Global Justice

Faculty Research Working Paper Series

Mathias Risse
Harvard Kennedy School

January 2011
RWP11-001

The views expressed in the HKS Faculty Research Working Paper

www.hks.harvard.edu
 Global Justice

Mathias Risse, Harvard University

December 21, 2010

1. Justice generates especially stringent claims. 1 In Shakespeare’s Merchant of Venice,

Shylock makes his demand to a pound of his delinquent debtor’s flesh in terms of justice.

Until the clever Portia finds a device for voiding the contract, the presumption is that it

must be granted. Conceptually, demands of justice are the hardest to outweigh or

suspend. Kant goes too far insisting that there is no point for us to continue to live on

earth unless justice prevails. 2 Still, justice plays its central role in human affairs precisely

because it enables persons to present claims of such stringency. A theory of distributive

justice explains why certain individuals have particularly stringent claims to certain

relative or absolute shares, quantities, or amounts of something whose distribution over

certain people must be justifiable to them. Alongside distributive justice there is also

rectificatory justice, and perhaps other kinds. Yet since our concern in all but sections 8

and 9 is with distributive justice, “justice” means distributive justice unless otherwise

noted.

Increasing political and economic interconnectedness draws much philosophical

attention to the question of the conditions under which such stringent claims arise. Do

claims of justice arise only among those who share membership in a state? Alternatively,

do they arise among all those who are jointly subject to the global political and economic

order? Or do they apply among all human beings simply because they are human?

1
This piece was prepared for the Oxford Handbook of Political Philosophy, edited by David Estlund. Many
thanks to Micha Glaeser, Gabriel Wollner, and David Estlund for helpful comments.
2
Kant (1996), p 473.

1
Inquiries into global justice differ from those into international justice precisely by not

limiting inquiry to what states should do. They may well also question the very moral

acceptability of states, and explore alternative arrangements.

When Hobbes devoted De Cive to exploring the rights of the state and the duties

of its subjects, he did something fundamentally new. Focusing on the confrontation

between individual and state, after all, meant not to focus on the relationship of the

individual with particular rulers or multiple authorities. It meant to assess a person’s

relationship with an enduring institution that made exclusive claims to the exercise of

certain powers within a domain. Centuries later Kavka (1986) could write that “the

relationship between the individual and the State forms the core of Western political

philosophy,” just as “the relationship between morality and prudence lies at the center of

Western ethics” (p 21). Due to its focus on the state (a reflection of the political realities

of the age), until recently modern political philosophy has done comparatively little to

theorize justice outside of states, or to assess political and economic structures other than

states in light of such inquiries.

Nowadays, however, we must ask with some urgency about the conditions under

which principles of justice apply. That inquiry is central to the burgeoning field of global

justice. This question about the applicability of principles of justice also generates

(renewed or new) interest in political structures such as a world state; a world with

federative structures stronger than the United Nations; with a more comprehensive

system of collective security; one where jurisdictions are disaggregated; or where border-

control is collectively administered or abandoned entirely. Reflection on such structures

is of great interest in a politically and economically increasingly interconnected world

2
where enormous differences in life prospects nonetheless persist. Depending on what one

thinks about the conditions under which demands of justice apply, one may favor the one

or the other of these structures. Even if one thinks that subverting the state system is

practically impossible, one may still explore moral objections to it, specifically from a

standpoint of justice. Such inquiries naturally assume the shape of determining the

conditions under which principles of justice apply.

Fleischacker (2004) shows that the modern conception of domestic justice

incorporates several premises. First, each individual has a good that deserves respect, and

individuals are due rights and protections to that end. Justice is not (merely) a matter of

realizing, say, a divine order. Second, some share of material goods is among the rights

and protections everyone deserves. Third, the fact that each person deserves such rights

and protections is rationally and secularly justifiable. Fourth, the distribution of these

goods is practical: it is neither a fool’s project nor self-undermining like attempts to

enforce friendship. Fifth, it is for the state (and conceivably other political entities) to

achieve justice. These commitments about how the fates of individuals are tied are

strikingly unusual by historical standards. It is such an understanding of domestic justice

that also generates the debate about global justice. If each individual has a good that

deserves respect, we must ask if corresponding duties expire at borders. If shares of

material goods are among the rights and protections everyone deserves, we must ask if

this depends on where people live. If protections require rational justification, we must

ask if such justification is available for (what below I will call) the normative peculiarity

of the state. If the distribution of these goods is practical, we must note that nowadays a

3
network of organizations seeks to make this true globally. Plausibly, entities other than

states too ought to strive for justice.

Let me explain what I mean by the “global (political and economic) order” and

hence by an “increasingly interconnected world.” Our current global society has emerged

from developments that began in the 15th century through the spread of European control,

continuing with the formation of new states through independence or decolonization.

While this order has no government, it comprises treaty- and convention-based norms

regulating territorial sovereignty, security and trade, some property rights, human rights,

and the environment. Politically, the United Nations Charter codifies the most significant

rules governing this system. Economically, the Bretton Woods institutions (International

Monetary Fund, World Bank, later also the General Agreement on Trade and

Tariffs/World Trade Organization) form a network intended to prevent war and foster

worldwide betterment. Jointly with more powerful states, these institutions shape the

economic order.

The term “globalization” describes processes that erode the political and

economic importance of national boundaries and increasingly affect life chances through

the system of rules that is constitutive of the global order. “Globalization” is not new. It

traces back to the spread of European control, a process accompanied by the emergence

of a state system whose central features were reflected in the doctrine of “sovereignty.”

