Environmental Obligations and the Limits of Transnational Citizenship

Concepts and Conceptions of Citizenship

Dobson defends an ecological model of citizenship, grounded in obligations of justice which extend beyond the boundaries of the state. His model emerges out of reflection upon the strengths and weaknesses of a number of different theories of citizenship, particularly in the light of concerns about environmental degradation. Before examining Dobson's specific ecological model, however, I shall in this section consider his typology for classifying different theories of citizenship, and then introduce my own favoured way of dividing up the terrain since this plays a role in the argument I develop for rejecting his model.

Dobson divides theories of citizenship into three types – liberal, civic republican and post-cosmopolitan – on the basis of four main criteria: whether the theory focuses on rights and entitlements or duties and responsibilities, and whether these are regarded as contractually based; whether it restricts citizenship to the public sphere or allows it to include activities in the private sphere; whether it incorporates virtues, and if so, what kind; and whether citizenship is regarded as territorially bounded or not. According to Dobson, the liberal theory and the republican theory both regard citizenship as a contractual relationship that is territorially bounded and focused on the public sphere. But the republican theory differs from the liberal theory because it analyses citizenship primarily in terms of duties, responsibilities and virtues, while the liberal theory analyses it primarily in terms of rights and entitlements. The post-cosmopolitan theory differs from both liberal and republican theories in denying that citizenship is necessarily territorially bounded or contractually based, and in extending it to activities that take place in the private sphere. Like the republican theory it focuses on duties and responsibilities rather than rights and entitlements, but because of its extension to the private sphere it incorporates a wider range of virtues and responsibilities.

In place of Dobson's classification scheme, I offer what is at root a simpler analysis, but one which I think enables us to understand better what is at issue between different theories of citizenship while at the same time identifying limits to the concept of citizenship. My analysis employs the distinction between concept and conception, and argues that there is a core concept of citizenship which constrains what can count as an adequate conception of it. The distinction between concept and conception is usually credited to John Rawls. Even though he insists that invoking it when theorising about justice ‘settles no important questions’ (Rawls, 1971, p. 6), the distinction is not without difficulties, some of which I shall discuss later in this section. But for the time being, I shall simply work with it. How then should we draw the distinction between concept and conception in the context of theorising about citizenship?

Consider the following proposal, which seems to me to be plausible: citizenship is a relationship in which the parties enjoy rights or entitlements as part of that relationship, and incur obligations, duties or responsibilities to each other as a result of that relationship or, more precisely, in virtue of some aspect of their common citizenship. (For example, it might be held that citizens enjoy a right to personal protection through legal institutions and owe special obligations to each other to obey the law because they benefit from the protection it provides.) Specific conceptions of citizenship may add further elements to those contained in this characterisation; for example, they may suppose that to be a citizen is to stand in a relationship of equality to fellow citizens and then spell out what this involves, and these other features might in some cases ground obligations, duties or responsibilities of citizenship. (In this context it might be maintained that standing in a relationship of equality requires citizens to treat each other as equals in what has traditionally been regarded as the private sphere.)

Competing conceptions will also arise from different accounts of the rights or entitlements that citizens enjoy and from different accounts of the obligations, duties or responsibilities they incur. For example, some maintain that these rights are all civil and political in character, while others contend that they include social rights. Some maintain that citizens have obligations only to respect the rights of others and to obey the law, while others argue that they also have obligations to participate politically and engage in various forms of public service. Some maintain that rights or entitlements that are part of citizenship may be conditional; for example, that welfare rights are to be enjoyed only by those who are willing to work when they are able, while others insist that all these rights and entitlements are unconditional. If we want a more complex classificatory scheme that, like Dobson's, allows us to distinguish a variety of different types of citizenship, we can do so in a number of different ways; for example, by distinguishing between kinds of citizenship rights in terms of their content, the domain in which they operate or what is thought to justify them.

