Justice in context: assessing contextualism as an approach to justice

Michael Buckley
Justice in context: assessing contextualism as an approach to justice

Justice in context: assessing contextualism as an approach to justice

Michael Buckley*

Philosophy Department, Lehman College, City University of New York, New York, NY, USA

Abstract

Moral and political philosophers are increasingly using empirical data to inform their normative theories. This has sparked renewed interest into questions concerning the relationship between facts and principles. A recent attempt to frame these questions within a broader approach to normative theory comes from David Miller, who has on several occasions defended ‘contextualism’ as the best approach to justice. Miller argues that the context of distribution itself brings one or another political principle into play. This paper examines this claim. It considers several plausible strategies for carrying out Miller's general project and argues that each strategy fails. Nevertheless, the author maintains that an investigation into why they fail paves the way for a philosophically plausible account of the relationship between facts and principles.

Keywords: justice; contextualism; global distributive justice; justification; John Rawls; David Miller

Published: 28 May 2012

*Correspondence to: Michael Buckley, Philosophy Department, Lehman College, City University of New York, New York, NY, USA. Email: michael.buckley@lehman.cuny.edu

©2012 M. Buckley. This is an Open Access article distributed under the terms of the Creative Commons Attribution-Noncommercial 3.0 Unported License (http://creativecommons.org/licenses/by-nc/3.0/), permitting all non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Citation: Ethics & Global Politics, Vol. 5, No. 2, 2012, pp. 71–94. http://dx.doi.org/10.3402/egp.v5i2.8970

 

An enduring problem in moral and political philosophy concerns the relationship between facts and principles. Every normative theory must, either implicitly or explicitly, address this relationship, because each theory's conception of justification, attribution, and objectivity hinges on an interpretation of it. Assuming, as many political philosophers do, that principles ‘must be justified by the conditions of life as we know it or not at all,’1 then the growing body of empirical data collected from the social and biological sciences might safely be thought to bear on these interpretations.2 However, determining exactly how empirical data figures into them remains a philosophical, rather than an empirical, undertaking, for empirical data can only help inform, but never resolve, this enduring problem.

A recent attempt to frame these issues within a broader approach to normative theory comes from David Miller, who has on several occasions defended ‘contextualism’ as the best way to approach justice.3 Contextualism holds that ‘it is the context of distribution itself that brings one or other principles into play.’4 In contrast to ‘universalist’ approaches to justice, which view principles as both justified independently of their context of application and universally applicable to all contexts, contextualism treats diverse political issues in terms of their unique contextual features rather than systematically as part of—and answered in terms of—an overarching theory of justice.5 It does this by ‘tracing systematic links between principles of justice and their contexts of application,’ and by exposing an ‘underlying logic that we can both grasp and use as a critical tool to assess prevailing conceptions of justice at any particular moment.’6 As a result, contextualism interprets the relationship between facts and principles as interdependent or mutually supportive.

Miller thinks that contextualism is intuitively appealing insofar as people tend to think a just or unjust action depends on the circumstances of the situation.7 In addition, it accommodates a plurality of principles, which he again believes is appealing insofar as it explains the persistence of diverse moral judgments across various political issues. And finally, contextualism supports the idea that different principles of justice are rationally justified in virtue of their being tethered, but not reduced, to facts. Consequently, contextualism comprises a number of related positions that bear on the relationship between facts and principles: it is a piecemeal, pluralistic approach that avoids intuitionism, relativism, and reductionism by defending normative judgments in terms of a link between principles and factual circumstances.

Miller frames his defense of contextualism in general terms, prompting the need for further evaluative scrutiny, which he invites. This paper takes up his invitation. Following Thomas Pogge, it begins by defending the claim that Miller's general view is best labeled critical contextualism.8 Critical contextualism traces systematic links between principles of justice and their contexts of application, as noted earlier, but makes clear what seems implicit in Miller's position, namely, that the delineation of contexts must also be justified. In the absence of further justification, moral disputes over principles slip too easily into disputes over competing contextual descriptions. If one person defends a particular response to global poverty by reference to context A, and another defends a different response by reference to context B, then it is difficult to see how references to contexts are anything but normative disputes themselves. Ideally, critical contextualism can resolve the normative dispute by encompassing a non-normative defense of the context. As Miller puts it, ‘clashes of principle will not, in general occur, once the context in which the distributive decision in question is being made is properly specified.’9 I develop this thought in ‘CONTEXTUALISM AS CRITICAL CONTEXTUALISM’ section, where I argue that Miller's contextualism must be interpreted as critical contextualism if it is to avoid, as Miller puts it, ‘selling the pass to skepticism, relativism, conventionalism or any other view that discredits justice as a workable idea for social criticism and social reform.’10

Assuming critical contextualism is an accurate interpretation of Miller's general position, it remains unclear how one might realize its several positions within a single approach, or whether such a realization is even possible. For example, can the delineation of a context admit of a clear and uncontroversial description? Is the account of pluralism sufficiently critical to avoid skepticism, relativism, or conventionalism? Do principles appropriate to one context ever conflict with another, and, if so, is there a contextual solution to the conflict?

Miller responds to these questions by drawing heavily on empirical research. I argue in ‘DELIMITING THE CONTEXT: MILLER'S STRATEGY AND ITS PROBLEMS’ section that this strategy is insufficiently critical, since it is too accommodating of the status quo and thus incapable of providing a workable idea for social criticism. I develop an alternative strategy in ‘THREE DISTRIBUTIVE CONTEXTS’ section, where systematic links between principles and contexts are presented in terms of a mirroring relationship between a principle's directive and the contextual facts that generate claims to resources. The idea is that if one abstracts from distributive contexts familiar to the philosophical literature on global justice, one can expose the formal structure underlying a person's claim to resources. Assuming these formal structures capture legitimate claims, plausible principles of justice must share the same formal structure in order to answer the claim.

In ‘JUSTIFICATION IN CONTEXT’ section, I argue that this formal strategy realizes many of the key features of critical contextualism. However, the strategy offers no contextual way of balancing conflicts between distributive contexts. Specifically, it cannot adjudicate situations where the defense of a principle in one context is conceptually linked to the defense of a principle in another context. Having failed to develop a contextual answer to what I frame as a priority problem, I argue that the several plausible strategies for explaining how contextualism might trace systematic links between principles and factual contexts fail to realize the distinctive features of contextualism.

These findings raise doubts about whether critical contextualism can realize its several positions within a distinct approach. However, I do not further conclude from these doubts that contextualism is without insight. Indeed, an investigation into why the several strategies fail paves the way for outlining a distinct, philosophically plausible account of the relationship between facts and principles in normative theory, which I identify in the conclusion. As a result, one might view this paper as working within the spirit of Miller's invitation to scrutinize, examine and develop the contextual dimensions of normative political theory.

CONTEXTUALISM AS CRITICAL CONTEXTUALISM

A contextual approach to justice reflects a number of related positions. It is pluralistic insofar as different contexts of distribution support the appropriateness of different principles; it is objective insofar as judgments based on principles find rational grounds of justification in a link with factual circumstances; and it is critical insofar as justified principles serve as independent criteria against which social practices and institutions are judged and found wanting. To illustrate the key idea, take the factual circumstances surrounding organ transplants. These circumstances are usually sufficiently robust to at least rule out as inappropriate certain distributive principles. For example, the principle first-come-first-served makes little sense if the biological match between a kidney and a recipient is crucial to the operation's long-term success. In this and other cases, the relevant distribution principle depends on facts about what is being distributed and to whom it is being distributed. References to these facts justify the principle.