Political philosophers of the 17th and 18th century, such as Grotius, Locke, Hobbes,

Wolff, Vattel, and Kant, explored questions about that stage of globalization. They

developed the doctrine of sovereignty, explored under what conditions one could acquire

non-European territories, or what kind of ownership there could be of the seas. The

4
spread of European control was complete by the end of the 19th century. By that time,

political philosophers such as Tocqueville and Mill had been busy justifying why non-

Europeans should endure political dependence. A period of devising rules for the spread

of “empire” gave way to a period of justifying its persistence.

After World War II “global governance” came into its own, in the form of the

international institutions mentioned above. Political philosophers at this stage of

globalization must worry about normative issues that such governance raises. Among

these issues is the question of whether principles of justice indeed apply only within

states. Perhaps such principles apply to the global order as such. Or, again, they might

apply independently of political and economic structures, and instead hold because of

common humanity and thus apply to all human beings. By “positions on global justice” I

mean views about the conditions under which principles of justice hold.

2. We can formulate positions on global (distributive) justice in terms of the distinction

between relationism and non-relationism. That distinction captures some of the most

important debates in contemporary political philosophy. “Relationists” think principles of

justice only hold among persons who stand in some essentially practice-mediated relation

to each other. “Non-relationists” think such principles may apply among those who stand

in no such relation.

A reference to practices keeps non-relationism from collapsing into relationism.

The relation of “being within 100,000 km of each other” is not essentially practice-

mediated, nor is, more relevantly, that of “being a fellow-human.” I talk about

“essentially” practice-mediated relations since there may be practices associated

5
especially with this latter relation, which, however, are dispensable to understanding its

content. Paradigmatic non-relationists base the applicability of principles of justice on

common humanity, relationists on shared political structures. Relationists and non-

relationists disagree about the grounds of justice, the norm-generating conditions or

considerations that render demands of justice applicable among persons for whom these

conditions and considerations hold. I use the term “grounds” to remain neutral between

relationists and non-relationists and thus arrive at a rather abstract formulation in terms of

“conditions or considerations.” I use the term “relationship” sufficiently broadly for

relationists and non-relationists to register as offering different accounts of what one

might call the “justice relationship.”

Relationists may hold a range of views about the nature of the relevant relation.

They may disagree about the scope of justice, the range of people in the justice

relationship. Globalists think the relevant relation holds among all human beings in virtue

of the existence of practices that relate all humans to each other within a single global

order. Internationalists think the relevant relation holds (only) among individuals who

share membership in a state. 3 They owe an account of what it is (exclusively) about

shared membership in states that generates demands of justice. Globalists owe an account

of what it is about involvement with, or subjection to, the global order that generates

demands of justice. Those who accept duties of justice that hold (only) among people

who share a state endorse “the normative peculiarity of the state;” those who do not,

reject it. Internationalists endorse the normative peculiarity of the state. Globalists and

non-relationists do not. So while globalists and internationalists are united by being

3
I disregard differences between citizenship and permanent residency and speak of shared membership.

6
relationists, globalists and non-relationists are united by rejecting the normative

peculiarity of the state. 4

Relevant versions of non-relationism take the scope of justice to be global,

including all of (and only) humanity. Yet non-relationists may determine the scope

differently. One could limit justice to a subset of humanity by insisting on the normative

importance of, say, sex or race. Or one may insist that justice must have all sentient

beings in its scope, at least higher animals and conceivably rational Martians. Yet the

former possibility is implausible, and the latter I set aside. Non-relationists owe an

account of how common humanity generates demands of justice. 5 Relationists think of

principles of justice as regulating practices that some persons share with each other. This

implies two things. First of all, for relationists principles of justice apply only to those

who respectively share the practices. Relationists are motivated by the moral relevance of

practices in which certain individuals stand. Such practices may include not only those

which individuals chose to adopt, but also those of which they have never chosen to

4
Some of my terminology draws on Sangiovanni (2007). Yet my usage deviates from his. For instance,
globalism, on my account, is by definition a relationist view. “Internationalism” often refers to political
movements advocating greater cooperation among nations, which is not what is mean. Often “statism” is
used for what I mean by “internationalism”. I prefer “internationalism” because the term captures an
understanding of global justice as composed of multiple entities within which principles of justice hold
respectively. Whatever moral principles apply globally would then differ from principles of distributive
justice, and hold between (“inter”) these entities. There are, of course, only so many terms to go around in
this domain.
5
Globalists and internationalists disagree about both grounds and scope, but agree that grounds are
relational. Relationists may also agree about the scope while disagreeing about the grounds. Below we
encounter coercion-based and cooperation-based internationalists. Both think the people who respectively
stand in the justice relationship are those who share a state. They agree about the scope of justice while
disagreeing about the grounds in the sense that they disagree about what it is about shared citizenship that
generates demands of justice. Distinctive of a ground is the account of the conditions and considerations
that are norm-generating. The term “scope” does not do much independent work. “Not much:” It would be
inappropriate to stipulate that grounds always be specified so precisely that the scope is uniquely fixed. Yet
once the grounds are fixed, disagreement about the scope should be relatively minor, of the magnitude of a
dispute about who exactly counts as a citizen given that this matter is largely fixed through legal rules. I
define “globalism” as a view about grounds, not as one about the scope that is consistent with a non-
relationist ground.

7
partake. Second, relationists think of principles of justice as only regulating those

practices, rather than every aspect of the lives of those who share them.

3. To illustrate what is at stake, note that John Rawls is a relationist. He famously calls

justice “the first virtue of institutions, as truth is of systems of thought” and talks about

“justice in social cooperation” ((1999b), p 3). (See the article, “John Rawls” by Leif

Wenar in this volume.) Justice here is a characteristic of institutions, which are practices.

