Abstract
I present a new challenge to the Rawlsian insistence on ideal theory as a compass orienting concrete policy choices. My challenge, focusing on global politics, consists of three claims. First, I contend that our global ideal can become more ambitious over time. Second, I argue that Rawlsian ideal theory’s level of ambition might change because of concrete policy choices, responding to moral failures which can be identified and resolved without ideal theory. Third, I argue that we currently face such potentially transformative choices. I conclude that these choices are analytically prior to, rather than derivative from, global ideal theory.
Keywords
Over the last decade, political philosophers have been spending considerable energy debating what political philosophy should be doing. Much of this debate has centred on John Rawls’s claim that political philosophy should focus on constructing ‘ideal theories’ – theories which present a complete vision of perfect justice for the domain they seek to regulate, and which assume that all agents comply or can be brought to comply with the demands of justice. 1 Rawls himself was early to recognise the point that his critics (such as Amartya Sen) would later stress: that problems of injustice ‘are … the things that we are faced with in everyday life … the pressing and urgent matters’ (1999a: 8). However, Rawls insisted that ‘ideal theory … provides … the only basis for the systematic grasp of these more pressing problems’ (1999a: 8).
One way to understand this Rawlsian claim is as follows. We might be able to recognise manifest injustices as injustices, even without an ideal theory offering a complete vision of perfect justice. Indeed, our condemnation of some specific injustices (slavery, for instance) comes before an ideal theory: this condemnation is not a result of ideal theory, but is a fixed judgment against which we evaluate any ideal theory. However, we cannot have a systematic response to injustice without an ideal theory. This is because the systematic response to specific injustice is the one that best leads us towards a vision of perfect justice, and only an ideal theory can provide such a vision. Call this Rawlsian position the ideal compass view. Defending Rawls against Sen, AJ Simmons (2010: 35) captures the ideal compass view as follows: We don’t need to know all that ideal justice requires in order to compare … our policy options here and now, any more than we need to know that Everest is the tallest mountain in the world before we can compare the heights of lesser peaks … [But] which of two smaller ‘peaks’ of justice is the higher … matters conclusively only if they are both on equally feasible paths to the highest peak of perfect justice. And in order to endorse a route to that highest peak, we certainly do need to know which one that highest peak is.
These claims have significant methodological implications, because they turn the ideal compass view on its head: they show that some concrete public policy choices we are currently facing are analytically prior to, rather than derivative from, global ideal theory, and that as long as these choices are not made, circumstances are not ripe for focusing on global ideal theory. In turn, this point should push us to re-evaluate many aspects of global political philosophy: if it is indeed the case, that as long as some concrete choices are not made, philosophers discussing global politics ought not focus on constructing an ideal theory of global justice, much of what global political philosophy has been doing for the last two decades is misguided.
In order to establish these claims, and show that concrete public policy choices can have such a dramatic effect on global ideal theory, I need to be specific from the outset about which particular choices I have in mind. I will highlight a set of choices that has rarely been thought to be a problem for the ideal compass view: those concerning natural resource trade between non-democratic regimes and corporations based in liberal democracies. Defenders of Rawls’s international theory have emphasised cases where such trade is obviously ruled out by global ideal theory (when it comes to what Rawls calls ‘outlaw’ 3 regimes); while Rawls’s theory itself suggests that customary commercial ties are to be retained with non-democratic regimes that qualify as ‘decent hierarchies’ (1999b: 59). But in either case, conventional Rawlsian wisdom holds that the issue of trade with non-democratic regimes does not undermine the ideal compass view. I will argue, however, that this conventional wisdom is mistaken. Whether to prohibit natural resource trade with non-democratic regimes is in some cases a complex moral question, 4 but this question is independent of global ideal theory. And because the answer to this question might also condition the ambitions of global ideal theory, the question has to be resolved before, not after, ideal theory discussions.
I advance this thesis as follows. In the first section, I introduce two premises derived directly from Rawls: that in order to function as a ‘realistic utopia’, every ideal theory must account for the distinct limits of practical possibility that prevail in the context for which it is formed; and that, when the limits of practical possibility change, the content of ideal theory as a realistic utopia may change as well. In the second section, I argue – again building on Rawlsian premises – that a world of liberal democracies will feature different limits of global practical possibility than those that obtain at present. The combined upshot of these opening two sections is that a world of liberal democracies might require a new ideal theory.
In the third section, I argue that liberal democracies might have a principled duty to prohibit natural resource transactions with non-democratic regimes, and that such a principled prohibition could potentially lead to global democratisation, even if only as a side-effect. I then elaborate several ways in which this potential makes present circumstances the wrong time for focusing on global ideal theory. In the final, fourth section, I anticipate objections.
1. Rawlsian ideal theory and the limits of practical possibility
A key aim of ideal theory on the Rawlsian understanding is to offer systematic normative guidance on fundamental political choices in the real world. 5 Rawls holds that in order to provide such guidance, any ideal theory must have an adequate account of what he calls ‘the limits of practical possibility’ (2001: 4). Without such an account, Rawls suggests, no ideal theory can give us a compelling picture of the best social ideal that we can reasonably hope for – no theory can offer a ‘realistic utopia’ (1999b: 12).
