Autonomy, Well-Being and the Order of Things: Gilabert on the conditions of social and global justice

Gilabert argues that the humanist conception of duties of global justice and the principle of cosmopolitan justifiability will lead us to accept an egalitarian definition of individual autonomy. Gilabert further argues that realizing conditions of individual autonomy can serve as the cut-off point to duties of global justice. I investigate his idea of autonomy, arguing that in order to make sense of this claim, we need a concept of autonomy. I propose 4 possible definitions of autonomy, none of which seem to necessitate Gilabert’s duties of egalitarian global justice. Instead, I propose that he may have in mind Autonomy 5, which requires that individuals have access to a maximum number of options and not simply a sufficient range of options to choose from. I criticize this premise as too demanding in the global world characterized by fundamental inequality. Second, I argue that if we were to endorse the preconditions for Autonomy 5, we would have to accept that Gilabert’s theory of global justice doesn’t provide for a cut-off point of duties of global justice

Pablo Gilabert's book aims to establish an argument for a humanist conception of global justice. Such a conception is built on the fundamental premise that duties of justice are egalitarian in nature, and global in scope. In the pursuit of his argument, Gilabert carefully clears some obstacles that are brought forward against such a premise. In my comments, I want to focus on the themes of individual autonomy and well-being. These themes run through the book, but are most importantly brought forward in chapters 5 and 6. I want to argue that Gilabert uses ideas of autonomy and well-being as measures of global justice without, however, providing a clear definition of the concept of autonomy or well-being that he has in mind. This, I want to show, leads at best to imprecision in his argument for a humanist account of global justice and the duties that flow from it; less sympathetic readers may further claim that without a definition of the concept of autonomy and well-being, Gilabert's claim that the conditions of autonomy are universal is unsubstantiated.
My discussion is structured as follows: first, I will briefly review Gilabert's account of the concern we should bring to conditions of autonomy and wellbeing. Second, I will aim to contextualize this concern in some of the cosmopolitan literature. Third, I will provide some points of critique to Gilabert's discussion.

I.
Gilabert argues that if we accept the principles of moral equality and cosmopolitan justifiability, we will necessarily arrive at the humanist principles of global justice he so well describes. The first such principle is that of moral equality that characterizes our relations in the world. This is to say that all human beings have equal moral standing in our moral deliberations. If we accept this, then we should also accept that "we should treat each other on the basis of principles of justice that no one, as free and equal persons, could reasonably reject" (p. 10). 2 Note that this clause serves both, as a definition of equal moral standing, and also as the definition of what Gilabert calls 'cosmopolitan justifiability'.
How does this refer to autonomy? Gilabert argues that the goal of our duties of global justice and, importantly, their cut-off point is to realize a level of wellbeing and individual autonomy. This formulation is important since it defines the good that humanist principles of global justice pursue. Methodologically, defining the realization of conditions of individual autonomy as the cut-off point of duties of global justice, moreover, helps to address one of the criticisms often levelled against proposals of global justice. Most notably raised by Rawls and repeated by others, the problem with such accounts is that they don't offer a cutoff point at which justice is realized. The distribution of goods, in other words, would be endless and non-satiable, and we would never be able to fulfill our duties of global justice. I will return to this point at the end of my comments.
The realization of conditions of individual autonomy is at the basis of Gilabert's claim that we have global duties to distribute access to important advantages equally among all individuals around the globe. A first, preliminary, definition of autonomy that Gilabert seems to suggest is thus that autonomy requires access to a set of advantages. Two examples of such advantages are access to medical care to realize the good of healthy lives; and access to opportunities for education. Both of these, we can say, are necessary for a certain conception of autonomy. Admittedly, I may be going out on a limb here since autonomy per se is not actually defined in the book -or not as explicitly as to make it into the index. My first task is thus to reconstitute the kind of idea of autonomy that Gilabert seems to rely on from the places in the book where it is mentioned.
We can agree that requiring access to medical care in order to lead healthy lives is fairly uncontroversial -we can simply accept that individuals can't really lead lives worth living if they lack a basic standard of health (however this may be defined). Call this Autonomy 1. Certainly, if individuals lack the means to protect themselves from easily preventable diseases, and actually die, then they obviously lack the means for autonomy since they are dead. Even less controversially, we can say that if persons suffer from such diseases, then they lack a basic level of well-being.
