworks within which such efforts proceed. For this reason, public debate about democracy and democratization needs climatic realism. Academic experts have not always offered sounder guidance on this question than could be gleaned from Gore or Lovelock, but we can and should do better.

Affirmative and Skeptical Views on Democracy and Climate

What do social scientists know, or think they know, about the impacts of different political regimes on climate change? This section considers how empirical and normative claims interact in the relevant scholarly literatures, yielding two major positions. The more common, more affirmative position holds that Gore is right (democracies mitigate climate change) and Lovelock is wrong. But a second, more skeptical position has also attracted a following.

Debate about whether democratic or autocratic regimes are more compatible with climate-change mitigation has inspired considerable empirical research in the social sciences, particularly economics and political science. This literature sometimes seems disconnected from normative scholarship on related issues, and vice versa. For example, the normative debate on climatic politics has recently been revitalized by Mittiga's argument that political institutions tending to mitigate climate change should be accorded a kind of “foundational legitimacy,” even if they derogate from the lesser, “contingent” legitimacy supplied by democratic rights and liberties. This argument, however, is explicitly predicated on setting aside the empirical question of whether democracies or autocracies are more likely, as a class, to cause meaningful mitigation.⁶

It is analytically legitimate to adopt such a posture of provisional agnosticism, for the sake of argument. A normative claim about the priority of some values over others, in case certain conditions apply, can be logically distinguished from an empirical claim about the probability that such conditions actually do apply. In practical terms, however, bracketing empirical issues in this way is unlikely to satisfy many readers. Whoever questions the inevitability, likelihood, or mere possibility that a more climate-friendly autocracy could exist in the real world can simply say that the normative argument in its favor is idle at best and dangerous at worst.⁷ This kind of objection offers reasons for disregarding normative theorizing on the subject, rather than refuting (after engaging) such theorizing.

What answers, then, can scholarship offer to the relevant empirical questions? There are two approaches to answering these questions in social-scientific literature.

The first, more affirmative approach is evident in claims that the empirical literature “almost uniformly” favors democracy's advantage on “environmental outcomes” in general and that it finds democracy “stronger” on climate change in particular.⁸ Such affirmative assessments of the literature overall even come from articles that reach somewhat skeptical conclusions themselves. Dissenting studies are typically characterized as part of a minority opinion, against a “majority … in favor” of democracy's superiority.⁹

The affirmative approach sometimes projects its own image onto future efforts to mitigate climate change. One of the classic studies in the empirical literature noted democracies’ superior record at remediating locally based sources of pollution, while adding that “it is hard to see why this pattern should not extend to global environmental problems, such as climate change, at least in the long run.”¹⁰ A more recent example of the affirmative line claims that “democracies possess advantages that make them more, not less, effective in managing the problem of global climate change.”¹¹

In the second, more skeptical approach, qualifications or reservations about the empirical record somehow assume greater weight. After noting “mixed results,” the authors of one recent study go so far as to suggest that “the relationship between democracy and environmental outcomes appears far-fetched.”¹² Other recent analyses, deeming prior results “mixed” and “inconclusive” on the specific issue of climate change, respond by testing “deliberative,” “egalitarian,” and “participatory” variants of democracy.¹³

A similar skepticism about the democracy/climate nexus has long been the norm in the subfield of environmental political theory. Here, acknowledging a democratic advantage over autocracy, in general environmental terms, is consistent with the reservation that “there are many different types of democracy, not all of which are equally adept at managing the complex challenge of climate change.”¹⁴ In this tradition, liberal democracy is one species in the democratic genus. Its weaknesses include the shortness of electoral time horizons and the limits of territorial jurisdiction.¹⁵ Because of these and other defects, liberal democracy's appeal is vulnerable to “an agenda that is more transformative, participatory, cosmopolitan, and ecocentric.”¹⁶ Autocracy is not the answer, in other words. Liberal democracy should instead give way to some new form of “ecological democracy.”¹⁷

Importantly, however, the more affirmative approach appears to have surpassed the more skeptical approach in penetrating the gray literature that turns scholarship in a public-facing direction, toward policymakers and public media outlets. For example, the Varieties of Democracy Institute announced in a policy brief that “scientific evidence shows democracy is critical for achieving SDG [Sustainable Development Goal] 13[,] ‘Climate Action’,” and “democracies are better at taking action” on climate change.¹⁸ This brief, in turn, was cited for its authority on the subject by a London newspaper.¹⁹ Less affirmative, more skeptical examples of gray literature also exist.²⁰ Genuine differences of opinion or interpretation, then, characterize academic research on the relation between political regimes and climate change.

There are two problems with the more affirmative approach to democracy and climate change, with its tendency to endorse a strong version of their causal connections (and normative compatibility). First, this view harbors theoretical weaknesses in so far as it relies on a causal logic that was developed for environmental issues in general but is substantially inapplicable to the realities of climate change. Second, it shows empirical weaknesses when viewed through the scholarly literature that is devoted to cross-national statistical tests of the democracy/climate nexus.