However, Rawls offers little guidance on how to identify limits of practical possibility. Rawls seemingly takes it for granted that we instinctively recognise (or at least, that many of us instinctively recognise) certain features of the world as limits of practical possibility – as fundamental constraints with wide-ranging implications that a practically oriented political philosophy cannot ignore. Thus, for example, Rawls seems to assume our recognition of human beings’ limited altruism, when he complains that utilitarianism imposes ‘strains of commitment’ on individuals that are not practically possible, insofar as these strains ‘exceed the capacity of human nature’ (1999a: 154). Similarly, entrenched disagreements regarding metaphysical questions within liberal democracies appear to Rawls as the sources of obvious limits of practical possibility. It simply is the case that, within modern liberal democracies, the ‘fact of reasonable pluralism’ about metaphysical questions ‘limits what is practically possible here and now’ (Rawls, 1999b: 12). Generalising from these examples, we may intuit that Rawls’s refusal to provide anything like a complete list of limits of practical possibility, or a guide for identifying these limits, derives partly from an implicit sense that we ‘know these limits when we see them’.
Rawls is explicit, however, that at least some limits of practical possibility can change over time, noting that what is ‘practically possible here and now’ is different from ‘whatever may have been the case in other historical ages’ (1999b: 12). And if it is the case that any ideal theory must take its bearings from the limits of practical possibility, then it follows that when the limits of practical possibility change, the content of Rawlsian ideal theory itself may change as well. In other words, it seems that from a Rawlsian perspective, different ideal theories may be appropriate for different ages. An ideal theory different from Rawls’s ‘justice as fairness’ might have been appropriate for previous ‘ages’, and yet another ideal theory might be appropriate for a future age. One issue that should be central to Rawlsian methodology is therefore what changes in the world – or what kind of changes in the world – will take us beyond what Rawls calls the ‘here and now’, to another age featuring new limits of practical possibility, and thus potentially requiring a new ideal theory. A related key issue is how our own agency might bring about such changes in the limits of practical possibility, and therefore in the content of ideal theory. In a sense, Rawls recognises this issue, only to put it aside: … there is a question about how the limits of the practicable are discerned and what the conditions of our social world in fact are; the problem here is that the limits of the possible are not given by the actual, for we to a greater or lesser extent change political and social institutions, and much else … I shall not pursue this deep question …(2001: 4–5)
The most plausible Rawlsian answer – which I shall assume as I develop my claims – is arguably the following. It is coherent to speak of limits of practical possibility that are action-sensitive – that human action might shift – because there could be circumstances where the relevant human actions are morally untenable. Take the example of reasonable pluralism again. ‘The pluralism of liberal democratic societies’, Rawls writes, ‘is best seen as the outcome of the exercise of human reason under free institutions’ (1999b: 122). 6 If this is true, then we may think that, at least in some circumstances, the only way to eliminate reasonable pluralism is to violently remove free institutions. But this violent strategy is plainly morally unacceptable – so much so that, for practical purposes, we should never even contemplate it. So once we have reasonable pluralism under free institutions, we should take this pluralism as a fundamental limit of practical possibility, and constrain our ‘realistic utopias’ accordingly.
This reasoning allows us to make better sense of the Rawlsian thought that at least some limits of practical possibility might change over time. Perhaps (for instance) a limit of practical possibility that can currently only be altered through morally untenable actions, will later on be alterable through other morally permissible actions, which will only become available in the future, and which we cannot even envision now. Or it could be that past limits of practical possibility have been altered as a result of actions that were themselves morally prohibited, but were nonetheless pursued. Or it could be that at least some past generations engaged in morally permissible actions that have transformed the most basic life circumstances of current generations, but that have also made it impermissible for present generations to pursue parallel actions (think, for example, of industrialisation and climate change).
Once we bear these variations in mind, we should broaden the way we understand debates about limits of practical possibility and ideal theory’s level of ambition. These debates are often understood in terms of relative pessimism or optimism: if one side adopts a more ambitious ideal than the other, this is often because of a more ‘optimistic’ view that identifies fewer limits of practical possibility. Yet debates about ideal theory’s level of ambition and limits of practical possibility can have other important dimensions, to which we should also be attentive. Thus for example, different sides may agree that a certain limit of practical possibility obtains at a given point in time, but disagree on whether available courses of action have the potential to shift this limit. Alternatively, different sides may disagree on which (if any) of the relevant courses of action are morally permissible. Or they may disagree as to whether a relevant course of action that would be impermissible if grounded in a certain rationale, would become permissible if grounded in another rationale.
Disputes of this kind have significant meta-theoretical implications, which I believe have been neglected by philosophers discussing Rawlsian ideal theory. My argument can be seen (at least in part) as an illustration of these implications. Focusing on Rawls’s global theory, I aim to establish that current choices we might make independently of ideal theory could shift what Rawls sees as current limits of global practical possibility. And these potentially transformative choices, I will argue later, mean that ideal theory cannot serve as a useful compass in existing global circumstances. This argument begins from a specific limit of practical possibility that Rawls identifies in the existing global order: the prevalence of non-democratic regimes. The task of the next section is to show that there are good Rawlsian reasons to think that a world of liberal democracies will feature different limits of practical possibility than those that obtain in the present world, and could therefore require a different global ideal theory.