His principles regulate the practices constitutive of the basic structure of society (the way

in which the major social institutions fit together into one system, and how they assign

fundamental rights and duties and shape the division of advantages from cooperation).

“Distributive justice,” says Freeman (2007) by way of expounding this approach, “poses

the general problem of fairly designing the system of basic legal institutions and social

norms that make production, exchange, distribution, and consumption possible among

free and equal persons” (p 305f). Many aspects of advantage and its distribution are

natural facts. But “what is just and unjust,” says Rawls ((1999b), p 87), are not these

facts, but “the way that institutions deal with these facts.”

Relationists can recognize duties to those with whom they do not stand in the

justice relationship. Alas, those duties would either differ relevantly from duties of justice,

or else in some other way differ from those duties of justice that hold among individuals

who share the relevant relation. Nagel (2005) adopts the former approach, insisting that

justice only holds within states. Rawls (1999a) adopts the latter. He implicitly

acknowledges a distinction between duties of distributive justice that hold within states,

and other duties of justice that may hold otherwise. The duty of assistance to “burdened

8
societies” in Law of Peoples is not one of distributive justice ((1999a), p 106, pp 113-120;

see also Freeman (2007), chapter 9). Duties of distributive justice are duties with regard to

shares in a system of economic production and exchange, which Rawls thinks presuppose

a basic structure.

Rawls is an internationalist, but his main goal is to offer principles of domestic

justice. Law of Peoples adds an approach to international justice, by way of sketching the

foreign policy of a society within which his domestic principles of justice apply.

Methodologically in the background is his political constructivism. Rawls begins with

domestic justice and works “outward” from there to the Law of Peoples, and “inward” to

local justice ((2001), p 11). These other subjects presuppose domestic justice. As

Freeman (2007) says:

The principles that appropriately regulate social and political relations depend
upon the kinds of institutions or practices to be regulated, and these principles are
to be ‘constructed’ on the basis of ideas that are central to the functioning of those
institutions or practices and people’s awareness of them. (p 270)

Freeman sees this political constructivism – a view about justification -- as integral to

Rawls’ rejection of global principles of distributive justice. The convictions and

intuitions that must be in reflective equilibrium to obtain a theory of justice concern the

practices and institutions in which we lead our lives. These convictions are less

developed outside of the domestic setting.

Wenar (2006) offers a plausible reading of Rawls that responds to critics, like

Caney (2002), who think Law of Peoples is incoherent with Rawls’ earlier work.

Crucially, both in the domestic and in the global case Rawls draws on ideas implicit in

the public political culture. Rawls believes “that humans should be coerced only

according to a self-image that is acceptable to them,” which means that “[s]ince ‘global

9
citizens’ cannot be presumed to view themselves as free and equal individuals who

should relate fairly to each other across national boundaries, we cannot legitimately build

coercive social institutions that assume that they do” (Wenar, p 103). Wenar rightly uses

this observation to explain why Rawls did not advocate global egalitarian ideals of a sort

that Beitz (1999) and Pogge (1989) found a natural extension of his principles. Global

public political culture is of a different (much thinner) nature than that of a constitutional

democracy.

4. Non-relationists object that, by making justice practice-mediated, relationists either tie

justice to properties of individuals that omit too much of moral importance (for practices

individuals have not selected), or over-emphasize some morally important aspects at the

expense of others (for practices in which individuals chose to participate). Parallel to what

I said about relationists, non-relationism implies two things. Non-relationists seek to avoid

the alleged arbitrariness of restricting justice to the regulation of certain practices. 6

Second, since non-relationists do not limit justice in this way, they will plausibly apply

principles of justice to the whole range of advantageous and disadvantageous events in a

life. (“Will plausibly:” recall that there is logical space for non-relationists to proceed

differently.) For non-relationists justice is a property of the distribution of advantage,

broadly understood. While for relationists individuals stand in the justice relationship if

they have special claims within particular practices, for non-relationists that relationship is

6
Miller (2007), p 32f, reminds us that “arbitrary” sometimes means “undeserved” and sometimes “should
make no difference.” Differences in needs are undeserved, but should make a difference. The manner in
which I have introduced non-relationism above seeks to characterize this position in a way that avoids
pitfalls from this ambiguity.

10
distinguished by the absence of special claims. 7

Note two things in addition. First, grounds differ from circumstances of justice.

“[T]he circumstances of justice obtain,” explains Rawls (1999b), following Hume,

“whenever persons put forward conflicting claims to the division of social advantages

under conditions of moderate scarcity” (p 110). Both circumstances and grounds tell us

“when demands of justice apply,” but do so in different senses. “Circumstances of justice”

specify those living conditions of human beings under which any principles of justice

apply in the first place. Unless we live under such circumstances, no principles of justice

apply to begin with. If we live under these circumstances, the grounds specify which

principles of justice apply to which people.

Second, consider Pogge’s (1994) widely-quoted definition of “cosmopolitanism:”

Three elements are shared by all cosmopolitan positions: First, individualism: the
ultimate unit of concern are human beings, or persons (…). Second, universality:
the status of ultimate unit of concern attaches to every living human being equally
– not merely to some sub-set, such as men, aristocrats, Aryans, whites, or Muslims.
Third, generality: this special status has global force. (p 89)

None of the positions I discussed (internationalism, globalism, non-relationism) denies

the moral equality of persons. Each of these positions has capacities to makes sense of

individualism, universality, and generality. A crucial issue for each of these positions is

how rich a notion of moral equality its advocates wish to endorse, and how the relevant

notion of moral equality relates to ideas of political equality and distributive equality. In

any event, one needs additional arguments to derive distributive equality of anything

7
Tan (2004) captures the non-relationist’s concern with arbitrariness: “At the foundational level of
deliberation about global justice, impartiality requires that we do not allow people’s nationality to influence
our views of what people’s baseline entitlements are. (…) A person’s nationality, a mere accident of birth,
cannot by itself be a reason for giving her greater consideration at the foundational level” (p 158; see also
pp 27f and 159f). See also Pogge (1989), p 247; Moellendorf (2002), pp 55f, p 79.