Causal Theory

If liberal democracy could exert special powers over policy problems related to climate change, by what causal mechanisms would they operate? This section describes a durable twentieth-century logic based on competitive elections and civil liberties as effective vehicles for environmental protection in general. Such an account of green (liberal) democracy fails to translate, however, into a theory of clean (climate-fixing) democracy because of at least three distinctive features of climate change as a political problem.

A wide variety of causal mechanisms has been proposed over the years for the expectation that democracies should systematically show better environmental performance than autocracies. Two perennial themes involve popular accountability and open information. These particular mechanisms roughly match the two conceptual components of a Dahlian conception of liberal democracy, or “polyarchy,” as a regime featuring competitive elections and civil liberties.²¹ Thus the causal logic of green democracy examined here offers support specifically for the liberal vision of democracy, as distinct from a more minimal version (elections without liberties) or a more maximal one (more participatory, egalitarian, or deliberative).

Elections enable mass preferences to prevail over elite interests, the argument goes, leading democracies to supply public goods such as environmental protection.²² Civil liberties further support democracies’ green credentials because ecological degradation is more likely to be exposed, and then reduced, where citizens enjoy free access to information and media.²³ Institutionalized freedoms also help to nourish a vibrant civil society of private associations that link pluralistic public spheres to policymaking mechanisms.²⁴

Occasionally, a third set of institutional mechanisms is considered essential to the nature and functioning of a liberal-democratic state: constitutional checks.²⁵ These include various “horizontal” constraints which one state agency is legally empowered to enforce over another—of judicial agencies over legislative agencies, of legislative over executive, and so on—to supplement the “vertical” constraint that mass publics are given, through elections, over government as a whole. Falling outside the Dahlian paradigm, this third potential component of democratic regimes is rarely included in causal theories of green democracy because of its tendency to aid status quo economic interests. In modern industrialized states, such interests have historically used their “veto player” positions to diminish rather than augment state regulation of environmental concerns.²⁶

The prime theoretic alternative to green democracy is not green autocracy; it is regime-neutrality. Notwithstanding the introductory sections of numerous empirical papers, the causal (as distinct from normative) theory of green autocracy is a chimera. There is nothing worth the name to confirm or reject. Several scholars have ably demonstrated that the most commonly alleged “eco-authoritarian” texts are either talking about something other than the causal impact of political institutions on ecological conditions or indulging in speculative gestures that do not remotely approach the level of causal theories.²⁷ Nor is Mittiga's recent contribution pro-autocratic in causal terms. He develops narrowly normative arguments after explicitly bracketing empirical issues,²⁸ as any instance of normative theorizing may logically opt (without being logically obligated) to do. His normative argument for the conditional, climate-oriented legitimacy of certain illiberal rules and practices is compatible with a posture of regime-neutrality on causal issues.

With its lean and compelling two-part structure, based on competitive elections and civil liberties, the causal theory of green democracy has attracted considerable interest and support in the social sciences. Arguably, however, events in the twenty-first century have overtaken its twentieth-century logic. Climate change now dominates other environmental issues, and it is a different type of problem in significant respects. This claim has been articulated many times before,²⁹ and it is worth clarifying and synthesizing the reasons behind it to show why the causal logic of green (environmentally protective) democracy cannot easily translate to clean (climate-fixing) democracy.

First is the problem of spatial diffusion. Climatic destabilization originates in the accumulation of greenhouse gases, principally carbon dioxide, circulating throughout the upper atmosphere. Through the greenhouse effect, exhaust pipes in China affect average temperatures and other weather patterns in Brazil, and vice versa. The causal logic of green democracy made sense for early environmental issues, like the disposal of toxic waste or the destruction of natural habitat, to the extent that discrete impacts within national boundaries could be addressed at discrete sources within those same boundaries. With climate change, however, voters and politicians cannot be presumed to have equal jurisdiction over the sources or direct stakes in the impacts. This aspect of the problem has been recognized in a few empirical analyses that find regime type to have no significant impact on “global” types of pollution, even when it does reduce “local” types of pollution.³⁰

It is true that other types of pollution crossed national boundaries well before atmospheric carbonization was widely recognized as a major problem in the 1990s. But the first truly global atmospheric problem in international politics, the emission of ozone-depleting chemicals known as chlorofluorocarbons, differed from climate change in several important respects.³¹

Second, then, is the problem of temporal diffusion. Greenhouse gases do their damage over longer time scales than ozone-depleting chemicals, to say nothing of sulfurous or other types of air pollution. New emissions or sequestrations of carbon today cannot make significant impacts on warming trends until decades or even centuries have passed.³² Such temporal lags between source and impact exacerbate the problems of spatial diffusion for political action. To continue with the example of China and Brazil, when exhaust pipes in the former create problems for the weather in the latter, they do so not for the current generation but instead for multiple future generations. Voters and governments in one country are thus doubly displaced from the problems that they are causing or are able to mitigate.³³

Third, the problem of climate change originates largely in processes related to the most fundamental activities in the global economy: the production and consumption of energy.³⁴ Humans obtain energy in the first instance from food, and land-use practices associated with agriculture affect both the emission and sequestration of carbon. More to the point, every sector of the global economy depends heavily on burning fossil fuels for basic productive processes as well as for transportation of products. Unlike the ozone problem, climate change's drivers are pervasive to the economy itself, not just a feature of one or two of its sectors.

