1 23 Ethical Theory and Moral Practice An International Forum ISSN 1386-2820 Volume 15 Number 3 Ethic Theory Moral Prac (2012) 15:369-386 DOI 10.1007/s10677-011-9300-0 Burdened Societies and Transitional Justice Lisa L. Fuller 1 23 Your article is protected by copyright and all rights are held exclusively by Springer Science+Business Media B.V.. This e-offprint is for personal use only and shall not be self- archived in electronic repositories. If you wish to self-archive your work, please use the accepted author’s version for posting to your own website or your institution’s repository. You may further deposit the accepted author’s version on a funder’s repository at a funder’s request, provided it is not made publicly available until 12 months after publication. Burdened Societies and Transitional Justice Lisa L. Fuller Accepted: 12 June 2011 /Published online: 30 June 2011 # Springer Science+Business Media B.V. 2011 Abstract Following John Rawls, nonideal theory is typically divided into: (1) “partial- compliance theory” and (2) “transitional theory." The former is concerned with those circumstances in which individuals and political regimes do not fully comply with the requirements of justice, such as when people break the law or some individuals do not do their fair share within a distributive scheme. The latter is concerned with circumstances in which background institutions may be unjust or may not exist at all. This paper focuses on issues arising in transitional theory. In particular, I am concerned with what Rawls’ has called “burdened societies," that is, those societies that find themselves in unfavorable conditions, such that their historical, social or economic circumstances make it difficult to establish just institutions. The paper investigates exactly how such burdened societies should proceed towards a more just condition in an acceptable fashion. Rawls himself tells us very little, except to suggest that societies in this condition should look for policies and courses of action that are morally permissible, politically possible and likely to be effective. In this paper I first try to anticipate what a Rawlsian might say about the best way for burdened societies to handle transitional problems and so move towards the ideal of justice. Next, I construct a model of transitional justice for burdened societies. Ultimately, I argue for a model of transitional justice that makes use of a nonideal version of Rawls’ notion of the worst-off representative person. Keywords Burdened societies . Transitional theory. Nonideal theory. Development . Human rights 1 Introduction John Rawls famously divides political philosophy into ideal and nonideal theory. Ideal theory “assumes strict compliance and works out the principles that characterize a well- ordered society under favorable circumstances,” while nonideal theory “is worked out after Ethic Theory Moral Prac (2012) 15:369–386 DOI 10.1007/s10677-011-9300-0 L. L. Fuller (*) Department of Philosophy, University at Albany (SUNY), 1400 Washington Ave., Albany, NY 12222, USA e-mail: lfuller@albany.edu Author's personal copy a conception of [ideal] justice has been chosen,” and “asks how this long-term goal might be achieved, or worked toward, usually in gradual steps” (Rawls (1999a) p. 216, hereafter “TOJ”; Rawls (1999b) p. 89, hereafter “LP”.) He further divides nonideal theory into two kinds: (1) “partial-compliance theory” and (2) “transitional theory” (Rawls LP, pp. 5, 90). Partial-compliance theory is concerned with those circumstances in which individuals and political regimes do not fully comply with the requirements of (domestic or international) justice, such as when people break the law or when states engage in wars of aggression. Transitional theory is concerned with circumstances in which background institutions are unjust or do not exist at all, due to the natural limitations of agents, historical contingencies, or unfavorable conditions such as extreme poverty (Rawls TOJ, p. 216; LP p. 90). This paper focuses on issues arising in transitional theory. In particular, I am concerned with what Rawls’ has called “burdened societies,” that is, those generally peaceable societies that “lack the political and cultural traditions, the human capital and know-how, and often, the material and technological resources needed to be well-ordered” (Rawls LP, p. 106). Burdened societies find themselves in “historical, social or economic circumstances” that make it “difficult if not impossible” to establish just institutions.1 The purpose of this paper is to investigate how such burdened societies can reasonably proceed towards a more just condition. When facing problems in nonideal theory, Rawls suggests that we must look for “policies and courses of action that are morally permissible and politically possible as well as likely to be effective [emphasis added]” (Rawls LP, p. 89). However, he offers little insight into how societies might go about meeting these requirements. This is not surprising, since his general approach does not depart sufficiently from the ideal to make these issues especially salient. I intend to “take up where he left off” so to speak, that is, to construct a model of transitional justice for burdened societies. The paper has four main sections. First, I outline the “terrain” and methodology appropriate to nonideal theory as I conceive it. Next, I explain how burdened societies fit into Rawls’ overall project, and how they ought to be understood. Third, I try to anticipate what a Rawlsian might say about the best way for them to handle transitional problems and so move towards the ideal of justice. Finally, I argue for a model of transitional justice that makes use of a nonideal version of Rawls’ notion of the worst-off representative person. 2 Notes on Non-ideal Theory Recent work in nonideal theory contains persuasive arguments for expanding its scope beyond the original definition introduced by Rawls. In addition, much of this work has specified in some detail both the subject matter and methods suitable to nonideal theorizing. Rather than giving an exhaustive overview of these developments, in this section I simply aim to explain those features of nonideal theory that will be relevant to my later discussion. On the traditional picture, nonideal theory presupposes ideal theory. According to Rawls, “until the ideal is identified … nonideal theory lacks an objective, an aim, by reference to which its queries can be answered” (Rawls LP, p. 90). Further, existing arrangements are to be judged in light of our understanding of what is ideally just, and “held to be unjust to the extent that they depart from [the ideal] without sufficient reason” (Rawls TOJ, p. 216). Ideal 1 Rawls, LP, p. 5. I leave aside cases of total impossibility, since if there are no possible reforms that a burdened society could implement that would lead to a more just condition, then I have nothing to discuss. However, given the ability of societies to change, especially with the input and assistance of other societies, it seems likely that these cases will be very few. 370 L.L. Fuller Author's personal copy John Simmons points out that in Law of Peoples Rawls claims that ideal theory has two “parts” or “domains.” The first part of ideal theory “is an account of justice appropriate to liberal democratic societies,” that is, principles for a perfectly just society as laid out in a Theory of Justice. The other part “extends the same ideas to the international community, deriving principles for a ‘society of peoples’ … that can bring about and make stable (‘well-order’) a just international ‘society’” (Simmons (2010), p. 10). Simmons identifies two corresponding parts of nonideal theory—one addressing failures of domestic institutions to live up to the ideal of justice, and the other addressing failures of societies to live up to the requirements of ideal international justice. Finally, Simmons, like others, notes “that there is third part of ideal theory [only briefly alluded to by Rawls] plainly requiring a corresponding third kind of nonideal theory.” 5 This third part of ideal theory will specify a set of ideal (moral) principles to govern individual conduct, and will have its counterpart in nonideal theory that will specify the requirements of individual conduct in nonideal conditions. In each “domain” the nonideal theorist will be confronted both by problems that arise from partial compliance and problems concerning the transition from a less-just to a more-just condition. Some further clarification is warranted with respect to transitional theory in particular. Perhaps most important is Gopal Sreenivasan’s observation that transitional justice, “as it emerges on the second branch of Rawls’ division, means something quite different from what it means in the political science literature burgeoning under the same name” (Sreenivasan (forthcoming)). Basically, issues of “transitional justice” or “transitional theory” in the context of nonideal theory are distinct from discussions of the aftermath of regime change that often go by the same name.6 As I understand it, the main areas of concern within transitional theory are fourfold: (1) costs (2) unfavorable conditions (3) feasibility and (4) path-dependence. I take them up in turn. 2.1 Colin Farrelly rightly observes that in the nonideal world, protecting even the most basic negative rights costs money. This means that trade-offs between these and other primary goods will need to be made. Sreenivasan notes that beyond the costs of merely maintaining the various agencies required to administer services in a just society—such as courts, police, environmental protection, social services, national defense, etc.