For this analysis to seem plausible, we need to distinguish between two different concepts of citizenship: a moral concept and a non-moral concept. The non-moral concept equates citizenship with membership in a polity, and sees it as consisting in whatever rights and entitlements are enjoyed in virtue of that membership and in whatever duties, obligations or responsibilities are attributed on the basis of it. These rights, entitlements, obligations and responsibilities will differ from one state to another, and hence what is involved in being a citizen of a state may vary. In contrast, the moral concept with which I am concerned specifies an ideal: it spells out the rights or entitlements that have to be enjoyed in order for a relationship to count as one of full citizenship, and the obligations, duties or responsibilities which are necessarily incurred in that relationship. The moral concept allows us to stand in judgement on the packages of rights that are delivered by particular states, and provides us with a basis for evaluating the behaviour of citizens within those states: we might conclude that a person is not really a citizen, or not a full citizen, because he or she lacks the complete range of rights which, morally speaking, are required for full citizenship, or we might conclude that an individual is a bad citizen on the grounds that he or she does not fulfil his or her obligations as a citizen.

One of the merits of the analysis I have given is that it allows us to diagnose what is really at issue between liberals and republicans. Dobson follows others in claiming that the liberal conception of citizenship treats it as a status that is exhausted by the enjoyment of rights while the republican conception treats it as a matter of discharging responsibilities. The liberal conception is therefore often viewed as passive; the republican conception as active. ³ The analysis I am proposing acts as a corrective to this way of drawing the distinction. First, it maintains that liberal and republican conceptions both acknowledge the existence of rights and obligations of citizenship; they differ primarily in terms of how extensive they suppose the obligations of citizenship to be. The liberal conception maintains that we have obligations to respect the rights of others and to obey the law. ⁴ (Indeed it cannot avoid positing the former because some of the citizenship rights it acknowledges are rights not just against the state but also against other citizens, and entail correlative obligations.) The republican conception, in contrast, posits a more extensive range of obligations of citizenship, including an obligation to participate politically. Second, it is misleading to claim that the republican conception of citizenship is active while the liberal conception is passive (Dobson, 2003, pp. 39–40). Even though the republican conception recognises a more extensive range of obligations of citizenship, it accepts that citizens may fail to live up to them. They do not cease to be citizens as a consequence, however; they are just bad citizens. Good citizenship consists (at least in part) in fulfilling the obligations of citizenship, so the liberal and republican conceptions of citizenship may simply offer divergent accounts of what it is to be a good citizen.

Both the republican and liberal conceptions can also make space for the idea of the virtues of citizenship. These virtues might be conceived as those states of character that dispose people to fulfil the obligations of citizenship (Dobson, 2003, pp. 66–7). Conceptions of citizenship need not give a reductive account of the virtues, however. They might suppose that the virtues of citizenship are those states of character that dispose people to perform acts that are praiseworthy from the point of view of citizenship, and they can hold that some acts may be praiseworthy from this point of view even when they are not obligatory.

In response it might be said that this way of accommodating the republican view of citizenship simply misunderstands it. David Miller, for example, argues that according to the republican view, citizenship involves more than the possession of rights and obligations. It also involves:

being willing to take active steps to defend the rights of other members of the political community, and more generally to promote its common interests. The citizen is someone who goes to the aid of a fellow citizen who collapses in the street, or who intervenes when he is able to prevent a criminal act being committed (Miller, 2000, p. 83).

Being a citizen also involves playing ‘an active role in both the formal and informal arenas of politics’ (Miller, 2000, p. 83).

I do not wish to deny that the term ‘citizenship’ has been used in this way historically but this usage now seems alien to us. We do not say that someone who does not come to the aid of a fellow citizen who collapses in the street is not a citizen, or even that someone who fails to participate politically is not a citizen, as we would have to do on the republican view to which Miller subscribes. This is not because we have simply abandoned the republican view. Far from it. The republican view is reflected instead in those accounts of citizenship which maintain that being a citizen involves incurring an extensive set of responsibilities to promote the common interests of the political community. Someone who does not discharge their responsibilities, for example, because he or she fails to come to the aid of a fellow citizen when that is demanded, does not cease to be a citizen, though he or she falls short of being a good citizen. Indeed if we insist that someone counts as a citizen only if he or she participates politically and comes to the aid of other citizens in need, we will lose the distinction between being a citizen and being a good citizen – a distinction that is central to ordinary usage. ⁵

