Abstract
An increasingly popular approach to global justice claims we should be ‘integrationist,’ where integrationism represents an attempt to unify our theorising between different domains of global politics. These political theorists have argued that we cannot identify plausible principles in one domain, such as climate justice, which are not sensitive to general moral concerns. This paper argues we ought to reject the concept of integrationism. It shows that integrationism is either trivial, or it obscures relevant disagreement by ignoring the distinctive methodological and substantive commitments held by its opponents. The paper then argues that the relevant disagreement is actually about the role of practices for political philosophy and, as such, should be framed in terms of the distinction between practice-dependent and practice-independent theory. Finally, I provide my own account of that distinction, identifying a practice-dependent claim that those concerned about the narrowness of prominent accounts of global justice should target.
When questions about global justice arise, they often do so in relation to specific political issues. This can lead us to perceive them as isolated, raising problems which are distinctive to a particular domain and which, as a result, ought to be addressed on their own terms. Climate change can seem isolated in this way, and some philosophers have thus devoted attention to questions of climate justice which arise apparently separately to questions about justice in global trade, say, or in migration. At the same time, though, we recognise that these purportedly distinct moral issues interact in complex ways: we know that climate change is in part driven by the production of goods for global trade, and we know its effects will prompt new claims for international migration. Recently, the idea that we can bracket off different political issues, treating them as separate spheres of global justice, has come under scrutiny. More specifically, a number of theorists have pressed on the idea that we can identify (plausible) principles that are isolated from broader concerns of justice. These theorists claim that we should instead be ‘integrationist’, where integrationism represents some kind of attempt to unify our theorising between global domains.
The debate about integrationism is an important one. If its proponents are right, theorists of global justice, by training their sights too narrowly, may have been arriving at incorrect principles. Although we will likely be concerned by the level of injustice that exists in global politics, perhaps we have mischaracterised it and misplaced our condemnation. There might be rather more injustice than we thought, or perhaps rather less. The critique from integrationism therefore offers a potentially revisionary perspective on these discussions. This perspective also represents a new frontier of the burgeoning debate about methodology in political philosophy. As we will see, proponents often frame their arguments in methodological terms, casting their view as making space for a new approach to global justice. Debates about methodology are important: they help us approach theorising more systematically and help us see where, and why, different perspectives depart.
This paper has two main aims. First, I show that integrationism is an unhelpful concept in debates about global justice. On one reading it is trivial, and so does not reveal a meaningful point of disagreement. On its other readings it smuggles in controversial substantive or methodological commitments, and so actually obscures genuine disagreement. Integrationism does not shed light on debates about global justice, so I will suggest we ought to reject it. Second, I will develop an alternative diagnosis of the disagreement between integrationism and the view against which it is counterposed, sometimes labelled ‘the internal approach' (Walton, 2020) or, as I will refer to it, ‘isolationism’ (Caney, 2012; 2018a). Importantly, my rejection of integrationism does not signal an affinity with isolationism; it is rather a rejection of the way in which the two views have been distinguished. My claim is that viewing this disagreement as one between practice-dependent and practice-independent theorists (Sangiovanni, 2008: 138) is a more productive way of framing what is at stake. But this distinction too needs some clarifying. I will further argue, against some alternative candidates, that the specific site of disagreement is a claim about the appropriate role of the interpretation of a social practice for political philosophy. By the end, I hope we will be left with a clearer picture of the conceptual terrain available to theorists of global justice. 1
The argument proceeds as follows. The first section will introduce integrationism in its general form. I will show that this integrationism is trivial, because it encompasses all reputable views about global justice. The second section will consider and reject Simon Caney's proposal, which I call methodological integrationism. His account does not follow from his arguments for it, which are not, as they are claimed to be, neutral between different substantive theories of distributive justice. The third section will consider a recent account of integrationism offered by Andrew Walton. I will argue that his substantive integrationism fails also, for related reasons: his argument does not appropriately engage with the distinctive methodological commitments held by his opponents. In the fourth section, I recast the disagreement. I argue that integrationists should focus directly on the question about how practices can matter for political philosophy, since it is likely these assumptions which lead some theories to have an objectionably narrow outlook. I consider three possibilities, concluding that only the claim about the necessity of contextual interpretation need be denied by the practice-independent theorist
The general reading of integrationism
The term ‘Integrationism’ was introduced by Simon Caney (2012) in an influential paper on the right to emit greenhouse gases.
2
He defines integrationism as the approach which treats climate responsibilities ‘in light of’ or in ‘conjunction with’ broader concerns about global justice (Caney, 2012: 259). In a recent article on trade justice, Andrew Walton (2020) echoes this general statement.
3
‘The essence of the integrationist approach’, he tells us, ‘is that judgments about… actor's claims to a share of the gains and losses of trade must be sensitive to their advantages and burdens in other domains’ (Walton, 2020: 54). From these brief remarks, we can state integrationism in its most general form: Integrationism: an agent’s claim to a benefit, or responsibility for a burden, in one global domain should be sensitive to broader concerns of distributive justice.
