International Political Theory and the Global Environment: Some Critical Questions for Liberal Cosmopolitans


In recent years the intellectual resources of mainstream liberal political theory have been increasingly applied to environmental issues, and, of late, especially to those associated with climate change. That such issues are no longer treated as marginal concerns with respect to established agendas of political theory is a very welcome development. Nevertheless, I believe it is worth asking whether contemporary environmental concerns should in fact occasion a more radical questioning of inherited liberal framings of political theory—especially as questions of international and global relations come to the fore.

In the face of global challenges—particularly widespread severe poverty and environmental threats—contemporary political theorists are seeking to formulate cosmopolitan responses. While endorsing this cosmopolitan impulse, my critical questions concern the continued formulation of these responses within established liberal framings of political theory. Two major changes of context have occurred since liberalism came to maturity. First, the theory and practice of liberalism developed before awareness arose of environmental constraints on economic activity, and on the basis of presuppositions which those constraints serve to undermine. Second, liberalism developed historically with nation-states and thus prior to serious consideration of global, cosmopolitan, politics—as distinct from the international relations between those states as the primary and usually sole locus of sovereign power. To refer to these developments as changes of context is perhaps to understate the issue. For they could be thought such as to challenge certain fundamental, even constitutive, presuppositions of liberalism.

This, then, is the first general question to consider: what reason is there to presume that liberalism provides a suitable framing for political theory relating to global justice and the environment? In section 1, I aim to loosen this presumption by highlighting how liberalism as a whole is rather ambivalent with regard to both sets of issues. Regarding global justice, specifically, there are strong liberal arguments against cosmopolitanism. I suggest that if there are stronger arguments to be advanced against these in favor of cosmopolitanism then they might have to be supported by premises other than liberal ones. To provide a focus for this suggestion, I point out how cosmopolitanism might better be promoted within a socialist framing.

Having thus questioned why liberalism would necessarily provide an appropriate framing for global justice, I then question whether it is any more appropriate for framing global environmental issues. In section 2, the specific focus is on the idea of sustainable development and liberal understandings of it. The aim of sustainable development has been widely endorsed because, I suggest, its usual interpretation conveniently implies the possibility of a win-win-win scenario for protecting the environment while at the same time securing economic development and promoting global justice. Yet the truth may be less convenient, given the evident tensions between these objectives in practice. As economic development continues, the global environment deteriorates and global inequalities intensify. For that reason, I argue, we need also to attend to the losses and costs already associated with this development model. On this basis, it is more appropriate to reframe the aims relating to global environmental justice in terms of ecological debt. This reframing represents a challenge to liberal understandings of “sustainable development.”

This argument leads us then to ask critical questions about the empirical assumptions made in liberal theory regarding the material, biophysical basis of global political economy. Section 3 thus begins by asking what must be assumed when liberal cosmopolitans propose, for instance, globalizing Rawls's difference principle. For the application of such a principle appears to presuppose that justice means sharing more fairly the proceeds of ever-increasing wealth as are generated by a globally liberalized economy. This fosters the same idea it presupposes: that liberalism will ultimately be good for everyone, providing only that we ensure some redistribution of surplus value. This general idea seems also to underpin various suggestions made by contemporary political theorists that relatively modest reforms of international institutions may suffice to secure basic justice for all, including the eradication of world poverty. Once we heed the ecological constraints on growth of the productive economy, however, the basis of this faith is called into question.

The arguments thus far highlight ambivalences and uncertainties about liberalism—which are in good measure due to its own internal tensions, particularly between its progressive ethical goals and its morally more troubling association with capitalism. In section 4 I raise some questions about how progressive in fact is a distinctively liberal construction of ethical goals in the contemporary circumstances of global justice when these importantly include recognition of ecological finitude. I question the extent to which peculiarly liberal values should be preserved by cosmopolitanism, and focus in particular on the argument that what present circumstances require is an ethos of restraint and that this, in key respects, is the antithesis of a liberal ethos.

In conclusion, I sum up how, taking human rights as the fundamental normative touchstone, we can develop an account of what global justice requires in the light of ecological constraints that is recognizably cosmopolitan but at variance from liberal understandings of cosmopolitanism in key respects.

1. Liberalism, Cosmopolitanism, and Global Justice

Today there is an influential current of thought that conceives of cosmopolitan justice as a globalization of liberal principles. As Anthony J Langlois notes, “much of what goes by the name of contemporary cosmopolitanism is liberalism envisioned at the global level.”1 What gives coherence to this vision, he states, is the core of values associated with human rights. But there is a tension within it: while affirming a basic liberal right to own private property as a way of protecting the individual, Langlois notes it is also important in the liberal tradition that individuals who have less as a consequence of economic structures should be protected from the power of those better situated by virtue of their private property rights. He conceptualizes the tension here as one between political liberals, who “have a focus on the commonwealth, the common good,” and economic liberals whose focus on individual self-maximization translates into a concern about market advantage. What is distinctive about political liberalism, then, is a kind of commitment that economic liberals would presumably struggle to recognize as liberal at all.

I think Langlois's view fairly reflects that of many liberal cosmopolitans. The tension within it is one that is familiar from the history of liberal thought. Political liberals have a moral vision, captured by what C. B. Macpherson saw as the essential meaning of liberal democracy: “a society striving to ensure that all its members are equally free to realize their capabilities.”2 This vision, he thought, could be dissociated from liberal democracy in the sense of “the democracy of a capitalist market society”: “a liberal position need not be taken to depend forever on an acceptance of capitalist assumptions, although historically it has been so taken.” What I think we can consider a distinctively liberal view, quite generally, is the belief that this dichotomy between ethico-political liberalism and economic liberalism is in some way satisfactorily manageable and not problematic in any radical sense. It involves a belief that whatever criticisms might be directed against economic liberalism as the ideological counterpart of capitalism, the values of “ethical” liberalism or principles of “political” liberalism are immune from any implications of those criticisms.