—there will also be one-time start-up costs associated with making the 2 The grievousness of an injustice is said to be a matter of the extent to which a given institution deviates from perfect justice. It is conceivable that a given reform my make a society more just right now, but will not be consistent with perfect justice-or may even impede its realization. Rawls, TOJ, p. 216. 3 For instance, Sen (2009) criticizes this approach in great detail. Sreenivasan (2007) also suggests an alternative conception of the relationship between ideal and nonideal theory. 4 It is not my intention to present an unorthodox interpretation of Rawls. Instead, I try to work within established interpretations of Rawls’ work as articulated by Simmons (2010) and Moellendorf (1996), among others. Any errors in this regard are purely my own. 5 See Simmons (2010), p. 11, and also Phillips (1985), p. 553. 6 Therefore, questions of retributive and restorative justice that are appropriate to this different notion will not be considered here. Burdened Societies and Transitional Justice 371 theory, then, specifies the end-point or long-term goal to be achieved. The job of nonideal theory is to work toward the ultimate goal of achieving perfect (domestic or international) justice by identifying and proposing solutions to existing injustices, beginning with those that are “most grievous,”—keeping in mind that no single reform should ever prevent us from reaching the end goal.2 This picture has been criticized, but I will not dispute it here.3 Instead, the traditional model will form the background for my discussion. For the sake of argument, I will also assume that a just society is one that approximates Rawls’ conception of justice as fairness.4 Author's personal copy transition from a society without the required institutions to one that has them.7 These are in addition to the personal cost to individuals of bringing their behavior, attitudes and expectations into line with what is required by the new institutions, when they have been socialized under vastly different conditions. Transitional theorists must attempt to assess the costs of various alternative sets of improvements in terms of their efficiency, effectiveness and also in terms of the size and distribution of burdens that they impose, both over time and across various affected groups. 2.2 Since ideal theory assumes reasonably “favorable” or “fortunate” conditions, we should expect that nonideal theory will take into account various types of human misfortune. These misfortunes are the types of occurrence which make it difficult for a given society to achieve justice. Unfavorable conditions can be historical, economic, or social, and we can understand them as largely “inevitable” and “unfortunate” rather than as coming about as a result of deliberate or “clearly avoidable” acts or policies (Simmons (2010), p.15.) Examples of unfavorable economic conditions might be extreme poverty (perhaps due to environmental disasters or lack of development), lack of natural resources, lack of technology or an unworkably low number of working-age citizens (perhaps caused by disease). Social conditions that might be considered unfavorable are certain cultural or social attitudes or norms that make more just conditions undesirable to much of the population (for instance, racism, sexism, ethnic tensions or prejudices, adaptive preferences, a culture of corruption and bribery, or lack of political will). Examples of unfavorable historical conditions might be: deep social or political divides that make cooperative and coordinated reforms very unlikely, or the negative after-effects of colonialism or war. While the precise combination of difficulties will be different for each society, and so no “one-size-fits-all” theoretical solution is possible, there is nevertheless a role for the nonideal theorist to play in developing principles by which we might better understand the most pervasive of these problems, and evaluate proposals for overcoming them. 2.3 When we theorize about the political feasibility of particular reforms we are essentially concerned with the obstacles that stand in the way of realizing a given ideal of justice. While non-compliance and unfavorable conditions are clearly related to feasibility, the focus here is on certain kinds of obstacles. Pablo Gilabert describes these obstacles as depending “not only on general facts about human psychology and social organization, but also on the specific ways these operate in different cultural and institutional settings” (Gilabert (2008), pp. 415). So, concerns about feasibility are, in large part, concerns about those obstacles to realizing justice that arise from human nature generally, as well as specific social contexts. Gilabert notes that arguments about feasibility are highly uncertain and imprecise. He also observes that what is feasible in a given society is “a moving target” since [t]here are moments in which political agents (including, in particular, inspired leaders) reshape and correct perceptions of what is politically possible by pursuing ambitious projects which turn out to be realizable despite initial general dismissal (Gilabert (2008), pp. 416–417). Some reforms may be infeasible due to the kinds of unfavorable historical or social conditions mentioned earlier, or they may arise from brute facts about human nature, such as the desire for power and wealth, or instances of irrationality. Worries about feasibility—when they are plausible—function as reasons why a given society may 7 See Farrelly (2007), p. 850, and also Sreenivasan (forthcoming). 372 L.L. Fuller Author's personal copy not be required to pursue a given course of reform toward a more just society. They act as a kind of excusing condition vis-à-vis certain desirable reforms. However, a plausible argument that a certain reform is infeasible must walk a difficult line, since treating too many elements of the status quo as fixed results in what Gilabert calls “conservative narrowness” while taking too few elements to be fixed results in recommendations that are mere fantasies. 2.4 The term “path dependence” is sometimes used in reference to problems associated with “how to get there from here”, that is, the route a society should take from where it is now to a situation where it approximates the relevant ideal of justice. Simmons notes that if we understand nonideal theory as fundamentally oriented toward achieving a particular end-point, we can see that the best path might in fact lead us to endorse a set of policies or actions that are less just than the status quo, in order to better facilitate arriving at the correct end-point. His basic argument is that we may need “to take one step back in order to take two steps forward,” and that this would be in line with Rawls’ understanding of the role of nonideal theory (Simmons (2010), p. 23). Upon reflection, we can see that not only is Simmons correct that the route to a perfectly just society is not necessarily linear; it is also likely that there are multiple routes that could be taken to the same end-point. Different routes will vary according to their costs, their general moral permissibility, the length of time they will take to implement, and the likelihood of their being realized (rather than thwarted mid-way). I take it to be a central task of nonideal theory to try to imagine and evaluate these routes. Such an evaluation should take account not merely of the efficiency and certainty of a given set of policies in arriving at the end-point, but also the processes that would be utilized to arrive there. As Ingrid Robeyns points out, not all processes are created equal. We might prefer processes that “are respectful and democratic” for instance, over those that are not (Robeyns (2008) p. 350). 