On this view, the benefits and burdens of global trade, for example, should be sensitive to the distribution of responsibilities for mitigating and adapting to climate change.
Stated generally, integrationism is so inclusive it is trivial. There are a range of ways in which an account might display a sensitivity to broader concerns about global justice. All who are committed to a basic rights constraint, or some other moral minimum, are integrationists in these terms. This is because in order to satisfy the constraint, we need to know whether benefits and burdens across domains combine in a way which leaves a particular agent below the minimal threshold. If they do, that agent should not be further burdened in the context of global justice under consideration. 4 But almost all contemporary theorists of global justice would endorse this constraint, which means that integrationism so understood does not reflect a meaningful distinction between views. 5
Consider first debates about climate justice. In Caney's view, isolationism has been the norm in this literature, which has been primarily concerned with allocating climate change mitigation and adaptation burdens. He claims that ‘[v]ery many philosophers have adopted an isolationist approach’ to this problem (Caney, 2018a: 678). It is true that by endorsing principles, such as Emissions Egalitarianism (EE) (Baatz and Ott, 2017) or the Polluter Pays Principle (PPP), 6 which refer only to internal features of the problem at hand, many theorists opt for what at first glance appear to be isolationist responses. But at the same time, there is what Stephen Gardiner (2017: 445) describes as a ‘ubiquitous’ ‘concern for the global poor in the climate ethics literature’, where it is assumed that climate justice must not thwart poverty alleviation. Thus it has become convention for theorists to constrain the PPP, or its rivals, in cases where the burdens flowing from the principle would fall on states below the poverty threshold (e.g. Page, 2008; Miller, 2009; see also, Shue, 1993). Indeed, it would be very difficult to find an account that would deny such a constraint – we might go further and say that a central aim of the climate justice literature has been to promote broader concerns of global justice by shielding the poorest states from the costs of mitigation and adaptation. Even when EE accounts do not explicitly affirm a basic rights constraint, they arguably have this character, because it is assumed that they will burden only rich states that currently emit more than their equal share and have done so in the past. On this general reading of integrationism, then, it seems that the (supposedly isolationist) climate justice literature falls under its ambit.
A slightly more complicated example is the literature on global trade. Walton claims that prominent accounts, such as those provided by Mathias Risse and Aaron James, are isolationist. Would they deny the basic rights constraint that would render their accounts integrationist, at least in the thin sense currently under consideration? The reason trade poses a complication is that it is often assumed to involve primarily mutual benefit: questions about global trade justice are therefore taken to be questions about the distribution of relative gains rather than burdens. If we assume trade does not entail net costs for different states, then it appears we will never run up against the basic rights constraint. That said, there is evidence that both James and Risse would qualify as integrationists.
James (2014: 181), for example, endorses a principle, International Relative Gains (IRG), which holds that trade benefits should be split evenly between states, except when ‘greater gains flow (e.g. via special trade privileges) to poor countries’. This principle seems to signal that the benefits of trade should be sensitive to broader concerns of global justice. Risse (2012: 216), meanwhile, takes his project to be one of determining ‘what follows about trade from… human rights–oriented principles’, before establishing whether any ‘new and distinct’ principles of justice emerge from the particular practice of global trade. Again, the first part of this aim clearly implies an integrationist constraint, sensitive to background disadvantage, on what the further trade-specific principles can demand.
The integrationist might offer two responses to my accusation of triviality. First, they might say that integrationism is a reaction to accounts that state their positions in isolationist terms. As Walton notes, for example, James frames his account by claiming that the normative features of trade can ‘generate… substantial fundamental principles in …[their] own right, without relying on moral claims’ deriving from other issues or more general commitments (James, 2014: 179). It is true that integrationists have identified a problem here, and the view may be non-trivial in the sense that it intervenes in debates about global justice by highlighting an inconsistency between how accounts are framed and their ultimate substance. Nonetheless, if integrationism is intended as a positive position, it should offer something other accounts do not. This is the sense of ‘triviality’ at stake in my claim, and given the difficulty of finding an account that denies a basic rights constraint, the charge remains valid.
Second, the integrationist might respond that the view discussed in this section is too thin. A thicker account of integrationism, they might say, would successfully locate a disagreement between views. The next two sections will consider and reject two such attempts.
Methodological integrationism
Though he initially defines integrationism in trivial terms, Caney later arrives at a more specific version. Integrationism, he argues, is the methodological approach we should adopt after we have seen the problems with EE, the view that people should have an equal right to emit greenhouse gases. Caney's criticisms of EE are important, but integrationism does not follow from them. Given Caney uses his critique of EE as a platform for his methodological integrationism, it will be necessary to state the key moves in his rejection of the view. His core complaint is that EE is too narrow in its focus, in effect treating emissions as separate from general considerations of distributive justice. He gives this concern theoretical substance in two main steps.