Does my analysis also enable us to distinguish what Dobson calls post-cosmopolitan citizenship from both liberal and republican conceptions? According to Dobson, post-cosmopolitan theories have a distinctive account of the duties and responsibilities of citizenship. They maintain that these duties and responsibilities are non-contractual, govern activities that take place in the private as well as the public sphere and extend beyond the borders of the state. In principle, the core concept of citizenship can accommodate each of these ideas. It does not place any explicit restriction on the content or domain of citizenship obligations, and does not rule out the possibility of obligations of citizenship that extend beyond state boundaries. But the core concept I have identified does impose constraints which pose a challenge to some accounts of post-cosmopolitan citizenship. Any adequate account of citizenship has to explain how the duties, obligations and responsibilities to which it is wedded are owed to fellow citizens in virtue of some aspect of their common citizenship. As I try to show in the next section, this is a demanding condition – and one that Dobson's favoured ecological model has trouble meeting.

Even though the analysis I am proposing offers a satisfying way of distinguishing between liberal, republican and post-cosmopolitan conceptions of citizenship, it may nevertheless strike some as misguided. The concept/conception distinction I am employing seems to be on shaky ground in so far as it assumes that we can identify an uncontested, and indeed uncontestable, core concept, which constrains the different conceptions or interpretations of citizenship that are possible. Why cannot the core concept of citizenship that I have identified be contested, and indeed is that not precisely what is happening when feminists, cosmopolitans or environmentalists advance conceptions of citizenship that are apparently incompatible with it? From this perspective, it can seem that I am simply stipulating a set of conditions that different conceptions of citizenship must meet in a way that lacks any authority to constrain the way in which the concept may evolve in the face of changing social, political and economic circumstances, and the intellectual developments that may occur in the light of these changes. ⁶

Put more philosophically, the approach I am using might seem to assume too rigid a division of statements into analytic or synthetic, that is, into statements which are true solely in virtue of the meaning of the terms that are used to express them, and statements which, if true, are true not only in virtue of those meanings but also because of the way the world is. ⁷ Those who reject the analytic/synthetic distinction, so understood, can nevertheless suppose that there are statements which we are reluctant to give up in the face of changing circumstances because of the intellectual costs this would impose on us, without maintaining that to do so would simply be to change the meaning of the terms we are using. From this perspective, insisting on my analysis of the concept of citizenship in the face of challenges to it that have arisen in the light of changes in our social, political or economic circumstances would be to refuse to engage in a discussion about the costs and benefits involved in changing our ways of thinking about citizenship.

There is some force to these points. In so far as the concept/conception distinction is tenable, it should not be regarded as drawing a line between what can be contested and what is beyond contest. It should not presuppose a rigid distinction between analytic and synthetic statements. But it can avoid falling into either of these traps. The analysis I have given can be interpreted as identifying deeply rooted assumptions about the nature of citizenship that might in principle come to be rejected without necessarily entailing that the term was being given an entirely new meaning in a way that would preclude the need to choose one usage rather than another. When new models of citizenship are proposed that abandon one or more of these assumptions, they should not be dismissed out of hand, but we need to assess the costs and benefits of rejecting the assumptions. If we are presented with models of transnational citizenship which abandon assumptions that are built into the core concept, we need to be persuaded that a radical change of this kind to the concept of citizenship is warranted.

Challenging Ideas of Transnational Citizenship

The core concept that I am claiming underlies different conceptions of citizenship is flexible and accommodating. For example, it allows the possibility that the obligations of citizenship extend to what is conventionally regarded as the private sphere and include obligations to perform one's fair share of domestic labour, such as the work involved in caring for children and other dependants (Bubeck, 1995; Lister, 1991; Prokhovnik, 1998). It allows the possibility that equal citizenship for all may require group-differentiated rights which give different packages of rights to different groups. ⁸ The core concept does, however, draw limits that some accounts of transnational citizenship transgress, in a way that, I shall argue, poses a challenge for those accounts rather than a case for conceptual change. Accounts of this kind, such as Dobson's model of ecological citizenship, extend the relationship of citizenship beyond the boundaries of the state, so that two people may lie in that relationship even though they are not members of the same state. The core concept of citizenship I have identified does not rule out the possibility in principle that the relationship of citizenship might be extended beyond the boundaries of the nation state in the way that Dobson's model does, but it places constraints on how this can be achieved – or, at least, demands that we weigh the costs that are incurred when this extension involves abandoning deeply rooted assumptions.