A defining feature of liberalism is thus that it does not have a constitutive objective of ending capitalism's endless pursuit of accumulation. It may aim to make capitalism more “benign,” for instance through policies geared to distributive justice; and some left-leaning liberals may even emphasize their non-objection to the ending of capitalism, but there is no sense in which they would hasten it.

Thus, what “liberalism” means can in part be expounded from the perspective of critics who would contrast it with socialism. On such accounts liberalism is a political ideology which serves to legitimate subservience to the capitalist market either nakedly (e.g., neoliberalism) or more insidiously by expounding moral principles that will never get traction. C Wright Mills, for instance, wrote: “The ideals of liberalism have been divorced from any realities of modern social structure that might serve as the means of their realization. . . . if the moral force of liberalism is still stimulating, its sociological content is weak; it has no theory of society adequate to its moral aims.”3

Now this charge may seem overstated in relation to more recent liberal theory. John Rawls, for instance, was very clear that “ethical principles depend upon general facts and therefore a theory of justice for the basic structure presupposes an account of these [political economic] arrangements.”4 Rawls believed that an adequate sociological understanding was available for his purposes. Accordingly, however, he was also clear that his normative theory applies to a well-defined context—that of a modern liberal democratic state.

Given that liberal cosmopolitanism precisely aims at transcending the nation-state context, Wright Mills's criticism regains pertinence: how is the moral theory related to socioeconomic or political realities? But now the question is posed from within liberalism. Liberal critics of cosmopolitanism argue broadly as follows:

  • (i) Normative theorizing and theories of justice presuppose an account of basic socioeconomic facts about the context to which the theory applies.
  • (ii) Liberal political theory presupposes facts of liberal democratic states.
  • (iii) Not all states are liberal democratic states, and the world as a whole is not a liberal democratic state.Therefore:
  • (iv) Normative liberal political theory cannot simply be applied or extended to the whole world without, minimally, reviewing the factual assumptions it originally presupposed.

Rawls himself was clear on this point and developed the quite separate normative theory of international relations presented in his The Law of Peoples.

So how do things stand with liberal cosmopolitanism? To answer this we need to observe the distinction drawn by Charles Beitz between “moral cosmopolitanism” and “institutional cosmopolitanism.” The former is a moral philosophy that presupposes little specifically about the socioeconomic facts of the world; rather, it takes its bearings from the moral status of individuals. Correspondingly, it implies nothing very particular about institutional arrangements. If cosmopolitanism is understood purely as a moral position, then, it is not even inconsistent with nationalism.5 As David Miller put it, if this “weak” interpretation is all there is to it then “we are all cosmopolitans now.”6

Let us, then, consider what is required of “strong” cosmopolitanism:

  • (v) Cosmopolitanism as a normative theory applies to the whole world at the level of basic institutions.
  • (vi) To apply, the theory must presuppose an account of the facts of the whole world, including, importantly, economic facts.
  • (vii) One possible account of the facts is that of liberal international relations theory.
  • (viii) But (at least according to Rawls and others) this account does not support strong cosmopolitanism.

It would follow, then, that either Rawls gets liberal international relations theory quite seriously wrong or else strong cosmopolitanism must depend on some alternative account of the facts.

Political theorists who are consistent in their liberalism—and these include not only Rawls and his followers, but also people like Miller or Michael Walzer—argue that there are no grounds for strong cosmopolitanism. Samuel Freeman captures the problem: “what bothers many cosmopolitans is that global capitalism has created ways to elude political control by the world's governments . . . and part of the problem is that there is no global basic structure to deal with it.”7 It is no solution to advert to the moral interpretation of cosmopolitanism, for the problem is precisely the want of adequately cosmopolitan institutions. “It is a serious failing of cosmopolitan accounts of distributive justice that they discount the significance of social cooperation and regard distributive justice as asocial and apolitical.”8 Leif Wenar criticizes cosmopolitans' “insistence upon radical distributive principles without a prior demonstration that they can validate the most fundamental norms of global stability.”9

But if critics of liberal cosmopolitanism seem clearer than its advocates about what the challenge is, this should not mean they have the last word. Since Wenar cites Brian Barry in support of his concluding message—that in the global arena “the problem of establishing a peaceful order eclipses all others”10—I shall defer to the same authority. What Brian Barry has contributed to global justice debates is, I believe, a salutary dose of realism. Contrary to what Wenar implies, Barry's position does not entail a rejection of cosmopolitanism but rather of the illusions of liberal constructions of it. In his essay “The Continuing Relevance of Socialism,” Barry writes that “the transformation from a society ruled by the tyranny of the market to one of freedom requires collective control over the economy.”11 Whereas the tyranny of the market under global capitalism is not in principle challenged by liberals, Barry articulates the redistributive cosmopolitan case in terms that challenge the basic assumptions of a liberal framing of international relations: “industrial countries have achieved their present prosperity by first using their own natural resources and then, when these began to get scarce, by using those of the rest of the world at relatively low cost to themselves . . . In effect, this bonanza has been turned into accumulated capital that is regarded by these countries as their private property to do with as they choose.”12 Poor countries meanwhile have been left at a significant disadvantage in international economic relations. Thus, on the question of securing a peaceful world order, it is not obvious that cosmopolitans should acquiesce in the liberal perspective. Darrel Moellendorf, for instance, offers an alternative perspective:

The problem is that constitutional democracies with capitalist economies contain a system of incentives for persons with business interests in other states to advocate the use of state power to protect those interests. In other words, such states contain a class of people who have an interest in imperialist wars. Imperialist wars are externalities of a system in which profits accrue to private individuals. The only way to remove these externalities is through the establishment of an economic system that is not primarily based upon the profit motive or one that institutes public control over profits.13

Thus, Moellendorf recognizes that “socialist states may be a desirable goal of an egalitarian world order.”14