Miller is not alone in thinking the context important to the justification of principles. Rawls expresses a similar idea. He notes, ‘it is the distinct structure of the social framework, and the purpose and role of its various parts and how they fit together, that explain why there are different principles for different kinds of subjects.’11 On this view, the distributive context plays a distinct justificatory role. Different principles are justified in virtue of their appropriate fit to different subjects or contextual circumstances. Miller describes the ‘fit’ as follows: ‘in context C it is always right to distribute the relevant goods and bads according to principle P.’12 Provided we can identify the unique features of context C, we can then rationally explain and defend the appropriateness of principle P without reference to a more fundamental or transcendental principle. Given a variety of distinct contextual situations, it further follows that there is a plurality of equally fundamental principles that is neither derived from, nor represents the substantive applications of, an overarching principle of justice.13

One might reasonably resist the claim that contextualism is rationally grounded and thus sufficiently critical by citing the work of Michael Walzer. Walzer's manner of identifying distributive spheres, or contexts, captures a factual component (the ‘factual’ social meaning of a particular good) and matches these factual elements to normative directives (the principles that inform the good's distribution). On Walzer's view, the ‘crucial medium of social relations’ is a shared conception of ‘what the goods are and what they are for.’14 It is this shared social meaning that determines the distribution of a particular good. Moreover, ‘[m]en and women take on concrete identities because of the way they conceive and create, and then possess and employ social goods.’15 As a result, both distributive principles and judgments about them are based on ‘people's deepest sense of how they ought to live,’ or how ‘we mean to live and don't yet live.’16 Since these meanings are rooted in culture, distributive principles—along with the operative conceptions of moral persons—are culturally relative.

While Walzer's line of reasoning could generate agreement on the ‘facts’ (in an anthropological sort of way), his contextualism fails to be critical in the following sense: principles fail to serve as independent criteria against which social practices can be assessed and found wanting. This is because the defense of principles is based on the very social values (and practices) the principles are meant to assess. Walzer readily concedes the point. He views the social critic as an interpreter of a society's history and culture—one who assesses the ‘fit’ between a society's institutions and ideals. Those best positioned to interpret this ‘fit’ are those raised in the society.17 Outside criticism is generally illegitimate, because knowledge about the social fabric of society—essential to the interpretation—is generally unavailable to foreigners.18

Walzer's ‘contextualism’ illustrates the following point. To avoid relativism or conventionalism, principles must be sufficiently independent of cultural bias so as to serve as independent criteria against which prevailing norms and the policies they support can be assessed. It is not enough to trace ‘systematic links between principles of justice and their contexts of application.’ A critical contextualism must trace systematic links between principles and contexts that are themselves justified.

It is not always clear whether Miller views his conception of ‘social criticism’ as being significantly different from Walzer's, because he sometimes cites Walzer's view as an illustration of contextualism.19 However, if contextualism is to avoid, as Miller puts it, ‘selling the pass to skepticism, relativism, conventionalism or any other view that discredits justice as a workable idea for social criticism and social reform,’ then Walzer serves as a poor model.20 We must instead follow Pogge, who suggests anchoring the delimitation of the context in a more objective defense.21 Supposing we can objectivley defend the delimitation, then the principles appropriate to the context are likewise firmly anchored, and thus able to serve as independent metrics of assessment.

Tracing systematic links between principles and contexts that are themselves justified suggests the following. In addition to justifying a principle as appropriate to some context, the delineation of the context requires a self-standing justification. In the absence of a self-standing justification, moral debates over principles slip too easily into debates over competing contexts.

How might the context be justified? As I have just described it, contextualism seems to imply a division of labor between the justification of a context and the justification of a normative principle appropriate to the context. This suggests achieving the former defense prior to the latter, which in turn suggests the arrows of justification point in one direction, from the facts comprising the context to the principle of distribution. In some cases, the contextual facts might imply aims that in turn move justification smoothly from facts to principles, as perhaps in the case of organ transplants. However, there are two problems with this approach. First, the arrangement of facts comprising a context might be complex and open to competing interpretations.22 Consider the question: why have some societies become rich while others have remained poor? Historical studies and development economics offer competing answers.23 For example, while there are good reasons to think domestic institutions—such as dependable property rights, law, bureaucratic capacity, and independent courts—are causally efficacious for economic growth, it is unclear whether these factors are sufficiently insulated from global structures to be independently efficacious. In the absence of a resolution among rival explanatory theories, the context cannot be described clearly and uncontroversially.

One might object to this by noting that while no dominant explanatory model currently exists, further empirical work could yield one. As with other areas of empirical investigation, the supporting data can mount in favor of one theory over another, at which point the context of distribution would be justified on familiar empirical grounds. But this leads to a second, more troubling philosophical difficulty; namely, that the very purpose of identifying certain facts in support of a normative principle is already guided by a practical enterprise influenced by a variety of social factors. Here the problem is not that of reconciling competing empirical accounts through further empirical research. Rather, the problem is establishing a clear and uncontroversial description of the context. This problem might result because people assign different weights to the facts agreed on. Since the assignment of weights is not itself determined by the facts (for if it were, then those further facts would be subject to competing assignments as well, suggesting an infinite regress), the different descriptions are incommensurable.24 Or the problem might result from a variety of more complicated reasons advanced by philosophers as diverse as Nietzsche, Gadamer, Quine, Davidson, and Rorty. These philosophers have problematized the conceptions of rationality and reality on which clear and uncontroversial descriptions of facts are normally based. In the process, they challenge the very possibility of developing a self-standing justification of a particular context based on empirical data. In the absence of an uncontroversial description, little justificatory force is gained by referencing contexts, because disputes over principles simply slip to disputes over contextual interpretations. This undermines Miller's claim that ‘clashes of principle will not, in general occur, once the context in which the distributive decision in question is being made is properly specified.’25

How might one respond to these notable and troubling philosophical challenges? There are at least three possibilities. First, one might concede their points by couching interpretations of contexts in terms of reflective equilibrium.26 Second, one might concede their points by specifying interpretations of contexts in terms of political problems in need of rectification.27 Third, one might jettison the idea of a self-standing justification by construing the relationship between principles and context as mutually supportive. Miller opts for the third strategy, which is sometimes specified as a modified version of the first.28 He claims that distributive principles both reflect and help constitute contextual circumstances.29 In the next section, I explore this idea and argue that Miller's execution of it is insufficiently critical.

DELIMITING THE CONTEXT: MILLER'S STRATEGY AND ITS PROBLEMS

Identifying an ‘interdependent’ relationship between principles and contexts is an exercise in justification—we are looking for principles appropriate to a context. Miller conceives political justification in terms of a link between the epistemic and practical dimensions of political philosophy. ‘A condition for a theory's being valid,’ Miller writes, ‘is that it should be possible for people to come to accept it and live according to its principles.’30 Principles cannot make impossible demands on us and still be publicly justifiable; nor can they demand a social order too far from our institutional capacities to be practicably feasible.31 Public justification and practical feasibility place limits on the appropriateness of principles, and these two limits predominately characterize Miller's approach to justification.

When fixing on these limits, Miller is acutely aware of the philosophical problems they create. For example, when identifying co-citizenship as a factual limit on the feasibility of global justice, Miller notes that this seems to ‘take for granted an institutional arrangement that might itself be regarded as unjust: the existence of separate states each delivering a separate bundle of rights, opportunities, and resources to its own members, but not to outsiders.’32 Miller seems to imagine a hypothetical critic arguing that co-citizenship, while indeed a fact, represents a historically contingent fact in need of justification, and it is a mistake to treat this contingent fact as a limiting condition on a principle's justification. To do so, the critic might continue, is to turn the critical relationship between principles and institutions on its head. Rather than principles serving as independent criteria against which institutions are assessed and found wanting, the institutions serve as fixed points against which principles are judged feasible. This undermines their critical function by privileging the status quo and placing it beyond reproach.