According to the core concept, if a relationship is one of citizenship, then we should be able to identify some set of rights that are enjoyed by fellow citizens in virtue of their citizenship. Is it enough here for various moral rights to be attributed to them in virtue of some property they share such as their common humanity or the fact that they are all persons? If so, provided we can make sense of the idea of natural rights or human rights, or something similar, then this constraint has been met. But enjoying a right is different from simply having it, in the sense in which a person might be said to have moral rights independently of social, political or legal arrangements. One does not enjoy a right unless that right is, at least in general, respected by others. We can make sense of a community in which there is no legal apparatus to enforce rights or to ensure that entitlements are received, but in which those rights are nevertheless enjoyed as members of that community and each has what they are entitled to (perhaps as a result of voluntary arrangements), but this will be relatively rare in practice. In general at least, enjoying rights as a member of a group will require some methods of enforcement or protection. Enforcement of rights need not require a full-blown state in the traditional sense, but it will normally require some network of legal and political institutions. So we can make sense of the idea that people might enjoy rights as transnational citizens, but, with the exception perhaps of European citizenship, it would seem that we are nevertheless a considerable distance away from realising these forms of citizenship in practice. ⁹

Now it might be replied that even this (largely practical or empirical rather than philosophical) problem with ideas of transnational citizenship can be set aside, on the grounds that we can and should simply abandon the common-sense but deeply rooted assumption that citizenship requires the enjoyment of rights, and instead hold that it merely requires that fellow citizens incur some set of responsibilities, obligations or duties to each other. In this way Dobson distinguishes between environmental citizenship, which is primarily concerned with the enjoyment of rights in the public sphere, ¹⁰ and his model of ecological citizenship, which is concerned with virtues and responsibilities that are equally at home in the private sphere and which demand as a matter of justice that one reduce the size of one's ecological footprint to a sustainable level (Dobson, 2003, pp. 88–90). The responsibilities of ecological citizenship extend beyond the borders of the nation state because individuals are bound together by obligations of justice, rather than obligations of mere charity, as a result of participating in practices (‘material activities’), the effects of which are felt beyond state boundaries. In particular Dobson argues that those in developed states incur obligations of justice (to act individually and collectively) as a result of participating in practices which create unsustainable ecological footprints, and which adversely affect others who do not participate in these practices to any great extent – and indeed who create much smaller and sustainable ecological footprints but have to endure, for example, higher temperatures and rising sea levels (Dobson, 2003, ch. 3).

But there are serious costs involved in travelling along this path. According to the model that I outlined, obligations of citizenship are owed to fellow citizens at least partly in virtue of some aspect of that relationship, that is, in virtue of some aspect of the common citizenship they enjoy. Yet the idea of ecological citizenship as Dobson understands it does not seem able to explain why the special obligations of citizenship that it recognises are owed to fellow citizens in virtue of their common citizenship rather than to others in virtue of our non-citizenly interactions with them. ¹¹ In so far as we detach the idea of ecological citizenship from the shared enjoyment of rights and entitlements, we make it difficult to explain how the responsibilities and obligations it involves can arise from (some aspect of) the relationship of citizenship.

Consider two possible responses that might be made on Dobson's behalf. First, it might be argued that there is no good reason to insist that the obligations which citizens owe to each other must be owed in virtue of their common citizenship. ¹² Second, it might be argued that the ecological account of citizenship can indeed make sense of the idea that the obligations which citizens owe to each other are owed in virtue of their common citizenship. ¹³ Consider the first kind of response, which in effect involves abandoning what I claim is part of the core concept of citizenship. In its favour it might be said that it completes the process that has already begun of disentangling the idea of citizenship from the idea of rights and entitlements. The most obvious way of construing the idea that the obligations of citizenship are owed in virtue of (some aspect of their) common citizenship is to take it to mean that they are owed in virtue of the shared enjoyment of rights or entitlements. ¹⁴ If a notion of citizenship is to dispense with rights and entitlements, then it cannot justify obligations of citizenship by reference to them, so it might seem that the next logical step is to deny that obligations of citizenship are owed in virtue of common citizenship at all. But can this further step be defended?