So cosmopolitans can share Wenar's view about the indispensability of a peaceful order and yet take a different view of what its accomplishment might entail. Rather than trust in the idealist tenets of liberal peace coming eventually to be shared by all peoples through the beneficent workings of a “truly free market,” cosmopolitanism potentially challenges the adequacy of liberal theory's construction of the facts of global political economy. Yet this is the break that liberal cosmopolitans in the tradition of Beitz and Pogge have refrained from completing.15 Their reason being, I think, that it involves entertaining explanatory hypotheses—such as those of world systems theory, for instance16—that are highly controversial within a liberal worldview and are strongly resisted by liberal critics of cosmopolitanism. At any rate, if “explanatory nationalism” is inadequate, this implies there must be a better explanatory theory to account for the contemporary circumstances of justice globally. But instead of showing how liberal theory is overoptimistic about the cooperative character of international political and economic relations, and thereby focusing on the necessary preconditions still to be achieved for a more just world order, the liberal cosmopolitans promote what is arguably an even more optimistic view of international relations as a “cooperative scheme” analogous to that of a modern nation-state. To suggest there is a “basic structure” globally equivalent to that presupposed by Rawls for a liberal democratic state—if this is understood as referring to a set of constitutionally and politically governed institutions—is tantamount to saying institutional cosmopolitanism is already a reality. This would be liberal political theory becoming once more a moral philosophy cut adrift from moorings in socioeconomic explanation.

However, the need to liberate cosmopolitanism from liberal assumptions becomes even clearer when we consider the global environmental context. Here we may find reason to appreciate Barry's claim that if the critique of capitalism implies socialism, then socialism, in turn, provides “the essential intellectual framework for environmental concerns.”17

2. Liberalism and the Global Environment: Sustainable Development or Ecological Debt?

As the other contributions to this special issue testify, appropriate political responses to global environmental threats have to appreciate their connection to matters of global justice. In light of this, I pose the question of why it should be anticipated that liberalism would be a necessary part of an appropriate response.

Various authors have discussed how liberalism can be compatible with environmental values. A major focus of their attention in arguing this is on how liberalism can—despite apparent presumptions to the contrary—accommodate noninstrumental value of nonhuman nature.18 I do not think we should be surprised to find that liberalism can accommodate a very wide range of ideas, including those of environmentalists and ecologists. The touchstone of liberalism as a moral philosophy is, after all, toleration. It is a relatively easy matter for reasonable people (at least when they find themselves in circumstances propitious to reasonableness) to reach broad agreement on desirable goals, and the recent liberal accommodation of environmental goals would be a case in point. But what is desirable has in practice also to be conditioned by what is possible, and this may provide a sterner test of the compatibility of liberal and environmental values—of how well liberal principles can be pressed into service of the aim of securing a decent environment, when that is understood, especially, as a global environment.

Liberal political theorists—indeed, like businesses and politicians of various stripes as well as publics—have had little difficulty in embracing the goal of “sustainable development” as articulated in the Brundtland Report some twenty years ago.19 Brundtland's idea of sustainable development held out the appealing aspiration of international cooperation being able to protect the environment while at the same time securing economic development and promoting global justice.

Yet the tensions between these objectives remain all too evident. As economic development continues, the global environment deteriorates and global inequalities intensify. The benefits of development carry serious environmental costs and are unjustly distributed. We have to recognize that the question of what it means to pursue sustainable development cannot be answered simply in terms of the benefits aimed at. We need to attend to the costs involved in achieving them and to the issue of their just distribution. For either the costs are borne equitably or those who are already victims of human rights deficits suffer worse.

The most fundamental norm proposed by Brundtland was that every person has a fundamental human right to live in an environment adequate for their health and well-being.20 The achievement of this right, for every person, would involve a more radical transformation of global relations than seems to be supposed in most discussions of sustainable development. For it cannot be achieved without also achieving a range of basic social rights; yet the environment also sets constraints on economic activity in the aggregate, and thus on the generation of the wherewithal to fulfill those rights. Considering the conditions of possibility of its achievement would thus suggest a profound challenge to the system of private property rights which allows some to draw immense profit from the world's natural resources while others are deprived of even the basic necessities of life.

This implication, that full respect for all human rights indivisibly involves a fundamental challenge to the existing order of property rights worldwide, is not generally foregrounded in discussions of sustainability. The dominant ethos remains that of liberalization, premised on the idea that we can all become better off—the rich raising standards so that the poor benefit too.

Yet the situation is that developed countries benefit from the use of more natural resources and environmental services than is either ecologically sustainable or arguably, especially in light of global inequalities, their due share. From the perspective of developing countries, we can be accused of running an “ecological debt.” The idea of ecological debt refers to the myriad ways that human societies today, both separately and in the aggregate, live, so to speak, beyond their ecological means. Ecological debt accrues whenever resources are taken out of their natural state at a rate faster than they can naturally be renewed, or when pollutants are emitted at a rate faster than they can naturally be assimilated. Allegations of ecological debt can be understood as claims that there is an unjust distribution of rights in the planet's various natural resources and environmental services. The allocation of rights is certainly haphazard: international law accommodates an array of property and sovereignty rights which have arisen historically as products of unregulated exploitation, wars, colonialism, power politics, ad hoc negotiations, and, in the best of cases, multilateral treaty agreements. Meanwhile, as international institutions create new rights—for example, carbon emissions rights or intellectual property rights in genetic resources—old rights, and particularly rights of territorial sovereignty, are being significantly modified. How just these regimes are, individually or in the aggregate, is a central question for assessing allegations of ecological debt. The magnitude of the problem is indicated by the estimate that to sustain the world's population at the current consumption levels of the affluent would actually require the resources of three additional Planet Earths. Even the current aggregate consumption level, which includes that of some billion people who exist in absolute poverty, is not sustainable. From this perspective, it appears that in myriad ways the affluent are “ecological debtors” who actively deprive the planet's poorer peoples of their “fair share” of the earth's ecological space.

This perspective is not a liberal one. Liberals, whether cosmopolitan or nationalist, envisage, at most, a more equitable sharing of economic benefits, but not any more radical questioning of the justifiability of the benefits themselves. To assess why it might be considered permissible to disregard such questions, we need to consider what can reasonably be presupposed about the general circumstances of justice in a globally liberalizing world.