Miller recognizes the problem, but rather than abandon reference to facts, he recasts the challenge as one about determining which facts count as contextual limits our theory must recognize as fixed, and which count as contingencies our theory's principles might alter.33 Assuming, as many political philosophers do, that principles ‘must be justified by the conditions of life as we know it or not at all,’ the challenge might be framed as follows.34 When defending a set of principles, one must avoid confusing cultural or historical contingencies with fixed limits, or else one unwittingly affirms the status quo. But one must also avoid confusing the fixed limits with contingent facts, or else one unwittingly creates a useless, utopian theory. Rawls identifies the same problem. For Rawls, there is a ‘question about how the limits of the practicably possible are discerned and what the conditions of our social world in fact are.’35 The problem is that ‘the limits of the possible are not given by the actual, for we can to a greater or lesser extent change political and social institutions and much else.’36

Like Miller, Rawls relates this problem to the practical roles of political philosophy. One such role is to orientate the public toward a possible way of viewing its political and social relations; the other is to reconcile the public with certain facts about human relations discovered over time.37 Pursuing these aims simultaneously implies constructing the most just theory possible. While feasibility places a constraint on justification, practical expedience must not be purchased at the price of leaving a more just, albeit less expedient, alternative on the table.38 As Rawls puts it, normative theory should identify a realistic utopia.

The question is whether Miller sufficiently meets this challenge. As characterized earlier, the ‘limits of the practicably possible’ imply a theory can err either by failing to reach the outer edge of the limit or by transgressing it. These two possibilities obtain because (1) facts serve as a ceiling beyond which our moral ideals become sheer fancy and (2) principles serve as a floor below which our practical observations expose mere expedience. The relationship between facts and principles portrayed by this challenge can be characterized as mutually supportive; facts and principles check each other. The challenge is to explain this relationship in a manner that preserves both the critical function of principles and the feasibility constraint of facts. Miller, I argue, employs a strategy that fails to preserve the critical function of principles.

To see this, consider Miller's public justifiability constraint. A publicly justified theory of justice is one that ‘appears no longer as an external imposition conjured up by the philosopher, but as a clearer and more systematic statement of the principles that people already hold.’39 In other words, a theory's principles should not transgress the factual limits of the practicably possible. This is important because political philosophy, as a branch of practical reasons, aims to inform and govern the judgments of citizens in their support and deliberation of institutional structures and social practices.40 A set of political principles cannot effectively serve this aim if it places foreign or idealistic obligations on those it is meant to govern. Miller, in an effort to guard against this error, looks ‘empirically at the judgments and behavior of people when they allocate resources in different contexts.’41 In the absence of empirical testing, Miller worries that the moral ideals embodied in a political theory risk becoming utopian in the pejorative sense of the word. A utopian theory is an invalid theory, since, as already noted, a condition for a theory's being valid is that it should be possible for people to accept it.

However, why should we suppose, as Miller's argument would have us suppose, that the empirical data relied on, which describes actual beliefs, set the limit for what people might possibly accept? Obviously, the beliefs and behaviors of Southern Americans prior to 1865 set public justifiability constraints on the abolition of slavery. It certainly was impossible for them to willingly accept emancipation. But to infer from this the invalidity of a slave-free society, and to further suggest the impossibility of accepting such a society at some point in the future, is to confuse historically contingent facts with contextual limits our theory must recognize as fixed.

Empirical data concerning actual beliefs cannot inform us as to the status of those beliefs. Here Rawls is instructive. For Rawls, correct normative judgments and the ‘facts’ to which they refer depend on our having first taken up some normative standpoint.42 To illustrate this, Rawls notes that slavery is unjust because it allows some to own others. It is this fact and not the fact that slavery occurred in some places or at certain times that makes slavery wrong. The first fact reflects the ‘so-called right-and-wrong-making characteristics,’ which we can identify because ‘a constructivist procedure is framed to yield the principles and criteria that specify which facts about actions, institutions, persons, and the social world generally, are relevant in political deliberation.’43 In the absence of a way of discerning contingent from non-contingent facts, a theory risks privileging the status quo at the price of a more just alternative.

A political theory developing an interdependent relationship between principles and facts must balance empirical data with a method for morally assessing the data's relevance. To some extent, Miller's reliance on a second type of factual consideration, ‘presuppositional’ evidence, is meant to achieve this balance, because it aims to draw analytical distinctions between social values that can then be used to interpret empirical findings. The general idea is that unless certain facts are presupposed to hold true for certain kinds of associations, the normative concepts ordinarily used to assess those associations become irrelevant or unfeasible.44 For example, it is sometimes argued that our ordinary concept of social justice requires, as a presuppositional condition, something like Hume's circumstances of justice.45 Likewise, Miller develops several ‘modes of association’ that explain certain distributive concepts commonly applied in contemporary liberal society. These include solidaristic associations such as the family, instrumental associations such as the corporation, and political associations such as the state. Each association has a corresponding distribution principle that ‘not only reflects the character of group relations but also helps to constitute those relations for the future.’46 For example, ‘need’ identifies the proper principle for solidaristic associations, ‘desert’ captures the appropriate principle for instrumental relations, and, within political associations, ‘equality’ determines just distributions. In the absence of presuppositional conditions, such as ‘common identity’ within solidaristic associations, certain distributive principles such as ‘need’ cannot hold. As a result, presuppositional facts place feasibility constraints on appropriate principles, but those principles also help isolated certain facts as relevant to normative assessment. Thus, there is a mutually supportive relationship between the moral concept and a fact.

While this approach risks reintroducing the sort of conventionalism associated with Walzer,47 I think an alternative way of carrying it out provides a promising strategy for establishing an interdependent relationship between principles and contexts, and I suggest such an alternative in the next section. Presenting a plausible version, however, faces significant challenges not only illustrated by the aforementioned risk of conventionalism but also by Miller's characterization of the global justice mode of association. He describes this relation as one among nations forming independent political communities with interests in determining the future of their own community, rather than ‘among people who are citizens of the same political community.’48 The presuppositional fact implicit in this mode, namely, the relations among self-determining political units, places a feasibility constraint on appropriate principles of global justice. But from where does this constraint arise? Miller again relies on empirical research, noting that philosophers working on global justice can ‘elaborate principles that are consistent with what can be regarded as common ground’ among social scientists working in the area.49 The ‘common ground’ includes the observation that ‘“domestic factors”… play at least a significant part in explaining the relative wealth and poverty of nations, and that this is enough to ground an investigation of national responsibility.’50 Apparently, the question is ‘not what institutions we ought to have, but what the institutions we currently have would have to do to be justified.’51

This is a perfectly legitimate question, but even a plausible response to it does not answer the challenge Miller sets himself. Recall, the challenge is to discern which facts count as contextual limits our theory must recognize as fixed and which count as contingencies our theory's principles might alter.52 Social science concerning existing associations cannot answer this question, since existing association do not exhaust the set of feasible associations. In order to know whether a theory has reached the feasible limits, we must distinguish between historically contingent institutions our theory's principles might alter and non-contingent institutions our theory must recognize as fixed. In the absence of this distinction, we risk purchasing a feasible theory at the price of a more just, albeit less expedient, alternative.

Miller's reliance on empirical data has been criticized for a variety of reasons in other places.53 I have focused my argument around Miller's own standard of criticism, arguing that Miller's reliance on empirical data does not avoid, as he puts it, ‘selling the pass to skepticism, relativism, conventionalism or any other view that discredits justice as a workable idea for social criticism and social reform.’54 It does not provide a workable idea of social criticism because its manner of testing normative principles is not balanced against a mechanism for testing the status of facts. Expressed in terms of the ‘limits of the practicably possible,’ it does not identify both the facts serving as a ceiling beyond which our moral ideals become sheer fancy and the principles serving as the floor below which our practical observations expose mere expedience. Rather, it risks purchasing feasibility and public acceptance at the price of expedience.