An approach of this sort need not involve denying that obligations of citizenship are special obligations, where special obligations are contrasted with general obligations: they may still be special obligations because they may be obligations we owe only to particular people with whom we interact in some significant way. ¹⁵ For example, we might hold (as Dobson seems to) that they are obligations incurred to particular others as a result of past or present interactions, justified by appeal to general principles of justice, such as the principle that people should be compensated for injustices they suffer or have suffered. But it is hard to see how we could make this sort of move without abandoning the crucial distinction we surely want to draw between obligations (of justice or otherwise) that are part of citizenship and those that are not. Once we allow that obligations of citizenship can be owed to someone simply in virtue of unjust past or present interactions, we lose sight of what it is that makes them obligations of citizenship as opposed to, say, merely obligations of justice.

The point here is not that citizenship is a contractual relationship, or that it requires the kind of reciprocity that is characteristic of contractual relationships, such that each citizen owes every other citizen the very same obligations that they owe to him or her. Dobson may well be right in suggesting that we should resist this idea (see Dobson, 2003, p. 47). The point is rather that, paradigmatically, an obligation of citizenship is one that is owed to another as a fellow citizen, and it is hard to see how this condition could be satisfied by an obligation unless it is owed to him or her in virtue of some feature of their common citizenship. This does not rule out the possibility that one citizen may owe obligations to other citizens (in virtue of their common citizenship) that they do not owe to him or her, perhaps because of the advantaged circumstances in which he or she lives. Indeed in principle one person may owe an obligation to others as fellow citizens even if he or she does not owe that obligation to each and every fellow citizen, and even if they do not each owe it to him or her as well.

Simon Hailwood denies that obligations of citizenship must be owed to others as fellow citizens. He maintains that it is possible for ‘citizens of one state to (correctly) recognise obligations as citizens of their own state to citizens of a different state, even though the latter are not fellow citizens’ and continues that ‘the stipulation that all citizen obligations are necessarily focused entirely on the “home affairs” requirements of one's own political community seems unjustified. Some citizenly obligations may be thought of on the model of being required qua good citizen to endorse some kind of “ethical foreign policy”’ (Hailwood, 2005, p. 202). Now a person can, as a citizen, argue the case for an ethical foreign policy, for in doing so he or she is exercising his or her civil and political rights – rights he or she has as a citizen – and perhaps even fulfilling an obligation to fellow citizens to participate politically. He or she may also be acting on general obligations has towards other human beings which s/he can fulfil by exercising his or her rights as the citizen of a particular state. But it is far from clear that s/he is thereby acting on obligations s/he owes as a citizen of one state to citizens of other states.

Let me concede, however, that it is intelligible, at least, to suppose that a citizen of one state may owe an obligation, as a citizen of that state, to citizens of another state. For example, he or she may owe that obligation because of the help that his or her state received from this other state in the past, or as a result of damage that his or her state inflicted on it. But these obligations are secondary, parasitic on there being a relationship in which citizens enjoy rights and incur obligations to fellow citizens in virtue of their common citizenship. They do not pave the way for a type of citizenship in which the obligations owed by one citizen to another are never owed in virtue of their common citizenship.

Can the obligations which advocates of ecological citizenship regard as citizenship obligations be interpreted in a way which shows that they are indeed owed in virtue of their common citizenship? Some scepticism is warranted here. Even if each of us has an obligation to act in ways that are environmentally sustainable, in so far as it is within our power to do so, why should we suppose that this is a special obligation that is owed to fellow ecological citizens in virtue of our shared citizenship, as opposed to a universal obligation that is owed to other human beings in virtue of our common humanity? To this it might be responded that these obligations are owed both to fellow ecological citizens and fellow human beings: we owe these obligations to each other partly in virtue of our common ecological citizenship and partly in virtue of our common humanity. They are both universal obligations owed in virtue of our common humanity and special obligations owed in virtue of our common ecological citizenship, where this consists, at least in part, in our interconnectedness which is such that each of us creates an environmental footprint, albeit of different sizes, because of the unavoidable impact we make on the environment as we go about the production and reproduction of our daily lives (compare Dobson, 2003, pp. 99ff.). But what do we add by introducing the idea of citizenship in this context? Dobson thinks it is crucial that what we have here are obligations of justice rather than merely charity, and that these obligations of justice arise as a result of pre-existing relationships and interactions between people that have created, and perhaps continue to create, injustice. ¹⁶ He writes, for example, that in his view ‘it is relations of systematic ecological injustice that give rise to the obligations of ecological citizenship’ (Dobson, 2003, p. 132). But we can acknowledge our environmental interconnectedness, and indeed maintain that we have an obligation of justice (rather than merely charity) to humanity, including to future generations, to use resources in a sustainable way and to compensate others when we harm them by violating this obligation, without invoking the idea of citizenship at all. ¹⁷