3. What Do Liberals Assume About Global Political Economy?

Historically, liberal political theory developed in conjunction with the theory and practice of liberal political economy. The latter provided important elements of empirical reference that could be assumed as the context of application for normative theory. Today, as liberal cosmopolitans attempt to adapt their political theory to changing contexts, the relation of that theory to any determinate understanding of political economy is becoming increasingly uncertain.

To take an illustration, when liberal cosmopolitans argue for a global extension of Rawls's difference principle, they tend to neglect what I think is a crucial question: If this principle is supposed to operate so that the affluent are incentivized to maintain economic growth provided that the poor can benefit too, and all this is subject moreover to a “just savings principle,” how is that growth in fact going to be maintained, let alone something saved, when the consequences of economic growth to date are already—as is now, belatedly, recognized—threatening the very biophysical basis of human life on this planet?

The question as posed is of course an empirical one, and normative political philosophers do not establish socioeconomic facts. But they do choose the empirical presuppositions which give meaning and application to their theories. Rawls himself, for instance, when unfolding his theory of justice, expressly assumed that persons who deliberate competently about principles of justice “know the general facts about human society,”“understand political affairs and the principles of economic theory,” and “know the basis of social organization.”21 In the domestic context of a modern democratic state with a relatively stable socioeconomic order such an assumption may have appeared quite serviceable. But what comparable assumption about taken-for-granted knowledge and understanding can we make in the global context today? In particular, should we choose to assume that continued economic growth, on the part of the West, plus all the developing countries, is possible indefinitely without causing any serious ecological disruption that would undermine its advance? Or should we choose to be skeptical about that possibility?

This choice can be characterized as one between a “cornucopian” assumption and an assumption that that distribution of access to the world's resources is somewhat closer to a “zero-sum” game.22

To date, most international political theorists have not been very explicit about whether they accept or deny the cornucopian presumption. One who is explicit is Loren E. Lomasky, who states that “The world's wealth is not zero-sum, and thus to consume more is not to visit a harm on those who consume less.”23 Problems of poverty on his account are due to incompetent and corrupt regimes. If rich states have some responsibility for global injustices, it is not due to “insufficient zeal in applying the difference principle beyond borders. Rather, the flaw is rooted more deeply in a transgression against the grounding theory of liberalism: denial of equal liberty to those with whom one transacts.”24 So, it emerges, the solution is the complete opening of borders to free trade and movement of people within a framework of rule of law that provides well-defined property rights. The universalist strand in liberal philosophy only needs to be put into practice with a more complete liberalizing of the world market. Lomasky's view is, from an ecological point of view, even more retrogressive than that of those who advocate globalizing the difference principle. The latter recognize at least that wealth distribution requires some institutional intervention; on Lomasky's it would appear to be expected as a natural consequence of further freeing up markets. This expectation does not appear to have any clear warrant in terms of evidence to date; moreover, regarding the question of presuppositions, it is important to note that in the initial statement quoted Lomasky's reasoning is fallacious. It is true that the world's wealth is not zero-sum; however, at any given time its amount is determinate and finite; and if we add to this fact the consideration that it is very unevenly distributed, then we can see that the opposite conclusion to Lomasky's can be drawn, namely, that the high consumers can very well visit harm on the worst off by marginal increases in their own consumption.

Now I am sure that many liberals will contend that Lomasky's libertarian view is not a view of liberalism one needs to accept. But my question is whether there is a comparably clear alternative view of liberal theory's assumptions regarding political economy. Others may place less faith in the market's capacity to deliver material equity, yet I am not aware of any significant strand of liberal thinking that decisively challenges the cornucopian presumption.

In fact, I believe there is an uneasy agnosticism about this question among normative theorists who for the most part, accordingly, simply avoid it. Yet the question goes to a very basic issue about the material circumstances of justice globally. In what follows, therefore, I consider two lines of resistance against the ecological critique of the cornucopian assumption. The first involves asking whether criticism of cornucopianism really is in fact relevant as a criticism of liberalism; the second is whether cornucopianism can anyway be defended against ecological criticism.

(i) Is criticism of cornucopianism relevant as a criticism of liberalism? One potential reason why not would be the following. It is not natural resources that matter most fundamentally for justice, but a well-ordered society; and the amount of natural resources required to support good social order is not so great as to require the indefinite demands implied by the critique of cornucopianism. Therefore, liberalism, conceived as primarily concerned with implementing a liberal theory of justice, does not need to make the cornucopian assumption. However, this line of reasoning does not do more than offer a defense of the agnosticism referred to. It leaves unanswered some questions that are salient even within its own terms of reference. If natural resources are assumed to be of limited relevance, where are the limits and how are they maintained, and what would happen if they were overstepped? Of wider significance is the point that this reasoning is deployed by those liberals who resist the cosmopolitan generalization of liberal principles. Liberal cosmopolitans, by contrast, do not accept that less advantaged nations should have to settle for significantly less in terms of resources than do the affluent. They cannot therefore consistently appeal to this argument.

A different line of reasoning, however, has been canvassed, for instance, by the liberal environmentalist Mark Sagoff.25 Sagoff sets out reasons to be skeptical about whether natural resources matter particularly for global justice. For not only is what is needed not necessarily natural resources, but, more crucially, to the extent that what is needed has a natural resource component, it is not scarce. On this view—which is that of mainstream economists—what is needed is ingenuity and knowledge; these overcome scarcity. Therefore, a biophysical perspective on the economy—that which generates the critique of cornucopianism—is largely beside the point:

Quantitative increase in the physical dimension of the economy is neither necessary nor sufficient for economic growth in the conventional sense, which has to do with the value of production rather than the physical size of whatever is produced or consumed.26

Now in response it has certainly to be agreed that the “value” of production does not have any uniquely determinable correlation with the biophysical dimension of the economy, and so we cannot claim that economic growth necessarily entails any particular increase in biophysical loads. Any such connection in practice would be contingent on the character of the economy rather than the manifestation of a necessary truth in any metaphysical or logical sense. It is thus indeed conceivable that an economy could be so predominantly service-based, for instance, that any such contingent connection was absent.27 It is also conceivable that an economy could be so ecologically efficient that the connection was absent. However, we should not disregard the difference between what is conceivable and what is actual. In actuality, economic growth is invariably accompanied by increased biophysical pressures. So empirical actuality is to date certainly different from what is implied by the hypotheses that Sagoff conceives. Certainly, the hypothesis that natural resources matter for justice is not refuted merely by showing that alternative hypotheses can be conceived. So the critique of cornucopianism is not irrelevant. Still, it might be argued to fail.