2. Why a world of liberal democracies will feature different limits of practical possibility
Throughout The Law of Peoples, Rawls claims that liberal democracies ought to accord equal respect to what he calls ‘decent hierarchies’ (1999b: 4–5) – regimes that are not liberal democracies, but that are non-aggressive, give a meaningful voice to their subjects in the political process through consultation, and have a ‘common good idea of justice that assigns human rights to all [their] members’ (1999b: 88). To support this claim, Rawls relies on the fact of reasonable pluralism that is at the heart of his political liberalism. Just as the liberal state ought to accommodate within its borders the pluralism of varying comprehensive moral doctrines, Rawls argues, it should also accept pluralism regarding forms of political organisation: ‘If all societies were required to be liberal … then the idea of political liberalism would fail to express due toleration for other acceptable ways (if such there are, as I assume) of ordering societies’ (1999b: 59). For Rawls, just as domestic pluralism ‘limits what is practically possible here and now’, within modern liberal democracies, so does international reasonable pluralism, regarding ‘ways of ordering societies’, limits what is practically possible in the ‘international political world as we see it’ (1999b: 83). 7
There are, however, several important points cutting against this analogy. For one thing, in the domestic context, Rawls assumes equal moral standing of the comprehensive doctrines that fall under reasonable pluralism. But Rawls makes no such assumption when it comes to international pluralism. Even while calling to accommodate ‘decent’ illiberal ways of ordering societies, Rawls is explicit that the ‘decent’ kind of ‘hierarchy’ he wishes to count as legitimate ‘does not treat its own members reasonably or justly as free and equal citizens, since it lacks the liberal idea of citizenship’ (1999b: 83). 8
Furthermore, as we have noted above, Rawls believes that domestic reasonable pluralism is bound to recreate itself at least as long as liberal institutions are in place, but makes no parallel suggestion at the international level. As long as ‘free institutions’ will exist, Rawls thinks that ‘the exercise of human reason’ will mean that domestic pluralism will exist as well. But Rawls makes no parallel claim at the international level: there is no claim that given certain international institutions or practices, ‘human reason’ will reproduce a pluralism that includes illiberal ways of organising societies.
Finally, very much related, Rawls does maintain a hope that the decent non-liberal regimes will liberalise over time. ‘The Law of Peoples’, Rawls proclaims, exhibits ‘confidence in the ideals of constitutional liberal democratic thought’. It accordingly ‘respects decent peoples by allowing them to find their own way to respect these ideals’ (Rawls, 1999b: 122). Decent hierarchies currently lack these ideals, since they ‘lack the liberal idea of citizenship’, but Rawls does think that there would be moral value to decent hierarchies adopting this idea: … if a liberal constitutional democracy is, in fact, superior to other forms of society, as I believe it to be, a liberal people should have confidence in their convictions and suppose that a decent society, when offered due respect by liberal peoples, may be more likely, over time, to recognize the advantages of liberal institutions and take steps toward becoming more liberal on its own. (1999b: 62)
These points are important for at least two reasons. First, they indicate that Rawls has no principled opposition to a world of liberal democracies. Second, if there is no principled opposition to a world of liberal democracies, then it makes sense to ask whether such a world will have different limits of practical possibility than those that obtain in the ‘here and now’. Taking up this question, I wish to suggest several Rawlsian reasons for why a world of liberal democracies might extend the limits of global practical possibility, and might accordingly require a fundamentally different ideal theory than a theory of the kind that Rawls sketched in The Law of Peoples (and that many Rawlsians have been defending).
The most general reason concerns what Rawls calls ‘public political culture’. A public political culture, reflected in the traditional public interpretation of society’s major institutions, and in famous historical texts bearing on these institutions, is not only a source of normative judgments that a theory of justice seeks to systematise. 9 It is also itself a determinant of the limits of practical possibility: a public criterion of justice that is widely out of kilter with the political culture prevalent in the contexts to which it is meant to apply, is not a realistic candidate as a criterion of justice for these contexts. Now, despite the fact that the public political culture even of ‘decent’ hierarchies is not as just as that of liberal democracies, Rawls clearly holds that a sufficiently realistic global vision must assume, as a given constraint, the prevalence of hierarchical political cultures in the world. This constraint is a key example of Rawls’s insistence that normative reflection ‘must always start from where we now are’ (Rawls, 1999b: 121). It is evident, for instance, in Rawls’s efforts to show that his international original position would be regarded as a model of fair cooperation not only by ‘you and I’ as ‘citizens of liberal democracies’ (Rawls, 1999b: 32), considering international practices from a liberal perspective, but also by ‘you and I’ as ‘members of decent hierarchies’ (Rawls, 1999b: 69) beginning from an illiberal political culture. Since it would not have to accommodate illiberal political cultures in this way, a realistic utopia for a world of liberal democracies could generally assume more ambitious limits of practical possibility.
There are also specific Rawlsian reasons to think that a world of liberal democracies would feature more ambitious limits of practical possibility. At the economic level, for example, Rawls makes very strong empirical assumptions about the decisive impact of a society’s political culture and institutions in shaping its economic performance. 10 These assumptions give a very direct sense in which societies’ political structures influence their members’ ‘life prospects, what they can expect to be and how well they can hope to do’ (Rawls, 1999a: 6–7). If we accept these assumptions, we might think that a world of liberal democracies will feature more ambitious limits of practical possibility already in a very concrete way. The threshold of minimal material resources to which all human beings are entitled, for instance, might become more ambitious as societies’ political institutions and culture become more conducive to economic growth: among other things, one may think here of the greater stability of markets through a reliable democratic rule of law, as well as the democratic embrace of free flow of information, yielding stronger knowledge-based economies that increase human capital. 11 Therefore, while under authoritarianism we might accept minimalism about human rights that focuses only on minimal resources needed for subsistence or the prevention of absolute deprivation, given more liberal political systems we might set the bar higher, so as to include resources that will secure or allow the development of more complex human ‘functionings’ or ‘capabilities’.
