Continuing the discussion of G. A. Cohen’s Rescuing Justice and Equality – sorry about the delay – chapter 7 is on “Constructivism.” Cohen argues against the view that fundamental principles of social justice can be identified by considering a selection procedure that addresses the question, “What rules of governance are to be adopted for our common social life?” (p.275) The selection of principles from the original position is his primary target, although the specific features of that choice situation are not at issue. The main objection that he presses is familiar from chapter 6: constructivism mistakenly identifies the principles of justice with all things considered judgment concerning rules of social regulation. This must be a mistake, for Cohen, because the all things considered perspective encompasses considerations other than justice (including other virtues), and asks how best, given certain circumstances, to achieve the optimal balance of these various considerations.
While the original position is not an appropriate basis from which to identify principles of justice, it, “or some variant of it, might be the right procedure for generating rules of regulation” (p.284) – although Cohen has independent reasons for doubting even that. Understood as selecting rules of regulation, the parties in the original position are properly influenced by considerations related to Pareto, publicity, and stability. These considerations, while relevant to the selection of rules of regulation, are all independent of the nature of justice.
The constructivism of the original position relies on other factual assumptions as well, and this shows – according to the argument of chapter 6 – that it cannot be identifying fundamental principles of justice. For example, it assumes that “there exist goods with which people need to be provided to pursue their life plans.” (p.293) To try to convince us that the fundamental principles of justice are not dependent on such a fact, Cohen has us “think of a being with a life plan that is internally fully provided from its inception with everything that it requires for whatever life plan it might choose.” (p.293) Similarly, a contingent fact which principles of justice ought not presuppose is that “human society is marked by both identity and conflicts of interests.” (p.294) These assumptions – that individuals are not completely self-sufficient and that there is both an agreement and disagreement about values – bring us very close to the assumption that the circumstances of justice hold.
When he turns to the discussion of the circumstances of justice, Cohen distinguishes several different questions that may easily be confused: “(1) Under what circumstances is (the achievement of) justice possible and/or necessary? (2) Under what circumstances do questions of justice arise?” For Cohen, the answers to these questions are independent of the nature of justice – the content of the fundamental principles of justice. This is just an application of his more general argument that the content of the fundamental principles of justice cannot be fact-dependent.
Rawls, on the other hand, when developing his principles of justice, assumes (among other things) that we are addressing members of a society who hold diverse and conflicting – though reasonable – values and who are not individually completely self-sufficient. I don’t know exactly what he would say about conditions in which these do not hold – whether justice would be impossible or wouldn’t be important – but, frankly, I have a hard time imagining such conditions.
The reason why Rawls makes these assumptions is because he thinks that identifying principles of justice is a matter of specifying fair terms of social cooperation. Principles of justice aim to answer the question: “what is the most acceptable political conception of justice for specifying the fair terms of social cooperation between citizens regarded as free and equal and as both reasonable and rational…?” (Justice as Fairness, pp.7-8) Contrary to Cohen’s suggestion, Rawls doesn’t identify justice with an all things considered perspective concerning how best to design social institutions. He thinks that justice is the virtue that responds to the fact of reasonable pluralism – the fact that reasonable people in a society have different and conflicting conceptions of the good, yet are not self-sufficient. And he claims that in such conditions, this virtue sets a necessary condition for the moral acceptability of basic institutions. This is not a trivial claim – as it would be if he identified justice with an all things considered perspective. (p.305) It amounts to saying that it is more important to design basic institutions that are fair to individuals conceived of as free and equal than it is to proclaim the glory of God, or to preserve a cultural heritage, or to achieve and maintain some particular distribution of welfare. Each of these may be permissible, but not if they conflict with the requirement of maintaining a fair system of cooperation among free and equal persons.
One more relatively minor but perhaps revealing point about Rawls. Cohen says that Rawls holds that Pareto improvement from equality are “required by justice, since principles mandating it would be chosen in the original position.” (p.318) Some passages do suggest this – including some statements of the difference principle itself – but I think Rawls’s considered position is that such inequalities are permitted by justice, not that they are required. It would be perfectly just for a society to maintain equality and forego an inequality that would satisfy the difference principle – and there can be good reasons for it to do so. In Justice as Fairness he points out that the difference principle “does not require continual economic growth over generations to maximize upward indefinitely the expectations of the least advantaged” and goes on to talk about “permissible” (as opposed to required) inequalities. (Justice as Fairness, pp.63-64) What ultimately matters to Rawls is not that the expected share of primary goods of the least advantaged position over a complete life be maximized. What matters is that nobody be subordinated (at a fundamental level) to anyone else. He argues that the inequalities allowed by the two principles of justice would avoid that. But so also – obviously – would greater equality, and (in general) he would have no objection to a society making such a decision.