The analysis of Miller's strategy, however, exposes contextualism to a more general difficulty, for if the identification of non-contingent factual limits first requires normative principles, then it cannot also be true that the justification of normative principles first requires the identification of non-contingent factual limits. Nor can one begin with a normative standpoint in order to inform the selection of contextual facts, because one would then require a non-normative justification of this standpoint.55 If each route assumes having first traversed the other, then how can a critical contextualism realize the otherwise appealing idea that principles are justified by virtue of a mutually supportive fit within different contexts?

I noted previously that Miller's ‘modes of association’ suggest a promising approach. Rather than provide a self-standing justification of contexts, or defend the connection between principles and modes of association on empirical grounds, or begin with moral concepts in order to isolate relevant facts, we might instead abstract from modes of distribution widely identified in the philosophical literature in order to analyze their formal structures. If the formal structures expose the manner in which certain claims to resources originate, then those structures include within them some minimal normative content—a person's claim to resources. Principles answering these claims can be said to ‘fit’ the mode of association and determine how resources should be distributed. This formal approach can then explain the ‘fit’ in terms of an underlying logical relationship between a principle's directive and the origins of a claim. It is this underlying logic rather than the moral concepts or empirical data that generates the critical leverage for assessing existing institutions and practices. I now turn to this alternative strategy to see whether it can realize the several positions associated with a critical contextualism.

THREE DISTRIBUTIVE CONTEXTS

In this section I examine the formal structures of three distributive contexts common to the global justice literature. The aim is to explain the idea of ‘an underlying logic [between principles and contexts] that we can both grasp and use as a critical tool when assessing the prevailing conceptions of justice.’56 The three contexts are not meant to be exhaustive. Rather, they are meant to illustrate an alternative way of explaining the interdependent relationship between principles and contexts. To avoid confusion, I note at the outset that I am not trying to defend these three contexts. Rather, I assume them in order to better explain what is meant by the ‘fit’ between principles and contexts, and how an ‘underlying logic’ might be used to generate the critical leverage expected of principles. This assumption should work in favor of critical contextualism. However, as I explain in the next section, this strategy also fails to realize all the distinct elements of critical contextualism.

Two of the three contexts are prevalent in the philosophical literature, including the global justice literature. Following Rawls, I will refer to one as an allocative association and the other as a pure procedural association.57 Allocative and pure procedural associations capture distinct relations between the goods delivered, the parties receiving or providing goods, and the manner in which claims to goods originate. I will argue that these contexts inform the structure of the normative directives meant to govern the associations, because principles responding to these distinct relations must be structured in particular ways if they are to properly respond, or reflect, the circumstances they are meant to govern.

Consider an allocative association, which involves the allocation of existing goods among non-cooperative agents, or agents having nothing to do with creating the goods distributed. The discernable features of an allocative context include: (1) the agents [states, individuals] whose needs, desires, capacities, and preferences are known, and between whom allotments of bundles of goods take place, (2) the fact that those receiving allotments have not produced the goods delivered, and (3) the absence of a shared institutional arrangement between donors and recipients. As a result, the recipients’ claims are not based on their productive efforts or on reasons citing shared institutional arrangements. Instead, claims are based on the recipients’ particular characteristics, such as their need for certain resources. Similarly, donors are identified by some characteristic, such as capacity, in virtue of which transfers might be deemed morally obligatory.

Given the discernable features of an allocative association, normative directives governing it must identify both the characteristics of recipients by which entitlements are determined and the characteristics of donor parties by which remedial responsibilities are assigned.58 For example, the principle to each according to need, from each according to ability reflects an allocative context insofar as it directs the transfer of some already made good from one group to another based on specific characteristics of each group. Rawls's ‘duty of assistance’ provides another example. For Rawls, the duty to assist obligates well-ordered societies to assist those states that could be ‘well-ordered’ save for their want of human, material, technological, or institutional capital.59 The underlying logical structure of an allocative principle is captured by the following decision-making rules:

Resource Entitlement Rule (RER)

  • Allocate resource m to agent a whenever a meets criterion x.

Remedial Transfer Rule (RTR)

  • If agent b meets criterion y, allocate resource m from b to a whenever a meets the previous RER condition.

Here x and y represent substantive criteria by which allocations are deemed just. Using Rawls's duty of assistance as an example of a substantive principle, we can fill in the variables as follows: If society b is well-ordered (y), then allocate resource m to society a if society a's needs results only from a want of human, material, technological, or institutional capital (x).

Rawls's duty of assistance has been criticized for failing to address the effects of an increasingly complex global marketplace.60 According to this criticism, a relevant contextual feature of global distributive justice is a global economic order comparable in extent and in distributional effects to the domestic economic order. As such, the moral dimensions of global poverty must be analyzed under the light of a global economic basic structure.

Those taking this perspective frame global distributive justice in terms of a pure procedural context. A pure procedural context involves the distribution of goods created by and distributed among cooperative agents. Its salient features include an institutional order within which production and subsequent distribution takes place. The principles appropriate to this context set background economic rules, and are thus ‘prior to both production and distribution and therefore involve neither the idea of an already existing pool of stuff to be doled out nor the idea of already owned resources to be redistributed.’61

Within a pure procedural association, a person's claim to a resource derives from one's having played by the rules governing the institutional order. Distributions are determined not by an allocative principle redistributing already produced goods from one party to another on the basis of their specific characteristics, but rather by a pure procedural principle governing the background economic structure. As in the case of a lottery, whereby each person is given an equal chance to receive a good, the outcome—whatever it might be—is appropriate in light of the rules and purposes of the distributive process.

In a pure procedural association, claims to resources originate with one's cooperative effort in creating the goods under a shared set of institutional arrangements. A key function of a pure procedural principle is to define appropriate distributions in terms of legitimate expectations. Capitalistic economies governed by private property rules illustrate the point: distributions result from various decisions made within a marketplace and no attempt is made to redistribute already produced resources on the basis of specific characteristics of particular individuals.62 Instead, supply and demand determines distributions through a price-coordinated system of exchange affected by the innumerable decisions of consumers and producers.

We can express the underlying logic of pure procedural principles more formally by noting the following decision-making rules:

Public Structure Rule (PSR)

  • If structure z is a unified structure of cooperation, then z is to be ordered by a set of public rules against which claims to resources can be made and judged legitimate.

Legitimate Expectations Rule (LER)

  • If a acts in accordance with the public rules identified by PSR, then a is entitled to whatever distribution follows from acting in accordance with those rules.

Notice that a pure procedural association is logically distinct from an allocative one. Whereas the latter identifies appropriate holdings on the basis of certain characteristics of particular individuals, the former identifies an appropriate set of public rules against which legitimate expectations are formed and holdings deemed just. A logically sound defense of a distributive principle depends on its fit with a distributive association. For those citing domestic factors as the cause of global poverty, the problem of global distributive justice (if there is such a problem) is to identify the relevant characteristics by which recipients are entitled and donors obligated to transfer existing goods. Pure procedural principles of distributive justice, such as Rawls's difference principle, are conceptually inappropriate to this context, since the manner in which they adjudicate conflicting claims depends on a unified structure of cooperation within which production and subsequent distribution takes place. In the absence of a unified structure of cooperation, pure procedural principles cannot perform their key function, namely, to order the public rules governing unified schemes and determining the legitimacy of claims. Conversely, those citing global economic institutions as the cause of poverty must find allocative principles logically inappropriate, because they merely redirect the holdings of preexisting goods. However, the problem does not rest on the redistribution of preexisting goods—the problem rests on the inappropriate rules governing the global institutions within which production takes place.