First, Caney (2012: 264–272) shows that EE cannot find support from any egalitarian theories of distributive justice. 7 Though he canvasses a range of examples, the reason for this is the same in each case: theories of equality advance a currency of justice, which can be framed in terms of a ‘total package’ of goods (Caney, 2012: e.g., 271), and they do not generally care about equal holdings of any one resource in particular. This is obviously true of ‘orthodox’ (Caney, 2012: 265; Bell, 2008: 248–249) egalitarian theories, which care about relative advantage broadly construed, but it is also true of views which are more restrictive, only seeking equality across natural resources. Even these accounts still strive for equality across more than just the resource of ‘emissions’ (Caney, 2012: 266–272). 8 Second, Caney (2012: 272–283) rejects the notion that emissions are the sort of good which would make for an exception to this tendency, warranting its own distributive principle. While he admits that such exceptions exist, he denies that emissions qualify, largely because they can be substituted with other resources that provide energy with no loss to the end user (Caney, 2012: 283–291).
For our purposes, what is important is not the plausibility of this critique of EE but rather how it is leveraged in articulating methodological integrationism. In moving to his positive position, Caney (2012: 299) stresses the importance of locating greenhouse gas emissions in a ‘more general theory of justice’, something which EE fails to do. The theory will then ‘bring in its wake a certain distribution of greenhouse gas emissions’ (Caney, 2012: 299). Caney's central claim is thus to insist on a coherence in our theorising about distributive justice between the general level and the applied: an agent will either be entitled to a good or not, and this will be determined by a general theory of distributive justice. We can state this as a conceptual claim about distributive justice. The Conceptual Claim: Whether an agent is entitled to good A will be determined by a general theory of distributive justice, unless A is exceptional in a way that emissions are not.
This claim captures his criticisms of EE. EE is not proposed as a complete theory of distributive justice, it does not follow from one and it does not make for a plausible exceptional case.
Caney's methodological integrationism proposes a five-step procedure (Caney, 2012: 291–300) as a way of expressing The Conceptual Claim. The idea is that we should start with our general theory of global and intergenerational justice (step 1), and then, while ensuring that this account makes sustainable demands on the natural world (step 2), move towards political practice (steps 3–4), finally allowing a participatory process to give content to an agent's entitlement to greenhouse gas emissions (step 5). Although the procedure is applied to emissions, Caney's arguments are clearly general; an adjusted version will also apply to other resources.
With these details in hand, we can identify two main problems with Caney's methodological integrationism. If correct, these problems not only show that Caney's integrationism is limited on its own terms, but also that it obscures rather than illuminates disagreement about global justice. Indeed, I will argue in Section IV that the assumptions Caney makes here about the (ir)relevance of practices for distributive justice are better viewed as what is at stake between integrationists and isolationists.
Practice dependence and methodological integrationism
The first problem with Caney's methodological integrationism is that his five-step procedure relies on a view about the relationship between distributive justice and political practice that is not motivated by his arguments against EE. Not only does he fail to provide an argument connecting his procedure to The Conceptual Claim around which his opposition to EE is organised, but he claims that none is necessary.
For Caney, his five-step procedure is a methodological intervention that remains neutral on substantive questions of distributive justice. His argument, which I have expressed as The Conceptual Claim, is that we should endorse a general theory of distributive justice rather than a theory for a specific good. He recognises that people will disagree about the content of that theory of distributive justice, but he sees no reason why they should disagree about how to move from that theory to an entitlement to a specific good, given that this is primarily a matter of resolving contingent details about the world we live in. To take one clear example, though there are many, he claims: ‘I have not provided a substantive answer to the question… [of emission rights], but, building on my criticisms of the equal per capita view, I have outlined the method by which one can ascertain what would constitute a fair distribution of greenhouse gas emissions’ (Caney, 2012: 300, emphasis original). 9
The five-step procedure, however, is not neutral between different substantive views about distributive justice. Crucially, Caney's procedure moves from a pre-articulated theory into a specific real-world context involving the distribution of goods and bads; a theory, on this view, is worked out prior to questions about benefits and burdens as they arise in practice. As a result, any view which describes itself as practice-dependent will be hostile to methodological integrationism. Andrea Sangiovanni (2008: 138) famously defines practice-dependence as follows. Practice-dependence Thesis: The content, scope, and justification of a conception of justice depends on the structure and form of the practices that the conception is intended to govern.
As we can see, this view holds that the content of principles of distributive justice are tied to political practices, and so denies that we can move from an abstract theory to a political practice, for this practice is integral for the articulation of the theory in the first place.
Practice-dependent theorists claim their view is a way of recognising the importance and distinctiveness of the different practices we are implicated in (see e.g. James, 2005a; Ronzoni, 2009; Sanngiovanni, 2013; Valentini, 2011a). In contrast to Caney's procedure, they begin the theoretical process by interpreting the point and purpose of a particular practice, as a way of uncovering its underlying normative structure. Importantly, though, practice-dependent theorists can still endorse The Conceptual Claim: they can claim that whether or not an agent is entitled to a specific resource will be determined by their general theory. More strongly, given their Rawlsian heritage (see James, 2005b) I think that practice-dependent theorists would endorse The Conceptual Claim. They will tend to view their principles as applying to the institutions that confer advantages in a given domain, not directly to the holdings of individual resources that will follow from these principles. 10 Practice-dependent theorists just deny that these general theories are invariant across different spheres of human activity.