Alongside these changes in the economic limits of practical possibility, changes in institutional limits might also come about in a world of liberal democracies. For one thing, the density and stability of global institutions can be expected to increase substantially in a world of liberal democracies, building on higher levels of mutual trust and respect among a larger number of liberal democracies. 12 More specifically, a world of liberal democracies might very well feature a significant growth of supranational political institutions, perhaps along the lines of the European Union (EU) (that is, after all, meant to be comprised only of democracies). Such growth could extend limits of practical possibility, and complicate ideal theory questions, in at least two ways.
One way concerns the arguments typically made by ‘statists’ like Rawls, who reject the extension of domestic egalitarianism to the global level and who often cite the existence of special political ties and a coercive framework between co-citizens as justifying the limitation of egalitarian concern to compatriots only. 13 Statists have some trouble already today, deciding whether a supra-national body like the EU with its own regulations and coercive apparatus produces special kinds of political obligations among all of its members. Do Germans have egalitarian duties towards Greeks? Are these duties identical to those that Germans have towards each other? Are these egalitarian duties, if they exist, stronger than the distributive duties of Germans towards Australians? The complexity of such questions is bound to be further accentuated in a world of liberal democracies, assuming (as seems likely) that it will witness further proliferation of supranational institutions.
The other way in which a world of liberal democracies might encourage transformation of global institutional architecture is even more ambitious in terms of the limits of practical possibility. If it is indeed the case that a world of liberal democracies will significantly facilitate the creation of EU-like institutions, and if there is reason to think that these institutions will (eventually) internally consolidate into something like a state, one may think that a world of liberal democracies could accelerate the historical trend of a decrease in the number of political units in the world. 14 In turn, if several blocs emerge in EU-like fashion, and if each of these blocs eventually becomes a state, then an idea that Rawls dismisses as lying outside current limits of practical possibility might seem eminently more possible: the idea of a world state. 15 If it is indeed the case that a world of liberal democracies will be much more hospitable to the creation of a global government, that is further reason to think that such a world will differ from ours in the limits of practical possibility it presents, potentially requiring a very different realistic utopia.
3. Customary trade and the ideal compass
Already in his domestic theory, Rawls insisted that ‘[c]onceptions of justice must be justified by the conditions of our life as we know it or not at all’ (1999a: 398). Yet non-democratic regimes are prevalent in international life ‘as we know it’. Why does it matter then, that a world of liberal democracies might require a different conception of justice?
The remainder of this essay can be seen as an answer to this question. I begin this section by arguing that liberal democracies might have a principled duty to prohibit natural resource transactions with non-democratic regimes, and that such a principled prohibition could potentially lead, even if only as a side-effect rather than as its main intended consequence, to global democratisation (section 3.1). I then elaborate several ways in which this potential for global democratisation makes present circumstances the wrong time for constructing global ‘realistic utopias’ (section 3.2).
3.1 Customary trade, public property and global democratisation
The argument that liberal democracies ought to prohibit natural resource transactions with (at least some) non-democratic regimes has become familiar over the last 15 years. 16 The argument’s starting point is the intuitive claim that state-owned natural resources, like all state property, belong to the people rather than to those wielding effective political power. 17 In order for a regime to legitimately sell state-owned natural resources, it must therefore claim proper authorisation from the people to engage in such sales. This authorisation comes about as citizens elect government officials who get to manage public property on behalf of the body politic. Yet, the argument goes, regimes that rely on bullets rather than ballots to secure political power cannot claim to enjoy such authorisation. Such regimes can therefore be said to be systematically violating their people’s property rights. Prospective purchasers of state-owned natural resources who know that such violations are taking place ought not to transact with the regimes committing these violations. Just as individuals have a duty not to purchase property that they know the vendor has illicitly seized from the real owner (person A ought not buy from B what she knows that B has illicitly seized from the real owner C), resource corporations have a duty not to purchase (national) natural resources that they know certain regimes have come to control in illicit ways. 18 And the democratic governments regulating these corporations have a duty to prohibit them from making such purchases.
Different political philosophers elaborating this argument have offered different accounts as to which regimes should be considered violators of their peoples’ property rights, and thus as illegitimate vendors of natural resources. In my view, the most coherent version of the argument includes all and only regimes that clearly lack free and fair elections and a rule of law binding upon those wielding effective political power. 19 In the absence of free and fair elections, de-facto rulers cannot claim proper authorisation from the people to manage the people’s natural resources. In the absence of a rule of law, de-facto rulers cannot be formally and legally accountable to the people as the owner of public property – there is no mechanism to sanction rulers who abuse their control over this property. 20
The guiding thought in applying the customary trade argument in this way is that we cannot take seriously a sovereign people’s ownership over the natural resources encompassed in its territory, unless we take seriously a people’s right to democratically authorise specific individuals to manage its resources – unless, that is, we take the Rawlsian notion of ‘property-owning democracy’ quite literally. 21 This emphasis on the need for democratic authorisation is compatible with retaining customary commercial ties with what one may call flawed democracies: regimes that clearly do take free and fair elections and rule of law seriously, even if imperfectly, should be considered legitimate vendors of their people’s natural resources. 22 At the same time, the emphasis on democratic authorisation generates a strong presumption against natural resource transactions with any regime that fundamentally violates the requirements of free and fair elections and rule of law, independently of what portion of natural resource revenues it distributes to the population.