While allocative and pure procedural contexts frame many philosophical arguments on global distributive justice, a third context is appropriate in some cases. I call this third context a transitive context, since it reflects claims arising from externalized burdens, that is, harms created by one group but experienced by another group having nothing to do with creating the harms. The question of a transitive context is this: to what extent must some individuals or states honor the claims of others with whom they share a common structure but no cooperative relation? A transitive relation is distinct from a pure procedural one, since the latter focuses on the cooperative efforts within a unified structure of institutions and identifies legitimate expectations in light of public rules. However, those rules say nothing about the effects of cooperative decisions on other parties through a shared but non-cooperative structure, such as a shared ecosystem. These shared structures provide conduits through which benefits created by one group pass over and take effect as burdens for another group having nothing to do with creating those burdens.63 As a decision made within a cooperative association might create a costly and coercive imposition on another non-cooperating party, the latter party may have legitimate claims against the first.

A key function of transitive principles is to identify the particular characteristics by which transferred burdens are assessed. Another key function is to identify the links through which remedial responsibilities are assigned. Given these functions, the normative directive must specify a point at which an externalized burden is deemed illegitimate. We can express its logical structure with the following decision-making rules:

The Externality Rule (ER)

  • If non-cooperative structure z can facilitate the externalization of burden x from a to b, then z must be constrained by criterion y, which identifies a point below which the externality is illegitimate.

Remedial Transitive Rule (RTR)

  • If burden x is created by a and realized in b through (ER) such that the burden causes b to fall below y, then b is entitled to resources reestablishing level y.

While a transitive context has a unique structure, it shares features with both pure procedural and allocative contexts. Like pure procedural contexts, it ties resource claims to shared structures that help determine and shape distributions. Like allocative contexts, it identifies a criterion for assessing resource holdings on the basis of specific information about particular agents. The similarities can sometimes lead to confusion over whether a distributive principle fits one or another context, but one way to contrast the three distributive contexts is to ask the following. (1) Are the distributed resources created through the cooperative efforts of those receiving the distribution? (2) Do parties share a structure that helps determine the distributions? (3) Is some evaluative criterion required to assess people's individual resource holdings? Allocative contexts answer ‘no’ to the first two questions and ‘yes’ to the last. Pure procedural contexts answer ‘yes’ to the first two questions and ‘no’ to the last, and transitive contexts answer ‘no’ to the first and ‘yes’ to the last two questions.

JUSTIFICATION IN CONTEXT

The distributive contexts identified previously are formal in that they identify how the structure of an association determines the manner in which claims to resources originate. The structure in turn informs the key functions of a principle responding to these claims. In order for the principle to answer a claim, its structure must reflect the form of association within which the claim originates. It is in virtue of this reflection that principles are said to ‘fit’ their contexts of application, and it is in the analysis of the context that this ‘fit’ can be said to express an underlying logic—for once the association is analyzed as an allocative, pure procedural, or transitive association, then the normative theorist will know instantly what kind of formal principle is required to answer the claim.

Moreover, the formal strategy informs the relevance of empirical data. That is, empirical data are relevant when it fills out the details of the formal association. We are to look for data determining which association is appropriate to the moral analysis of political issues, such as, poverty, public health, resource conflict, and human migration. In some cases, the data might not sufficiently determine the appropriateness of one association. In other cases, the data might mount in favor of one association over another. Nevertheless, the formal strategy provides a perspective from which to interpret and clarify the relevance of empirical data to moral analysis.

To give these abstract considerations some content, consider the effects of global warming on the Himalayan glaciers and Tibetan plateau. These glaciers are among the world's largest freshwater catchments and the source of fresh water for over 1 billion people.64 Stable glaciers act as seasonal water towers, locking fresh water in the winter and releasing it gradually throughout the warming months. According to the Intergovernmental Panel on Climate Change (IPCC), ‘changes in snow and glacier melt, as well as rising snowlines in the Himalayas, will affect seasonal variation in runoff, causing water shortages during the dry summer months. One-quarter of China's population and hundreds of millions in India will be affected.’65 In order to correct for the quickening pace of glacier melt and compensate for increasingly drier summer months, China, India, and other countries must invest in new water storage facilities now. In the absence of investment today, people will suffer terribly tomorrow.

How should the normative dimensions of this issue be framed, and what principle should govern a response? If variations in the earth's surface temperature are naturally caused, then claims to resources cannot originate from either cooperative efforts or coercive impositions among groups. Supposing this is the case, then questions of justice cannot be framed in terms of pure procedural or transitive contexts. They must instead be framed in terms of an allocative context (if they are to be framed at all), since the normative principle appropriate to this analysis must identify: (1) the particular characteristics by which people are entitled to resources and (2) the particular characteristics by which donors are assigned remedial responsibilities. If the principle does not identify these features, it cannot answer the claim in a manner fitting its origin. And if it cannot answer the claim in a manner fitting its origin, the solution offered will appear inappropriate, arbitrary, and unjustified. By contrast, an appropriate principle might direct wealthy countries to aid poorer countries facing serious fresh water shortages, as Rawls's duty of assistance suggests; or it might direct today's residents of the affected areas to allocate savings toward investments for the sake of tomorrow's potentially water-starved residents. In each case, existing resources are allocated from one group to another based on specific characteristics of the groups.

Now suppose that global warming is anthropogenic, as the IPCC argues.66 Claims to fresh water resources will have originated from a shared structure through which coercive impositions are facilitated between groups. One might argue that because the cooperative efforts of one (or several) party have generated pollutants passing over and taking effect as burdens in another party having (at least until recently) nothing to do with creating the burden, the first party is obligated to offset the burden. In this case, the principle must reflect a transitive context; its directive must identify (1) the appropriate characteristics by which transferred burdens are deemed illegitimate and (2) a link through which remedial responsibilities are assigned. Appropriate holdings might be framed in terms of an adequate network of water storage towers, and remedial responsibilities might be assigned to those burning fossil fuels, perhaps through a global carbon tax on fossil fuel consumption.

These two competing analyses depend for their resolution on empirical data. But, given the formal strategy, we now possess a perspective shaping the fact patterns drawn from raw data. Once the data mounts in favor of one pattern, the association it favors will determine the form of the principle governing obligations and entitlements. This illustrates how a distributive principle's directive ‘fits’ the context of application. It also illustrates how one might trace ‘systematic links between principles of justice and their contexts of application’ by exposing an ‘underlying logic that we can both grasp and use as a critical tool when assessing the prevailing conceptions of justice.’67 The underlying logic is captured by the mirroring relationship between the principle's directive and the facts filling out the structure of the association. In other words, once a context is analyzed as allocative, for example, then its governing principle must—as a matter of logical structure—allocate existing goods between parties based on the characteristics of the parties. Expressed the other way around, the rules reflected in allocative, pure procedural, and transitive principles require certain things to be true of their contexts of application if each is to function properly. In the case of allocative principles, goods must already exist, and donors and recipients must exhibit clear characteristics. By contrast, pure procedural principles function in a context where what matters is neither the existence of already produced goods nor the specific characteristics of persons, but rather a unified scheme of cooperation within which production and subsequent distribution takes place. And finally, transitive principles require for their proper functioning a context characterized by externalized harms. In the absence of these factual features, a distributive principle cannot function properly, it cannot answer claims people make. If it cannot function properly, it cannot be justified as appropriate to the context.

The formal strategy also explains Miller's claim that ‘clashes of principle will not, in general occur, once the context in which the distributive decision in question is being made is properly specified.’68 Once the analysis of the context identifies it to be allocative, for example, the distributive principle must take an allocative form. It cannot take the form of either a pure procedural or transitive principle. As a result, the formal strategy provides the critical leverage expected of critical contextualism, because the underlying logic exposed by the analysis determines which formal principle is appropriate. If one applies a pure procedural principle to an allocative context, then the error can be explained and rationally corrected.

One notable drawback to this strategy is that it is unable to resolve clashes between substantive principles sharing the same form. This is because the strategy only exposes the formal relationship between principles and contexts. It does not further discriminate among substantive principles and, as a result, produces a rather weak account of justification. Normative political judgments are justified when they are based on principles whose form fits the context of application. The form can be expressed by more than one substantive principle. For example, the substantive principles to each according to need, from each according to ability and Rawls's duty of assistance are each allocative. The formal strategy offers no way to determine which of the competing principles better serves the claims resulting from allocative associations. As a result, it cannot determine which of the two is uniquely appropriate to a context. In the absence of further determination, the formal strategy's critical leverage is limited.