Dobson might reply that the special obligations we owe to fellow ecological citizens are much more specific and depend on our circumstances; for example, whether or not we are in a position ‘to compromise or foreclose the ability of others in present and future generations to pursue options important to them’ (Dobson, 2003, p. 120), and what are the practices in which we are involved or implicated. When our protests can make a difference, we may have special obligations to oppose (individually or collectively) the building of a factory that will increase the air pollution near a town, and this obligation may be owed primarily to those who live in that town given that it is their interests that are worst affected by the decision to do so. It might then be maintained that this is an obligation owed by some to others in virtue of their common ecological citizenship. Indeed, reflecting upon the wider effects of pollution on climate change, Dobson argues that obligations of compensatory justice can be obligations of ecological citizenship. Those who are principally responsible for global warming may owe obligations of compensatory justice to fellow environmental citizens who suffer a greater share of the burdens of global warming, such as those living on islands that are threatened by rising sea levels (Dobson, 2003, p. 31). Again, however, we need to ask what theoretical illumination is gained in this context by speaking of obligations that are owed by some to others in virtue of their common ecological citizenship, rather than speaking of obligations of justice that are owed by some to others because of the ways in which they have acted, and continue to act. Why does the fact that some cause or have caused injustice to others mean that one lies in a relationship of citizenship with them? Are those who benefited from colonial exploitation thereby in a relationship of citizenship with those who suffered from it? There are many to whom we owe special obligations of compensatory justice because of what we did in the past, and indeed because of how we continue to act in the present, but there is no reason to think of them as fellow citizens, or to think of these obligations as obligations of citizenship. Intellectually speaking, we gain nothing by invoking the idea of citizenship, and we lose what is distinctive about that relationship.

To this it might be responded that even if there is an intellectual cost incurred by invoking the idea of citizenship in these contexts, this cost is more than outweighed by the political gains: by presenting the obligations of global environmental justice as demands of ecological citizenship, we promote compliance with these obligations. People are motivated by the thought that they are acting as global citizens – or being good global citizens – in ways in which they would not be motivated by the mere thought that they are acting justly or virtuously. I doubt this is true, but it is an empirical claim, a claim of moral psychology, that cannot be refuted (or indeed defended) except by appeal to empirical evidence. If we are going to speculate about the effects of using different forms of expression, we might also wonder whether introducing the language of ecological citizenship could result in the substance of what is being claimed getting lost in the rhetoric, with the result that movements for environmental justice lose their focus.

Justifying Obligations of Citizenship

Further light is cast on these issues by considering how, in the more familiar context of the nation state, special obligations to fellow citizens are justified, in order to see whether these strategies, suitably revised and extended perhaps, might be employed in the context of ecological citizenship. According to the core concept I have identified, citizens incur special obligations to each other in virtue of some aspect of their common citizenship, but there are different ways in which this justificatory link has been made, especially in relation to our obligation to obey the law. I shall consider what I think are the four most promising approaches to see whether they might be extended in order to defend the kind of obligations of ecological citizenship that Dobson has in mind. ¹⁸ First, some have appealed to a principle of fair play, in particular, the idea that citizens have special obligations to obey the state (and contribute to its maintenance) because they have enjoyed various benefits that have been provided by it (Dagger, 1985; 1997). Second, some have argued that these obligations are part of the good of citizenship and are justified by that good (Mason, 1997; 2000, ch. 4). Third, some have maintained that citizens incur obligations to obey the state and contribute to its maintenance in various ways because of their common identity (Horton, 1992, ch. 6; 2006; 2007). Fourth, some have argued that the state has special obligations to its citizens, and that citizens have special obligations to each other, because allocating obligations in this way is the best method of protecting the vulnerable, globally speaking (Goodin, 1985; 1988).