The most coherent way to defend a prohibition on natural resource trade with non-democratic regimes is to ground it in a principled respect for the property rights of the peoples living under them. Just as colonial powers had a duty in the past to end colonial rule simply as a matter of principle, liberal democracies might have a duty at present to end the neo-colonial practice of massive-scale transactions with rulers who control resources without the authorisation of their peoples. On this non-consequentialist stance, even if political and economic circumstances in most non-democratic regimes will remain the same as a result of liberal democracies prohibiting natural resource transactions, such a prohibition would still be morally required – perhaps just as ending colonialism would have been morally required even if one could not prove that doing so will make the victims of colonialism better off.
Consequentialist considerations, on this view, feature not as the grounds for reform, but rather at the level of side-effects to which we should be attentive. On one hand, we need to be attentive to extreme negative consequences that reform to customary trade might yield. In any case where the result of ending natural resource transactions with a given regime would be absolute deprivation for its population, there will clearly be strong reasons in favour of continued transactions. On the other hand, we also need to consider the chance that the side-effects of a prohibition on natural resource transactions with non-democratic regimes will be – at least in the longer run – much more positive. Specifically for our meta-theoretical inquiry, it matters that such a prohibition might have the side-effect of triggering global democratisation.
One way to see this potential is to consider the social-science debate on the ‘resource curse’. Despite important disagreements, both social scientists who believe that natural resource abundance causes political maladies, and their critics who insist that natural resources have no independent causal force, agree on the following key point: once a situation exists where a non-democratic regime controls sizeable amounts of natural resources traded as commodities in global markets, a self-reinforcing dynamic sets in, with the regime perpetuating its hold on power through its control of natural resource wealth. 23 A prohibition on natural resource purchases from non-democratic regimes interacts with the very heart of this dynamic. At least if carried out by a critical mass of democracies, this prohibition would take away much of the wealth on which a sizeable portion of the world’s non-democratic regimes rely to fund their rule. It is not implausible to think that a withdrawal of natural resource wealth could destabilise many of these regimes. And it is therefore not implausible to think that a prohibition on natural resource purchases from non-democratic regimes could trigger global democratisation. 24
It might be helpful to spell out further some of the causal mechanisms that are relevant here, especially when it comes to specific political moments in which political power is contested. Coups provide one key example. In many countries where democratic procedures are either non-existent or extremely fragile, the threat and reality of coups represent an integral part of the political landscape. And wherever a coup is a well-tried method for gaining political power, the prospects for democracy depend to a considerable degree on the odds that a coup will occur at any given time, as well as on the kind of regime that is likely to follow in the aftermath of a coup. It is therefore significant that recent empirical work finds that various economic measures taken by foreign democracies in response to coups both precipitate post-coup competitive elections, and reduce the overall incidence of coups against elected governments. 25 This empirical finding, of course, supports the thought that reform to customary trade might alter the incentive structure pushing various actors to seek political power by brute force, and to retain power through such force.
Another key example concerns rigged elections. A large number of non-democratic regimes hold elections only to massively rig them. 26 But such elections, despite not being remotely free and fair, can still be meaningful political events, insofar as they can provide focal points for popular protests against the regime. And in this context, a refusal by foreign governments to grant the regime certain customary privileges can have special significance. These customary privileges can aggravate the collective action problems facing opposition movements, by entrenching the sense that the odds of defeating the regime are overwhelmingly small. 27 Correspondingly, denying these privileges means generating less headwind for potential opposition movements as they try to rally ordinary citizens – especially in the context of rigged elections – to demand genuine democratisation.
Notwithstanding these potential mechanisms, however, it should be clear that there is no certainty regarding the prospects of democratisation following reform to customary trade. One key source of uncertainty is our inability to compare our world to any world in which all was held constant aside from the ‘small’ global difference of all liberal democracies prohibiting natural resource purchases from non-democracies. Another issue is the opacity of many non-democracies’ internal power structures and ideational divisions: this opacity makes it difficult to predict which elites will be strengthened in relation to others as a result of a withdrawal of the regime’s trading privileges in natural resources, and which results of internal power struggles will be most conducive to internal reforms. 28 And, last but not least, there is the crucial ‘predictable unpredictability’ of massive uprisings against non-democratic governments. 29 For all of these reasons, any potential causal mechanism we sketch as leading to democratisation, however sensible in theory, has to be accompanied by a heavy dose of humility about our ability to predict political developments in practice.
3.2 Three ways to move away from global ideal theory as a necessary compass
We can now start to pull together the argument’s different threads, and see how they bear on the Rawlsian view of ideal theory as a necessary compass. In section 1, I sought to establish that from a Rawlsian perspective, new limits of practical possibility might require a new ideal theory. In section 2, I argued that from a Rawlsian perspective, a world of liberal democracies will feature new limits of global practical possibility, and might therefore require a new global ideal theory. So far in this section, I have argued that liberal democracies might have a principled duty to prohibit natural resource transactions with non-democratic regimes, and that this prohibition has the potential to trigger global democratisation. The thought resulting from these claims is that such a prohibition could yield a world with different limits of global practical possibility, for which a new global ideal theory could be needed. Moreover, I believe that if the previous claims are granted, there are multiple paths to a crucial methodological conclusion: that the concrete political choice whether to reform customary trade must be analytically prior, rather than derivative from, global ideal theory.