While this is a serious shortcoming, I do not find it fatal. On the contrary, it helps explain the open texture of moral debate, and provides a rationale for why rival substantive principles might gain wide consensus among opposing parties. At the same time, it significantly narrows the field of plausible rivals by eliminating as unfit those principles failing to reflect the manner in which claims to resources originate. So while the formal strategy cannot reconcile competing substantive principles sharing the same form, its inability to do so actually explains a persistent feature of moral discourse.

A much more serious problem with this strategy is that it offers no contextual way of balancing conflicts between distributive contexts. Specifically, it is unable to resolve situations where the defense of a principle in one context is conceptually linked to the principle of another context. Such conflicts seem likely to occur between pure procedural and transitive principles, because the legitimate expectations established by pure procedural principles depend in part on whether the unified structure of cooperation they support create illegitimate externalities. Rawls provides an instructive case when discussing the relation between the difference principle and the just saving principle.69 The difference principle is a pure procedural principle. It governs legitimate expectations by specifying the public rules against which claims among cooperative members are made and judged legitimate. The just savings principle is a transitive principle. It identifies a criterion against which the present generation can assess its rate of accumulation so that future generations do not fall below a just level of wealth. ‘The just savings principle acts as a constraint on the rate of accumulation;’ the difference principle arranges economic inequalities so that they are ‘to the greatest benefit of the least advantaged, consistent with the just savings principle.’70 As a result, the defense of the difference principle is conceptually linked to the just savings principles. Future generations’ claim to a minimum level of wealth takes priority over the present generation's claim to wealth (beyond that minimum) resulting from their cooperative efforts within a unified scheme of institutions.

Rawls's case is an instance of a priority problem: how should competing claims and their respective principles be prioritized? In A Theory of Justice, Rawls discusses the priority problem in relation to intuitionism, faulting the intuitionist with providing no systematic assignment of weights for adjudicating competing claims and principles.71 Instead, intuitionism relies on intuitive trade-offs between conflicting claims. Critical contextualism attempts to avoid intuitive trade-offs by rationally grounding the justification of principles in distinct contexts of deployment. However, the formal strategy fails in this regard, for there are at least some cases when a contextualist approach will fail to provide such grounds. Take, for example, a global analog to the relationship between the difference principle and the just savings principle, namely, how the distributive claims among co-nationals are balanced, or prioritized, against the claims of foreigners affected by the cooperative efforts among co-nationals. Here the conflict is between a pure procedural and transitive context. To avoid intuitionist trade-offs, critical contextualism must offer a reasoned response to this priority problem, and, to do that, it must resituate the conflict within a broader context. But what might be that broader context? Further abstraction could fail to capture the manner in which claims to resources originate, thereby severing the link between normative directives and their contexts of application. So, while a more abstract context, such as ‘resource allocation,’ includes both pure procedural and transitive contexts, it is too abstract to expose a systematic link between principles and contexts, and thus too abstract to solve the priority problem in a manner consistent with critical contextualism.

While my inability to provide a contextual answer does not entail the absence of any such answer, it seems almost certain that at some point contextualism will face a priority problem that falls between contexts. Contexts are not always autonomous. They are often conceptually linked. Yet if this is the case, then the formal strategy runs up against a limit, and, in order to circumvent the limit, contextual arguments must rely on another approach (intuitionism, constructivism, and utilitarianism) to determine the priority between conflicting principles.

Interestingly, David Miller's approach to the priority problem includes a non-contextual set of priority rules (he calls them sub-duties), which he derives from our duty to respect human rights. Miller identifies four priority rules when considering conflicts between global and domestic duties.72 The first rule—refraining from infringing on basic human rights—entails that each person's basic human rights are to be given equal consideration irrespective of nationality. For example, ‘the government of a nation whose members are starving would not be justified in seizing resources from another nation if this meant that some of that nation's members would fall below the threshold for adequate nutrition.’73 Not unlike Rawls, Miller thinks cooperative activities among one group are constrained by a criterion identifying illegitimate externalities. In this case, the priority rule coordinates pure procedural and transitive contexts, giving priority to the latter. This is an intuitively plausible answer, but it is not a contextual answer. Miller does not justify basic human rights by reference to an interdependent relationship between facts and principles.74 Rather, he advances a humanitarian strategy along the lines of the first strategy considered and dismissed at the outset of this paper—the stand-alone justification of context. He grounds basic human rights on ‘facts’ about what humans ‘need.’75 This creates a number of philosophical challenges, but it also suggests a departure from contextualism in at least this case. Interestingly, Miller does not ‘claim that there are no context-independent principles of justice… [and] it may be that contextualism alone can provide no decisive argument for prioritizing one context and its associated principles.’76 But if this is so, where does it leave contextualism and its unique perspective concerning the relation between facts and principles? Why not suppose, as some philosophers suggest, that context-independent principles are logically prior to, and ground the justification of, context-dependent principles?77

CONCLUSION

Such questions take us well beyond the scope of this paper, which already raises tough questions for critical contextualism. Nevertheless, recalling a point made earlier, I think critical contextualism offers meaningful insights into normative political theory. I will close this paper by briefly underscoring the insights that, I think, move normative political theory in a productive direction.

The first is that critical contextualism tries to identify a role for factual considerations in the defense of principles. It does this by taking a piecemeal approach to theory, one that treats diverse political issues in terms of their unique contextual features rather than systematically as part of—and answered in terms of—an overarching theory of justice. I think this piecemeal approach to normative theory is promising because it is connected to a methodological approach that covers all normative issues. On this point, Miller again converges with Rawls, who took an increasingly piecemeal approach in his later writings, noting:

Each time [the subject changes] the constructivist procedure is modified to fit the subject in question. In due course all the main principles are on hand, including those needed for the various political duties and obligations of individuals and associations. Thus, a constructivist liberal doctrine is universal in its reach once it is extended to give principles for all politically relevant subjects.78

Critical contextualism, like Rawlsian constructivism, is methodologically universal in its reach. It specifies a systematic approach for addressing diverse normative issues. It does not identify a set of universal principles in terms of which diverse subjects can be assessed. Instead, each subject is taken up separately and assessed in terms of its contextual features.

This methodological approach helps explain the persistence of moral pluralism without invoking mysterious intuitions or surrendering to blind relativism. This is the second important feature of contextualism, namely, that the structure of justification explains the existence of rival principles in terms of ordinary factual circumstances that can be rationally analyzed. Different contextual circumstances require different principles of justice. Analyses of the context can help determine which principle is appropriate to which context. In the case of the formal strategy, although the analyses cannot resolve all conflicts—such as conflicts between substantive principles sharing the same form or conflicts between contexts—they nevertheless might resolve many conflicts. Moreover, for those conflicts left unresolved, the formal strategy at least provides a reason for why certain disputes cannot be resolved to fewer than several plausible rivals.

In virtue of rationally explaining pluralism, contextualism provides an account of what a correct normative judgment looks like. Correct normative judgments reflect a relational property, rather than a non-natural or supervenient property.79 This is the third important feature. With respect to the formal strategy, the relational property obtains between a normative directive and the manner in which claims to resources originate. The relation admits of an underlying logic whereby the appropriate principle is the one that answers the claim. In order to answer the claim, the structure of the principle must reflect the structure of the association within which the claim to resources originates. When these structures match, normative judgments based on the appropriation principle are correct; they reflect the right form.