This third feature of climate change is particularly important because the most fundamental processes of modern economic life do not vary with political regime type. Events or policies that raise the price of fuel generally raise the price of food as well, and unaffordable food and fuel are reliable triggers of mass unrest in any political system,³⁵ whether in autocratic Egypt (in 2011) or democratic France (in 2018). If anything, democratic leaders should be even less tolerant of risks related to economically motivated unrest, given the distinctive constraints on liberal democracies’ options for defusing or suppressing it.

Partial validation of this critique of extending the logic of green democracy to climate change comes from empirical and theoretical studies of long-term policies that impose significant short-term costs. Climate change is one of several such “investment” problems. The most significant pattern in the global evidence is that democracies are most likely to make long-term investments when the government of the day is relatively insulated from electoral competition—a situation arising from contingent features of the balance of ideological or factional power in a democracy, and presumably from rather less contingent factors in an autocracy.³⁶

Yet short-termism cannot be regarded as the special property of liberal democracy. On the contrary, the five-year plans of one-party states have created similar time horizons to the four- or five-year cycles of multiparty electoral regimes.³⁷ Even without five-year plans, autocracies still have to anticipate the troubles that popular dissent can cause in the short term. Above all, every public treasury on the planet operates in a political tempo of months rather than years, in which temporal milestones are geared to the publishing of accounts by publicly traded companies and the adjustment of monetary policy by central banks. Impatience is endemic to every domestic political system because it is built into the global economic system.

Because of its heightened levels of spatial diffusion, temporal diffusion, and energetic-economic entanglement, climate change is prima facie unlikely to interact with political institutions on the same logic as other environmental problems. The tendency of competitive elections and civil liberties to make effective inroads on atmospheric carbonization seems more speculative and less compelling, theoretically, than the weight of twentieth-century thinking would lead us to expect.

Empirical Evidence

If theoretic expectations for climate-fixing democracy should be low, in principle, what does the evidence indicate? This section reviews cross-national statistical analyses published in the past two decades, with a focus on physical outcomes related to climate-change mitigation since 1990. This body of evidence yields ambiguous results that are consistent with a regime-agnostic position.

Several previous reviews of empirical literature on environmental policy have shed some light on the theory of clean democracy. Fiorino reviewed about a dozen studies and found generally pro-democratic results on environmental protection but with reservations about climate change in particular.³⁸ Similarly, Dasgupta and De Cian found in favor of democracies’ general green credentials but, in a sub-sample of eleven analyses of carbon emissions, a mixture of four pro-democratic results, five null or ambiguous results, and two anti-democratic results.³⁹ Bernauer reviewed climate-change studies in relation to a wide variety of domestic and international factors, finding among thirty-four sources cited on the effects of national-level institutions that regime type had “no robust significant effect” on greenhouse-gas emissions.⁴⁰ In a recent and more focused review of “the democracy-climate nexus,” the seventy-two sources consulted found either an ambiguous or a negative association between liberal democracies and policy outcomes that mitigated climate change more than twice as often as they found a positive association, though much stronger links were identified between democracy and the adoption of new climate policies.⁴¹

Here, I offer the most thorough review to date on three areas of policy related to climate-change mitigation: greenhouse-gas emissions, deforestation, and energy use.⁴² My procedure was to search peer-reviewed, English-language venues from 2003 to 2023, including disciplinary journals in economics, sociology, and political science, as well as topical journals on environment, energy, and natural resources.

The focus here is on cross-national (large N) analyses using quantitative methods, which have particular virtues and limitations for empirical analysis. Statistical associations in global datasets can neither confirm nor reject a causal theory on their own. Instead, they offer tools for comparing the relative strengths of multiple factors—political institutions, economic structures, demographic characteristics, and so on—in relation to observable features of climate-change policy. The question to ask is whether a given pattern of statistical results looks consistent with a relevant causal theory. If the answer is yes, further analysis with different methods may establish causality; if the answer is no, the causal theory itself may be reasonably doubted.