(ii) So what of the second line of argument, that cornucopianism can anyway be defended against ecological criticism? The argument here is that even if natural resources might matter, they are nevertheless not finite, or at least not as limited in availability as the ecological approach has to suppose they are to support a principled critique of liberalism's presuppositions. This argument goes beyond the claim that economic growth could conceivably be sustained within natural limits to challenge the idea of natural limits itself. Sagoff locates this claim at the core of liberal economic thought: “Mainstream economists, such as James Tobin, Robert Solow, and William B. Nordhaus, typically state that nature sets no limits to economic growth. Trusting to human intelligence and ingenuity as people seek to satisfy their preferences and achieve well-being, these economists argue that people can choose among an indefinitely large number of alternatives.”28 This view as stated amounts simply to an article of faith—“trusting to human intelligence and ingenuity.” Can it draw on more persuasive support?

Sagoff notes three arguments offered by mainstream economists to show that knowledge and ingenuity are likely always to alleviate resource shortages. First, reserves of natural resources “themselves are actually functions of technology. The more advanced the technology, the more reserves become known and recoverable.” However, while there is some truth in this, the new reserves themselves stand to be consumed, and we cannot expect that ingenuity will create more reserves once all have been discovered and exploited, which is what would need to be the case to support the mainstream position. This, though, is where the second argument comes in: “advances in technology allow us not only to increase available reserves but also to employ substitutes for resources that may become scarce.” Again, there is some truth in this, but not enough: for it is debatable whether all resources can or should be substituted; it is also often the case that the substitution itself involves heavy use of other natural resources—as for instance, when plastics or chemical fertilizers requiring heavy use of fossil fuels substitute for naturally occurring materials. More decisively, as a supposed refutation of ultimate natural limits, it again does not appear to be an argument as such, since it merely asserts that substitutes will always be available, the thought underlying it being encapsulated in a restatement of faith “that in the aggregate resources are infinite, that when one flow dries up, there will always be another, and that technology will always find cheap ways to exploit the next resource.”29 Yet although this sentence uses the word “always,” the actual reports cited in evidence refer to a rather shorter time horizon: “the world is not yet running out of most nonrenewable resources and is not likely to, at least in the next few decades.”30 Clearly, the ecological economists' claim that resource depletion cannot continue indefinitely is not refuted by the observation that it can nevertheless continue for some decades yet. While the amount of ecological space available is partly contingent on the efficiency with which natural resources are exploited, this only means that the “natural limits” to resource exploitation cannot be fixed with precision once and for all. It would be a non sequitur to conclude from this, however, that there are no such limits.

Before turning to the third argument, it is worth emphasizing that the question of natural limits is not merely a technical one but also has profound implications in the sphere of global justice. For it is not just a question, as some suppose, of whether “we confront an age of scarcity in the near or, at best, the medium term.” The actual problem is both more mundane and more immediately pressing. Natural limits to economic growth are mediated by complex socioeconomic arrangements that produce wide disparities of outcomes. We only need to consider the plight of those millions who have lost lives, livelihoods, and homes through famine, disease or environmental displacement to recognize that these vast populations have already encountered natural limits. Thus, the claim that there is a problem of natural limits is not refuted by the observation that the rich can for the time being stave off the most serious effects of it for themselves.

This point is highly salient regarding the third argument offered by mainstream economists against the idea of natural limits. This is “that the power of knowledge continually reduces the amounts of resources needed to produce a constant or increasing flow of consumer goods and services.”31 As evidence, Sagoff notes that “Societies with big gross domestic products, such as Sweden, protect nature, while nations in the former Soviet bloc with much smaller gross domestic products, such as Poland, have devastated their environments.”32 However, even leaving aside the point that Sweden's territorial endowment of resources per capita is something like three times that of Poland, a fact which somewhat compromises this particular comparison, there are more general critical observations to make about the claim that economic growth ultimately serves to protect the environment.

In support of this claim there has been much attention devoted in the literature to the hypothesis of an “Environmental Kuznets Curve.”33 According to this hypothesis, during the process of economic development, countries tend to increase consumption of energy and materials at the same rate as growth in income, until a certain level of income is reached; beyond that point further increases in the level of output will no longer be followed by increases (at the same rate) of energy and material consumption. However, while there is evidence that a level of economic growth can bring with it a reduction of material and energy inputs per unit of output, this stops a long way short of supporting the general claim that growth is “good for the environment,” as does evidence that richer countries tend to protect their territorial environments better than poorer countries. For one thing, ecological efficiency gains per unit of output do not translate into greater environmental protection if growth of aggregate consumption more than offsets them, which it tends to for reasons inherent in the dynamics of growth. For another, even to the extent that a de-linking of growth and environmental degradation does occur, it is only after the country has reached a threshold of income and consumption of energy and materials per capita which is such a high one that it could not be emulated by all countries in the world without provoking likely ecological collapse in the meantime. Moreover, recent research has suggested that even if developed countries may go through a dematerialization phase, they can also then go through a re-materialization phase—so the inverted U-shaped curve would then appear as an N-shaped one, just depending on the time window used for observation. But the most fundamental criticism of this hypothesis is that it simply disregards how countries benefiting from economic growth can effectively export their environmental problems. When account is taken of how the rich nations can “externalise” the most polluting factors of production, the very basis of the Environmental Kuznets Curve hypothesis is called into question.

Empirical evidence showing that developed countries have improved their environments while growing their economies could only be held to support the economists' conclusion if it could also be shown that no appreciable resource depletion, pollution or increased entropy had occurred as a result of that same economic activity elsewhere in the global ecosystem. The evidence does not show this.