The existence of the conceptual space practice-dependence occupies is sufficient to undermine Caney's methodological integrationism. Practice-dependent views can endorse The Conceptual Claim but reject the five-step procedure that is supposed to follow from it for any view about distributive justice. It is important to add that I am not endorsing practice-dependence. The point is that these views cannot be ruled out by fiat. As we will see, Caney has elsewhere addressed practice-dependence, but his account of integrationism relies on over-simplified picture of the relevant methodological terrain. In any case, my second criticism of integrationism suggests that many practice-independent theorists also ought to reject methodological integrationism.
Pluralism and the conceptual claim
My first criticism attacked methodological integrationism by severing the link between The Conceptual Claim and Caney's five-step procedure. But this assumes we should endorse The Conceptual Claim, an assumption I want to exert some pressure on now. To start with, note that The Conceptual Claim posits a relation between a theory of distributive justice and a particular resource. But this relationship might be more or less direct. A direct view holds that a theory of distributive justice tells us who ought to be in possession of the particular resource. An indirect view holds that a theory of justice will provide part of the answer to the question of who should ultimately be in possession of a particular resource. The indirect view might, for example, be couched in terms of a pro tanto claim to the resource in question (e.g. Armstrong, 2017). These two views differ in the strength they attribute to distributive justice when it comes to determining actual resource holdings. 11
Caney appears to hold the direct view, which explains my use of the word ‘determined’ in The Conceptual Claim. First, he deploys the language of ‘entitlement’ throughout, which suggests he thinks people can demand, in the strong sense, that they should have access to the resource in question (see also, Caney, 2005: 104). Second, and more importantly, Caney's five-step procedure is offered as a way of translating our theory of justice into actual holdings of emissions. As he puts it, his aim is to provide ‘specific outcomes’ that ‘realize people's entitlements’ (Caney, 2012: 298).
Caney, then, seems to think there is a direct relationship between a theory of distributive justice and a particular resource like greenhouse gas emissions. In my view, however, most contemporary egalitarians – including those who share his commitment to practice-independence – will reject this picture. They will hold that who ought to possess a particular resource, bear a particular burden or enjoy a particular benefit is an all-things-considered judgment which will reflect a number of further, non-distributive considerations. A distribution of emissions might contain injustice, by egalitarian lights, but nonetheless be judged all-things-considered permissible by virtue of the way it trades different values against one another. 12
To put the point formally, egalitarians are generally pluralists about value and do not claim that equality is all that matters from a moral point of view. And pluralist egalitarians will not endorse The Conceptual Claim, insofar as I am right about it expressing the strong view about distributive justice, because it will preclude them from holding that other moral considerations are relevant for determining the quantity of emissions an agent ought to have access to. As such, the procedure Caney claims is neutral between views of distributive justice would also exclude many contemporary egalitarians. Given the significance of this point, it is important to be sure that Caney is denying pluralism. Upon closer inspection of his procedure, we might doubt he is.
Specifically, we might think that Step 5, Indeterminacy and the Role of Institutions, incorporates a procedural value, such as that of legitimacy, into the procedure. Here, Caney (2012: 298) makes room for ‘political processes… in which the participants can select from the various different combinations available and agree among themselves precisely how people's entitlements are secured’. 13 I am sceptical that Caney had pluralism in mind when articulating this step. But even if he did, this would build further unargued for substantive commitments into the procedure, relating now to non-distributive considerations. Specifically, it would suggest that political processes have a role in the regulation of greenhouse gases only as a means for choosing between different available options for realising agents’ entitlements, as worked out by a theory of distributive justice. This would assert a priority for distributive over procedural justice that egalitarians need not accept. Further, unless we can find other non-distributive moral considerations in Caney's procedure, 14 it would suggest that he thinks legitimacy is the only other value consideration that has a bearing on emissions rights. Caney has not given us reason to accept this claim, which strikes me as implausible; but anyway, to the extent that it is not obviously true, Caney's procedure obscures disagreement about the relevance of practices for global justice.
Substantive integrationism
In the previous section I argued that Caney has not offered a plausible but non-trivial account of integrationism. Walton has recently provided a version of integrationism which, he claims, departs from Caney's. He labels this view substantive integrationism. Walton raises some important criticisms against isolationism, and I will suggest later that they can be separated from his positive view. As with Caney, however, I think Walton's integrationism mislocates the relevant disagreement and, as a result, fails to properly address his opponents.