I initially identified eighty-one peer-reviewed sources from the past two decades in which the authors conducted at least one analysis of the relationship between national regime type (the independent variable) and climate-change mitigation (the dependent variable). I then narrowed this sample according to criteria of chronology, geography, and economics. Chronologically, global warming through the accumulation of greenhouse gases in the atmosphere was not a major item on the agendas of national policymakers before the later 1980s,⁴³ and the first international treaty on climate change was negotiated in 1992. Before about 1990, therefore, political institutions were generally not in a position to affect greenhouse-gas sources or sinks in intentional rather than incidental ways. Geographically and economically, it is important to consider analyses that cover diverse regions of the world and that include poor, middle-income, and rich nations alike. These three criteria of restriction leave seventy-four distinct analyses from thirty-three peer-reviewed articles.⁴⁴

Of these seventy-four analyses, 46 percent produced a normatively positive result for democracy, 34 percent yielded a null result, and 20 percent found democracy to have made matters worse for climate change. These slightly positive results require further interpretation in light of the distinction between policy outputs and policy outcomes.⁴⁵ Whereas an output is a promulgation of a new policy related to climate change, an outcome is an observable change in the relevant physical processes. The passage of a law and the signing of a treaty count as policy outputs, whereas the physical states of carbon sources (e.g., tailpipes) and sinks (e.g., trees) count as policy outcomes. Of the seventy-four analyses under review, twenty-three involve outputs and fifty-one involve outcomes. The difference in results between these two types of analysis is notable: for outputs, 74 percent of findings (n = 17 of 23) were positive on democracy, compared with 17 percent null and 9 percent negative; for outcomes, 33 percent (n = 17 of 51) were positive, compared with 41 percent null and 25 percent negative. In other words, nearly three out of four analyses of climate-change outputs found in favor of democracy, whereas one out of three analyses of climate-change outcomes did the same.

These results confirm a point that is familiar to experts on climate-change policy: democracies’ superiority lies in enacting policies more than achieving results.⁴⁶ This distinction appears to support a “signing ceremony” theory of the democracy-climate nexus: democratic institutions systematically cause political leaders to make formal efforts or public commitments to mitigate climate change, to an extent that autocratic regimes cannot rival. The problem is that democracies’ weakness at realizing their climate-policy objectives seems to defeat the presumed normative goal behind efforts at climate-change mitigation, which is to alter the reality of atmospheric carbonization and destabilization.

When we narrow the sample of studies to those that use emissions of greenhouse gases as the dependent variable, the success of extending the twentieth-century logic of green democracy to climate change becomes more questionable rather than less. These are arguably the most relevant to climate-change mitigation as a political problem, whereas deforestation and energy use represent important but less direct measures. Forests aid the absorption of atmospheric carbon but can remove only a fraction of new emissions. Energy use and energy efficiency are causally related to carbon emissions, but their practical significance for mitigation efforts depends on intervening variables. For instance, the mitigative impact of proportionally greater use of non-fossil (“clean”) energy has historically been offset by absolute increases in energy use overall.⁴⁷ From this perspective, greenhouse-gas emissions are the most conceptually appropriate indicator of successful mitigation.

When six analyses of deforestation and twelve analyses of energy consumption are excluded from the fifty-one analyses of post-1990 policy outcomes, the remaining thirty-three analyses of greenhouse gases show fewer positive results and more negative ones: 27 percent positive on democracy, 42 percent null, and 30 percent negative. Two of the sixteen articles in this greenhouse gas sub-sample report a mixture of positive and null results, four of them report a mixture of positive and negative results, seven report only null results, and three report only negative results.⁴⁸

Statistical associations in global datasets cannot amount to causal proof. Their analytic value is different. If the causal logic of liberal democracy's general green virtues could be successfully extended to climate change, it would be relatively easy for researchers to find statistically significant and positive associations and relatively difficult to find significant negative associations. In fact, however, analyses of outcomes that represent real-world mitigation have found it equally challenging to produce pro- or anti-democratic results. The plurality winner is the null finding. The empirical record, therefore, is difficult to reconcile with a causal theory of clean democracy and is more consistent with climatic regime-neutrality.

This understanding of the empirical record requires, in turn, a change in how we think about normative issues involving climate change and political regimes. Claiming that democracies are normatively superior to autocracies because they do better on climate change is a move that, on its intellectual merits, deserves to start an argument rather than end one. Ethical deliberation should not lean on causal theory when the latter is a wobbly stick. Even people who agree about the value of both maintaining liberal democracy and mitigating climate change must be prepared to deal with situations in which the two goals come into conflict. Normative theorizing, in short, is one form of disaster preparedness.

Normative Responses to Regime-Neutrality: Intrinsic Value

If normative theory cannot rely on the causal theory of climate-fixing democracy, what else can it use to support a preference for democracy over autocracy in light of climate change? This section considers, and rejects, both liberal and republican accounts of the superior intrinsic value of democratic freedom relative to values under threat from climate change. In the process, it reveals that hybrid regimes, embodying a mixture of democratic and autocratic features, have rarely been explicitly incorporated into such normative arguments.

The causal theory of clean democracy says nothing about normative priority directly. But it does imply that, empirically, situations of incompatibility between democratic freedom and climatic stability should be rare. Ethical debates about the priority of one norm over the other, therefore, gain added interest whenever such causal crutches are weakened.