11
from ideas of moral or political equality. Internationalists deny a close link between

moral and political equality: all human beings are morally equal, but it is only in the

presence of certain practices (those of shared citizenship in a state) that ideas of political

equality even apply. Internationalists may or may not find inequality among individuals

in one country (in terms of outcomes, resources, or opportunities) morally problematic.

But they do not find inequality among countries as such morally problematic. Globalists

and non-relationists may endorse a global maximin or sufficiency criterion, or yet other

criteria. A range of views on inequality is open to them, but none that depends on shared

membership in states.

We have learned the basic cosmopolitan lesson: moral equality is an essential part

of any credible theory of global justice. In the domain of distributive justice, the term

“cosmopolitan” has become the victim of its own success. Therefore we should conduct

the philosophical debate in the terms discussed in this chapter, rather than in a way that

thinks of “cosmopolitanism” as a distinctive position on global justice. We live on a

“cosmopolitan plateau” in the way in which Kymlicka (2002) (following a suggestion by

Ronald Dworkin) claims that plausible political theories populate an “egalitarian

plateau.” All plausible theories of domestic justice define “the social, economic and

political conditions under which the members of the community are treated as equals”

(Kymlicka (2002), p 4). Similarly, all plausible theories of global justice ascribe

significance to moral equality.

5. Non-relationists insist that relations in which particular individuals stand with each

other cannot have the kind of moral importance that would imply that moral obligations

12
(or anyway, a broad range of obligations) only apply among those who do so. Caney

(2005) offers a version of this approach. He offers the following thought regarding the

relevance of economic interaction, and a similar thought would apply to all practices:

Consider a world with two separate systems of interaction that have no contact but
are aware of each other and suppose that one of them is prosperous whereas the
other is extremely impoverished. Compare, now, two individuals – one from the
prosperous system and the other from the impoverished system – who are identical
in their abilities and needs. The member of the prosperous system receives more.
But it is difficult to see why – concentrating on any possible and reasonable
criteria for entitlement – this is fair. Ex hypothesi, she is not more hard-working or
more gifted or more needy. In all respects they are identical (bar one, namely that
one is lucky to live in the prosperous society and one is not) and yet an
institutionalist approach confers on one more benefits. (p 110) 8

To develop his case, Caney’s strategy is to identify a moral argument of sorts, and then to

argue that that argument appeals to properties that everybody has. Limiting such

arguments to particular groups means to commit what Caney, following Black (1991),

calls the “fallacy of restricted universalism:” “A distributive theory, that ascribes rights

and claims on the basis of certain universal attributes of persons, cannot at the same time

restrict the grounds for those claims to a person’s membership or status within a given

society” (p 357). Attempts to derive principles of justice from universal attributes that

nevertheless are supposed to be limited to certain groups (e.g., compatriots) commit this

fallacy. Consider the way in which Caney applies this strategy to civil and political

liberties. He argues for the “scope1 claim:” “the standard justifications of rights to civil

and political liberties entail that there are human rights to these same civil and political

liberties” (p 66). The scope1 claim holds

because the standard arguments for civil and political rights invoke a universalist
‘moral personality.’ That is, the relevant aspect of persons is the right to be
subject to principles to which they can reasonably consent (for contractarians), or

8
See also Moellendorff (2002), p 79, and Tan (2004), p 158.

13
 their use of moral language (for Habermas), or their humanity and status as
persons (for deontologists), or their ability to lead a fulfilling life (for
perfectionists). As such, it would be incoherent to adopt any of these lines of
reasoning for a particular right and then ascribe that right only to other members
of one’s community. (p 77)

Arguments of this sort at the very least pose a challenge for relationists to explain just

what it is about their preferred relation that, in the case that interests us, generates

demands of justice that would only apply among those who share that relation. Without

further ado, let us see whether they succeed at this task and so can respond to non-

relationists like Caney.

6. Internationalists and globalists disagree about what relation is relevant for the

applicability of principles of justice. Nonetheless, they are both relationists, resting

claims of justice on nationally or globally shared practices, respectively, and thus to some

extent use similar arguments to defend their views. Defenses of relationism may enlist

two strategies.

The first strategy draws on the fact that it is (conceptually) difficult for us

successfully to press demands upon each other at all, especially the stringent demands of

justice. Relationists are well-equipped to deal with this difficulty. They can help

themselves to considerations that arise from within the practices they consider central to

the applicability of justice. They need not deny that there can be natural rights and duties

of justice at all. But arguments in their support can enlist only features of shared

humanity. Crucially, derivations of transactional and associational duties (the kind of

arguments relationists offer) can enlist a larger set of considerations. They can use claims

about persons having undergone certain transactions under specific conditions (e.g.,

14
promises, contracts), or about them living in certain human-made arrangements that

could often assume different shapes, and that put demands on those involved in them.

Claims of justice cannot succeed merely based on references to the significance of

something for the claimant. We need reasons why others ought to provide what is

significant. Non-relationists can most readily meet that challenge if they restrict

themselves to establishing rights and duties pertaining to elementary human concerns,

such as basic needs satisfaction. Relationists are better equipped to make such a case. In

particular, duties pertaining to relative, rather than absolute, economic status are (at the

very least) easier to establish, and are more demanding, if we can resort to shared

practices to make that case.