On the mainstream view, national economies are looked at in isolation and there is assumed to be no limit to their potential growth. This is because external impacts are simply disregarded, just as future impacts are discounted.34 This reflects the traditional liberal political perspective which sees the international order in terms of relations between discrete nation-states, each rationally pursuing its own interest.

Cosmopolitans do not share this perspective, and thus have little reason, as far as I can see, to accept the cornucopian presumption or rely on its futurevindication, given the evident need here and now to begin addressing—for reasons of both ecology and justice—the overarching problem of ecological debt. Yet contemporary liberal cosmopolitans appear to envisage relatively modest reforms that involve a greater sharing of the benefits of a liberal economy, rather than take a more probing approach to the sustainability of those benefits themselves. If I am right about this, then cosmopolitanism represents a challenge to liberalism of a more thoroughgoing kind than liberal cosmopolitans believe.

4. Critical Questions About the Liberal Ethos

Historically, the development of liberalism attended and served the development of the modern nation-state. Today, there is a widely held view that liberalism provides the normative principles not only for international moral and political cooperation, but for a cosmopolitan world order. However, I think we need to consider more carefully the extent to which cosmopolitanism really can be seen as an extension of liberal normative principles.

One of the few voices raising questions about this assumption is that of Katrin Flikschuh: “If liberal economic policy has significantly contributed to the production of extreme wealth in some parts of the earth and extreme poverty in other parts, can we continue to assert with confidence liberalism's ‘moral universalism’?” She believes this discrepancy should lead us at the very least “to ask whether liberal theory is capable of delivering, politically and economically, on its universalistic moral aspirations.”35 Liberal cosmopolitans tend to treat these moral aspirations as separable from their “contingent” material conditions of possibility. Flikschuh thinks we should find this more surprising than we have become accustomed to do when discussing proposed reforms to the global institutional order:

if the current global order is a major contributor to global poverty, and if the existence of that order turns out also to have been an essential prerequisite to the achievement of the liberal good life in mature liberal societies, how can the proposed reforms be expected to result in the widening of access to a life-style the possibility of which presupposes the global status quo?36

She thinks liberals may rather have lost sight of the material presuppositions of liberalism, and the implications of these:

Arguably, what is depicted in current liberal theorising as the individual quest for the critically examined good life, is observed by the global poor as an attractively high standard of living. . . . It is arguably these more material aspects of the liberal good life, rather than the promise of sharpening one's powers of critical reflection, that very many of the materially deprived global poor not only aspire to, but that they may also—given global terms of production—claim to have a just stake in.37

The trouble is that the liberal good life is expensive, she observes, and once the economic costs of making it more widely available are counted, liberal thinkers, mindful of the “strains of commitment,” frequently shrink back from promoting its universalization. But if liberals are partial and selective about the values and obligations to be universalized, this raises critical questions about the extent to which cosmopolitanism should be conceived in terms of the universalization of liberal values at all: “. . . is it not more reasonable to accept that institutional reform is likely to impose new moral constraints—the constraint upon citizens of the West, for example, to lower their material expectations so as indeed to enable distant others ‘to meet their basic needs with dignity’?”38 This suggestion can be reinforced by considering the ecological constraints on the required resources.

To meet global challenges, it is arguable that we—those of us currently over-consuming—require the inculcation of an ethic, an ethos, of restraint to accompany and underpin the recognition of our obligations to reduce our demands on resources. With this ethos we would “tread lightly” on the earth, and live in a way that is “simple in means, rich in ends,” with a focus on satisfactions derived from being as opposed to having. This characteristically “green” conception of the good life is arguably in humans' own enlightened interest, entailing the happiness that comes with exercise of our faculties and capacities rather than sacrifice. An ethos of restraint can foster the resourcefulness that makes such activities their own reward rather than something experienced as a cost or sacrifice.39 An ethos of restraint, though, has in turn to be deeply grounded in a sense, and principles, of justice. It certainly should not be confused with the self-righteousness of the affluent ascetic, who, from a position of material security, can demonstrate how easy it is to “dematerialize” their interests. From the point of view of the have-nots, it is a perverse mockery to be shown by well-off people how they can “do without.” The imperative of restraint, then, is a matter both of the good and of the right.40

Liberalism in many respects is an ethos of non-restraint.41 The way the liberal emphasis on the value of individual freedom is placed, in particular, does not clearly or directly tend to promote the kind of collective restraint that global environmental justice would seem to require. In pursuit of their traditional socioeconomic goals, liberals have embraced a constitutive encouragement of the pursuit of individual self-interest as the engine of aggregate economic development, something whose benefits in due course supposedly will (or can be made to) trickle down to all. We are accustomed to thinking of our liberties as achievements, but the fact that they were achieved by people in the past and at some cost to others tends to be passed over in contemporary versions of the implicit idealism manifest in various “win-win” ideas such as that of the possibility of adequate trickle down of wealth and of “sustainable development.” It is as if there is a line tacitly drawn in the sand: we want the rest of the world to be able to achieve what we have achieved; but we are not going to enter questions about relinquishing any of what we have.

This position is defensible with regard to human rights relating to security and respect of the person—rights to non-zero-sum goods that we should not give up at all. Indeed, the one clear principle of restraint provided by liberalism (and which protects such rights) is the harm principle, and this—in virtue of its negative tenor—is readily universalizable and can unproblematically be extended globally. But what of the non-non-zero-sum goods, those directly carrying costs, that are also required for the more complete fulfillment of human rights—including subsistence, livelihoods and adequate living conditions, including an adequate environment? Some theorists believe these can also be secured by applying the harm principle, albeit in novel ways, including as a guide for international institutional reform: both with regard to environmental harms (as, for instance, some libertarian defenses of using property rights for environmental protection do) and with regard to global injustices (as Pogge has recently argued regarding the avoidable harms caused under existing international institutional arrangements).