In order to develop substantive integrationism, Walton presents the following case. Substantively Unequal Trade: Country A and Country B agree to participate in a structured trade system creating gains that would not exist without the arrangement…. These gains are divided such that Country A receives 90 per cent and Country B receives 10 per cent. In addition, because the new arrangements lead to restructuring domestic economies, some individuals in Country A have more income than before while some individuals in Country B have less than before. (Walton, 2020: 57)
Walton’s claim is that we cannot come to a plausible judgement about this case without further information. Specifically, it makes a difference whether the relative losers (in Country B) are ‘wealthy farmers previously supported by agricultural subsidies’ or ‘workers in fledgling industries in developing countries’ (Walton, 2020: 58). If it is the latter, the exchange appears problematic; but it is not clear that we should object if the wealthy farmers are set to lose out from the trading arrangement, despite the significantly unequal outcome. For Walton, we need to know the background levels of (dis)advantage before we can come to a plausible judgment. He further claims that such information is unavailable to the isolationist, who is ‘pressed to hold that the loss is (pro tanto) similarly problematic whether they are advantaged farmers in wealthy states or fledgling producers in developing countries’ (Walton, 202: 63). As a corrective, Walton (2020: 59) claims we should endorse substantive integrationism, which holds that an agent's general level of advantage plays a constitutive role in our claims to the benefits and burdens in a particular domain. Substantive integrationism would therefore hold, for instance, that an agent who is disadvantaged would have a stronger claim to the benefits of global trade, or that an advantaged agent would have a weaker claim to the benefits of emissions.
Walton is right to say that it makes a difference in Substantively Unequal Trade whether the losers are subsidised farmers in a developing state or workers in fledgling industries in a developing one. Our question, though, is whether his argument motivates a distinctively integrationist approach. Two considerations suggest not. First, substantive integrationism appears trivial in the sense expounded in Section I. In order to know whether a particular distribution of benefits and burdens violates a basic rights constraint, we need to know the background levels of advantage of the different agents involved.
Second, consider again practice-dependence. It is not true that the practice-dependent theorist is compelled to arrive at an implausible judgement in Substantively Unequal Trade based only on the unequal outcome of the trade, but nor is it clear that to avoid this fate they must appeal to background advantage. Practice-dependent theorists aim to produce principles which regulate particular practices. These principles draw on the aims of a practice and on the normative commitments that inhere in them; they are not developed by testing our intuitions about particular cases. Rather, when confronted with this case, the question the practice-dependent theorist will ask is ‘does this outcome distribution follow from principles which appropriately cohere with a normative reconstruction of the given practice?’. If it does, they will rule the outcome permissible; if it does not, they will object to it. But this does not entail a direct appeal to background advantage. Walton's case, then, does not establish a case for an integrationist over an isolationist approach. To do this he would need to show that the plausible judgement in Substantively Unequal Trade – that it is not objectionable when the losers are subsidised farmers in developed states – cannot be arrived at via a practice-dependent approach.
Walton does attempt to develop an argument to this effect. He claims that while James’ practice-dependent view arrives at plausible conclusions in cases like Substantively Unequal Trade, it does so by illicitly appealing to considerations external to the practice of global trade. Walton's (2020: 64) argument is that James – and, we can assume, isolationists in general - should advocate a principle of strict equality, regardless of the relative levels of advantage
For instance, James (2012: 224; see also James, 2017) explains that when we view global trade as a series of exchanges ‘embedded within a cooperative practice of market reliance that extends indefinitely into the future and substantially shapes people's whole life prospects, especially the world's poor’, we will attach a greater priority to the needs of the disadvantaged, as a way of recognising their more urgent development priorities. In James’ view we should understand this as a principle of fairness appropriate for that specific context, and we need not construe it in terms of a general principle of global justice aimed at alleviating background disadvantage. Perhaps this is implausible. But the point is that unless Walton shows this, Substantively Unequal Trade does not motivate a distinctly integrationist account, since the isolationist will respond that they can arrive at the desired conclusion without a direct appeal to disadvantage. As we will see, some of Walton's criticisms of practice-dependence are powerful; but, like Caney, his positive account of integrationism is not sufficiently sensitive in its handling of the distinctive methodological commitments at stake in the debate.
This concludes my critique of integrationism. I do not think its main incarnations are successful. Moreover, I have argued that proponents of the view face a dilemma. Either integrationism is plausible, but it is also trivial; or integrationism is not trivial, but the integrationism/isolationism distinction obscures, rather than illuminates, relevant disagreement between views.
Recasting the disagreement
Integrationists are concerned with the way contemporary theorists of global justice tend to bracket off problems in global politics and treat them as isolated. The theories that emerge are, according to integrationists, too narrow, and they overlook how interconnected political issues raise the same general moral concerns. But I have shown that integrationism fails to recognise the distinctive methodological significance attributed by many to political practices. As such, my proposal is that we focus our disagreement on the distinction between practice-dependence and practice-independence. Although this distinction is clearly related to that between integrationism and isolationism, the disagreement about the status of political practices in political philosophy seems more fundamental, and locating the disagreement at this deeper level will help avoid the sort of confusion outlined above. In my view, we should be sceptical of practice-dependence. Indeed, I think Caney and Walton have raised some powerful objections to the view, but these are muted rather than amplified by their focus on integrationism.