Does Autonomy 1 provide reason to accept the humanist conception of global justice? Not necessarily. Instead, I would argue that it is readily endorsed by most authors debating global inequality: access to the basic means of living is part of the kind of basic duties of humanitarian justice that Miller, Rawls and many other 'associationists' accept. 3 Note here that the emphasis is on humanitarian justice, and not egalitarian justice. Gilabert's aim, however, is to justify egalitarian principles of justice, seemingly necessitated by a humanist conception of justice and cosmopolitan justifiability. This suggests, then, that Gilabert must have something else in mind when he makes the link between autonomy and the specific humanist duties of global justice he wants to argue for. In fact, he does accept articulations of autonomy that go beyond sheer existence, and that are meant to express our concern for individuals to be able to make choices in their lives. Call this Autonomy 2, which differs from Autonomy 1 in that the former requires the possibility of choice in individual lives, rather than sheer existence.

II.
This brings us to the more narrow and applicable definition of autonomy. In chapter 5, Gilabert addresses Michael Blake's argument for the associationist basis of egalitarian duties of justice based on his argument for the value of individual autonomy. Blake's principle of autonomy holds that "all individuals, regardless of institutional context, ought to have access to those goods and circumstances under which they are able to live as rationally autonomous agents, capable of selecting and pursuing plans of life in accordance with individual conceptions of the good" 4 . In other words, Blake accepts Autonomy 3. This definition of autonomy is characterised not simply by choices, but by a conception of a good life that we assume every rational person wanting to pursue.
What is important for my argument is that according to Blake, Autonomy 3 can be secured through principles of global sufficientarianism. This is to say that if all human beings have sufficient resources, options and advantages to lead what a reasonable person would consider a good life, then the duties of global justice have been satisfied. Global justice does not require egalitarian redistribution of these goods. Instead, to Blake, such principles are only necessary in the realm of the coercive state, where egalitarian justice plays a compensatory function for the kind of coercion the state exercises on its members. Blake seems to say that when the coercive state thwarts individual autonomy, it can only compensate for this by providing egalitarian distribution. Some of the goods of justice that the egalitarian state should distribute are for example, access to health care and education; and while these goods should be sufficiently accessible to all human beings, they don't have to be accessible to all to the same extent above a level of sufficiency.
Blake justifies his stance with the argument that sufficiency would provide conditions of autonomy, since what we need to be autonomous is a reasonable set of options, rather than a maximal set of options. Gilabert challenges the underlying assumption in Blake's argument that there are two standards that apply to the provision of options. Why, Gilabert argues, would there be a justification for providing people with 'more economic options' in the realm of the state that is different in the global sphere? Would anybody adopting the principle of cosmopolitan justifiability actually agree to this? And why is state coercion the only route to high justificatory burdens and egalitarian concern, as Blake would have it?
We can see that Gilabert is critical of Blake's account of the distinction between global and social goods of justice. In fact, Gilabert questions if the realization of sufficientarian principles only at the international level would allow for the realization of Autonomy 3, which he seems to endorse in his discussion of Blake. He asks instead whether Autonomy 3, when taken seriously, couldn't "also yield stronger demands" on the kinds of principles of justice we ought to adopt on the global level. Note here that Gilabert does not say that his concept of autonomy demands stronger global distributive principles, although I think that this is in fact his premise. I will return to this point below.
There are several points that are important here. First, what is the nature of the choices that make people autonomous, besides the kind of basic necessities of life, like being able to lead healthy lives. What, in other words, does Autonomy 3 demand? Gilabert makes a convincing case that education is necessary to put individuals in a position of choice of the good life. He provides us later on in chapter 6 with good arguments why we should accept equal opportunity in education. And I don't think that Blake would necessarily disagree with the argument that access to education is necessary for autonomy, as I argued above.