The logical relation between claims and principles has no practical import without either the actual facts that give rise to claims or the particular principles that help address them. The fourth and final important feature of contextualism is that real world facts help shape and shift our moral horizon. Following Rawls, I referred to this as the limits of the practicably possible, and understood these limits as providing both a ceiling beyond which our moral ideas represent sheer fancy and a floor below which factual considerations represent mere expedience. Some facts are morally relevant, and the challenge is to discern which facts. I think a plausible response to this challenge takes a piecemeal approach to normative theory, where issues are framed in terms of their (formal) contextual features. This returns us to the first insight noted previously.

Unfortunately, none of the strategies canvassed in this paper fully realize all four features. As a result, these insights seem to go unfulfilled. In light of this, those who find Miller's general view of contextualism appealing might develop its features by contextualizing an already existing approach rather than further developing critical contextualism. In my view, the leading candidate is Rawlsian constructivism. I noted earlier in this paper that, in response to a variety of reasons for thinking objective characterizations of contexts impossible, one might specify interpretations of contexts in terms of political problems in need of rectification. Certain elements of Rawlsian constructivism can be developed along these lines.80 As the other two strategies for addressing these arguments—reflected in Miller's manner of proceeding—seem to fail, a contextualized constructivism could very well be the best strategy for realizing the four aforementioned insights within one distinct theoretical approach.

NOTES

1. John Rawls, A Theory of Justice, rev. ed. (Cambridge: Harvard University Press, 1999), 398.

2. A notable exception is G.A. Cohen, who argues that fundamental principles of justice are fact-insensitive. G.A. Cohen, Rescuing Justice and Equality (Cambridge: Harvard University Press, 2008).

3. David Miller, ‘Two Ways to Think About Justice.’ Politics, Philosophy & Economics 1, no. 5 (2002): 5–28; David Miller, Principles of Social Justice (Oxford: Oxford University Press, 1999), Chapter 2; David Miller, ‘Political Philosophy for Earthlings,’ in Political Theory: Methods and Approaches, eds. David Leopold and Marc Stears (Oxford: Oxford University Press, 2008).

4. Miller, ‘Two Ways to Think About Justice,’ 10.

5. The ‘contextualism’ advanced by Miller and examined in this paper is different from what goes under the name ‘moral contextualism,’ which refers to a semantic thesis about truth values of moral attribution being relative to the context of the attributor. See John Greco, ‘What's Wrong with Moral Contextualism.’ Philosophical Quarterly 58, no. 232 (2008): 416–36. Berit Brogaard, ‘Moral Contextualism and Moral Relativism.’ both in Philosophical Quarterly 58, no. 232 (2008): 385–409.

6. Miller, ‘Two Ways to Think About Justice,’ 19, 13.

7. David Miller, National Responsibilities and Global Justice (Oxford: Oxford University Press, 2007), 14.

8. Pogge writes, ‘Miller and I may converge then upon an intermediate view—critical contextualism—defined by the rejection of monistic universalism on the one hand and dogmatic contextualism on the other. We both envision different fundamental moral principles applying in different contexts, and we both seek justifications for the delimitation of contexts and the formulation of fundamental principles appropriate to them.’ Thomas Pogge, World Poverty and Human Rights, 2nd ed. (Cambridge: Polity, 2008), 110.

9. Miller, ‘Two Ways to Think About Justice,’ 20.

10. Ibid., 25.

11. John Rawls, ‘The Law of Peoples,’ in Collected Papers, ed. Samuel Freeman (Cambridge: Harvard University Press, 1999), 533.

12. Miller, ‘Two Ways to Think About Justice,’ 12.

13. Ibid., 10.

14. Michael Walzer, Spheres of Justice (New York: Basic Books, 1983), 6–7.

15. Walzer, Spheres of Justice, 8.

16. Michael Walzer, The Company of Critics: Social Criticism and Political Commitment in the Twentieth Century (New York: Basic Books, 1988), 232, 230.

17. Brian Barry, ‘Social Criticism and Political Philosophy.’ Philosophy and Public Affairs 19, no. 4 (1990): 360–73.

18. Walzer thinks outsiders can legitimately assess the fit between a government and its people in at least some circumstances. Michael Walzer, ‘The Moral Standing of States: A Response to Four Critics.’ Philosophy and Public Affairs 9, no. 3 (1980): 209–29.

19. Miller, ‘Two Ways to Think About Justice,’ 10–12. Miller, Principles of Social Justice, 25. Moreover, Walzer does not always fit my depiction. Elster notes, ‘[a]t time, one gets the impression that he [Walzer] advocates a principle of goods-specificity according to which the nature of the good requires a certain principles of distribution … At other times, it is more natural to read him as proposing a principle of country-specificity, according to which the just allocation depends on the meaning of the good in the society in question.’ Jon Elster, Local Justice: How Institutions Allocate Scarce Goods and Necessary Burdens (New York: Russell Sage Foundation, 1992).

20. Miller, ‘Two Ways to Think About Justice,’ 25.

21. Pogge, World Poverty and Human Rights, 110.

22. This represents the first of Rawls's ‘burdens of judgment.’ John Rawls, Political Liberalism, 2nd ed. (New York: Columbia University Press), 55–6.

23. For some of the literature representing these competing answers, see Mathias Risse, ‘What We Owe the Global Poor.’ Journal of Ethics 9 (2005): 81–117; Thomas Pogge, ‘Eradicating Systemic Poverty: Brief for a Global Resources Dividend,’ and ‘Moral Universalism and Global Economic Justice,’ both in World Poverty and Global Justice, 118–21; 205–7. In addition to the domestic and global institutional theses, there are geographic and environmental theses. For an example of each, see Jared Diamond, Guns, Germs and Steel: The Fates of Human Societies (New York: W.W. Norton & Co., 1999); and Ian Morris, Why the West Rules—For Now (New York: Farrar, Straus and Giroux, 2010).

24. This represents the second of Rawls's ‘burdens of judgment.’ Rawls, Political Liberalism, 55–6.

25. Miller, ‘Two Ways to Think About Justice,’ 20.

26. Rawls, A Theory of Justice, 18; and Norman Daniels, Justice and Justification: Reflective Equilibrium in Theory and Practice (Cambridge: Cambridge University Press, 1996).

27. Rawls, Political Liberalism, 3; Christine Korsgaard ‘Realism and Constructivism in Twentieth-Century Moral Philosophy.’ Journal of Philosophical Research, APA Centennial Supplement (2003), 99–122; Michael Buckley, ‘The Structure of Justification in Political Constructivism.’ Metaphilosophy 41 no. 5 (2010): 669–89; and Sharon Street, ‘What is Constructivism in Ethics and Metaethics.’ Philosophy Compass 5, no. 5 (2010): 363–84.

28. Miller, Principles of Social Justice, 53–60; and Miller, ‘Two Ways to Think About Justice,’ 17–18.

29. Miller, Principles of Social Justice, 35, 43, 65; Miller, ‘Two Ways to Think About Justice,’ 6; Daniel M. Weinstock, ‘Miller on Distributive Justice,’ in Forms of Justice: Critical Perspectives on David Miller's Political Philosophy, eds. Daniel Bell and Avner de-Shalit (Oxford: Rowman and Littlefield Publishers, Inc.), 272.

30. David Miller, ‘A Response,’ in Forms of Justice, eds. Daniel Bell and Avner de-Shalit (Oxford: Rowman and Littlefield Publishers, Inc.), 352. For a longer treatment on this question, see Miller, ‘Political Philosophy for Earthlings.’

31. Miller, Principles of Social Justice, 34–35.

32. Miller, National Responsibilities, 17.

33. Miller, National Responsibilities, 18; and Miller, ‘Political Philosophy for Earthlings.’

34. Rawls, A Theory of Justice, 398.

35. John Rawls, The Law of Peoples (Cambridge: Harvard University Press, 2001), 12. This has not been a widely discussed issue in the secondary literature on Rawls. For some remarks, see Miller, ‘Political Philosophy for Earthlings’; and Mark Jensen, ‘The limits of practical possibility.’ The Journal of Political Philosophy 17, no. 2 (2009): 168–84.