Without a climatic kind of instrumental value for support, normative defenses of liberal democracy could turn toward intrinsic value. The classic form of such arguments sees conflict between security and liberty, necessity and luxury, and living and living well—and finds reasons to prefer the second option in each dichotomy. For security, read “climatic stabilization”; for liberty, “democratic freedom.” If and when climatic stabilization can be purchased at the price of democratic freedom, the reasoning goes, the cost is simply too high to justify payment.

For example, a long tradition going back to the Enlightenment holds that political freedom must be considered foundational rather than contingent. Rousseau memorably asserted the priority of liberty over security when rejecting (in The Social Contract) the option of collective submission to despotism in exchange for “civil tranquillity,” on the grounds that “there can be no possible compensation for someone who renounces everything.”⁴⁹ Whereas liberty was everything for Rousseau, climatic stability could be regarded as analogous to civil tranquility: a matter of mere physical safety. Paine used a similar logic, but with explicitly institutional overtones: “slavery consists in being subject to the will of another, and he that has not a vote in the election of representatives is in this case [read: condition].”⁵⁰ On this view, to be unfree but secure is an inherently untenable position. It is, in fact, to be unfree and insecure, because being subject to alien powers entails that they, not you, decide how secure you should be.

This posture of heroic Enlightenment liberalism, as in the refrain of “give me liberty or give me death” which is associated with the American Revolution, may seem old-fashioned to many scholars today. But there is a more genteel version of the same assumption in modern moral philosophy, bearing the name of “the priority of liberty.” This phrase has been memorably and controversially invoked in Rawlsian theories of justice. In this more sophisticated formulation, Rawls argued not that freedom may never be abridged politically but, instead, that the rationale for any political abridgment of freedom must always be the expansion of freedom.⁵¹

The operational meaning of the formula is difficult to interpret. When challenged on specific issues of “food supply” and “restrictions on the use of private property (e.g., automobiles) designed to protect the environment,”⁵² Rawls appeared to double down on the priority of liberty. No considerations of “a greater net sum of social and economic advantages for society as a whole” could normatively override individuals’ “basic liberties.”⁵³ Moreover, liberties of speech, religion, and voting must be constitutionally enshrined in any just regime, unlike norms supporting access to natural resources and to medical care.⁵⁴ Security issues, so far from being foundational, are subject to everyday policy debate within a liberal-democratic institutional framework.

Rawlsian justice, of course, tends to be associated with the liberals’ side of a debate—republicans occupying the other side—over the meaning and institutional implications of political freedom. Nonetheless, the republican school also upholds the priority of liberty vis-à-vis security. Paine articulated the principal reason already in the 1770s: losing the vote means losing everything else. Similarly, the late-modern republican principle of non-domination highlights the structural potential for people to be hamstrung by arbitrary power as more consequential than the momentary exercise or abeyance of such power.⁵⁵ In this view, liberals’ tendency to conceptualize freedom in terms of non-interference to preserve “options,” rather than non-domination to preserve “agency,” is just a different approach to the same foundational norm.⁵⁶

Some scholars regard non-domination, or agency-freedom, as more compatible than options-freedom with restrictive environmental regulations.⁵⁷ But republicans still justify such restrictions on freedom for the sake of more (or different) freedom. In effect, if not in name, they still accept Rawls's formula about the priority of liberty. It would be normatively legitimate, on this approach, for environmental regulations to “reduce some people's liberties” in exchange for a resulting “increase in the overall amount of liberties,” with the goal of preventing “irreversible damages caused to what supports freedom and life in general.”⁵⁸ As long as the state is enforcing laws over which citizens have equally distributed control, it seems, ecological crises can be met with all manner of policies without loss of freedom as non-domination.⁵⁹

Liberals' and republicans' kindred assumptions about the priority of liberty seem facially compelling but tend to be undermined by a surprising factor: their exponents' own concessions. On the liberal side, the early Rawls admitted that, “until the basic wants of individuals can be fulfilled, the relative urgency of their interest in liberty cannot be firmly decided in advance.”⁶⁰ Accordingly, he identified a range of “common” interests—including “public health and safety,” “public order and security,” “a certain level of wealth,” and even “the quality of civilization”—as possessing sufficient force to justify restrictions on liberty for the sake of something other than liberty.⁶¹ Even a pro-autocratic realist would find it challenging to articulate a more impressive roster of exceptions to the priority of liberty. Rawls, perhaps unsurprisingly, preferred to describe these exceptions as preconditions of liberty itself,⁶² rather than as trumps over it. Republicans make a similar move. They describe “social and ecological communities as structural preconditions for the flourishing of individuals,”⁶³ and they admit that access to “critical resources” is essential to a person's enjoyment of freedom, understood as non-domination.⁶⁴

Such concessions weaken the normative priority of liberty, whether republican or Rawlsian. The point of establishing normative priority, theoretically, is to manage trade-offs in hard cases. Imagine a straight choice between a climate-wrecking democracy and a climate-fixing autocracy. The priority accorded to political freedom by both liberals and republicans implies that climate-wrecking democracy is the better alternative, but the notion that freedom has preconditions complicates that choice. Does climatic stability figure among these preconditions?