To illustrate, consider Scanlon’s (2003) influential discussion of objections to

inequality, and hence to differences in relative economic status. Scanlon identifies five

reasons to pursue greater equality: (1) to relieve suffering or severe deprivation; (2) to

prevent stigmatizing differences in status; (3) to avoid unacceptable forms of power or

domination; (4) the preserve the equality of starting places which is required by

procedural fairness; and (5) procedural fairness sometimes supports a case for equality of

outcomes. (2) and (5) are the clearest expressions of egalitarianism. (4) is consistent with

considerable inequalities and so is only weakly egalitarian. (1) and (3) are not egalitarian

at all. Scanlon argues that Rawls uses (2) - (5), and perhaps (1) as well, to argue in

support of his principles of justice. So those principles are supported by reasons that are

distinctly egalitarian, but also by reasons that are distinctly not.

Crucially, however, even the force of (3) and (4) depends on the practices (if any)

that the relevant individuals share. For instance, to explain what counts as unacceptable

15
forms of power it helps to explore how individuals respectively contribute to the

maintenance of an economic system, and hence also what the economic ties among them

are to begin with. Undoubtedly, some exercises of power are unacceptable regardless of

what relations individuals stand in. But the more ties there are among individuals, the

more possibilities there will be for them to contribute to the maintenance of relations,

which in turn generate rationales for them to complain about certain exercises of power.

Similarly, to assess how much reason there is to preserve the equality of starting places

on behalf of procedural fairness it is essential to assess to what kinds, and range, of

procedures the individuals are jointly subject. Thus if we seek to argue for obligations

pertaining to relative standing without making use of relations, there is little we can say

in the first place. We can derive more demanding obligations if we can resort to relations.

The second strategy in defense of relationism appears in Scheffler’s (2001)

account of the link between special relations and responsibilities. Relations create

responsibilities because having reason to value relations non-instrumentally just is to

have reasons to see oneself under, and actually have, special obligations. As Scheffler

puts it, to attach non-instrumental value to a relationship with somebody means “to be

disposed, in contexts which vary depending on the nature of the relationship, to see that

person’s needs, interests, and desires as, in themselves, providing me with presumptively

decisive reasons for action, reasons that I would not have had in the absence of the

relationship” (p 100). (To call reasons “presumptively decisive” means to grant that in

principle they could be outweighed although they present themselves as reasons upon

which agents must act.) Skepticism about such responsibilities succeeds only if we have

no reasons at all to value our relations non-instrumentally. The case is clearest for family

16
ties and friendships, but less clear for political relations, and presumably internationalists

can more readily make it than globalists. Beitz (1999), for instance, does not use such

reasoning to support his globalism but wonders instead about the relevance of such

arguments even for shared membership in a state (p 212).

Internationalists must develop their version of relationism in a way that supports

the normative peculiarity of the state. They could first use the two strategies in support of

relationism to rebut non-relationism, and then offer an account of the normative

peculiarity of the state to rebut globalists (and thus settle the intramural debate between

relationists once relationism as such has been accepted). Two proposed accounts of the

normative peculiarity of the state are coercion-based internationalism (e.g., Blake (2001),

Nagel (2005)), according to which what distinguishes membership in a state is its

coerciveness; and reciprocity-based internationalism (e.g., Sangiovanni (2007)),

according to which it is its intense form of cooperation.

These views, however, face the challenge that forms of coercion and cooperation

also hold within the global order as such, which makes it problematic to argue that

principles of justice only govern the relation among those who share a state.

Internationalists can respond by arguing that the normative peculiarity of the state is

based on its particular kind of coerciveness or cooperativeness. Risse (2006), for

instance, accounts for the state’s coerciveness in terms of legal and political immediacy.

The legal aspect consists in the directness and pervasiveness of law enforcement. The

political aspect consists in the crucial importance of the environment provided by the

state for the realization of basic moral rights, capturing the profundity of this relationship.

But assuming that something like Risse’s (2006) account succeeds in explaining

17
what is morally special about shared membership in states, one must still wonder whether

this account matters for justice, that is, can explain why principles of justice apply only

among those who share a state. That is the point that globalists push. Capturing globalist

resistance to internationalism, Beitz (1999) argues that global interdependence

involves a pattern of transactions that produce substantial benefits and costs; their
increased volume and significance have led to the development of a global
regulative structure. (…) Taken together, these institutions and practices can be
considered as the constitutional structure of the world economy: their activities
have important distributional implications. (pp 148f)

It does not matter precisely what the nature of international economic interdependence is.

The dispute between internationalists and globalists already arises for a loose sense of

interconnectedness.

Beitz argues that in an interdependent world, limiting justice to domestic societies

means taxing poor nations so that others may live in “just” regimes (p 149f). Beitz’ target

is Rawls. He argues that if Rawls’ case for his principles of justice succeeds their content

should not change as a result of enlarging the scope of the original position to include the

global order. Beitz considers two objections (pp 154-161). The first insists that

interdependence is necessary, but insufficient for the applicability of justice. The global

order lacks any effective decision making mechanisms, as well as any real sense of

community, and these, the objector says, are also necessary for an order’s being subject

to standards of justice. Beitz responds that these differences fail to show that principles of

justice do not apply globally. Instead, they show that it is harder to implement the

principles.

According to the second objection, features of cooperation within states override

requirements of global principles even if justice applies globally. Rich countries may

18
deserve their advantages because of differences in organization, or technology. Beitz

responds that this means to base entitlements on morally arbitrary factors like those that,

as Rawls insisted, ought not to affect one’s share of social primary goods. Thus he rejects

this move much as Rawls rejects principles of justice drawing on undeserved social or

genetic characteristics.