3 Burdened Societies Having outlined the key issues facing transitional theory more generally, I’ll now briefly outline Rawls’ concept of a burdened society. Rawls’ general objective in Law of Peoples is to specify an ideal conception of international justice, that is, rules that specify the proper conduct of domestically just states with respect to one another. He generates these rules through a version of the “original position” in which representatives of domestically just peoples come together to deliberate about fair and reasonable rules to regulate their interactions behind a “veil of ignorance.”8 The result of these deliberations, he claims, would yield eight principles of international justice—the details of which need not detain us here—except to note that the sixth and eighth principles require that “Peoples are to honor human rights,” and already well-ordered peoples “have a duty to assist other peoples living under unfavorable conditions that prevent their having a just and decent political and social regime” (Rawls LP, p. 37). Rawls also emphasizes that the “aim of the Law of Peoples will be achieved when all societies have been able to establish either a liberal or a decent regime, however unlikely 8 In this version of the original position, “rational representatives” of peoples are “guided by appropriate reasons,” “situated symmetrically,” and do not know “the size of the territory, or the population, or the relative strength of the people … who they represent… the extent of their natural resources, or the level of their economic development.” Rawls, LP, pp. 32–33. Burdened Societies and Transitional Justice 373 Author's personal copy that may be” (Rawls LP, p. 5). As noted above, for the purposes of this paper, I take liberal societies to be those which approximate Rawls’ model of justice as fairness. By contrast, “decent peoples” are considered well-ordered even though they depart from the ideal model of a liberal society. Decent societies are organized around a particular (religious or secular) conception of the good, and place “duties and obligations on [their] members on the basis of this conception” (Moellendorf (1996) p. 144). John Tasioulas nicely characterizes their additional characteristics as follows: A decent society “has no aggressive aims in foreign policy and respects the political independence of other societies,” “it has a ‘common good idea of justice’ that secures the human rights of all its members, though not necessarily as extensively or as equally as they would be in a liberal society; its legal system imposes bona fide moral duties, beyond those arising from human rights, on all persons within its territory,” and “it affords its members a level of political participation that, although not democratic, suffices to underwrite meaningful collective political self-determination” (Tasioulas (2005), Note 6, p. 15). For simplicity I will assume that decent societies roughly resemble the example Rawls discusses, that is, I will assume that they are “decent consultation hierarchies.”9 Importantly, a decent society is only required to fulfill a minimal schedule of human rights, which includes an entitlement to the basic means of subsistence and physical security, a right to personal property, a right to “sufficient liberty of conscience to ensure freedom of religion and thought” and a right to formal equality understood as the requirement that like cases be treated alike (Rawls LP, p. 65). This minimal schedule does not include fully equal freedom of religion, freedom of speech, or the right to democratic participation. Burdened societies must take steps towards becoming (at least) decent societies, where this is possible. However, they are not required to proceed unaided. In the final section of Law of Peoples, Rawls explains that well-ordered societies have a duty to assist burdened societies so that they may become either liberal or decent. This duty is not an open-ended or ongoing duty of distributive justice. It is merely a duty to assist burdened societies up to the point where they become well-ordered. Thus, the duty has a “cut-off point” after which assistance may be discontinued. Rawls gives some general guidelines that should inform the way the duty of assistance is carried out. First he notes that, The levels of wealth and welfare among societies may vary, and presumably do so; but adjusting those levels is not the object of the duty of assistance. Only burdened societies need help. Furthermore, not all such societies are poor, any more than all well-ordered societies are wealthy. A society with few natural resources and little wealth can be well-ordered if its political traditions, law, property and class structure with their underlying religious and moral beliefs and culture are such as to sustain a liberal or decent society (Rawls LP, p. 106). Next he notes that when carrying out the duty of assistance, well-ordered societies should realize that burdened societies really need help changing their “political and social cultures” rather than simply help amassing more funds (Rawls, LP, p. 108). He cites Amartya Sen’s work on famines as evidence that lack of access to the means subsistence usually results from a failure of political and social institutions rather than mere lack of food or funds. While Rawls asserts that “money is often essential” he also maintains that merely 9 Technically not all liberal societies must endorse justice as fairness, and not all decent peoples will necessarily be decent consultation hierarchies. On Rawls’ account, these are each merely one of the possibilities within each category. (I thank Jon Mandle for bringing this point to my attention). 374 L.L. Fuller Author's personal copy “throwing funds” at the problem is “usually undesirable”, and that “there is no society anywhere in the world—except marginal cases—with resources so scarce that it could not, were it reasonably and rationally organized and governed, become well-ordered (Rawls, LP, pp. 108–109, 110, 108). Finally, well-ordered societies must not act paternalistically in carrying out their duty of assistance, since the goal is that burdened societies will become equals with other well-ordered societies and be able to freely manage their own affairs.10 Rawls locates his discussion of burdened societies within nonideal theory. However, it has been observed by Laura Valentini that Rawls only partially “enters” the nonideal world here. He gives up the assumption of favorable conditions but retains the assumption that there are some reasonably well-ordered societies in existence and that they will (at least sometimes) comply with their duties of assistance. He also seems to overlook the fact that the societies meant to be providing assistance might be implicated in worsening the condition of burdened societies in various ways, such as trading under an unjust international scheme of rules (Valentini (2009), p. 348). I agree with Valentini that failing to bring in these types of real-world considerations “distorts the reality of contemporary international relations,” (Valentini (2009), p. 348). Therefore, I will now attempt to describe the situation of burdened societies in a manner that reflects, at least to some extent, their characteristics and political position in the context of current international relations. It is my view that we will likely make more progress identifying satisfactory transitional principles if we at least try to describe burdened societies in a way that takes account of the kinds of injustices that commonly occur both within and between societies. Basically, in order to develop an account of transitional justice, we need to have some sense of what burdened societies look like in real life. Accordingly, for my purposes here burdened societies are to be understood as having (at least some of) the following sorts of difficulties and characteristics: & They are not aggressive toward other societies, and their governments do not deliberately violate or disregard the human rights of their people. Nevertheless, they do not presently have the ability to fulfill the human rights specified as necessary for becoming even a decent society. & Their economy, for various reasons having to do with technological expertise, lack of education, unfair international rules or agreements, etc., is not as strong as it could be. That is, they make some money but not enough to simply create government entitlements to fulfill the human rights that remain unfulfilled. & A good number of their citizens are very poor-they live at or below the $2-a-day international poverty line.11 Members of the same group are often illiterate and have little formal education. & The society is highly stratified, since there are also very wealthy, educated and powerful individuals within it. (I take this to be true of almost all societies). & Some (but not all) broadly liberal societies provide burdened societies with either bilateral, multilateral or NGO aid, which takes various forms: technical expertise on finance, governance and other policy issues, food, medical and educational programs, etc. However, these programs are only weakly regulated, and the terms are largely 10 Rawls, LP, p. 111. Also note that the use of force as part of an effort to assist burdened societies is ruled out since the societies providing aid are well-ordered and so do not act aggressively towards other societies. 