By ‘subject of justice’, I mean ‘what claims of justice are about’. So the title question, then, is: what constitutes the content of claims about justice?
Famously, Rawls (1971) answered the question as follows:
The primary subject of the principles of social justice is the basic structure of society, the arrangement of major social institutions into one scheme of cooperation. We have seen that these principles are to govern the assignment of rights and duties in these institutions and they are to determine the appropriate distribution of the benefits and burdens of social life. The principles of justice for institutions must not be confused with the principles which apply to individuals and their actions in particular circumstances. These two kinds of principles apply to different subjects and must be discussed separately. (TJ, sec. 10)
Following Rawls, a distinction has been made in the political philosophy literature between an institutional view and an interactional view. Pogge’s formulation of this distinction in ‘Cosmopolitanism and Sovereignty’ (1992) is representative:
An institutional conception postulates certain fundamental principles of justice. These apply to institutional schemes and are thus second-order principles: standards for assessing the ground rules and practices that regulate human interactions. An interactional conception, by contrast, postulates certain fundamental principles of ethics. These principles, like institutional ground rules, are first-order in that they apply directly to the conduct of persons and groups. (p. 50)
(If you’re interested, two prominent rejections of this distinction are G.A. Cohen, ‘Where the Action Is’ (1995) and L. Murphy, ‘Institutions and the Demands of Justice’ (1998). Pogge, ‘On the Site of Distributive Justice’ (2001) responds. Strictly speaking, Pogge characterizes the interactional view as a view concerning ethics as opposed to justice. But, as I’ll try to make clear below, there could be a way in which both the interactional and institutional views are views of justice.)
To illustrate the distinction, consider the following scenario:
The contract of employees at Company A is about to expire and contract negotiations with the employer are underway. The employees demand that their pay be increased to keep up with inflation, that they receive health benefits, and that certain safety measures are put in place. The employer responds with the following offer: `No pay increase, no health benefits, and no additional safety measures. Take it or leave it. If all of you reject the offer, then I’ll have no choice but to concede your demands. However, if any of you takes it, then those of you who don’t take it will be fired.’ Add to the case the following parameter: Employees lack rights that would make securing cooperation easier, i.e., they have no right to bargain collectively or to meet together under the auspices of a labour union. (Or, if you prefer, the employer has rights that enable it to protect its interests by breaking up union meetings, isolating employees in negotiations, etc.)
So, to clarify: Each employee can either accept or reject the employer’s terms. If all employees reject the employer’s terms, then each employee retains her job and gets a better contract. If an employee accepts the employer’s terms, then she retains her job but gets a worse contract. If an employee rejects while just one other employee accepts, then she loses her job.
Now, suppose we agree that there’s some injustice involved in this scenario. At least two distinct sets of claims aim at locating the injustice:
(1) The employer’s terms are unjust. Employers ought to pay their employees wages that keep pace with cost of living adjustments; they should provide access to affordable health care (in the absence of publicly provided health care); and they should guarantee their employees a safe work environment. By failing to meet these just demands, the employer is being unjust.
(2) This bargaining situation is unjust. Employees should have rights that grant them a more equitable bargaining position vis-à-vis their employers, i.e., a bargaining position that enables them to influence the terms of their employment. (Or: employers shouldn’t be granted rights that give them a decisive bargaining advantage over their employees, i.e., a bargaining position that enables them to unilaterally dictate the terms of employment.) The injustice here is a failure to situate employers and employees vis-à-vis each other in a way results in a fair bargaining situation.
I think there’s a genuine distinction between (1) and (2). Up til now, I’ve followed Rawls and Pogge in calling type-1 claims ‘interactional’ and type-2 claims ‘institutional’. (At least, I’ve thought I’ve been following Rawls and Pogge in distinguishing between (1) and (2) in this way. I should leave open the possibility that Pogge, at least, carves up the relevant space differently.) Intuitively, this seems plausible. The claims in (1) prima facie address the employer’s interactions with its employees; those in (2) prima facie address the rights and duties that are institutionally allocated to the employer and employees respectively. But I can see how this might be misleading for at least several reasons:
First, the motivation for (2) could be a worry about a certain class of interactions, viz., unjust bargaining interactions. The thought underlying the response could be: ‘Employers ought not be permitted to fire their employees and hire replacements when the employees refuse to accept the employer’s terms. This type of employer-employee interaction ought to be circumscribed.’ If this is right, then the supposed institutional worry would actually be an interactional one.
Second, the solution proposed to remedy the worry in (1) could be institutional: ‘We ought to establish institutions that will insure that employees wages keep pace with cost of living adjustments, that they are provided with access to affordable health care, and that they are guaranteed a safe work environment. In other words, we ought to establish institutions that constrain the contract terms any employer is able to offer.’ If this is right, then it might be tough to see how (1) couldn’t be institutional.
Third, the claims in (1) could be claims about which rights employees ought to have, i.e., they ought to have rights to affordable health care, a living wage, etc. If so, and if institutionalism is supposed to be characterized by a distinct worry about institutional schemes of rights, then (1) is ostensibly an institutional response.
Finally, institutions are often thought of as the net result of numerous individual interactions over time. If this is right (I must admit, I find this view plausible), then a worry about the justice of institutions is, at bottom, nothing other than a worry about interactions.
Perhaps the first three worries show that (1) and (2) don’t actually fall on opposite sides of the institutional/interactional divide; in that case, I’ve misconstrued the original distinction. (But the fourth worry would still have traction.) Or perhaps (1) and (2) capture the distinction just fine. In either case, and despite the foregoing worries, I still think there’s a genuine distinction between (1) and (2) and it’s this distinction I’m interested in clearly capturing. As a first approximation, I’ll say that (1) is concerned with non-structural features of the case, whereas (2) is concerned with structural features of the case. By calling the concerns in (1) ‘non-structural’, I mean to indicate that the content of ‘just contract terms’ is not dictated by constitutive features of the bargaining situation. The structural desiderata of a just bargaining situation don’t speak to the question of whether employees should be paid wages that keep up with inflation, or whether access to affordable health care ought to be provided to them, or whether they ought to be guaranteed a safe workplace. Not surprisingly, then, the concerns in (2) are ‘structural’ because they are concerns about the way that the bargaining situation is constituted; in particular, the sets of rights that are granted to employers and employees respectively when they enter bargaining situations. So instead of a distinction between interactional and institutional views, perhaps the relevant distinction is that between structural and non-structural views.
Admittedly, all of this is a little undercooked. I could say more, but then the post would be unbearable long. So here are the main questions I have:
1. Do (1) and (2) adequately capture the institutional/interactional distinction introduced by Rawls and defended by Pogge? If not, suggestions on what sorts of claims would fall on either side of the divide?
2. Is the structural/non-structural distinction I’ve tried to sketch here tenable? Does it capture a genuine distinction between (1) and (2)?
To foreshadow: I plan to add at least two more installments to this thread over the next few weeks. Pt. 2 will attempt to offer some reasons for thinking that the structural/non-structural distinction isn’t coextensive with the institutional/interactional distinction (while acknowledging that there is some significant overlap). In addition, I’ll try to offer some reasons for thinking that the former offers a more helpful way to carve up the relevant space (and perhaps better captures the intuitions Rawls and Pogge were trying to capture with the institutional/interactional distinction). Pt. 3 will try to offer some reasons for thinking that we ought to be structuralists about the subject of justice. I might also say something about what I think the more substantive implications of structuralism might be for global justice problems, but that sounds too ambitious at this point.