Refraining from harm, however, is only part of an ethos of restraint. For restraint in our use of ecological space means not only taking care not to harm others, but also reevaluating the nature and extent of the benefits we ourselves draw from our activities, even when these do not immediately or obviously precipitate some harm on others. If the rich are to retain what they already have, then the only way the poor can become better off is by increasing aggregate pressure on the world's resources; yet its ecology will not sustain this pressure. Taking this insight seriously, its normative implications should be deeply unsettling for those who envisage cosmopolitanism entailing the globalization of liberality. Unless they simply resort to faith in the possibility of indefinite material expansion through technological innovation, they have to face up to the implications of recognizing that, from an adequate normative perspective, what we today have is not, in fact, non-negotiably ours.

If any aspect of this is found “illiberal,” we should not assume, without further ado, that simply finding it so amounts to an objection. I have suggested elsewhere we should be cautious about any commitment to substantive liberal values when these conflict with ecological ones.42 This is not to suggest rejecting all the values liberal espouse. For instance, as Andrew Dobson argues, it may be that a conception of political citizenship embracing an ecological orientation has as its principal virtues “the liberal ones of reasonableness and a willingness to accept the force of the better argument and procedural legitimacy.”43 But while reasonableness and a willingness to accept the force of the better argument may be values liberals typically espouse, I am not sure they are peculiarly liberal values. To think that liberalism has a monopoly of reasonableness could prove, in today's world, to be a dangerous illusion. Liberalism has indeed played an important historical role in establishing and institutionalizing such principles as tolerance, freedom, civil rights, procedural equality, and non-discrimination. Yet one can hold these principles to be dear while also believing, say, that all property should be communally owned, or that economic growth should be severely constrained, or that citizens have as full and dense a set of obligations as they have rights. Reasonableness does not entail an obligation to tolerate the intolerable. Non-restraint is, from a green point of view, intolerable. The planet cannot tolerate it; the worst off cannot tolerate its effects; future sufferers from humanly induced ecological crises will not be able to tolerate it either.

In short, then, it seems to me there is good reason to challenge more actively and determinedly the assumption that cosmopolitanism represents a seamless continuation of liberalism.


Liberalism has fostered vitally important norms with its commitment to democracy and human rights, and I do not assume that democracy and human rights can necessarily flourish in just any non-liberal political context. What I do suggest is that our thinking about how such norms can be promoted in the altered context of a globally interdependent, geopolitically re-orientated, and ecologically constrained world should take its bearing from a more radically open-spirited assessment of the new context. The project of political theorists to see about extending domestic theories has of course needed to be tried—since not all questions are new and inherited wisdom may have much to teach. But I think we also need to free ourselves from any assumption that this approach will suffice. We need to be open to the more radical questions thrown up. This means some “thinking outside the box” of liberalism, and a readiness to reframe questions in new ways. So while I am not implying that all the values and assumptions of liberalism should be negated, I do think we need to be open to the possibility that some might. Approaching them in a spirit of immanent criticism, we can consider the kind of direction in which this might lead.

The idea that an immanent critique of liberalism leads in the direction of socialism is to be found in the work of influential political theorists of the twentieth century. If socialism can be arrived at as an immanent goal of liberalism—because the realization of individual freedom for all actually requires socialized material justice as a precondition—then we now also need to recognize that material justice for all has its own precondition, namely, the integrity of the compendious resources of the biophysical world.44 Thus, if the immanent critique of liberalism leads in the direction of socialism, then the immanent critique of socialism leads in the direction of what may be called ecologism—or, given that this term may be understood in different ways—ecological socialism.

We need to recognize rather than evade the extent to which existing international institutions are complicit in the logic of capital accumulation.45 For it is this which so structures global economic imperatives as to render the human calamities associated with ecological debt a systemic outcome and not only a moral failure. The injustices which existing institutions preside over should not be assumed to be simply incidental failings of an otherwise just system.

Furthermore, if liberals rightly emphasize the value of peace and stability in the global order, then we should also be concerned about the serious potential threats to global stability posed by “ecological insecurity.” For the same circumstances that can be condemned as “ecological debt” in a discourse of justice are liable increasingly to occasion security threats, as already evidenced in growing conflicts over resources. Liberalism—with its focus on “ecological modernization” and the win-win-win interpretation of sustainable development—would not appear to have an adequate theoretical apparatus for recognizing why such conflicts are becoming more pervasive.

Yet departing from liberalism is not to depart from the goals of cosmopolitanism. In particular, we can readily affirm a recognizably cosmopolitan conception of universal rights and global justice. The basics of justice, on the conception referred to here, include a universal right of access to the necessary means for a decent life. I take it as axiomatic that there is this fundamental right: for if there were not, then the very idea of human rights would be hollow; and if we could not rely conceptually and normatively on the idea of human rights as a touchstone for ideas of justice, I doubt we could talk both cogently and persuasively about global justice at all. As a material premise, I take it that the means of life necessarily and importantly include biophysical resources; biophysical resources, compendiously, can be referred to by the term “ecological space.” From these premises it follows that a right of each human to a sufficient allocation of ecological space is a human right.46

This right is what I believe cosmopolitanism commits us to as a matter of justice. Whether liberalism does, I am not so sure. Perhaps what most crucially has to be debated is whether taking this right seriously means according it priority over any mere right of property that conflicts with it.47

Earlier versions of this article were presented at seminars of International Politics and Political Theory research groups at the University of Edinburgh, and at the Centre for the Study of Global Ethics at the University of Birmingham. I am grateful for comments from participants, as well from Carol Gould and Piki Ish-Shalom.


  • 1

    Anthony J. Langlois, “Human Rights and Cosmopolitan Liberalism,”Critical Review of International Social and Political Philosophy 10, no. 1 (2007): 29–45, at 29.

  • 2

    C. B. Macpherson, The life and Times of Liberal Democracy (Oxford: Oxford University Press, 1977).

  • 3

    C. Wright Mills, “Liberal Values in the Modern World,” in Power, Politics and People: The Collected Essays of C. Wright Mills, ed. Irving Louis Horowitz (New York: Oxford University Press, 1963), 189.

  • 4

    John Rawls, A Theory of Justice (Oxford: Clarendon Press, 1972), 265.