11 The World Bank estimates that 2,561.5 million people lived below the $2 a day poverty line in 2005; As such, it is not an unreasonable assumption that most burdened societies would have a substantial percentage of people living at or below this line. See World Bank, “Prospects for the Global Economy” (2009) at http:// siteresources.worldbank.org/INTGEP2009/Resources/10363_WebPDF-01Chapter1-w47.pdf . Burdened Societies and Transitional Justice 375 Author's personal copy http://siteresources.worldbank.org/INTGEP2009/Resources/10363_WebPDF-01Chapter1-w47.pdf http://siteresources.worldbank.org/INTGEP2009/Resources/10363_WebPDF-01Chapter1-w47.pdf dictated by the donors. Here we are assuming that only some percentage of well-ordered societies will comply with the duty to assist burdened societies. This is one of the unfavorable conditions in which they find themselves. & Their society has its share of historically disadvantaged groups and so will have to contend with some (if not all of) sexism, racism, homophobia and discrimination against certain ethnicities or members of certain religions. & They may also have to contend with their members’ adaptive preferences, and/or corruption among civil servants, police officers and other people in a position to wield power over others. & They may have weak bargaining power in the international arena. & As in Rawls, they may not already have in place a legal structure guaranteeing the full and equal civil and political liberties that are in place in most liberal democracies. They may be proto-democracies (that is, they officially have the form of democratic government but their elections and other practices are not up to par). Alternatively, they could be proto-decent-consultation-hierarchies (in which they have a political system based on a particular conception of the good but that is also somewhat tainted by considerations of personal interest, say, among elites and the military). Here then, is a significantly more detailed picture of burdened societies than the one given by Rawls. What we now want to know is how such a society should approach the task of making itself more just. We shall suppose for the sake of argument that a particular society accepts that it must take steps toward becoming either liberal or decent (as set out by ideal theory). What should this society—including its government—do now? Is there anything general we can say about the way it ought to proceed? It is to these questions that I now turn.12 4 What Would Rawls(ians) Say? Rawls himself says relatively little on the subject of how a society should handle the transition from a less-just to a more-just condition. Much of what he does say appears in Theory of Justice, and so presupposes that the ideal to be reached is a liberal democratic society rather than merely a decent society. As we have seen, however, some burdened societies may only aim to become decent given their particular comprehensive doctrines. These considerations lead me to divide this section into two distinct parts. First, I will do my best to give a broadly Rawlsian account of transitional justice where the goal is to become a perfectly just liberal democratic society. Next, I attempt the same task where the goal is to become a decent society. In both cases I assume that the society begins as a burdened society, and so it has, roughly, the characteristics and difficulties described in the previous section.13 12 I do not mean to suggest that these questions are ones Rawls either did, or should have, addressed. They are not his questions, since he was mainly concerned with outlining a duty of assistance on the part of well- ordered societies. However, note that the answers to these questions must be part of a complete transitional theory, especially given that the duty of assistance is meant to be non-paternalistic and so its content must be in large part concretely determined by burdened societies themselves. (Many thanks to Steve Vanderheiden for this last observation). 13 Of course, transitions from a less-just to a more-just condition will be necessary for societies that are not “burdened” but are still unjust. Whether a society is burdened or not depends on the extent of the unfavorable circumstances in which it finds itself, particularly economically. Societies that are not burdened, but still have some injustices to correct, should be understood to be unjust societies in favorable conditions. 376 L.L. Fuller Author's personal copy 4.1 Transitioning to a Liberal Democratic Society The Rawlsian model of justice as fairness specifies that a perfectly just liberal democratic society is one where the basic structure is organized in conformity with the following two principles: 1. Each person is to have an equal right to the most extensive total system of equal basic liberties compatible with a similar system for all [the Liberty Principle]. 2. Social and economic inequalities are to be arranged so that they are both: (a) to the greatest benefit of the least advantaged … [the Difference Principle] and (b) attached to offices and positions open to all under conditions of fair equality of opportunity.14 The first principle is lexically prior to the second, and fair equality of opportunity is lexically prior to the difference principle. Further, for Rawls the “ranking of the principles of justice in ideal theory reflects back and guides the application of these principles to nonideal situations. It identifies which restrictions need to be dealt with first” (Rawls TOJ, p. 267). Clearly he thinks that guaranteeing the basic liberties should be the first priority of unjust societies. Reforms that restrict basic liberties are permissible, but only for the sake of liberty, which, on Simmons view, means that “restrictions on liberty are acceptable only if they are part of a process aimed at achieving the fullest possible societal justice” (Simmons (2010), p. 15). For instance, liberties such as free speech or association can be limited or made unequal if this is needed for the general order of society and the security of citizens, provided that these are temporary measures, and are aimed at the end of perfect justice. On Simmons’ interpretation, such temporary measures “do not really count as injustices” so long as they are “motivated by and directed toward” the correct end (Simmons (2010), p. 16). Thus, the priority of liberty is the main principle of transitional justice that Rawls spells out in Theory of Justice. However, Rawls also notes that “in the more extreme and tangled instances of nonideal theory… [the] priority of rules will no doubt fail; and indeed, we may be able to find no satisfactory answer at all” (Rawls TOJ, p. 267). Simmons comments that for those unlucky societies where social conditions are such that the basic liberties cannot be established, Rawls favors a more piecemeal or “relaxed” approach to transitional issues. Simmons observes, In this kind of case—Rawls appears to be thinking, primarily at least, of cases involving severe societal poverty—“one can concede their [i.e. rights] limitations,” but only where this limitation is “necessary to prepare the way for a free society by “raising the level of civilization” … In this case, Rawls suggests, society should be governed primarily by the general conception of justice, which lacks the lexically ordered principles of ideal theory, until it becomes reasonable, as a result of “improvement in economic well-being” to be guided by the two separate principles of ideal theory in their serial order. The general conception dictates that over the long run lesser liberties for all must benefit all, while unequal liberties must benefit those with lesser liberties (Simmons (2010) pp. 13–14). Rawls also allows that in such dire conditions, policies which violate fair equality of opportunity or depart from a system of equal basic liberties, “have the right form when they 14 Rawls, TOJ, p. 266. The basic liberties Rawls lists as “important” are the right to vote and hold office, freedom of speech and assembly, the right to bodily integrity and freedom from psychological oppression, freedom from arbitrary arrest and seizure, freedom of conscience and thought, and the right to hold personal property. Rawls, TOJ, p. 53. Burdened Societies and Transitional Justice 377 Author's personal copy claim … that the opportunities of the least favored sectors of the community would be still more limited if these inequalities were removed” (Rawls TOJ, p. 265). Some inequalities can be tolerated in very unfavorable conditions, provided that it makes sense to leave them in place from the perspective of the worst-off. Now, we might well think that on Rawls’ account burdened societies are extreme cases in which the priority of the liberty principle does not apply, and/or societies in which the basic liberties are not realizable. The problem with conceiving of burdened societies on this more “relaxed” model is that, as we have seen, Rawls thinks that virtually all existing societies have enough resources such that they can become well-ordered, and so must realize the most central liberties first where this is feasible (Rawls TOJ, pp. 217–218). Basically, while in Theory of Justice he