It is a well-established assumption among scholars who consider physical ecology as essential to political life that “ecosystem integrity is a precondition for individual and collective human well-being.”⁶⁵ Climatic stability, by this logic, bears a close relationship to the physical safety and health of human bodies and minds because it affects access to material prerequisites such as water, food, and shelter.⁶⁶ Climatic destabilization endangers such access in multiple ways. This type of argument against the priority of democratic freedom is based on a realist ontology: politics is embedded with, and in important ways dependent on, variable conditions of physical reality.

Mittiga's analysis is consistent with this line of attack on the priority of liberty. His contribution is to associate the value of climatic stability with a “foundational” legitimacy that stands above the “contingent” legitimacy supplied by democratic freedom.⁶⁷ Set against certain basic necessities, the individual rights associated with liberal democracy (e.g., expression, association, voting) take second place. The second-class values may be considered tremendously important, but the first-class values are indispensable. Importantly, Mittiga also relates the abstract normative question to some concrete political institutions. Thus the force of foundational legitimacy applies, under certain circumstances, to aspects of “illiberal,” “undemocratic,” and “authoritarian governance” within an otherwise democratic state: “limitations of rights to free movement, association, and speech.” Mittiga also mentions enforced “lifestyle changes,” censorship of factual misinformation, and confiscation of private property as potentially implicated in his normative argument.⁶⁸ The reasoning here is quite clear that no one should curtail a precious value like democratic freedom except in exigent circumstances. Perhaps Mittiga's main achievement is to have underlined the intuition that a norm can be precious without being primary.

In terms of regime type, it is equally clear that Mittiga's normative argument does not contemplate climate-fixing autocracy but, instead, a clean hybrid regime. The concept of hybridity applies when a regime is neither democratic nor autocratic but bears some characteristics of both. In this light, causal regime-neutrality does not logically imply any normative preference but does put hybrid regimes closer to the center of attention. Underdeveloped theorizing about hybrid regimes and climate change in academic political research makes arguments such as Mittiga's difficult to assess. For example, is there a significant difference between elections without liberties and liberties without elections? These are two distinct types of political regime, but Mittiga foregrounds the former. In transitional terms, is there a significant difference between a liberal democracy that becomes more like a hybrid regime, as Mittiga's argument contemplates, and an illiberal autocracy that adopts some democratic features?

The normative case for the priority of democratic freedom over climatic stability may be weak, but this weakness need not be fatal. It simply means that the democracy-first side has greater need of new arguments from instrumental value as opposed to intrinsic value. Indeed, the significant attention that causal issues have attracted in research on climate change in general could be read as endorsing a tacit acceptance that climate prevails over democracy on intrinsic value alone. Yet several other pro-democratic responses are available that seek new theoretic purchase on instrumental value.

Normative Responses to Regime-Neutrality: Instrumental Value

If arguments from intrinsic value are too weak to sustain the normative superiority of liberal-democratic regimes in light of climate change, could novel arguments from instrumental value come to the rescue? This section explains how the most promising options of this kind come with significant limitations.

My analysis, so far, has woven together two strands of realism, one more causal and empirical and the other more ethical and normative. The causal strand views institutional forms as relatively weak determinants of political outcomes, and the ethical strand views liberal ideals as having relatively weak claims on normative priority. Both strands of realism respect the influence of physical, material factors on basic human interests. But they are not anti-liberal per se. At its best, realism is motivated precisely by its materialist and empiricist orientation to be open-minded about the potential of new observations or novel understandings to unsettle old dogmas.

For instance, regime-neutrality on climate change could still be compatible with the normative priority of democracy over autocracy on non-climatic grounds. This is liberals' path of least resistance, theoretically. Many other norms seem to recommend liberal democracy. One recent meta-analysis revealed a positive statistical association for democracy with many development outcomes (e.g., promotion of public health, administrative transparency), but there are also null or inconclusive results on several others (e.g., quality of infrastructure, reduction of inequality).⁶⁹ Other research suggests that, in poorer countries, even the most basic amenities of the modern state (e.g., rule of law, governmental effectiveness) are not empirically strong in democracies by comparison with autocratic or hybrid regimes.⁷⁰

Rather than range beyond the goal of climatic stability, I now consider three other pathways that remain open for defending liberal democracy even while keeping climate change at the forefront of normative concern. For example, scholars have long recognized that global warming tends to elude efforts to corral it through domestic, jurisdictional solutions.⁷¹ Given the global and trans-boundary dynamics of atmospheric carbonization and climatic destabilization, international interactions among states are arguably more important than internal dynamics within them. Academic realists, of course, traditionally emphasize the systemic level of analysis over the domestic, on a whole range of issues. But prominent strands of thought within liberalism and constructivism do the same for climate change in particular. Domestic regimes, democratic and autocratic alike, seem relatively powerless in the face of this utterly planetary problem. On this perspective, state sovereignty cannot be the solution.