Three versions of this challenge to the ideal compass view are important. According to the first, minimalist challenge, so long as the dilemmas of customary trade are not resolved, we simply cannot know which ideal theory we need to work with as a ‘realistic utopia’ – whether an ideal theory constructed for a world in which illiberal regimes remain prevalent, or a theory constructed for a world of liberal democracies. The dilemmas of customary trade are analytically prior to global ideal theory, simply because until these dilemmas are resolved, and (in case they are resolved in favour of ending customary trade) until the consequences of their resolution are clear, we cannot know which of our global ideal theories should we employ. We therefore need to focus first on the concrete dilemmas of customary trade, and then, if we settle these dilemmas in favour of reforming customary trade, wait until it becomes clear whether this reform has indeed triggered global democratisation, shifting limits of global practical possibility in a way that requires a shift to a different global ideal theory.
The second, intermediate challenge holds that focusing on global ideal theory at present is problematic in a deeper way than the minimalist version suggests. The minimalist challenge presupposes (however implicitly) that we do not have epistemic difficulties in designing, ‘here and now’, an ideal theory for a world of liberal democracies. But (according to the intermediate challenge) this presupposition is overly optimistic about our epistemic capacities. If a world of liberal democracies will feature different limits of practical possibility than those that obtain in the present world, then it is far from clear that we in the ‘here and now’ are in an epistemic position to construct an ideal theory for such a world. This is because changes to the limits of practical possibility might deepen the complexity of ideal theory questions, in ways that are too hazy for us to see at present. 30
The boldest challenge to the ideal compass view picks up this intermediate notion, and takes it even further. Here the thought is that our epistemic position with regard to a world of liberal democracies is even worse than the intermediate challenge suggests. This is because such a world might feature not only added complexity to familiar ideal theory questions, but also fundamentally new ideal theory questions, whose basic content is too hazy for us to see ‘here and now’. Therefore, we have even deeper reasons to avoid global ideal theorising, as long as we have not settled concrete dilemmas whose resolution might produce a world of liberal democracies, with limits of global practical possibility different from those that obtain in our world. To try to theorise about such a world here and now would truly be to jump ahead of our time.
At least some philosophers should find congenial this kind of Hegelian warning. Yet my position does not require endorsing any specific version of the challenge to the ideal compass view. I assume that all three forms of this challenge have at least prima-facie plausibility. This assumption suffices for my thesis: that a concrete and fundamental political choice we are facing at present is analytically prior to, rather than derivative from, global ideal theory, and that as long as this choice is not made, current circumstances are not ripe for a focus on global ideal theory.
4. Anticipating objections
It is time to turn to some objections. I begin by considering objections that focus on reform of customary trade, holding that this reform is dependent upon global ideal theory (section 4.1). I then consider the objection that current global ideal theory might be helpful in tackling other ongoing global issues (section 4.2).
4.1 ‘To prohibit or not to prohibit’: Not an ideal theory question
The first likely objection that I wish to address is that reform to customary trade cannot be independent of global ideal theory, because the public property rights that are supposed to ground this reform themselves depend on global ideal theory.
Even if it is true that some ideal theories will treat the idea of public property as more central than others, this objection is ultimately implausible, as a domestic analogy quickly shows. We do not normally think that the morally appropriate response to domestic theft or robbery, for example, depends on the ideal theory question of whether, in a perfectly just world, the victim would have title to the good in question. Suppose, for instance, that the mafia forcibly seizes the property of an ordinary individual. Few would say that the proper response (by the legal authorities, or by morally decent citizens) depends upon whether the individual who claims a right to what the mafia has seized would have this right in a perfectly just society. The same point applies to the global case: there is no obvious reason to think that the proper response to violations of public property rights depends on whether these rights will exist (or on what these rights will entail) in a perfectly just world.
Let us move, then, to a second likely objection, which follows Simmons’ argument mentioned at the outset. According to this objection, global ideal theory is necessary for deciding how to respond to violations of public property rights, because the proper response will be one that is most conducive to transitioning towards a complete ideal of perfect global justice.
In evaluating this objection, it is best to begin by noting the sacrifices it entails. The objection holds that if the broader goal of overall perfect justice will ultimately be better realised by accepting or even deepening a specific injustice at present, then, however paradoxically, this is what justice requires. Simmons’ Rawlsian response to Sen, which with which we began, pivots on this kind of transitional strategy: There might, for instance, be only one action or policy (P) that is morally permissible, politically possible, and likely to be successful in reaching a social condition (C) from which the ultimate achievement of perfect justice is more likely than it is from the status quo. But if so, then in order to endorse P … [we] need not first determine that C is a ‘more just’ social state than the status quo (or than any other imaginable social state). C might well be less just than the status quo in terms of their relative resemblances to a perfectly just basic structure and still be more just in transitional terms. (2010: 23)
Now, what might we say about this transitional objection? At the most immediate level, I believe we should say the following. One does not have to reject the ideal compass view in all circumstances, in order to think that issues such as massive scale violations of public property rights cannot be reduced simply to a transitional point on the way to perfect justice. One can accept the ideal compass view with regard to many concrete normative dilemmas of public policy, while still thinking that some specific dilemmas have sufficient intrinsic weight such that it would be wrong to treat them merely as steps en-route to perfect justice. Even those who are normally firmly attached to the ideal compass view would arguably hesitate (or at least, should hesitate) to say that the normative judgment on whether liberal democracies had a duty to divest from Apartheid, for example, was reducible to how divestment related to perfect global justice. Apartheid was clearly sufficiently morally disturbing in and of itself, and divestment from Apartheid, with the social-scientific uncertainty concerning its impact on the regime, sufficiently complex, that even a defender of the ideal compass view should arguably be willing to admit that the normative dilemmas surrounding divestment were sufficiently weighty to stand on their own, and to merit adjudication independently of any global ideal theory. The same point applies to the question of whether liberal democracies ought to disengage from the natural resource dealings of certain regimes.