Consider the first approach, which appeals to the principle of fair play. This principle justifies special obligations to fellow citizens by maintaining that the willing receipt of benefits from the state (such as the enjoyment of law and order, and perhaps also the receipt of welfare benefits) places citizens under an obligation to each other to bear the burdens necessary for the maintenance of the state, such as those which are involved in obeying the law and paying one's taxes. In its standard form, this strategy appeals to the rights and entitlements that citizens enjoy in virtue of the state. But it could, in principle, justify special obligations beyond the bounds of the nation state, for it would have potential application to any cooperative scheme for mutual advantage that extended in this direction. However, it does not seem to offer a promising means of underpinning notions of ecological citizenship, at least in circumstances where there is not yet any cooperative scheme of the required kind (Barry, 1982, especially pp. 231–4). The existence of a scheme of global trade governed by international institutions may give rise to special obligations to obey whatever rules and conventions govern these markets, but it does not justify special obligations of environmental justice. This is not to deny that those committed to environmental sustainability might be understood as involved in a cooperative venture, one that benefits them as well as free-riders outside the scheme who continue to degrade the environment. Under the principle of fair play, those involved in this cooperative scheme would have special obligations to each other to bear the burdens of acting in environmentally sustainable ways, but those outside the scheme would not. This provides us with a possible model of environmental citizenship, but it is not one that is well suited to Dobson's aims, for it would not allow us to count the obligation that requires wanton polluters to mend their ways as an obligation of environmental citizenship; obligations of environmental citizenship would be incurred only by those who were already committed to sustainable practices.

Each of the approaches I am considering treats the obligations of citizenship as special obligations, but not all regard them as associative obligations, that is, as obligations which are incurred simply by belonging to groups defined by social practice (Dworkin, 1986, p. 196; Scheffler, 2001, p. 49). ¹⁹ However, the next two defences that I shall explore do understand these obligations in this way. Consider the idea that the obligations of citizenship are both part of, and justified by, the good of citizenship. This approach attributes non-instrumental value to citizenship, on the grounds that in virtue of being a citizen a person is a member of a collective body in which he or she enjoys equal status with its other members (and is thereby provided with recognition of his or her equal worth) and is provided with opportunities to participate directly and indirectly in the formation of law and policy by which he or she will be governed. According to this account, part of what it is to be a citizen is to incur special obligations that are justified by the good of the wider relationship to which he or she contributes. Again, this approach in its standard form appeals to the rights and entitlements that citizens enjoy in virtue of the state, but it might be extended beyond the bounds of the state to any collective decision-making bodies whose members enjoy equal status and possess the opportunity to participate in the formation of policy. However, it does not seem to provide us with a particularly promising way of justifying the obligations of ecological citizenship that Dobson has in mind, in the absence of any collective body in which those who nevertheless make different-sized ecological footprints are vested with equal power to formulate global environmental policy.

The third approach maintains that identity is intrinsically obligation generating, that part of what it is to have an identity is to incur special obligations to those who share that identity, and that since fellow citizens share an identity, they possess special obligations to each other. Although this approach may be able to justify the idea that those who are part of the same transnational movements (including those committed to acting individually or collectively to promote environmental sustainability) are bound together by associative obligations, it is an unpromising approach for justifying the obligations of ecological citizenship, for these are not owed to those who share the same identity or commitments; they are owed to human beings more generally, or at least to those who are at the sharp end of practices that are unsustainable environmentally speaking. Of course, people may have a strong sense of being bound to each other as fellow human beings or fellow inhabitants of the earth, to which those who defend cosmopolitan conceptions of citizenship might appeal. But even if this generates obligations to fellow human beings in virtue of their shared humanity or to others as fellow inhabitants of the planet, it is hard to see why we should suppose that these obligations are owed in virtue of a common citizenship. It is only when identity is combined with something else, such as the shared enjoyment of rights, or shared participation in a cooperative scheme, that the language of citizenship gains a foothold.