admits the conceptual possibility of a society that could not effectively establish basic rights, in Law of Peoples he seems to suggest that there is no real society that is in so dire a condition. Of course, whether or not any society is sufficiently impoverished that it cannot effectively establish the basic rights is an empirical question. From a theoretical perspective then, the more “relaxed” approach remains an option for societies in this predicament. So it looks like burdened societies still must assign lexical priority to the basic liberties, at least when this is not completely infeasible. Presumably, where this is completely infeasible in a given set of circumstances, all possible “lateral moves” should be pursued that would put a burdened society in a position to apply the priority-of-liberty rule. This would mean that where a society can move from one unjust condition to another that is equally unjust but that opens up feasible reforms that would bring them closer to realizing the basic liberties, then they should do so. This is because the priority-of-liberty rule effectively establishes the “grievousness” of any particular injustice, and so the urgency with which a society should respond.15 Beyond what can be extrapolated from Rawls emphasis on the priority of the basic liberties, we are left with his three further requirements: that transitional policies and courses of action must be (i) morally permissible, (ii) politically possible, and (iii) effective. We are unfortunately left wondering both how we are to interpret each requirement, and how we are to weigh them against one another, since Rawls does not tell us. Of the three requirements, moral permissibility seems the most problematic to interpret. Simmons provides three plausible understandings: a. Policies are morally permissible when they do not require or condone clearly immoral acts by public officials in order to be instituted. b. Policies are permissible when they do not require a more grievous injustice to be instituted in order to remedy a less grievous one. c. Policies are permissible when they do not impose excessive costs or burdens on individual citizens in order to be instituted, as in the case where people suffer substantial losses because they have organized their affairs around the previous (less just) set of institutions and expectations (Simmons (2010), p. 21). It seems to me that a broadly Rawlsian account might endorse any and perhaps all of these understandings depending on the circumstances, and I will adopt these requirements as part of my own approach. It also seems that if we combine Rawls’ emphasis on the priority of liberty with the guidelines for assistance to burdened societies, we can infer that 15 By lexically ordering the principles, Rawls establishes that violations of the first principle are more grievous than violations of fair equality of opportunity or the difference principle, since the priority is based on the importance of liberties to deliberators in the original position. 378 L.L. Fuller Author's personal copy burdened societies will accept aid that will help them guarantee basic liberties, except when the assistance is paternalistic in nature, since this can undermine the point of the assistance. The difficulty here is that in a genuinely nonideal world, much assistance is likely to be paternalistic in nature, or at least have some paternalistic conditions attached to it. Thus, it looks like there is a tension between the requirement that a burdened society work to become more just domestically and the requirements of international justice. Farrelly argues that there are deep difficulties with a model that assigns lexical priority to the liberty principle, and I tend to agree with him. His main argument is that Rawls fails to take the fact of scarcity seriously. He notes that even in a reasonably well-off society, guaranteeing basic liberties could turn out to be “a never-ending aim” since it is possible for a society to allocate all their wealth to making streets and borders secure, ensuring food and drugs are safe, protecting citizens from police brutality, etc. (Farrelly (2007), p. 852ff). In essence, Farrelly observes that trade-offs will be necessary in every society, and so the lexical priority of basic liberties over opportunities and other social and economic entitlements is unhelpful. He also questions Rawls’ contention that trade-offs between liberties and social and economic advantages are impermissible even when the advantages go to the same individuals whose liberties are restricted. It certainly seems plausible that citizens in burdened societies might sometimes choose to have fewer civil liberties in return for more economic advantages, if trade-offs must be made. And there is no distributive difficulty here when it is the same individuals who bear the burdens and obtain the benefits. The problem here is that people may prefer to guarantee fewer liberties, or leave them less secure, in order to allocate funds to other areas, even when total funds are sufficient to guarantee all the basic liberties but not much else in addition. For many burdened societies, scarcity is likely to be a much larger concern than for societies in more favorable conditions. This means that it is more likely that there will be situations in which guaranteeing civil liberties leaves very few resources left over to institute the difference principle and fair equality of opportunity. As such, opting for the lexical priority of the liberty principle seems to unfairly disadvantage those people most likely to desire trade-offs in favor of more economic advantages, namely, the very poorest in society.16 A more flexible approach to both the kind of protections and benefits that will be prioritized as well as who will receive them seems desirable, given that there is no choice but to make difficult trade-offs. As such, assigning lexical priority to basic liberties does not seem like a viable overall strategy for burdened societies—given their unfavorable conditions—even if the society in question did aspire to become a liberal democracy. 4.2 Transitioning to a Decent Society As we have seen, some societies may not aspire to become liberal, but instead may merely aim to become decent, such that they may be admitted into the Society of Peoples. Since those in a decent society do not embrace a liberal conception of justice, but instead organize their society on the basis of a particular comprehensive doctrine and a “common good idea of justice,” they are not required to institute the full range of basic liberties. Hence, they also are not required to proceed towards their goal by prioritizing these liberties. Instead, they 16 To be fair, Rawls allows that “the first principle covering the equal basic rights and liberties may easily be preceded by a lexically prior principle requiring that citizens’ basic needs be met, at least insofar as their being met is necessary for citizens to understand and to be able to fruitfully exercise those rights” (Rawls (1996), p. 7). Rawls is not here recommending that people prioritize the protection of civil liberties over their very survival. Burdened Societies and Transitional Justice 379 Author's personal copy must work towards a situation in which they can guarantee the fulfillment of a minimal list of human rights (rights to subsistence and security, freedom from slavery, serfdom and forced occupation, freedom of worship, formal legal equality and the right to personal property), and are subject to the three requirements of moral permissibility, political possibility, and effectiveness (Rawls LP, p. 65). They must accept assistance in this effort, (when this assistance is not paternalistic), especially in changing their political and social culture, since Rawls claims that “the causes of the wealth of a people and the forms it takes lie in their political culture and the religious, philosophical and moral traditions that support the basic structure of their political and social institutions, as well as the industriousness and cooperative talents of its members” (Rawls LP, p. 108). The specific elements of the political and social culture he describes as crucial to a society’s ability to become both wealthier and more decent are as follows: the political culture and civic society of the country, it members’ probity and industriousness, their talent for cooperation, their capacity for innovation, and their population policy. Rawls admits that there is “no easy recipe” for providing assistance that will help a burdened society make the required changes, and so perhaps this is why he does not attempt to prescribe which human rights should rank highest in terms of priority. He also does not indicate how burdened societies should go about realizing their aims, although he observes approvingly that educating women has been shown to be a good, non-coercive way to institute population control (Rawls LP, p. 110). Rawls also agrees with Henry Shue that subsistence rights are basic, in the sense that the “sensible and rational exercise of all liberties … as well as the intelligent use of property, always implies having general all-purpose economic means” (Rawls LP, Note 1, p. 65). We might infer from this that guaranteeing subsistence rights should be the top priority for societies aspiring to become decent, which seems like a reasonable recommendation for burdened societies. This general approach to becoming a decent society can be summed up as follows: Starting with subsistence rights, a burdened society should implement effective policies in morally permissible ways that are ultimately aimed at fulfilling a minimal schedule of human rights for its people. It also should accept non-paternalistic assistance in this effort from well-ordered societies. This approach looks more promising for burdened societies than the former one. However, it may be Rawls’ position that beyond this point we simply have to employ our intuitions and arguments in reference to the concrete situation of each real society; Perhaps he would assert that there is nothing more that we can say, in general, about how burdened societies should make the transition to decent or liberal societies. I disagree. In the next section, I attempt to modify and flesh out this second approach in more detail. My hope is that this modified approach will provide resources for both prescribing and evaluating the transitional efforts of actual societies approximating the condition of burdened societies. A central advantage of the approach is that it will be able to provide some direction to the well-ordered societies carrying out their duties of assistance with regard to which societies they ought to support. Where not all burdened societies can be assisted (or at least not to the same degree), my account provides a guideline for identifying those societies that are not only moving in the right direction, but also doing so in the right way, and so are most worthy of outside assistance. 5 Transitional Justice for Burdened Societies: A Proposal The approach I sketch out here is arguably Rawlsian “in spirit” since Rawls’ notion of the “worst-off representative person” is key to the structure of the model. Before I can describe 380 L.L. Fuller Author's personal copy my approach in detail however, I need to explain how I think this notion should be understood in a nonideal context, and why. A difficulty for both accounts of transitional justice discussed above and also for many accounts of human rights is that they are too neat. No doubt for conceptual and explanatory purposes, they treat rights as distinct entities and assess injustice on the basis of the extent to which a society fails to fulfill each particular one. Even when it is recognized that guaranteeing a right entails both “negative” and “positive” duties (such as setting up social institutions) this kind of approach leads us to think of rights as separate from one another in an artificial and misleading way. 17 This merely conceptual separation, in turn, causes us to think in terms of fulfilling human rights by first completely fulfilling the most morally important right (or set of rights), and then the next most important, and then the next, and so on. The problem with this is that injustice is not typically experienced and embodied as a set of distinct rights violations or lack of rights fulfillment, each one having little influence on the others. Instead, a victim of injustice is usually subject to a confluence of factors that produce a very particular condition of vulnerability. It is often observed that when human rights are fulfilled they are mutually reinforcing, and so it stands to reason that injustices are too. It is also not news that the way a black woman experiences poverty in a racist and sexist society is not the same as the way a white man experiences it. Nevertheless, I think the significance of these observations for a theory of transitional justice has not been properly appreciated. Paul Farmer suggests that insofar as we are concerned with alleviating injustice, we need to “identify the forces conspiring to promote suffering, with the understanding that these are differentially weighted in different settings” (Farmer (2010), p. 344). He furthers asserts that if we can identify these forces, then “we stand a chance at discerning … the forces that put some at risk for human rights abuses, while others are shielded from risk” ” (Farmer (2010), p. 344). The notion of multiple, intersecting “vectors” of injustice is illuminating. The forces, or vectors that he specifically mentions are: race or ethnicity, gender, poverty, immigration status (such as refugees, illegal aliens or stateless people), and sexual orientation.18 He notes that being subject to more than one such vector increases a person’s overall vulnerability, and those people who are not subject to them have certain characteristics that serve a protective function. For instance, being white can protect a young male from being regularly stopped and searched (for largely opaque or spurious reasons) by the police in most US cities. Even if a white man is identically placed with respect to other characteristics (such as income level) his whiteness serves to protect him from this particular sort of injustice. Farmer argues that no single factor can fully capture the level of vulnerability of a particular person. Poverty, for instance, might cancel out many of the protective effects of characteristics such as being of the dominant race, religion or sexual orientation in your society, but these factors are still relevant to where you are located within the social hierarchy. So how are these vectors of injustice relevant to issues of transitional justice? This understanding of how injustice manifests itself suggests that we need to alter the way we think about the worst-off person in a burdened society. What we need is a nonideal account of the “worst-off representative person.” First, we should recall that the society in question 17 In his classic book Basic Rights: Subsistence, Affluence and US Foreign Policy (1980) Henry Shue argues persuasively that the distinction between negative and positive rights is neither sharp nor significant. 18 Clearly there are vectors other than these. Also, “vector” is my terminology. Farmer prefers the term “axis”. He also cites some material from liberation theology in this connection, but it should be understood that I am not invoking that tradition here. See Farmer (2010), pp. 337–341. Burdened Societies and Transitional Justice 381 Author's personal copy is already highly stratified, and so there are people highly placed in the social hierarchy whose human rights are already fulfilled due to the protective characteristics they possess, such as great wealth, political influence, a good education, etc. Next, rather than understanding the worst-off representative person—as Rawls does in Theory of Justice— as someone whose basic rights are, in large part, already guaranteed but who must authorize any departures from equality in the social and economic realms, instead we should understand the worst-off person as the person who is living at the intersection of the most numerous/weightiest vectors of injustice in her society. This is the person (or group of people) whose unfulfilled rights should structure efforts to make the society more just. By identifying this representative person, we identify the problems that should be prioritized, since these people are suffering the most from injustice. And in a society where trade-offs must be made, policies should be organized with a view to eliminating the difficulties that typically threaten the lives and livelihoods of the worst-off people. In particular, policies should be organized such that the common threats and difficulties that distinguish them from the next-worst-off group should be the first problems that are tackled in the context of very limited resources. My suggestion here is that transitional justice requires burdened societies to prioritize a group of people and their urgent, concrete vulnerabilities, rather than their “rights” in some abstract and comprehensive sense. Clearly, much social scientific work will be needed in order for the relevant vectors in any particular society to be correctly identified and weighted. But once we see that groups of people within each burdened society can be ranked from the worst-off to the best-off on the basis of how they fare with respect to the vectors, then we can see whose lives are the most marred by adverse events that occur due to the mutually- reinforcing multiple injustices that they suffer. And we can propose solutions to their problems that take account of the fact that multiple rights violations are experienced by people as a single condition—a condition in which they are exposed to various harms and setbacks for a host of related reasons. Consider the following example: “B” grows up in a rural area, where most people engage in subsistence farming for a living, but the yields are not adequate to support