36. Rawls, The Law of Peoples, 12; John Rawls, Justice as Fairness: A Restatement (Cambridge: Harvard University Press, 2001), 4–5.

37. Rawls, Justice as Fairness: A Restatement, 2–5.

38. Jensen, ‘The limits of practical possibility,’ 168–184.

39. Miller, Principles of Social Justice, 51.

40. Miller, ‘Political Philosophy for Earthlings,’ 44.

41. Miller, Principles of Social Justice, 25, 34; and Miller, National Responsibilities, 14–16.

42. Street, ‘What is Constructivism,’ 366.

43. Rawls, Political Liberalism, 121–22.

44. Miller, ‘Political Philosophy for Earthlings,’ 41.

45. David Hume, A Treatise of Human Nature, 2nd ed., L.A. Selby-Bigge (Oxford: The Clarendon Press, 1978), 484–501; H.L.A Hart, The Concept of Law (Oxford: The Clarendon Press, 1961), 189–95; and Rawls, A Theory of Justice, 109–12.

46. Miller, Principles of Social Justice, 65.

47. Alternatively, one might associate this strategy with what T.M. Scanlon calls the ‘descriptive’ interpretation of reflectively equilibrium. T.M. Scanlon, ‘Rawls on Justification’, in The Cambridge Companion to Rawls, ed. Samuel Freeman (Cambridge: Cambridge University Press, 2003), 139–57.

48. Miller, National Responsibilities, 15.

49. Miller, ‘A Response.’ Critical Review of International and Social and Political Philosophy 11, no. 4 (2008): 554.

50. Ibid., 554.

51. Michael Blake, ‘Distributive Justice, State Coercion, and Autonomy.’ Philosophy and Public Affairs 30, no. 3 (2001): 262.

52. Miller, National Responsibilities, 18, and ‘Political Philosophy for Earthlings.’

53. See Adam Swift, ‘Social Justice: Why Does it Matter What the People Think,’ in Forms of Justice: Critical Perspectives on David Miller's Political Philosophy, eds. Daniel Bell and Avner de-Shalit (Oxford: Rowman and Littlefield Publishers, Inc.), 13–28; Brian Barry, ‘Statism and Nationalism: A Cosmopolitan Critique,’ in Global Justice, eds. Ian Shapiro and Lea Brilmayer (New York: NYU Press, 1999), 12–66; Daniel Weinstock, ‘National Partiality: Confronting the Intuitions.’ The Monist 82 (1999): 516–42; Veit Bader, ‘Global Justice in Complex Moral Worlds. Dilemmas of Contextualized Theories,’ Critical Review of International Social and Political Philosophy 11, no. 4 (2008): 539–52.

54. Miller, ‘Two Ways to Think About Justice,’ 25.

55. A common criticism of Rawls's constructivism argues that the constructivist procedure embodies dense and richly articulated yet unjustified moral concepts, and therefore begs important philosophical questions. Some examples include: Thomas Nagel, ‘Rawls on Justice,’ in Reading Rawls: Critical Studies on Rawls’ ‘A Theory of Justice’, ed. Norman Daniels (Stanford: Stanford University Press, 1975); Cohen, Rescuing Justice and Equality; David O. Brink, ‘Rawlsian Constructivism in Moral Theory.’ Canadian Journal of Philosophy 17 (1987): 71–90. Of course, Rawls relies on wide and general reflective equilibrium, but it is doubtful this reliance can avoid the very criticisms leveled against Miller's contextualism.

56. Miller, ‘Two Ways to Think about Justice,’ 13.

57. Rawls, A Theory of Justice, 76–77; and Rawls, Justice as Fairness, 50.

58. David Miller has discussed the problem of assigning remedial responsibilities in ‘Distributing Responsibilities.’ Journal of Political Philosophy 9, no. 4 (2001): 453–71.

59. Rawls, The Law of Peoples, 105–20.

60. Allen Buchanan, ‘Rawls's Law of Peoples: Rules for a Vanished Westphalian World.’ Ethics 110, no. 4 (2000): 697–721; and Thomas Pogge, ‘An Egalitarian Law of Peoples.’ Philosophy & Public Affairs 23, no. 3 (1994): 195–224.

61. Pogge, World Poverty and Human Rights, 182.

62. Although he does not use the same terminology, it is clear that Friedrich von Hayek's argument against social justice is based on the mistaken idea that all principles of social justice are allocative principles, and thus unsuitable for a pure procedural association such as markets. Friedrich von Hayek, Law, Legislation and Liberty: The Mirage of Social Justice, vol. 2 (Chicago, IL: University of Chicago Press, 1978).

63. The Trail Smelter Case is just one of many such cases. In this case, sulfur dioxide emissions from a zinc and lead smelter in Trail, British Columbia, damaged a 30-mile stretch from the international boundary to Kettle Falls, Washington, affecting logging, farming, and cattle grazing industries crucial to the area. Legal arbitration resulted in Canada having to pay the US for damages. United Nations Treaty Collection, ‘Report of International Atribral Awards: Trail Smelter Case,’ 2006. http://untreaty.un.org/cod/riaa/cases/vol_III/1905-1982.pdf (accessed December 15, 2011).

64. Kevin Watkins, United Nations Development Program Human Development Report 2006: Beyond Scarcity: Power, Poverty and the Global Water Crisis (New York: Palgrave Macmillan, 2006): 165.

65. Bates, B.C., Z.W. Kundzewicz, S. Wu, and J.P. Palutikof, (eds). ‘Climate Change and Water: Implications for Policy and Sustainable Development,’ 129; Intergovernmental Panel on Climate Change. Technical Paper VI, 2008. http://www.ipcc.ch/pdf/technical-papers/ccw/chapter7.pdf (accessed December 15, 2011).

66. The IPCC notes that ‘[d]etection and attribution studies consistently find evidence for an anthropogenic signal in the climate record of the last 35 to 50 years,’ and that ‘warming over several decades has been linked to changes in the large-scale hydrological cycle such as: increasing atmospheric water vapour content; changing precipitation patterns, intensity and extremes; reduced snow cover and widespread melting of ice.’ ‘Summary for Policymakers of the Synthesis Report of the IPCC Fourth Assessment Report,’ chap. 2. http://www.ipcc.ch/pdf/assessment-report/ar4/syr/ar4_sry.pdf (accessed February 1, 2012) and ‘Climate Change and Water: Executive Summary,’ 3; Intergovernmental Panel on Climate Change, ‘Climate Change Synthesis Report: Summary,’ 2007. http://www.ipcc.ch/pdf/technical-papers/ccw/executive-summary.pdf (accessed December 15, 2011).

67. Miller, ‘Two Ways to Think About Justice,’ 13, 19.

68. Ibid., 20

69. Rawls, A Theory of Justice, 263–5.

70. Ibid., 263, 266.

71. Ibid., 30–40.

72. Miller, National Responsibilities, 47.

73. Ibid., 48.

74. For a critical look at Miller's account of human rights, see Leif Werner, ‘Human Rights and Equality in the Work of David Miller.’ Critical Review of International and Social and Political Philosophy 11, no. 4 (2008): 401–11.

75. Miller, National Responsibilities, 178–200.

76. Miller, ‘Two Ways of Thinking about Justice,’ 24–5, 21.

77. Cohen, Rescuing Justice and Equality.

78. Rawls, ‘The Law of Peoples,’ 532.

79. For a relational account of moral objectivity, see Nicholas Rescher, ‘Moral Objectivity.’ Social Philosophy and Policy 25, no.1 (2008): 393–409.

80. Buckley, ‘The Structure of Justification’; and Street, ‘What is Constructivism.’

About The Author

Michael Buckley
Lehman College, City University of New York
United States

Assistant Professor

Philosophy Department