In order to exert pressure on practice-dependent views, we need to get clearer about what is distinctive about them. Recent contributions to the literature have revealed this to be more complicated than previously thought (Sangiovani, 2016; cf. Erman and Möller, 2019). As we have seen, practice-dependence is standardly understood in terms of the claim that the ‘content, scope and justification of a conception of justice’ depends on the nature of the specific practice it is intended to regulate. This specification is too vague: there are a range of ways in which a principle or conception of justice might ‘depend’ on the nature of a particular practice (see Brännmark, 2019; Caney, 2018b; Jubb, 2016: 82–89). In this section I will assess three candidate claims that might be thought to be distinctly practice-dependent. I will show, however, that the practice-independent theorist need only reject one of these, specifically, the claim which holds that a principle of justice should be derived from an interpretation of the ‘point and purpose’ of the practice it is intended to regulate.
While we refine our understanding of the practice-dependent view, I will work from the assumption that ‘nonrelational global egalitarians’ in debates about global justice are practice-independent theorists. 15 A nonrelational global egalitarian position is one which claims that people's common humanity alone gives us reason to constrain global distributive inequalities. 16 If a characterisation of practice-dependence cannot exclude nonrelational global egalitarians, then I claim it fails to locate an interesting disagreement between views. 17 This should not be a controversial assumption: nonrelational global egalitarians are cast (Jubb, 2016: 78; Sangiovanni, 2008: 147–149), and self-identify as (Caney, 2018b: 109), opponents of practice-dependence, and it is hard to see how practice-dependence would mark out anything distinctive if it did not exclude these views. Despite reservations, I will continue to use the term ‘practice-independent’ as the antonym of practice-dependent. The below discussion, however, will reveal that this is possibly not the best label, since practice-independent theorists, with the current stipulation in hand, are only practice-independent in quite a restrictive sense.
Justification
First, it is important to see that it is not enough to appeal to the simple claim that for practice-independent theorists practices are irrelevant for the justification of a principle. Practice-dependent and practice-independent theorists do disagree about how practices should feature in the justification of a principle, but the idea that practice-independent theorists justify principles ‘entirely in the abstract’ (James, 2012: 27) is a caricature.
Sangiovanni (2016: 7–15; see also Jubb, 2009) has recently emphasised a helpful distinction here between the grounds of a principle and its justification. The grounds of a principle ‘tells us in virtue of what a principle holds, or explains why it holds’, whereas the justification of a principle ‘is an epistemic relation between two or more beliefs about [it]’ (Sangiovanni, 2016: 11, original emphasis). Though nonrelational egalitarians will claim that distributive equality is grounded in moral personhood and not in facts about specific political practices, the justification for this principle will tend to do more than appeal to its grounds. In particular, global egalitarians often appeal to intuitions or considered convictions, and these can be held in relation to a judgement about a practice. 18 There are of course a number of ways in which these convictions can feature in the justification of a principle (e.g. McMahan, 2013). For our purposes, though, we can proceed by example. Consider two arguments Caney develops in support of his nonrelational global egalitarianism.
First, Caney (2001: 114) motivates his global egalitarianism by invoking our conviction that historical practices which allocated opportunities on the basis of race and social class are seriously unjust. He claims that practices which allocate opportunities on the basis of nationality are similarly unjust, given they the share the relevant moral property of arbitrariness. Second, arguing for the same global egalitarian principle, Caney (2005: 122) trades on our conviction that the ‘postcode lottery’ is objectionable in the domestic context. Here, his strategy is to try and persuade us that if geographical location is not a suitable distributive criteria in a democratic state, it should also be shunned in the context of the practices which structure the international order.
In both cases, Caney appeals to a conviction about a particular practice in order to justify a general principle which is grounded independently of that practice. Practice-dependent theorists will surely view this type of engagement as insufficient. My contention, however, is that this is because they view the interpretation of a practice as a necessary part of the justification of a principle. Nonetheless, it is important to note that practice-independent theorists can claim that practices are relevant in the justification of a principle. 19
Application
In His recent work, Sangiovanni has indicated a shift in his position. He now thinks the best way of drawing the distinction between practice-dependent and practice-independent views ‘is by focusing on two different modes of applying higher-level principles and values’ (Sangiovanni, 2016: 15, note 32, emphasis original). The two modes of application he terms Instrumental and Mediated Deduction. Sangiovanni thinks practice-independent theorists will endorse Instrumental – and I believe something like this view is widely held – so showing they need not will establish that this new distinction cannot do the work intended for it. Instrumental can be summarised as follows. Instrumental: 1) practices are only instruments for achieving higher-level values, and 2) these values should be promoted by rules of regulation, where a rule of regulation is a ‘device for having a particular effect, which we adopt or not in light of an evaluation of [its] likely effects’ (Sangiovanni, 2016: 16).
On this view, practice-independent theorists do not care about practices in and of themselves, only about how they bring us closer to further goals, such as morally superior states of affairs. At first glance, this might appear a plausible candidate for marking out a distinctly practice-independent view, but I think nonrelational global egalitarians can, and do, deny Instrumental, and can endorse an alternative mode of application in its place. Given the criteria I set out above – that an account of practice-independence must capture nonrelational global egalitarians, who have been paradigmatically cast in those terms – Sangiovanni's recasting of the distinction does not therefore capture the relevant disagreement between theorists of global justice.