everyone in his family. Once he is a teenager he moves to the city to find work and to relieve his family of the burden of providing for him. However, since he belongs to a minority group that is not universally accepted by employers of a different ethnicity, and because has no formal education, he is only able to secure a very low-paying job. This means he must live in a shantytown or slum outside the city center, where he does not have to pay rent. But since he is, legally-speaking, a squatter on the land where he lives, he is subject to repeated harassment and abuse by the police.19 This, in turn, causes him to miss work sometimes and to be unwell. He is then not always able to make enough money to purchase food that can provide adequate nutrition, which makes him worse-off still… Assume for the sake of argument that B is part of the worst-off group in his society. We can see that there are many ways in which his life can be made more secure from physical abuse and injury, and his ability to provide for his own subsistence can also be improved (by for example, making his living arrangements legal). We can also see that solutions to those problems that are specific to people in the worst-off group are not going to require that the society institute a comprehensive system of protections against assault, hunger, sickness etc., since many people in his society are already protected from these threats anyhow. Solutions to B’s specific set of vulnerabilities should be tailored to the risks that 19 Many of the global poor rank ending police brutality and harassment very highly on a list of improvements they would like to see in their lives. See Narayan et al. (2000), pp. 35–36 and 27. 382 L.L. Fuller Author's personal copy those in his group actually face. This means that these solutions will not necessarily require fully guaranteed rights in the sense set out by Shue, in which a right is only guaranteed when the relevant agents have fulfill the following three kinds of corresponding duties: (1) the duties to avoid depriving someone of their right, (2) duties to protect from deprivation, and (3) duties to aid those who have been deprived of the exercise or object of their right (Shue (1980), p. 52). Instead, sometimes a given source of vulnerability can be most effectively, and/or most feasibly be eliminated or substantially diminished by instituting a policy that does only one (or two) of the three. This is the correct strategy in terms of effectiveness because the way to genuinely improve the circumstances of the worst-off representative person is by taking a holistic approach, that is, by instituting policies that are aimed at the set of factors which make him worse-off than the next-worse-off representative person. Simply securing one right in full leaves this person open to it being sabotaged by the difficulties and adverse events correlated with other rights violations. For example, say a society-wide policy of improved policing and other measures are put in place to guarantee everyone’s security of the person, or freedom from physical assault. If someone cannot find work and must resort to illegal activities (prostitution, selling drugs, trafficking in stolen goods) in order to survive, then that person’s vulnerability to physical assault (from pimps, rivals, etc.) is simply “driven underground” rather than eliminated. This demonstrates why policy-makers should take a coordinated approach to making the worst-off group better off, which means doing what is necessary to bring the worst-off people up to the level of the next worst-off group. Of course, once this has been accomplished, it will be necessary to begin addressing the problems of that group, since its members are now the worst-off (and so on up through the social hierarchy). I should emphasize at this point that when I suggest that policies should address the common threats and difficulties experienced by the worst-off group of people, I do not mean that they should somehow be protected from all possible threats and vulnerabilities. Instead, I have in mind a notion similar to Shue’s notion of “standard threats,” which he defines as “ordinary and serious but remediable” (Shue (1980, p. 32). Protecting someone from “standard threats” according to Shue, is a matter of providing social guarantees (in the form of policies, legislation and social programs) that vastly reduces or eliminates their vulnerability to predictable difficulties that are susceptible to determinate (and not unreasonably costly) solutions. He rightly observes that no one is entitled to protection from “every conceivable threat,” or “ineradicable threats” such as untreatable serious illnesses, accidents, or death, since this would be impossible.20 What count as standard threats for any particular group will be determined by social scientific study, but these threats will no doubt show up in answer to questions such as: “What are the common causes of death in this group?” and “How do members of this group typically become deprived of the means of subsistence?” Having set out the basic framework, I can now comment on its justification. If we consider Rawls’ requirement that transitional reforms be morally permissible, we can see that prioritarian arguments give this approach a solid moral grounding. Dennis McKerlie characterizes prioritarianism as follows: [Prioritarianism] holds that, at least in general, badly off people have stronger claims than better off people. … It does not claim that as a general rule we can give the 20 Shue also notes that the “mixture of the analytic and the empirical in an element like standard threats is obviously difficult to characterize with any precision. …But which threats are pervasive, which are serious, and which can feasibly be resisted must be discovered from particular situations,” (Shue 1980 Note 22, p. 184). Burdened Societies and Transitional Justice 383 Author's personal copy 384 L.L. Fuller badly off larger benefits, it claims that benefits for them are morally more important than benefits for better off people. So this view can say that it is more important to give a smaller benefit to someone badly off than a larger benefit to someone better off (McKerlie (1994), pp. 25–26). Not only is it morally permissible on this view to help the worst-off, it is morally more important to do so. Provided that the reforms do not call for the unjustified use of coercive force for their implementation, it is plausible to think that addressing direct harms to the worst-off that result from injustice is a morally sound method by which to tackle injustice generally, since inevitably some criterion must be used to prioritize reforms. For those not persuaded by a prioritarian view, there is an alternative justification to consider. We can imagine a kind of nonideal “original position” in which all the facts about a real burdened society are known to the deliberators, but they remain ignorant of the characteristics they will have once they join the society. If asked—before being transformed into the selves they will be in society—which injustices they would choose to remedy if they could not remedy them all, a maximin strategy would suggest that they would choose the injustices that affect the worst-off people, since they could turn out to be those people. Even if the improved condition of those people is still not very good, they will probably prefer to minimize risk, since the risks to the worst-off people in a burdened society are very great, and they include many risks to survival itself. Finally, this model conforms to the requirements of moral permissibility introduced earlier, since it starts with the most grievous injustices—understood as the condition of injustice the worst-off experience as a result of the intersecting vectors to which they are subject. In addition, the holistic nature of the approach allows for the worst-off to be “lifted up” to the level of the next-worst-off and so there is no danger of one group being subject to a more grievous injustice in order to rectify another group’s unjust condition. Further, in this model the best-off members of society bear the greatest burden, in the sense that they will benefit the least from reforms and will wait the longest for reforms that benefit them to be instituted. This is as it should be, since members who are located near the bottom of the social hierarchy already bear the heavy burden of diminished welfare and liberty, relative to those at the top. By contrast, one might imagine a transitional process in which a burdened society becomes more just by first guaranteeing the rights it is easiest to guarantee, namely those of the middle and upper classes, after which they make slow incremental gains in securing some rights to the bottom economic quartile of society. Such a society, even if it were to ultimately become just or decent by such a process, would not have arrived there in a transitionally just manner, since it placed the burdens of transition on those who were already the most vulnerable. The