A focus on the systemic level of analysis leads to the first possibility for recovering liberal democracy's instrumental value relative to climate change. For many liberals and constructivists, global civil society (non-governmental and multinational organizations) must drive the social and economic changes needed for mitigation of climate change. But a global perspective can still tout the instrumental benefits of certain domestic arrangements. What makes liberal democracy essential to climate-change mitigation and other cross-border policy challenges, on this account, is that it offers an incubator for global civil society to flourish and expand.⁷² Representative, non-profit, trans-boundary pressure groups that hold the keys to co-ordinated, networked changes toward a global energy transition happen to start up and scale up better within democratic states than other states. In addition, and thanks in part to effective civil liberties and the interest groups fostered thereby, democracies can better lead the way in crafting effective diplomatic agreements and other co-ordination mechanisms on a trans-national scale.

There is promise as well as peril in this argument. Empirical studies routinely conclude that democracies have more often adopted international environmental agreements.⁷³ The frequently null and negative results on democracy and actual emissions of greenhouse gases could be explained away, on this account, by problematizing the use of the country-year as the unit of analysis in virtually all statistical studies of the climate-democracy nexus. Through global civil society and international agreements, democracies could be doing the work of mitigating this trans-boundary problem even when their own country-level emissions seem to indicate otherwise.

The theoretic challenge for this kind of reasoning, however, is identifying causal mechanisms. States (especially armies) and firms are the direct polluters, after all. International and non-governmental organizations (NGOs) attempt to influence states and firms but must do so with limited leverage.⁷⁴ However compelling the notion that democracies surpass autocracies in nurturing para-state and non-state groups, then, it must compete with the reality that states and firms have overwhelmingly greater causal agency.

Because of the complexities of energy transition and decarbonization, a second compelling possibility invokes liberal democracy's epistemic and technologic strengths. Climate change may be a peculiarly devilish problem, but democracies may be peculiarly angelic at problem-solving. The free flow of information, thanks in large part to effective civil liberties, allows democratic institutions to ask the right questions and to find the right answers.⁷⁵ Autocracies, by contrast, are structurally incapable of getting as many eyes on a problem and therefore tend to stumble in the search for solutions. This kind of epistemic logic has been suggested in the specific context of climate change by scholars who stress democracy's advantages in technology and innovation.⁷⁶

The pro-democratic logic of technology, however, also faces significant obstacles. Technical solutions to complex problems likely depend more on informational flows among privileged economic and political elites than among ordinary citizens, and institutionalized electoral pressure appears to play a diminished role. In short, hybrid regimes with sufficiently vibrant civil societies, but without competitive elections, should do as well on climate change as liberal democracies. Both technologic champions and global civil society can do without domestic electoral competition, without popular representation in legislative and executive agencies, and without public accountability in basic policy-making. Taken together, these two instrumental arguments do not hinge on what makes democracy a distinctive regime type. Perhaps, to coin a phrase, “NGOs plus tech bros equal mitigation”—but they do not equal democracy.

A third option for reimagining democracy's instrumental value for climate-change mitigation turns to the broad issue of time or temporality. The twentieth-century logic of green (liberal) democracy was not developed with processual or transitional mechanisms to the fore, and climate change is a more slow-moving and time-lagged process than most traditional environmental issues. There is some evidence that modelling longer time lags can make the theory look more empirically sound, but not many statistical studies use “democracy stock” or accumulated regime-type experience as the independent variable when greenhouse-gas emissions are the dependent variable. Those that do have produced pro-democratic results with significant time lags, including one estimate that a low-income state needs an average of 17 years of democratic experience before achieving significant reductions of carbon dioxide emissions.⁷⁷ Other studies model temporal accumulation in relation to climate-change outputs rather than outcomes, including one estimate that regime-type experience for up to 60 years is associated with the enactment of significant policy changes related to mitigation.⁷⁸ In most cases, these empirical analyses resort to democracy stock to achieve a statistically significant result when measures of the level of democracy fail to do so. Subject to further and better tests, the concept of democracy stock might end up saving the causal theory of climate-fixing democracy. Preferably, temporal dynamics would be treated in more subtle fashion, both theoretically and empirically, in future qualitative and quantitative research.⁷⁹

If time is democracy's secret weapon in the fight against global climatic destabilization, however, it is also a significant weapon for the greenhouse effect itself, whose dynamics operate over generations and centuries. Climatic time is a double-edged sword. Because of time lags in the physical and chemical processes of the upper atmosphere, many scholars have concluded that time is running out: not on the damaging processes themselves, of course, but on hopes to constrain future damages by limiting current emissions. This is the meaning of “climate emergency.”⁸⁰ Emergencies do not—by definition, cannot—wait. If liberal democracy is a regime that needs time to mitigate climate change, it is a regime that cannot satisfy those for whom time is already running out.⁸¹

In summary, resuscitating the instrumental value of democracy for climatic stability is far from an implausible task. But even the most promising pathways feature major roadblocks, theoretically speaking. It may be a worthwhile endeavor to accrete further scholarly effort toward assembling the logic and evidence necessary to clear the route for a new causal theory of climate-fixing democracy. In this way, perhaps, we might aspire to pre-empt genuine normative debate about conflicts among values associated with democratic freedom and climatic stability. But changing the conversation about climate and regime types in more fundamental ways might prove more fruitful.