Noting three observations will help us see the point. First, as discussed earlier, global egalitarians are generally pluralists about value, and so while they hold that equality is important, they do not claim that it is the only moral consideration which should factor into judgments about international politics. Second, there is no reason why global egalitarians must view each of the plurality of relevant considerations as standing in the same relation to particular practices. Even if they thought that practices are an instrument for realising an egalitarian distribution, we have not been given reason to generalise this methodological claim. Third, following from this, it seems clear that global egalitarians do claim that practices matter for their own sake or, perhaps better, are constitutive of relationships which matter in this way.
The fact that egalitarians would more likely frame their claim in terms of relations, not practices, does not alter the point. A key practice-dependent insight is that these two notions are deeply interconnected. For one thing, practices place us in relationships which would not exist in their absence. For example, the practices of citizenship are what make us stand in the relation of co-citizens in a contemporary democratic state. More than this, though, practices mediate relations (e.g. Sangiovanni, 2008: 139): they allow us to explain what is important about them and are how we enjoy the specific goods that relationships make possible. 20 When relations are practice-mediated, it does not make sense to view them in non-instrumental terms while viewing their mediating practices as instrumental to some higher-level value.
The type of examples we need to vindicate these conceptual observations are those where global egalitarians attribute non-instrumental value to particular practices and attempt to reconcile this with their distributive ideals. I will call this alternative mode of application Accommodation. I will only be able to sketch this possibility now, but that is enough to trouble Sangiovanni's claim that he has arrived at a more precise statement of the distinction between practice-dependence and practice-independence.
To start with, Accommodation requires us to identify a practice that we care about non-instrumentally. Interestingly, the sort of practices relevant for Accommodation are often presented as counterexamples to distributive egalitarianism. Accommodation, however, starts from the conviction that the apparent tension will be resolved, or at least significantly diminished, once we get clearer about the cases where the unjust inequalities which follow from a practice are closely connected to some non-distributive moral value it realises. Doing so will reveal the space in practice where the tension does arise – where we have to either curtail important features of the practice or compromise on our distributive egalitarianism – but will also reveal the equality-disturbing goods produced by the practice which are incidental to, or further from, the reasons we have to value it. In this sense, Accommodation also diverges from part 2) of Instrumental by making evaluative claims about trade-offs in contemporary international politics rather than articulating rules of regulation expressing all-things-considered judgements. 21
Consider the value of national self-determination, which, roughly, holds that a national community should be able to ‘steer itself according to a common set of values and ideals’ (Armstrong, 2010: 318). Here we do not see practices as helping us achieve some further goal; rather, self-determination attributes value to shared political practices and to the decisions that emerge from them. Against the tendency to present self-determination as incompatible with global egalitarianism (Miller, 2007: 68–76), Chris Armstrong has pursued Accommodation. He claims, on the one hand, that global egalitarians need to be ‘more circumspect about their argument that inequalities which track national membership should be condemned as “morally arbitrary” in nature’, given the fact that adult citizens can, in many cases, participate in making collective decisions (Armstrong, 2010: 314). On the other hand, he argues that defenders of national self-determination must recognise that egalitarian standards can be applied to inequalities emerging from national policy very differently (Armstrong, 2010: 319–325). For instance, global egalitarians can advocate that national communities should retain the ability to make choices about their collective life, which is after all what is central to the value of self-determination, while at the same time holding that the costs attending these decisions must be to at least some extent constrained by egalitarian standards (Armstrong, 2010: 323). 22
We can find further examples of this mode of application outside of the literature on global justice. Consider the value of the family. In this case too, a crude view, holding that the complete parental partiality which follows from a proper respect for family autonomy will undermine our pursuit of equality, has been challenged by egalitarians who have sought to provide a more detailed account of the sort of interactions which are central to the value of the family. Brighouse and Swift (2009) in particular have developed a much more nuanced account of the sort of valuable familial interactions which exert a tension with equality. Some interactions, such as reading bedtime stories, are an important way of maintaining the sort of intimate relations we should value within a family, and so do exert a tension with equality. Others, though, such as sending a child to an elite private school, do not have this character, and so cannot be shielded from egalitarian interventions by appealing to the value of the family. Even if we disagree with their substantive conclusions, Brighouse and Swift's account reveals that we can attribute non-instrumental importance to the practices of family life but also consider how they thwart or promote the realisation of the independent value of distributive equality.
More could be said about the details of Accommodation: my aim has been to cast doubt on Sangiovanni's reformulation of the distinction between practice-dependence and practice-independence. In Accommodation, practices are not merely instruments to the realisation of some further goal. Instead they can be viewed as supplying reason to compromise our general distributive commitments. Moreover, Accommodation arrives at evaluative judgments, not rules of regulation, which provide a more fine-grained account of where the relevant tension arises in practice. Under closer inspection, the tension will tend to be diminished, and this in turn will reveal a greater space for policy interventions. Even without a thoroughgoing account of Accommodation, then, we have grounds to reject the idea that the status of practices in the application of a principle, as Sangiovanni understands it, captures the disagreement between practice-dependent and practice-independent theorists.