  • 5

    See, for example, Kok-Chor Tan, Justice Without Borders: Cosmopolitanism, Nationalism, and Patriotism (Cambridge: Cambridge University Press, 2004), 193–94.

  • 6

    David Miller, National Responsibility and Global Justice (Oxford: Oxford University Press, 2007), 28.

  • 7

    Samuel Freeman, “Distributive Justice and The Law of Peoples,” in Rawls's Law of Peoples: A Realistic Utopia? ed. Rex Martin and David A. Reidy (Oxford: Blackwell, 2006), 258.

  • 8


  • 9

    Leif Wenar, “Why Rawls is Not a Cosmopolitan Egalitarian,” in Rawls's Law of Peoples: A Realistic Utopia? ed. Rex Martin and David A. Reidy (Oxford: Blackwell, 2006), 107.

  • 10

    Ibid., 110.

  • 11

    Brian Barry, Democracy, Power and Justice: Essays in Political Theory (Oxford: Clarendon Press, 1989), 529.

  • 12

    Ibid., 522.

  • 13

    Darrel Moellendorf, Cosmopolitan Justice (Boulder, CO: Westview Press, 2002), 174.

  • 14


  • 15

    I raised this issue in relation to Pogge in Tim Hayward, “On the Nature of Our Debt to the Global Poor,”Journal of Social Philosophy 39, no. 1 (2008): 1–19. Pogge has certainly focussed attention on crucial facts about the contemporary world order, including facts about international institutions. But I am not aware of any serious dispute about these facts on the part of an “explanatory nationalist” like Miller. The challenge posed to Pogge by Miller, I think, is to fill in the layers of social theoretic explanation between “our” individual moral responsibilities and the effects of global capitalism (see, for example, Miller, National Responsibility, 238–47).

  • 16

    I single out this kind of theory (which, incidentally, Barry also considers worth taking seriously—see Democracy, 476–90) because of its prima facie coherence with a global ecological perspective: see Tim Hayward, “Global Justice and the Distribution of Natural Resources,”Political Studies 54, no. 2 (2006): 349–69.

  • 17

    Barry, Democracy, 532.

  • 18

    See, for instance, Derek Bell, “Liberal Environmental Citizenship,”Environmental Politics 14, no. 2 (2005): 179–94; and, in the same issue, Simon Hailwood, “Environmental Citizenship as Reasonable Citizenship,” 195–210.

  • 19

    World Commission on Environment and Development, and United Nations Environment Programme (The Brundtland Report), Our Common Future (New York: The Commission, 1987).

  • 20

    In Tim Hayward, Constitutional Environmental Rights (Oxford: Oxford University Press, 2005) I argued in support of that right and in favor of its implementation in states' constitutions as well as in international law.

  • 21

    John Rawls, A Theory of Justice (Oxford: Oxford University Press, 1971), 137.

  • 22

    See, for example, Alf Hornborg, “Cornucopia or Zero-Sum Game? The Epistemology of Sustainability,”Journal of World-Systems Research 9, no. 2 (2003): 205–16.

  • 23

    Loren E. Lomasky, “Liberalism Beyond Borders,”Social Philosophy and Policy 24, no. 1 (2007): 206–33, at 214.

  • 24

    Ibid., 208.

  • 25

    Mark Sagoff, “Carrying Capacity and Ecological Economics,” in Ethics of Consumption, ed. David A. Crocker and Toby Linden (Lanham, MD: Rowman & Littlefield, 1998).

  • 26

    Ibid., 37.

  • 27

    By service-based I am thinking of personal services provided by labor without significant resource usage; I am not thinking about the “service” sector of Finance Insurance and Real Estate, which has a much more complicated relationship with productive sectors of the economy.

  • 28

    Sagoff, “Carrying Capacity,” 29.

  • 29

    Ibid., 31.

  • 30


  • 31

    Ibid., 30.

  • 32

    Ibid., 38.

  • 33

    For an overview, see for example, David I. Stern, “The Rise and Fall of the Environmental Kuznets Curve,”World Development 32, no. 8 (2004): 1419–39.

  • 34

    See, for example, Caney's article in this issue.

  • 35

    Katrin Flikschuh, “The Limits of Liberal Cosmopolitanism,”Res Publica 10, no. 2 (2004): 175–92, at 176.

  • 36

    Ibid., 191.

  • 37

    Ibid., 183.

  • 38

    Ibid., 189.

  • 39

    See, for example, Tim Hayward, “Ecological Citizenship: Justice, Rights and the Virtue of Resourcefulness,”Environmental Politics 15, no. 3 (2006): 435–46.

  • 40

    As such, it would also depend more widely on a sense of human solidarity and an inculcation of perspectives and values of the kind indicated, for instance, by Andrew Dobson, Citizenship and the Environment (Oxford: Oxford University Press, 2003).

  • 41

    A somewhat contrasting view is presented by Marcel Wissenburg, “Sustainability and the Limits of Liberalism,” in Debating the Earth: The Environmental Politics Reader, 2nd ed., ed. John S. Dryzek and David Schlosberg (Oxford: Oxford University Press, 2005). His suggestion that liberalism can be consistent with the principle of restraint seems to depend on the premise that “liberalism is ultimately about taking one's individual responsibility seriously” (p. 180) and a stipulation that “[t]he critique of economic liberalism must . . . be judged on its own merits and cannot reflect on political liberalism” (p. 181).

  • 42

    Hayward, “Ecological Citizenship,” 440.

  • 43

    Dobson, Citizenship and the Environment, 89.

  • 44

    There are of course reasons for concern about the nonhuman world over and above its function as a resource for humans. I assume their efficaciousness will correlate positively with the inculcation of restraint.

  • 45

    This point is forcefully argued by John Bellamy Foster and Brett Clerk, “Ecological Imperialism: The Curse of Capitalism,”Socialist Register (2004): 186–201.

  • 46

    For more on this argument see, for example, Hayward, “On the Nature.”

  • 47

    In saying this I am drawing a distinction which some liberals will challenge on the ground that some property rights can actually be justified as human rights. A discussion of this question not being possible here, I simply note that my phrasing “any mere right of property” leaves open that possibility.