There is, however, a further, deeper problem with the objection. This objection has to presuppose precisely what I have been assuming is not the case – namely, that there is a single fixed ideal of perfect global justice, against which concrete political choices can be evaluated (and that the most compelling ideal theory will capture this single fixed ideal). However, if there is no such single fixed ideal – if different ideal theories might be appropriate for different normative eras with different limits of practical possibility – the objection cannot get off the ground. The reason is that, once we make the assumption that different ideal theories might be appropriate for different normative eras with different limits of practical possibility, the objection has to take the following form: that an ideal theory whose content presupposes given limits of practical possibility can be used to adjudicate concrete political choices that might shift these very limits. But formulated in this way, the objection cannot work.
This is a complex response, and so it will be useful to illustrate it through an analogy with another kind of limit of practical possibility. Let us make, with Rawls, the plausible assumption that human nature, as we know it (however it is defined and identified), poses limits of practical possibility that any ‘realistic utopia’ must take into account. Now consider how this limit might change due to genetic enhancement. ‘Biotechnologies already on the horizon’, as Allen Buchanan puts it, can change some elements that we normally see as fairly fixed features of human beings – ‘can enable us to be smarter, have better memories, be stronger, quicker, have more stamina, live longer, be more resistant to diseases, and enjoy richer emotional lives’. 32 So it is not entirely implausible to think that at some point in the future, biotechnologies could become available which could also improve human capacity for altruistic behaviour, thus reducing the ‘strains of commitment’ that Rawls (to recall) sees as a key reason counting against utilitarianism. But Rawls’s theory cannot adjudicate concrete policy choices regarding such technologies: insofar as the content of this theory presupposes as limits of practical possibility, limits of human psychology or capacities that bio-enhancement technologies might modify, the theory cannot be a useful guide in telling us what to do about these technologies. The analogous conceptual point applies to the case before us here: a global ideal theory whose content presupposes the prevalence of non-democratic regimes as a limit of practical possibility, cannot serve as a useful guide in telling us whether to pursue policies that might take us past this limit.
With these points in mind, we can turn to the last objection concerning the relationship between ideal theory and reform of customary trade. According to this objection, there is a simple reason why we cannot discuss reform of customary trade independently of ideal theory: there is a specific global ideal theory – Rawls’s own theory – which explicitly rules out such reform, because it rules out curtailment of customary trade with certain regimes that satisfy Rawls’s criteria for decency.
A necessary condition for the objection to succeed would be to show that a significant portion of the non-democratic regimes with which natural resource transactions would be prohibited actually fulfil Rawls’s criteria for ‘decency’. There is ample reason to doubt that this is the case. It is extraordinarily difficult to find in the current world non-democratic regimes that uphold all of the human rights essential for Rawlsian ‘decency’ while manifestly lacking the free and fair elections and rule of law constitutive of liberal democracy. 33
However, even if we were to put this basic empirical point aside, the objection would still fail, since it misrepresents what Rawls’s global ideal theory rules out. As we have seen, Rawls does not have a principled objection to a world of liberal democracies: he could agree that such a world will constitute moral improvement in comparison to the global status quo. What Rawls’s theory rules out is coercive efforts by liberal democracies to impose liberal democracy on peoples who have certain illiberal preferences. But the proposal to prohibit natural resource transactions with non-democracies does not have to involve any such coercion. A liberal democracy can enact a prohibition on natural resource purchases from a given regime without necessarily trying to coerce the people living under this regime to reform it. 34 As Leif Wenar notes, a liberal democracy ‘can say that the political leadership of a foreign country is ‘none of our business’, while saying that in current conditions its resource exporters qualify for ‘none of our business’’ (2011: 34). That is a key reason why I elaborated earlier the understanding of global democratisation as a potential side-effect of a prohibition on natural resource transactions, rather than as its grounding justification. 35
A particularly trenchant critic might insist at this point that even if a prohibition on natural resource purchases will not have as its stated intent the coercive imposition of liberal democracy, there is nonetheless a dominant understanding of a refusal to trade as a coercive sanction, different only in degree, but not in kind, from aggressive war. 36 And this understanding is bound to overshadow any attempt to explain that this particular refusal to trade is not grounded in the same coercive intent.
But this rejoinder similarly fails. For one thing, it will not do to simply point out that those living under non-democratic regimes might interpret reform to customary trade as having coercive intent. It is (at the very least) an open question, whether the chance that someone will (mis)interpret a given action generates a duty to avoid this action. This chance might generate a duty to be particularly explicit about the reasons for one’s actions, in order to anticipate potential misunderstanding. But that is not the same as a duty to avoid the action that one would otherwise have compelling moral reasons to undertake.