Interpretation
What Ultimately seems to distinguish practice-dependent from practice-independent theorists is the role that the interpretation of a particular social context plays in the justification of a principle of justice. More specifically, practice-dependent theorists claim that a principle of justice should be drawn from an interpretation of the ‘point and purpose’ of the practice it is intended to regulate (Sangiovanni, 2008: 141–145, 2016: 18–20). It is this claim which imbues practice-dependence with a narrow focus, because it ties their theories to existing institutions, and so it is this claim which integrationists should target. Practice-independent theorists, in turn, deny that the interpretation of a social practice is a necessary starting point for the justification of a principle: though intuitions or considered judgements might be elicited in relation to a practice, they do not deem it necessary to engage in the interpretive work advocated by practice-dependent theorists. 23 This disagreement does not map perfectly onto the debates integrationists have been preoccupied with. 24 But my claim is that focusing our attention here will make for a more fruitful debate among theorists of global justice, given that it starts from a point of genuine disagreement.
I think practice-independent theorists will have at least two issues with the sort of interpretation practice-dependent theorists engage in. First, they will hold that the normative principles derived from interpretation should be constrained by external moral principles beyond the basic rights constraint I suggested (Section I) all reputable views will affirm. Thus, although Sangiovanni (2008: 146–148) articulates his favoured practice-dependent approach in a way that is committed to an independent basic rights constraint, my claim is that the practice-independent theorists will affirm at least one further general moral constraint, such as distributive equality. Second, practice-independent theorists take issue with the sort of information which practice-dependent theorists include in their interpretation. To date, this latter concern has been given voice in two more specific types of objection.
The status quo objection holds that given the fundamental role interpretation plays in the justification of a principle, it is a problem that the practice-dependent approach draws so substantially on empirical information, like legal documents, conventions and institutional design, which are, and have historically been, connected with injustice. In other words, the sort of information which practice-dependent theorists draw on when engaging in interpretation is tainted by injustice. Caney (2018b: 108) has advanced this objection. He further claims the worry will not be allayed by the practice-dependent theorist's insistence that their method is critical as well as descriptive, such is the depth of connection between the injustice associated with a practice and its point and purpose.
The delineation objection claims that practice-dependent theorists do not have a plausible rationale for what information to include in the interpretation of a practice and what to leave out. Walton has advanced this objection with most force. On the one hand, when practice-dependent theorists attempt to delineate a practice by appealing to some normative criteria, they cannot identify one which is specific enough to mark out a particular practice but which is also coextensive with it (Walton, 2020: 69–71). On the other hand, the practice-dependent theorist might try to avoid this problem by delineating the practice through description itself; but then they encounter the problem of having to adjudicate between competing descriptions of that practice (Walton, 2020: 69–71). It is unlikely we will agree on how to describe the international order. Practice-dependent theorists themselves will likely diverge on this question, while integrationists will claim (e.g. Caney, 2018b: 115) that such is the interconnectedness of different spheres of global activity, it does not make sense to describe the landscape as one made up of numerous separate domains. Walton argues that it is hard to see how the practice-dependent theorist can settle this matter of competing descriptions without appealing to some practice-independent standard.
Of course, I have not shown that either of these objections ultimately undermine practice-dependence. In my view, however, they are promising lines of inquiry, and they are both better situated within the disagreement between practice-dependence and practice-independence, than between integrationism and isolationism.
Conclusion
In My view, if we are concerned about the narrowness of certain contemporary approaches to global justice, we should focus on the way those positions construe the relevance of political practices. Indeed, I think that elaborating on and refining the two objections outlined in the previous section would be a good place to start. In many ways, the fragmentation of debates about global justice, no longer reducible to a disagreement between statism and cosmopolitanism, is welcome. It reflects a more sophisticated picture, where we recognise that our competing substantive principles are undergirded by, and sometimes in close proximity to, competing methodological claims about how to develop principles of global justice. But as these debates have become more complex, it is important we make sure we continue to address our arguments in the appropriate way to those we disagree with. I have argued that an increasingly popular approach, integrationism, fails to do this. I think a clearer picture of the conceptual terrain of debates about global justice is revealed if we focus instead on a distinction between practice-dependent and practice-independent theory.
Footnotes
Acknowledgements
This paper started life as a chapter in my PhD thesis, and I thank Catriona McKinnon and Rob Jubb for their extremely generous supervision of that project. I would also like to acknowledge my PhD examiners, Chris Armstrong and Alice Baderin, for discussion of these issues in my viva. For comments and suggestions on recent drafts of this paper, thanks to Chris Armstrong, Faye Bird, Simon Caney, Jamie Draper and Rob Jubb. I presented this paper at the REAPP work in progress seminar, and I am grateful to the audience for their helpful questions.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the British Academy (grant number PF20 100026, DS-2014-002).