7. One way of making progress in light of the debate among internationalists and

globalists that I have presented in section 6 is to deny that there is a single justice

relationship in which any two individuals either do or do not stand. One may use

“principles of justice” as a collective term for different principles with their respective

ground and scope. Let us call non-graded or monist internationalism the view that

principles of justice either do or do not apply, that they do apply within states, and thus

among people who share membership in a state, and only then. Non-graded or monist

internationalism is what I have so far taken internationalism to mean.

Graded internationalism holds that different principles of justice apply depending

on the associational (i.e., social, legal, political, or economical) arrangements. Graded

internationalism allows for associations such as the WTO, the EU, or the global order as

such to be governed by principles of justice, but endorses the normative peculiarity of the

state. Among the principles that apply within other associations we find weakened

versions of principles that apply within states. For this reason I talk about graded

internationalism in this case. I am lacking the space to motivate the graded view in detail.

Suffice it to say that all those who live, say, under WTO are tied to each other much more

loosely than individuals who respectively share a state. It is therefore plausible to think

19
that the principles of justice that hold within the WTO are weakened versions of those that

hold within a state.

However, now that we have introduced a non-monist view, we also must take

seriously the idea that some grounds could be relational, whereas others would not be.

We must consider the possibility that there is no deep conflict between relationism and

non-relationism. Perhaps advocates have respectively overemphasized facets of an

overall plausible theory that recognizes both relationist and non-relationist grounds.

Integrating relationist grounds into a theory of justice pays homage to the idea that

individuals find themselves in, or join, associations and that membership in some of them

generates duties. Integrating non-relationist grounds means taking seriously the idea that

some duties of justice do not depend on the existence of associations. One obvious non-

relational ground to add is common humanity. One view that develops these ideas could

be called pluralist internationalism. This view would endorse the normative peculiarity

of the state, but recognize multiple other grounds of justice, some of them relational (e.g.,

subjection to the global trade regime) and others not (e.g., common humanity).

Respectively different principles of justice would be associated with these different

grounds, all of which would be binding, say, for states and international organizations.

Pluralist internationalism transcends the distinction between relationism and non-

relationism.

This view offers one way of preserving the plausible aspects of non-relationism,

globalism, and internationalism. Needless to say, making this view credible, and proving

its fruitfulness, requires detailed discussions of its implications for a wide range of areas.

The theoretical costs of making such a move are considerable because one would give up

20
on the uniqueness of the justice relationship. One would also have to meet the challenge

that such a pluralist view does not, one way or another, collapse into one of the original

views. Other ways of making progress in this debate are possible as well, including those

that abandon the state’s normative peculiarity. We have now reached the research frontier

in this field. 9

8. So far we have only talked about distributive justice. However, perhaps the main

theme to explore in an assessment of global justice is whether the global order wrongfully

harms some people, presumably the weakest, the global poor. If so, we should think of

global justice primarily in terms of moral obligations to rectify harms (where the harms

are not simply defined by a conception of distributive justice). There are different ways

of articulating that thought. First, one may say the global order wrongfully harms human

beings because of the sheer existence of borders. Perhaps frontiers are inconsistent

especially with the value of freedom and with liberal justice. Second, perhaps the global

order wrongfully harms the poor (i.e., individuals who are unable to meet basic needs) by

imposing an institutional framework that is not as advantageous to them as some

alternatives. Finally, one may argue that the extents of poverty and inequality themselves

reveal that the global order wrongfully harms the poor. Let me elaborate.

One way of making the point that national borders can function as unjustified

restrictions of freedom draws on Amartya Sen and Martha Nussbaum’s approach to

9
I develop pluralist internationalism in my book The Grounds of Justice, to be published by Princeton
University Press.
10
See Sen (1985) and (1999) and Nussbaum (2006) and (2000).

21
life with dignity. “Bodily integrity” is on her list, of which one instantiation is “being

able to move freely from place to place” ((2006), p 76). The challenge is to explain why

this capability can be limited by frontiers. However, appeals to the value of freedom to

assess the legitimacy of borders cut both ways. It is true that frontiers limit choice, but so

does any immigration policy. Permissive immigration policies in countries where many

people wish to live are likely to constrain some who already live there.

Some have argued that immigration barriers are unjust owing to liberalism’s

commitment to moral equality. Liberalism, Carens (2003) notes, condemns the use of

morally arbitrary facts about persons to justify inequalities. Examples include race, sex,

and ethnicity. Political communities that treat people differently on the basis of such

features are illiberal and unjust. Yet citizenship seems as arbitrary as any of those.

Maintaining borders, to him, is as offensive as other perhaps more obvious cases of

injustice because it differentiates rights based on origin. Carens is correct that moral

equality and the value of common humanity cannot stop at borders, but it has been argued

that this does not mean shared citizenship is as morally irrelevant as race or ethnicity. The

fact that shared citizenship arises in a manner for which individuals deserve neither credit

nor blame does not make it morally irrelevant. On this view, there is no inference from

moral equality to political equality regardless of what structures persons share. 11

In a related vein one may argue that a system of states cannot properly consider

all affected interests and so they automatically wrongfully harm persons. The actions of

states affect many who have no say in the design of policies. For instance, Mexicans who

wish to enter the US never consented to there being an American people that can

11
For elaboration, see Blake (2001).

22
unilaterally constrain immigration. Yet interests can be properly considered in domain-

specific ways. To the extent that trade policies affect interests, for instance, we should

ensure that such policies are fair. But that does not mean one must enfranchise everybody

connected to a country by trade arrangements. We also need morally acceptable

immigration policies that spell out how many people from country A have a claim to

immigration to B. But once these claims are met, no additional demands, especially to

enfranchisement, arise. Moral acceptability here does not require consent.