The disagreement derives instead from their premise: Gilabert seems to hold that providing for the necessary provisions of autonomy (such as access to edu-cation, reasonable options and the like) require principles of egalitarian distribution, whereas Blake believes that sufficiency is, well, sufficient to guarantee conditions of autonomy. Put differently, while Blake argues that the sufficientarian provision is enough to assure conditions of autonomy, egalitarian principles of distribution of goods within the state are required because of the coercion state measures impose on its members. Gilabert, on the other hand, argues that egalitarian distribution of access to goods is a requirement for global conditions of individual autonomy.
Gilabert's premise is important here and should have been underlined more strongly: it is this premise that deals Blake's distinction between obligations of egalitarian justice and respect for individual autonomy the kind of deadly blow it may deserve. Gilabert quite correctly asks why else we would adopt principles of egalitarian justice in the realm of the nation-state if not to support individual autonomy. The causal arrow in Blake's argument seems to go the wrong way: Blake doesn't argue from what the conditions of autonomy require, instead he argues from the perspective of the obligations of the state towards its members. One of the great merits of Gilabert's argument, then, is to argue from the perspective of autonomy, and in a pragmatic way; he starts his investigation with a look into the necessary conditions for autonomy to prevail. This seems to me the more appropriate way of going about things.
Both Blake and Gilabert rely on Joseph Raz' conception of individual autonomy 5 , so let me briefly summarize this view here 6 . Call this Autonomy 4 or the most encompassing definition of autonomy so far. Raz proposed a concept of autonomy as 'self-authorship'; a concept, in other words, that requires us to enjoy a range of valuable or adequate options that may inform our choices in life, to create our own moral world 7 .
Raz explicitly acknowledges that place and culture may fundamentally shape the kinds of decisions we make -we may adopt some of the values our culture has promoted -we may believe in equality, solidarity and fraternity, as much as we may believe in the justice of the American dream. Options become real if we can refer to a context of choice, "some already accepted principles", i.e. principles that are adopted and endorsed by others around us. Our options also need to be adequate options if we are to carry out our lives autonomously. Both, longterm options that carry pervasive consequences as to the direction our lives will take, and short-term options that apply mostly to trivial decisions in our lives have to be available and open in order for us to be authors of our own lives 8 . We not only need to have a range of options; furthermore, these need to be viable for us. In other words, only if we are in a position where the options available can also become actual and concrete can they make sense to us and can they serve our autonomy. We can hence hold that options are necessary in order to make decisions in the most basic sense of what making decisions means, and that the original point from where we draw the options in our lives are those that are conveyed to us through our society, our social context like parents, family, friends and the societies we live in. So if this were the case, then we could say that associativist arguments such as those proposed by David Miller 9 and Michael Blake 10 have some traction -that in fact, societies translate the kind of possibilities available into options.
Finally, and discussing the variety of options we need to have at our disposal when making choices, Raz stipulates that the adequate choice "for most of the time…should not be dominated by the need to protect the life one has: A choice is dominated by that need if all options except one will make the continuation of the life one has rather unlikely" 11 . To put things in a more contemporary light, we can use Raz' argument to show that there is "an important connection between the view …that threats undermine freedom and the view that poverty undermines freedom." 12 This, we could say, supports the argument Blake makes -that a conception of justice on the global level requires sufficient options; this does not support Gilabert's claim, however, that autonomy requires an egalitarian amount of options.
Autonomy 4, then, indeed calls for sufficientarian duties of global justice, but not necessarily for the kind of humanist framework Gilabert wants to establish. In fact, I would argue that the response Gilabert proposes to Blake -namely, that the reason why the coercive state has obligations for egalitarian justice is simply because reasonable people would presumably want more options or the maximum number of options -may be correct; but that it is not necessarily what they would need in order to enjoy conditions of individual autonomy

III.
This brings me to my penultimate point: it may be important here to make a distinction between the conditions of autonomy, and the conditions of well-being. Earlier, I discussed briefly basic access to medical services that Gilabert promotes as a necessary right flowing from a humanist conception of global justice. There, I accepted that such access constitutes one of the conditions of individual well-being and autonomy. Reflecting on medical care, however, illustrates that well-being and autonomy as the two values we pursue in policies of global and social justice may not require the same conditions: health may be required for well-being, but autonomy may not be required for well-being. In fact, much of the current literature on global justice and health inequalities aims to grapple with the question of how to promote conditions of health. And some authors participating in the debate have come to the conclusion that recourse to paternalism might be the most effective way of improving well-being, even if paternalistic intervention is hardly ever justifiable from a perspective of individual autonomy, particularly if we accept Autonomy 4 which stipulates that we ought to be "part author of our own lives". Suffice to say, then, that conditions of autonomy may demand one set of duties, while conditions of well-being may demand another.