The final approach maintains that special obligations to others in virtue of common citizenship derive from a universal obligation to others, as fellow human beings, to protect the vulnerable. In other words, special obligations to fellow citizens are viewed as a way of distributing responsibilities among human beings: our general moral obligations are most effectively fulfilled by assigning particular people (and the collective entities to which they belong) with special responsibilities and the means to discharge them. This does not seem particularly promising in the context of ideas of ecological citizenship, however, for these do not seem to be concerned with distributing responsibilities to care for the vulnerable in an analogous way. It is true that, for Dobson, different material circumstances may mean that different people will be vulnerable in different ways to environmental degradation and that they have different obligations as a result. But on Dobson's scheme this does not involve distributing particular groups of environmentally vulnerable people and the responsibility to care for them to specific collectives, in the way that the analogy would require. On the ‘distributing responsibilities’ model, the relationship of citizenship would hold within these collectives, not across them in the way that Dobson's account of ecological citizenship requires.

Concluding Remarks: From Ecological Citizenship to Global Citizenship

Although I have focused on Dobson's notion of ecological citizenship, I think that the difficulties I have raised for it also arise for many accounts of cosmopolitan or global citizenship. My account of the core concept of citizenship does not entail that the very idea of global citizenship is incoherent. The idea that, say, each and every human being is a citizen of the world is not unintelligible. But if a model of global citizenship is to be faithful to the core concept I have identified, it must explain how global citizens are, or could be, bound together by special obligations that are owed, at least in part, in virtue of particular features of their relationship to one another. Now there are ways of grounding obligations of global justice which would allow us to derive special obligations, and it might be thought that these could underwrite ideas of global citizenship. For example, the kind of interdependence that arises from the existence of a global economic order might be able to justify special obligations that are derived from universal obligations of global justice to aid others or refrain from harming them, in conjunction with empirical facts concerning the ways in which our acts (and omissions) can cause poverty and suffering beyond the borders of the state in which we live (e.g. Pogge, 2002). But just as our special obligations to compensate others for the effects of the environmentally unsustainable practices in which we participate do not give reason to invoke the idea of ecological citizenship, neither do these special obligations to aid, or refrain from harming, those beyond our borders who are vulnerable to our actions, give reason to invoke the idea of global citizenship. Nothing would be gained in terms of our understanding by doing so, for nothing would be added to the thought that we are bound by special obligations of justice that extend beyond our borders, and which are derived from universal obligations of justice through the addition of empirical premises.

Are there any other routes to the idea of global citizenship? It might be supposed that a system of global trade and investment, governed in part by international institutions and conventions, is sufficient for us to be regarded as participants in a cooperative scheme for mutual advantage of a kind that would engage the principle of fair play at the global level. Could this then justify the existence of special obligations between the world's inhabitants in a way that would enable us to give content to the notion of global citizenship? There is reason to be sceptical here since, as I pointed out earlier, the kind of special obligations it could justify would be limited to requiring that those who trade with others beyond the borders of their state accept their fair share of the burdens of doing so, such as obeying whatever laws and conventions govern international trade (Barry, 1982, especially pp. 232–4). It is hard to see why this would warrant talk of a shared global citizenship.

Some have tried to motivate the idea of global citizenship simply by appealing to the existence of demanding obligations of justice that we owe to fellow human beings which may outweigh any special obligations that we owe to our compatriots, and to obligations (perhaps also of justice) to establish, strengthen and participate in embryonic transnational institutions, involving wider communities of discourse, which enable us to fulfil these universal obligations (Dower, 2002, p. 40; Linklater, 1999). But if we make this move, we again lose what is distinctive about the idea of citizenship, and the rhetoric of citizenship adds nothing to the idea that we owe obligations of justice to fellow human beings, and duties of various kinds to promote and participate in institutions that enable us to fulfil these obligations. Nigel Dower suggests that those who are sceptical about notions of global citizenship face a dilemma: ‘either one's ethical theory is robustly global so that commitment to global citizenship follows naturally from it, or one's denial of global citizenship has to be the denial of serious global obligation’ (Dower, 2000, p. 564). But this is a false dilemma. It is possible to be sceptical about the idea of global citizenship because one thinks that obligations of citizenship must be special obligations that are owed in virtue of some feature of common citizenship: the language of global or cosmopolitan citizenship requires more than global obligations, or an acknowledgement of them, to motivate it. The mere existence of demanding obligations to other human beings, even if these obligations are obligations of justice, is not enough to give the idea of global or cosmopolitan citizenship a foothold.