Let us proceed to the argument that the global order harms the poor because there

is a feasible alternative under which situation of the global poor is improved, as presented

by Pogge (2002). Pogge seems to think “feasibility” is primarily a matter of allocating

money. It would just take 1.2% of the income of rich economies, $312 billion annually,

to bridge the aggregate shortfall of those living on less than $1 per day to the $2 line (p

7). Pogge’s proposal for raising some of those funds is the Global Resource Dividend,

which taxes extraction of resources. Yet while Pogge’s calculations show that abject

poverty could be surmounted if closing such a gap is a matter of transferring money, it is

doubtful whether financial transfers are enough to seriously mitigate poverty. Suppose in

situation S1 we have the funds to cover the financial shortfall. We still need reliable ways

of distributing funds to individuals who do not simply have bank accounts that they can

securely access. We would also need an environment where individuals can actually

spend the money on available goods and services. Both times this involves institutional

improvements, especially if one wants the changes to be lasting. Similar points apply if

one wishes to support medical and educational advancements. One cannot simply start to

“work on AIDS,” but must build and maintain medical infrastructure. One cannot

23
improve education by building a few school houses, but must invest in teacher training,

provision of books, supplies, family support, and much else.

That sustainable measures for enduring changes require good institutions has

become a guiding insight for many at the intersection between the social sciences of

development and its practice. Having funds to close the aggregate financial shortfall

between S1 and S2 is at best necessary, but not sufficient for S2 to be feasible. Pogge

may respond that while it is true that money does not automatically educate or cure

anyone, the money would be used precisely to take care of these things in appropriate

ways. Money is not enough by itself, but the other necessary conditions normally require

money. This is fair enough, but the relevant points now are these. We have found a sense

in which the poor are wrongfully harmed, namely, if not enough effort goes into

exploring possibilities for implementing appropriate institutional change. But while this

is presumably a valid charge, it makes it much harder than it appears to be on Pogge’s

proposal to ascertain how to go about creating a feasible alternative and who is guilty of

what failings in this regard.

9. Let us consider the import of poverty and inequality statistics. Statistics do not show in

any obvious way that the global order harms (let alone wrongfully harm) the poor. For

instance, while indeed 1.2 billion people in 1998 lived below the poverty line of $1.08

PPP 1993 per day (Pogge (2002), p 98), 12 there is now less misery than ever before, as

measured in terms of any standard development indicator. The progress made over the

last 200 years is miraculous. In 1820, 75% of the world population lived on less than $1 a

12
PPP means Purchasing Power Parity: the poverty line is fixed at what $1.08 bought in the US in 1993.

24
day (appropriately adjusted). Today, in Europe, almost nobody does, in China less than

20% do, in South Asia around 40%, and altogether slightly more than 20% do. The share

of people living on less than $1 a day fell from 42% in 1950 to 17% in 1992. Historically

almost everybody was poor, but that is no longer true.

What conclusion such statistics warrant depends on (a) the period considered

(“Sub-Saharan Africa has made progress over a 200-year horizon, but not for the last 20

years”), (b) whether one looks at absolute or relative quantities (“the number of

abysmally poor has remained unchanged for 15 years, but their share of the world

population decreased”), and (c) whether one looks at individuals or countries (“the

median developing country has experienced zero growth over the last 20 years; still,

inequality between two randomly chosen individuals has fallen” – because of growth in

India and China). Still, what is remarkable is not that so many now live in poverty, but

that so many do not; not that so many die young, but that so many do not; not that so

many are illiterate, but that so many are literate. If one looks at the last 200, 100, or 50

years, things have improved dramatically for the poor. The 200-year and the 50-year

horizon (roughly speaking) are especially significant. The former captures the period

when the industrial revolution has perfected the system of the division of labour, which

led to technological advancements across the board, advancements originating largely in

what are nowadays industrialized countries but that have worked to everybody’s benefit.

The 50-year horizon captures the period when the global order has come into its own.

Historically speaking, the global order at least seems to have benefited the poor

dramatically.

25
 Surely, one may say, developing countries are better off than 200 years ago, but

should we not assess if wrongful harm has occurred by asking what things would be like

had European supremacists never invaded the rest of the globe? The trouble is that it is

impossible to say anything about this benchmark. What are we to make of the idea that

the world would be a better place if states had never emerged, or if colonialism had never

happened? Such questions defy sensible answers. It is hard (if not impossible) to assess

when agents came close to deciding differently, or natural events may readily have

occurred in other ways. It is equally hard to assess what alternate course would then have

emerged. Had Europeans not colonized Africa, alternate political structures may have

allowed indigenous peoples to exploit their continent’s wealth to build prosperous

civilizations. It is also conceivable that war would have thwarted such efforts. According

to Herbst (2000), physical geography in Africa impeded the emergence of powerful

states.

There is yet another way of articulating the thought that developing countries are

being harmed by the global order: a benchmark of fairness, the reference point being a

state of nature where resources are distributed fairly. “’Worldwide 34,000 children under

age five die daily from hunger and preventable diseases.’ Try to conceive a state of nature

that can match this amazing feat of our globalized civilization!” writes Pogge (2004), p

274. Yet state-of-nature references cannot distinguish between the view that the global

order does harm, and any other view explaining how such poverty could arise. Such

references only show that things are not as we would have hoped. Whatever else is true,

among the three benchmarks we have considered, the historical benchmark is the only

26
one which we can make sense of. In terms of that benchmark, the global order has

brought tremendous advances.

10. There are many questions about global justice that I have been unable to discuss here.

The philosophical foundations of human rights have attracted much attention recently.