general justification for my approach is further supported by additional arguments pertaining to feasibility, cost and path dependence. Clearly reforms need to be politically and practically feasible in order for them to be implemented. It wouldn’t make sense to require that a given society implement policies aimed at the difficulties of the worst-off, if these policies were genuinely impossible to implement at particular time. This situation could arise either because the reforms have no popular support or because political disagreement among factions has reached an impasse. However, it seems implausible that all available methods of addressing all the vulnerabilities of the worst-off group will be blocked by these kinds of circumstances. Certainly creativity will be vital for generating political agreement and garnering support for programs from citizens. And in cases where a given Author's personal copy problem cannot be addressed in the short-term, it must be considered an ongoing priority in the long term, and so should be addressed when better circumstances are in place. Moving on to cost issues, the model requires that funds be directed first towards those difficulties that distinguish members of the worst-off group from the next-worst-off. Clearly it makes the most sense to choose those that are also least costly to solve, since this is more efficient, but cheaper coercive measures should not be favored over non-coercive ones purely for this reason. Policies that do not seek to coerce members of the worst-off group are to be strongly preferred because they are already so vulnerable to those who are more powerful that it is unreasonable to subject them to more risk of abuse. In any case, policies that require immoral conduct (such as unjustified coercion) on the part of officials are already ruled out. The question everyone wants answered when discussing the start-up and maintenance costs of transitional policies and programs is “Where will all the money come from, when we are discussing a society that has a substantial group of very poor members?” This question often leads people to suggest that what is needed is a “growth-led” policy, in which growing the economy is made the priority, and transitional policies are put-off until a society is richer overall and so can afford to roll out large, expensive programs. This strategy typically also recommends de-regulating trade, labor standards, and offering other incentives in order to attract foreign investment. While there is some reason to think that it is not necessary to put-off reforms in order to first generate funds, even without resolving this controversy we can say that any foreign aid that a burdened society receives can be used in a manner consistent with the model. This would mean, however, that foreign aid which was offered with conditions that would cause additional difficulties for the worst-off group would have to be rejected or re-negotiated in order to accord with this notion of transitional justice. This is likely to be a “sticking point” for some well-ordered societies since they most likely do not wish simply to aid the burdened society but also to take advantage of the cheap labor or other conditions prevailing there in order to make money themselves. But such exploitative international practices would not help a given society to reach its goal of becoming decent, and so it is not morally permissible to accept such offers. Additional funding for reforms could come from taxation of wealthy members of the society, and from NGOs. This will require burdened societies to monitor NGOs and approve the content of the work they seek to do before they do it. This monitoring will itself be a cost, but it will also allow the government of a given society to serve a coordinating function for NGOs within its borders that is sorely lacking in the aid system overall. This lack of coordination is one reason that NGO aid has not historically been as effective as expected, and so taking a holistic approach to serving the worst-off may also create overall efficiency gains for NGO aid. Finally, a note on path dependence: Even someone well-disposed toward the general account I have put forward might wonder whether the path of reform it recommends will lead burdened societies towards either decency or liberal democracy in the long run. A critic might suggest that while my approach seems justifiable in the short-run, a piecemeal, bottom-up model of this kind is unlikely to result in anything approximating the standard of decency or the model of perfect justice. Instead, it might result in a society getting “stuck” in a second-, or third-, or fourth-best condition with no further routes for improvement. In response to this concern, I will first observe that such claims are highly speculative and imprecise. I have difficulty conceiving of what it would mean for someone to “know for sure” that a given path of reform could never lead to the chosen end-point (whatever it is). It is also difficult to understand how we could know that of the very many routes that could lead to the chosen end-point, that not even one would approximate the model I have Burdened Societies and Transitional Justice 385 Author's personal copy endorsed. But supposing that we could know this, I think my approach might be defended on the grounds that one of the reasons we care about justice is that we care about how well- off people are, and this approach puts the right people first. In addition, it matters very much how close we could come to the desired end-point while at the same time prioritizing the worst-off. If we could come reasonably close, then this might well be better all things considered than achieving perfect justice or decency at the expense of those whose lives are already collapsing under the weight of multiple injustices. 6 Conclusion I have attempted two distinct tasks: (i) to re-construct Rawls’ conception of transitional justice in both Theory of Justice and Law of Peoples, and (ii) to expand and modify the latter conception in order to apply it to burdened societies. I have also introduced the notion of multiple, intersecting “vectors” of injustice in order to generate an account of the worst- off representative person for nonideal theory. While this notion is no doubt relevant to all societies that are not perfectly just, it seems to me it is especially crucial for charting a defensible course towards a more just condition for those societies that must begin this process in unfavorable circumstances, such as extreme scarcity of resources, pervasive corruption, deeply rooted prejudice and discrimination, and severe social inequality. Acknowledgements I am grateful to audiences at both the North American Society for Social Philosophy Conference (July 2010) and the Rocky Mountain Ethics Congress (August 2010) for their excellent feedback. I would also like to thank Steve Vanderheiden, Helga Varden and Jon Mandle for their detailed and insightful comments on earlier drafts of this article. References Farmer P (2010) On suffering and structural violence: social and economic rights in the global era. In: Saussy H (ed) Partner to the poor: a Paul Farmer reader. University of California Press, Berkeley, pp 328–349 Farrelly C (2007) Justice in ideal theory: a refutation. Polit Stud 55:844–864 Gilabert P (2008) Global justice and poverty relief in nonideal circumstances. Soc Theor Pract 34:411–438 McKerlie D (1994) Equality and priority. Utilitas 6:25–42 Moellendorf D (1996) Constructing the law of peoples. Pac Phil Q 77:132–154 Narayan D et al (2000) Voices of the poor: crying out for change. Oxford University Press for The World Bank, New York Phillips M (1985) Reflections on the transition from ideal to non-ideal theory. Noûs 19:551–570 Rawls J (1996) Political liberalism. Columbia University Press, New York Rawls J (1999a) Theory of justice, Revised Edition. Harvard University Press, Cambridge Rawls J (1999b) Law of peoples. Harvard University Press, Cambridge Robeyns I (2008) Ideal theory in theory and practice. Soc Theor Pract 34:341–362 Sen A (2009) The idea of justice. Belknap, Cambridge Shue H (1980) Basic rights: subsistence, affluence and US foreign policy. Princeton University Press, Princeton Simmons AJ (2010) Ideal and nonideal theory. Phil Pub Affairs 38:5–36 Sreenivasan G (2007) Health and justice in our nonideal world. Polit Philos Econ 6:218–236 Sreenivasan G (forthcoming) What is non-ideal theory? In: Williams M, Elster J (eds) Transitional justice. New York University Press, New York. Tasioulas J (2005) Global justice without end? Metaphilosophy 36:3–29 Valentini L (2009) On the apparent paradox of ideal theory. J Polit Philos 17:332–355 386 L.L. Fuller Author's personal copy Burdened Societies and Transitional Justice Abstract Introduction Notes on Non-ideal Theory Burdened Societies What Would Rawls(ians) Say? Transitioning to a Liberal Democratic Society Transitioning to a Decent Society Transitional Justice for Burdened Societies: A Proposal Conclusion References