Assume that Gilabert accepts the sufficientarian argument for autonomy (Autonomy 3)-he could then still hold that concern for individual well-being demands humanist duties of global justice. But in order to sustain this proposal, it seems to me that we need to know more about the nature of well-being. If we accept, for example, a description of well-being along the lines of Sen's and Nussbaum's capabilities approach, then a closer look at what kind of capabilities individuals should have access to may again support the kind of duties of basic justice Blake and Miller endorse; but capabilities as a definition of well-being may not actually yield humanist principles of global justice as Gilabert proposes them. Sen, for example, is particularly vocal against what he calls 'transcendental principles of global justice' because he believes that they divert us from the kind of contextual interpretation those concerned with individual well-being need to employ when describing our duties of global justice. 13 To be sure, many have criticized Sen for what they perceive as a neglect of the important role that principles of justice play when designing duties of justice, and I believe that Gilabert would probably agree with much of the criticism. Suffice it to say here, though, that if our aim is to define well-being, and if in doing so, we make reference to Sen, we may actually have to agree with those who propose that a lot of what determines our well-being is contextual to the society we live in (or culture or something alike). If we accept that, however, then Miller's proposal that the conditions for and principles of justice, like the value of autonomy, for instance, are contextual to the society we live in, gains in plausibility. In other words, if we accept an account of well-being that builds on Sen's work, we may also have to accept associationist accounts of the duties of global justice.
So let's leave aside the definition of well-being and return instead to the concept of autonomy. Earlier, I argued that a Razian definition of autonomy (Autonomy 4) would lend itself to support sufficientarian principles of global justice but not necessarily yield the humanist principles that Gilabert wants to defend. This suggests that Gilabert has yet another concept of autonomy in mind (Autonomy 5). What, then, is Gilabert's concept of autonomy? Remember that the principles put forward here are submitted to the cosmopolitan justifiability test, which is to say that all are treated as moral equals. To test whether or not principles satisfy this test, Gilabert proposes to think about actual realizations of moral equality, for example through provisions of political equality. Early on in the book (pg 148 and section 2.4.1. page 44ff), Gilabert discusses the idea of individual autonomy in the context of political self-determination and political participation. He identifies enabling such participation as a humanist duty of global justice, and provides a detailed discussion of possible global institutional structures and the role individuals could play in these. If we accept this proposal, then it seems to me that we accept, also, that individual political autonomy can only take shape within an institutional framework. Autonomy, in other words, could possibly be construed as only realizable in specific contexts that enable the realization of autonomous agency. 14 But if we accept that, and to return to the associativist critique of duties of global justice, the fact that autonomy requires a specific context doesn't tell us which contextual conditions autonomy requires. An associativist proposal building on Sen could for example refer to David Miller's argument proposed in 'Principles of Social Justice'. 15 There, Miller argues that the social context provides us not only with the necessary context of choice (see above, my discussion of Autonomy 4), but also the reasons for this context of choice as the result of decisions we as a society have taken. 16 The framework of society, put differently, only makes autonomy plausible. If we accept Autonomy 4, it is plausible to argue that we need a set of reference points along which we are actually able to define options. It is not clear that institutions can provide us with this reference point. And I believe that Gilabert would agree that institutions are simply the means we employ to implement certain choices we have made as a society about the good life. As a society, we afford ourselves public institutions of higher learning, the arts and sciences but also institutions of health care and early childhood care because our society has decided that these are goods that are part and parcel of the good life that all members should have access to. Institutions can't, however, provide the definition of the goals of justice. That discussion has to precede the establishment of institutions.