And this brings us to the question I want to ask of Cohen: what is justice about? I’m looking for an answer analogous to what I take Rawls to hold, which is something like this: justice is about recognizing the diversity of reasonable values among the members of a society and designing basic institutions that won’t subordinate any one to any other one. Let me elaborate my confusion – and please let me know in comments if you have any ideas about how Cohen would reply. First, Cohen holds that justice is more than distributive justice. He refers, for example, to “considerations of nondistributive justice” (p.7) and he mentions that sometimes we must “trade one form of justice off against another, at the level of rules of regulation.” (p.367) Perhaps he has in mind something like political justice – say, the requirement that individuals have an opportunity for democratic participation. My question is why, on his view, are distributive justice and political justice both forms of justice? According to Rawls, they are both requirements of a fair system of cooperation. I don’t see why Cohen would say – if he would – that they are part of a single virtue.
But I’m even confused about how Cohen understands the more limited virtue of distributive justice. I’m not talking about the principle that he advocates (although I admit to being confused about that, too). I’m talking about what we might call the concept of distributive justice. On the one hand, he insists that “it would be insane for the intuition about distributive justice on which [luck egalitarians] focus to be the only norm to be taken into account in organizing a society.” (p.301) From an all things considered point of view, sometimes distributive justice should give way to other principles and virtues. I assume this means that sometimes, in determining who should get what, there are good reasons to do what conflicts with the requirements of distributive justice. But on the other hand, when describing the topic of distributive justice, Cohen says that by “distributive justice” he “uneccentrically mean[s] justice (and its lack) in the distribution of benefits and burdens to individuals.” (p.126) Alternatively, it is “giving each person her due.” (p.7) If this is interpreted as who should get what, this is an all things considered perspective. Any consideration that bears on what a person’s benefits and burdens should be appears relevant from the point of view of distributive justice.
I think this lack of clarity about the nature of justice – and even of distributive justice – shows up in an example that Cohen gives. That a change would be a Pareto improvement is a reason for favoring the change, but not a reason of (distributive) justice. (p.315) It may, for example, be preferable on “ground of human flourishing” (p.319) and this may be sufficient to recommend it, but it may nonetheless be unjust. Consider a state of nature “in which manna falls from heaven and gets shared equally because the sharers think that’s the right way to deal with manna from heaven. Now suppose an extra piece of irremovable but destructible manna falls on Jane’s plot.” (p.317) Jane, being a luck egalitarian, destroys the manna because she believes that it is not fair for her to have more than others for no good reason. Far from thinking that Jane is foolish, Cohen holds that “she is simply a remarkably just person.” (p.318)
My own reaction is that she may or may not be foolish in this case. It is possible, for example, that she is expressing a high degree of solidarity with others. If so, she is still no more or less just than if she simply consumed the manna. It is Cohen, I claim, who is here confusing the virtue of justice with something else – perhaps with solidarity. The case does not yet isolate an intuition about distributive justice (as Cohen conceives it). After all, for Cohen, justice is not about showing others that you are just – publicity is irrelevant to what justice is. So, suppose that nobody except Jane knows that she has this windfall. And suppose (what I take to be already implicit in the example) that it cannot be traded or used to enhance her social status. The example still doesn’t isolate the right intuition because Cohen has us imagine the people sharing the (ordinary) manna. So imagine that they are all completely isolated from each other, and it happens that they all receive equal amounts of manna from heaven. One day, Jane receives more. There is no way for her to share it or even to communicate with the others, but she somehow knows that she has received more than the others. Now I would say that if she destroys the stuff, she is simply being foolish. The principles of justice give her no reason to do so. Without at least even the possibility of interactions – and, indeed, without other factual conditions obtaining – the principles of justice give us no reasons to act one way or another.