What if we flesh out the rejoinder in a different way? What if we say that a true commitment to respect decent hierarchical political cultures requires avoiding not only policies meant to democratise them but also even policies that could generate democratisation as a side-effect? This variant of the rejoinder does no better, because it rules out various policies that hardly anyone thinks should be ruled out. To give just one example: the global spread of internet technology, for instance, seems to have the potential to destabilise at least some non-democratic regimes (among other things, as the events of the Arab Spring attest, this technology increases the regime’s difficulty in controlling the flow of political information, while lowering obstacles towards collective action connecting individuals opposing the regime). But hardly anyone would say that because the global spread of internet technology might destabilise ‘decent’ non-liberal regimes, liberal democracies ought to refrain from contributing to this spread.
4.2 Ideal theory for the present world?
So far in this section, I have argued that reform to customary trade with non-democratic regimes should be seen as independent of global ideal theory. However, some Rawlsians may be willing to grant this point, and even grant that reforming customary trade might trigger a need for different global ideal theory, while still refusing to concede my thesis. Such Rawlsians might use the following thought to reject my contention, that current global circumstances are not ripe for ideal theory. Even if reform of customary trade might yield a different world for which a different global ideal theory might be needed, and even if the moral judgments about this reform are independent of ideal theory, our current global ideal theory still has a key role to play, because it still can and should guide our moral analysis of other current policy questions. After all, if we have a certain ideal theory designed for the ‘world as we see it’, why should we not continue to use it to reflect on as many present moral questions as we can until a different world emerges?
The main flaw of this objection stems from the Rawlsian willingness to make local moral sacrifices for the sake of realising a broader vision of perfect justice. Recall, one last time, Simmons’ Rawlsian insistence that these kinds of local sacrifices are a key part of what makes ideal theory a necessary compass: when deepening certain local injustices is the best way to eventually realise an ‘integrated ideal’ of perfect justice, we must be willing to take ‘one step backward’ in order to eventually make ‘two steps forward’. Yet, whatever appeal this transitional strategy has in a context where we assume that the limits of practical possibility are stable, this appeal is significantly undermined where we have reason to believe that relevant limits might shift. For in such circumstances, the willingness to deepen local injustices for the sake of progress towards a complete ideal of perfect justice seems questionable on at least two counts.
First, it is far harder to justify such a transitional strategy to the victims of the relevant local injustices. The Rawlsian transitional strategy wishes to reconcile victims of (at least some) local injustices to the continuation (or even deepening) of their predicament, if this is most conducive to justice’s broader aims. But what if the relevant victims reject this demand to ‘take the broader view’ by pointing out that, before too long, there might be a different point in the horizon – a different broader ideal – to which we might shift our gaze? Whatever sense it might make to demand local sacrifices for the sake of reaching a broader ideal that is reasonably stable, it makes much less sense to demand such sacrifices for the sake of reaching an ideal that, later on, could very well cease to be our ideal.
The second problem goes even deeper. An insistence on local sacrifices for the sake of a broader ideal, even when there is a non-trivial probability that this very ideal will change, does not align with the core commitments of Rawlsian meta-theory itself. After all, it might very well be that transitional sacrifices conducive to the realisation of our current ‘integrated ideal’ will turn out to set back efforts to realise our later ideal vision. But as long as there is a non-trivial chance that such a danger will materialise, then by the lights of Rawlsian meta-theory itself, it would be wrongheaded to let our current ideal guide our ‘local compromises’. Doing so would conflict with the Rawlsian call to keep our attention fixed on the ideal of perfect justice, and to avoid prioritising local improvements that might move us away from this goal in the longer run.
5. Conclusion
Let us take stock. If my claims above have been cogent, then it follows that the orthodox Rawlsian view of ideal theory as a compass is ill-suited to current global realities, given the commitments of Rawlsian meta-theory itself. I have sought to establish this conclusion through the key example of reform to customary trade with non-democratic regimes, as a ‘non-ideal’ problem that – I have argued – must be seen not only as more urgent than visions of perfect global justice, but as analytically prior to such visions. More generally, I hope to have shown that Rawlsian ideal theorists must be attentive to the ways in which concrete policy choices might – at least in some cases – shift the limits of practical possibility that condition ideal visions. If the Everest of global ideal theory might move, and if it might move because of how we climb in the journey towards a more just world, then Rawlsian meta-theory itself should lead us to focus on climbing, rather than on where Everest happens to be at the moment.
Footnotes
Acknowledgements
Thanks to Chris Armstrong, Seyla Benhabib, Simon Caney, Tom Donhaue, Stefan Eich, Eva Erman, Burke Hendrix, Lisa Herzog, Jeff Howard, Aaron James, Robert Lepenies, Matt Lindauer, Niki Marinov, Florian Ostmann, Avia Pasternak, Thomas Pogge, Michael Ross, A.J. Simmons, Jiewuh Song, Nic Southwood, Paul Weithman, Leif Wenar, David Wiens, Scott Wisor, Lea Ypi, and Alex Zakaras, for conversations and comments on this essay and its themes. Thanks also to the editors and two anonymous reviewers for the European Journal of Political Theory, for their helpful critiques and suggestions.
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) received no financial support for the research, authorship, and/or publication of this article.