Conclusion: Climatic Realism

The conventional wisdom of academic research about liberal democracy's distinctly favorable relation to ecological values is based on theory and evidence that fit poorly with the twenty-first century realities of climate change. The alternative is not eco-authoritarianism but regime-neutrality. In other words, the pro-democratic view personified by Al Gore and the anti-democratic view personified by James Lovelock appear to be equally misguided, at least where climate change is concerned. We should move forward without them.

The weakness of theory and evidence for climate-fixing democracy plays into the hands of a general kind of political realism. Regime-neutrality is consistent with the realist assumption that significant differences among states tend to arise less from their political institutions than from their differential geopolitical positions, domestic interests, and state capacities. On this account, national policies on climate-change mitigation are bound to follow primarily from how greenhouse-gas emissions, deforestation, and energy production and consumption affect states' military security and economic prosperity. In comparison, domestic institutional arrangements have a secondary role to play. Realism of this sort could look above or below the domestic level of analysis: above to international institutions, below to sub-national institutions.

But going above or below national regimes is not the only set of options. The kind of realist approach taken here might instead maintain the domestic regime as a useful level of analysis while reorientating our conceptual map of how its institutional variations relate to climate change. If democracy and autocracy make a poor dichotomy for managing the political and intellectual challenges, we might instead think in terms of clean hybrid regimes and post-liberal forms of democracy. The first category appears to be related to substantial empirical experience but meager theoretical reflection. The second category has a similar problem, but in reverse: much theory, little experience. Both categories could stand some reimagining in relation to climate change. A realist can look “down” from liberalism to a cleaner hybrid regime or “up” to a cleaner form of democracy. The step up seems an odd option for realism but should not be. Maybe liberalism's problem is not its insufficiently autocratic approach to climate change but its insufficiently democratic approach. This framing, at least, would be consistent with previous scholarship that reveals conventional assumptions about realism's inherent hostility to democracy to be historically and conceptually faulty.⁸²

But people understandably wonder what post-liberal democracy could mean in practice. There is not a lot of it, at least at the national level, for comparativists to measure. In environmental political theory and in other circles, post-liberal alternatives have been discussed with such terms as “participatory,” “associational,” “discursive,” or “deliberative” to the fore. Such discussions often have an anti- or at least non-institutional cast, privileging instead the alteration of norms, cultures, and discursive practices.⁸³ The problem, of course, is that formal institutions are easier to change in a deliberate and timely fashion. In this connection, the most promising practical option in climate-change policy-making currently appears to be “deliberative forums,” “citizen assemblies,” or “policy juries.”⁸⁴ These are randomly selected assemblies of lay citizens who, after focused and intensive study of particular problems, may variously exercise policy-recommending, agenda-setting, constitution-interpreting, or politician-sanctioning powers. In conceptual terms, a novel move would be a reconciliation between political realism and policy juries. In operational terms, existing trends should be elevated toward relentless practical experimentation with policy juries, at whatever levels of analysis become available.

Unless defenders of liberal democracy can find a new causal theory that is equal to the realities of climate change, defenders of alternate institutional arrangements are poised to seize the platform. The quest for wiser and faster options already holds considerable interest. Such options may or may not be practically available today or tomorrow. But theory is not exclusively for action today; it is also for preparedness to act next year, next decade, or next generation. In that spirit, the seriousness of climate change requires abandoning the antinomy of liberal democracy versus illiberal autocracy and expanding our domestic-regime toolbox.

Supplementary Material

Supplementary material is available at Political Science Quarterly online.

Acknowledgements

My thanks are due to numerous colleagues for their comments and insights on previous drafts of this article, including Baris Alan, Samuel Bagg, Joe Clare, Pearce Edwards, Zachary Elkins, John Gerring, Yann Kerevel, Gregory Koutnik, Joseph Lane, Alexander Orwin, James Stoner, Daniel Tirone, Mark Warren, Rachel Wellhausen, and Mary Witlacil. The Department of Government at the University of Texas and the Department of Political Science at Louisiana State University each hosted me in person for a presentation of some of the material published here, and I am grateful to the faculty and staff of both departments for making those visits both possible and productive. I owe my students thanks for making me realize that “regime-neutrality” is a better term than “regime-agnosticism.”

Data Availability

Data used in this article are all drawn from publicly available sources, as discussed in depth in the Supplementary Material.

Footnotes

Author notes