(See the article by Alan Buchanan in this volume.) Topics that are both in need of much

more, and accessible to, philosophical analysis include immigration and fairness in trade.

In need of but less accessible to philosophical analysis is the question of how to distribute

the burdens from climate change. Deserving of more attention are also political and

economic structures other than states. We must explore, for instance, what demands of

justice, but also what types of accountability, apply to entities like the World Trade

Organization or the European Union.

One topic whose more systematic resurrection political philosophy would benefit

tremendously from is humanity’s collective ownership of the earth. (See Peter Vallentyne

on “Left Libertarianism” in this volume for more on this subject.) To illustrate, suppose

the population of the US shrinks to two, but they control access through border-

surveillance mechanisms. Nothing changes elsewhere. Surely these two should permit

immigration since they are grossly under-using their area. We can best explain this view

by the fact that all of humanity has claims to the earth. Immigration is but one topic

which such theorizing could illuminate. Humanity’s ownership of the earth was the

pivotal theme of 17th century political philosophy. In an age in which global problems

have become central, we have much to gain from reinvigorating that standpoint.

27
 Literature

Beitz, Charles. 1999. Political Theory and International Relations (revised

edition). Princeton: Princeton University Press.

Black, Samuel. 1991. “Individualism at an Impasse.” Canadian Journal of

Philosophy 21 (3): 347-377.

Blake, Michael. 2001. “Distributive Justice, State Coercion, and Autonomy.”

Philosophy and Public Affairs 30: 257-296.

Caney, Simon. 2005. Justice Beyond Borders: A Global Political Theory. Oxford:

Oxford University Press.

Caney, Simon. 2002. “Cosmopolitanism and the Law of Peoples.” Journal of

Political Philosophy 10: 95-123.

Carens, Joseph. 2003. “Who Should Get In? The Ethics of Immigration

Decisions,” Ethics and International Affairs 17: 95-110.

Feinberg, Joel. 1973. Social Philosophy. Englewood Cliffs: Prentice Hall.

Fleischacker, Samuel. 2004. A Short History of Distributive Justice. Cambridge:

Harvard University Press.

Freeman, Samuel. 2007. Justice and the Social Contract. Essays on Rawlsian

Political Philosophy. Oxford: Oxford University Press.

Herbst, Jeffrey. 2000. States and Power in Africa. Princeton: Princeton

University Press.

Kant, Immanuel. 1996. “The Metaphysics of Morals” In Mary Gregor (ed.),

Immanuel Kant: Practical Philosophy. Cambridge: Cambridge University Press, 353-

605.

28
 Kavka, Gregory. 1986. Hobbesian Moral and Political Theory. Princeton:

Princeton University Press.

Kymlicka, Will. 2002. Contemporary Political Philosophy: an Introduction.

Oxford: Oxford University Press.

Miller, David. 2007. National Responsibility and Global Justice. Oxford: Oxford

University Press.

Moellendorf, Darrel. 2002. Cosmopolitan Justice. Boulder: Westview.

Nagel, Thomas. 2005. “The Problem of Global Justice.” Philosophy and Public

Affairs 33 (2): 113-147.

Nussbaum, Martha. 2006. Frontiers of Justice: Disability, Nationality, Species

Membership. Cambridge: Harvard University Press.

Nussbaum, Martha. 2000. Women and Human Development: The Capabilities

Approach. Cambridge: Cambridge University Press.

Pogge, Thomas. 2004. “’Assisting’ the Global Poor.” In Deen K. Chatterjee (ed.),

The Ethics of Assistance: Morality and the Distant Needy. Cambridge: Cambridge

University Press.

Pogge, Thomas. 2002. World Poverty and Human Rights. Cambridge: Polity.

Pogge, Thomas. 1994. “Cosmopolitanism and Sovereignty.” In Chris Brown

(ed.), Political Restructuring in Europe: Ethical Perspectives. London: Routledge, 89-

122.

Pogge, Thomas. 1989. Realizing Rawls. Ithaca: Cornell University Press.

Rawls, John. 2001. Justice as Fairness: A Restatement. Ed. by Erin Kelly.

Cambridge: Harvard University Press.

29
 Rawls, John. 1999a. The Law of Peoples. Cambridge: Harvard University Press.

Rawls, John. 1999b. A Theory of Justice. Revised Edition. Cambridge: Harvard

University Press.

Risse, Mathias. 2006. “What to Say about the State.” Social Theory and Practice

32 (4): 671-698.

Sangiovanni, Andrea. 2007. “Global Justice, Reciprocity, and the State.”

Philosophy and Public Affairs 35 (1): 3-39.

Scanlon, T.M. 2003. “The Diversity of Objections to Inequality.” In Scanlon, The

Difficulty of Tolerance. Essays in Political Philosophy. Cambridge: Cambridge

University Press; 202-219.

Scheffler, Samuel. 2001. Boundaries and Allegiances. Problems of Justice in

Liberal Thought. Oxford: Oxford University Press.

Sen, Amartya. 1999. Development as Freedom. New York: Random House.

Sen, Amartya. 1985. Commodities and Capabilities. Amsterdam: North Holland.

Tan, Kok-Chor. 2004. Justice Without Borders. Cosmopolitanism, Nationalism,

and Patriotism. Cambridge: Cambridge University Press.

Wenar, Leif. 2006. “Why Rawls is Not a Cosmopolitan Egalitarian.” In R. Martin

and D. Reidy (eds.), Rawls’ Law of Peoples: A Realistic Utopia? Oxford: Blackwell.

Williams, Bernard. 2005. In the Beginning was the Deed. Realism and Moralism

in Political Argument. Ed. by Geoffrey Hawthorn. Princeton: Princeton University Press.

30