Finally, I believe that an egalitarian principle that demands maximum options for autonomy raises an objection that Gilabert himself anticipates. Here, I simply want to raise a flag as to his answer to the objection. Towards the end of chapter 5, Gilabert discusses to what extent one could argue that global egalitarianism restricts individual autonomy of the well-off. (p. 179). The question raised, in particular, is whether or not the duty to set up 'domains of cooperation' imposes unreasonable costs that might go against individual preferences of the welloff, thus thwarting their autonomy. Gilabert acknowledges that in the current (unequal) world, the demands on the well-off that would follow from a humanist duty of global justice would indeed be high, and one might speculate to what extent such demands would impose 'unreasonable' costs: "The challenge is based on the important claim that there is a strong personal prerogative to be able to pursue one's good without unreasonable interference from others, or demands to promote their interest…the challenge may be more biting for humanism because it may be less ready than various forms of associativism to see any current contingent factual limits to the depth of globalization as automatically warranting normative limits to global distributive demands" (pp. 179-180).
Gilabert accepts that if agent-neutral duties derive from humanist global justice principles, then "they may turn out to be quite demanding" (p. 180) -however, Gilabert posits that if such duties are meant to enable somebody else's autonomy, then "the charge of demandingness looses some force" (p. 180) since "a protection of one's personal autonomy should not be seen as a passport for exiting the moral space of responsibilities" (p. 181). The context that is required for autonomy to be realized, then, is in this instance circumscribed by the moral duties we have to others to enable their autonomy.
This brings me to the question raised at the end of this section in the book, which is left somewhat hanging. Gilabert accepts that the demandingness objection would become particularly problematic in non-ideal contexts, where, often, the conscientious rich do more than their fair share and where this duty to do more (established earlier) might "threaten their ability to pursue their own personal projects unless they fuse such projects (as nobody should be demanded to) with the pursuit of justice." (pp. 181-182). Recall here the definition of Autonomy 4, which demands that our options need to be viable for us, and also that we can actually implement them. What to do, in other words, if we care for egalitarian conditions of autonomy for all, but when in the pursuit of realizing this goal, the autonomy of some is curtailed? How do we justify this?
Justifying sufficientarian conditions of individual autonomy in this context should come easily: if we accept the principle of moral equality, then it is immediately plausible and reasonable to argue that the maximal options of the rich may be justifiably curtailed to guarantee access to sufficient options of those who would otherwise lack them. I am not certain, however, that an equal concern for individual autonomy would equally be able to justify the kind of maximization of options that Gilabert calls for in his response to Blake. Put differently, if adopting egalitarian principles to provide conditions of autonomy in a world of fundamental inequality would indeed entail high costs to individual autonomy of the well-off, we need to ask why we should adopt the maximization concept of autonomy rather than stick with the sufficient and adequate conditions (Autonomy 3).
The balancing act that considerations of justice have to fulfill in negotiating between the legitimate autonomy concerns of the worst-off and those of the welloff seems to easily accept basic conditions of autonomy, but I don't see how we can ground egalitarian principles of distribution of options on the concern for providing conditions of autonomy -unless, of course, egalitarian principles are not simply the way we achieve autonomy, but are a goal in themselves. We could say, in other words, that our concern is simply to treat all individuals equally since we assume that this is the principle that cosmopolitan justifiability would yield -without aiming to buttress this claim with concern for individual autonomy. If Gilabert were to choose this route, however, I suspect that he would encounter the response that this doesn't provide us with a cut-off point at which we can say that we have satisfied our global duties of justice. Recall that providing a cut-off point is one of the advantages of cosmopolitan justifiability. What I am trying to say here is that Gilabert uses autonomy to support the claim that humanist egalitarian principles of global justice are plausible because, contrary to other proposals of cosmopolitan justice, he accepts the need for a cutoff point for our duties of global egalitarian distribution. This point is reached once we have distributed what we need to realize conditions of autonomy. My reading of his account of autonomy -as we move from Autonomy 1 to Autonomy 5 -however, suggests that somewhere along the way, his goal morphs from realizing conditions of autonomy to maximizing options as a condition for autonomy. And that, I believe, is not a clear cut-off point at all. If a book stimulates the reader to reflection, then it has certainly achieved a great deal. Gilabert's treatment of the humanist conception of global duties of justice is such a book and we can only hope that he will pursue his exploration of the consequences of applying the humanist account to questions of global justice.