Part I of Rescuing Justice and Equality consisted in a series of chapters designed to rescue equality from the arguments of Rawlsians who sought to dilute an underlying egalitarian commitment with the incentives argument, the Pareto argument, the restricted focus on the basic structure, and then the difference principle itself. In each case, the structure of the argument was a kind of imminent critique. As far as I recall, Cohen nowhere directly defended the egalitarian commitment itself. Rather, he pointed to alleged tensions in the Rawlsian edifice and submitted that they should be resolved in the direction of greater egalitarianism than Rawls’s position recommends.
Part II aims to rescue the concept of justice itself, and the argument is structured very differently. The critique does not proceed from tensions within Rawls’s work. Rather, we get an argument in defense of a certain meta-ethical position. Cohen remarks that “the meta-ethical literature says very little about the question pursued in the present chapter. But a notable exception is the work of John Rawls, who argued that fundamental principles of justice and, indeed, ‘first principles’ in general, are a response to the facts of the human condition” – which is exactly the position that Cohen rejects. (pp.258-259) Rawls is simply mistaken, Cohen thinks, because he confuses “the first principles of justice with the principles that we should adopt to regulate society.” (p.265)
First, definitions. A “principle” is “a general directive that tells agents what (they ought, or ought not) to do.” (p.229) And a “fact is, or corresponds to, any truth other than (if any principles are truths) a principle, of a kind that someone might reasonably think supports a principle.” (p.229) (Cohen recognizes a distinct sense of “fact” that might be suggested by the phrase “the moral facts,” but the meta-ethical question that this possibility raises is not his concern.)
Next, the thesis – “a principle can respond to (that is, be grounded in) a fact only because it is also a response to a more ultimate principle that is not a response to a fact: accordingly, if principles respond to facts, then the principles at the summit of our conviction are grounded in no facts whatsoever.” (p.229) Cohen is apparently using “respond to” and “grounded in” synonymously – and later he uses “grounds for affirming,” “reflects,” “represents a reason to endorse,” and “justifies” in similar ways. Cohen claims that his thesis applies “to anyone’s principles, be they correct or not, so long as she has a clear grasp both of what her principles are and of why she holds them.” (p.233)
And then the argument, which contains three premises:
1. “whenever a fact F confers support on a principle P, there is an explanation why F supports P, an explanation of how, that is, F represents a reason to endorse P. That first premise rests upon the more general claim that there is always an explanation why any ground grounds what it grounds.” This general claim, Cohen says, is “self-evidently true.” (p.236)
2. “the explanation whose existence is affirmed by the first premise invokes or implies a more ultimate principle, commitment to which would survive denial of F, a more ultimate principle that explains why F supports P.” (p.236) Cohen challenges anyone who denies this to “provide an example in which a credible and satisfying explanation of why some F supports some P invokes or implies no such more ultimate principle.” (p.236)
3. the process of citing more ultimate principles in the second premise cannot continue indefinitely. Cohen believes it is simply “implausible that a credible interrogation of that form might go on indefinitely.” Furthermore, such a process would require “something like an infinite nesting of principles, and few will think that there exist a relevantly infinite number of principles.” And finally, if such an infinite sequence were necessary to justify some fact-based principle, this would imply that the anyone affirming the fact-based principle did not have “a clear grasp of what her principles are and of why she holds them.” (p.237)
Hence, the conclusion: “every fact-sensitive principle reflects a fact-insensitive principle.” (p.237)
The picture is supposed to be something like this: Suppose we endorse a principle only on the assumption that certain factual conditions hold – say, assumptions about human nature or about natural or social circumstances. There is an explanation as to why these factual conditions are relevant, and the explanation is going to invoke another more general principle that is not dependent on those conditions holding. Repeating this procedure, we will eventually reach a principle that is not conditional on any factual assumptions at all – it is fully general in that sense. Although Cohen says that the more general principles explain why the facts are relevant, it seems to me that they also justify the more specific, conditional principles (in conditions where the relevant facts obtain). The issue is confused by talk of factual conditions themselves justifying principles. (That’s what calls for explanation, according to Cohen.) The correct question should not be why some fact justifies some principle, but why some principle is justified when some fact obtains. And when we ask that question, it becomes more clear that Cohen has an unduly narrow understanding of justification. This may also be revealed by his use of “grounding,” “reflects,” etc. as synonyms for “justifies.” What seems to be excluded is some kind of process of reflective equilibrium, in which justification goes not only from more general to more specific, but potentially in either direction. (See note 19 on p.243.) Perhaps Cohen holds that such a justification would not be sufficiently “satisfying.”
The reason this matters for Cohen, I think it is safe to say, is because it allows him to distinguish “fundamental principles” from “rules of regulation.” I think the next two chapters will be better places to discuss this distinction, but here I’ll just record what he says about the distinction and make one comment. A rule of regulation is “a certain type of social instrument, to be legislated and implemented whether by government itself or within social consciousness and practice. A rule of regulation is ‘a device for having certain effects,’ [quoting Nozick] which we adopt or not, in the light of an evaluation, precisely, of its likely effects, and, therefore, in the light of an understanding of the facts. And we evaluate those effects, and thereby decide which fact-bound principles to adopt, by reference to principles that are not devices for achieving effects but statements of our more ultimate and fact-free convictions [i.e., fundamental principles].” (p.265) (Elsewhere, e.g. p.21, he says that rules of regulation can be responsive to values other than justice.) My comment is simply this: given this characterization, the rules of regulation are not simply fact-dependent (non-ultimate) principles. And fundamental principles are not simply maximally general (non-fact-dependent) rules of regulation.
One reason this distinction is important is that Cohen thinks that philosophers should be especially concerned to identify fundamental principles. He complains that there has been “insufficient effort to identify” fact-free principles (p.269), and that “the question for political philosophy is not what we should do but what we should think, even when what we should think makes no practical difference.” (p.268) This may seem somewhat peculiar since a principle is practical by definition – it’s “a general directive that tells agents what (they ought, or ought not) to do.” (p.229) His point, I think, is that philosophy should still be concerned to distinguish between principles that differ in their recommendations only under circumstances that do not and will not obtain. The choice between two such principles makes no “practical difference” but is still important for philosophers. It is only when we know what our principles recommend under all conceivable factual circumstances that we will fully understand them.
But if Cohen wants us to focus on ultimate principles, how does he think they should be justified? Officially, he remains agnostic. He specifically rejects the implication that “ultimate principles cannot themselves be justified.” (p.238) He believes he is only committed to the claim that “ultimate principles cannot be justified by facts.” (p.238) But facts are anything other than principles. So if principles are not justified by facts, they can only be justified by other principles. And ultimate principles must somehow be self-justifying or not justified at all. Rawls’s characterization of this position as a form of “rational intuitionism” seems appropriate.
An alternative model of justification, I suggested above, is some kind of reflective equilibrium. If that’s right, then justification proceeds from where our considered judgments are most secure. And as Cohen recognizes, our most secure moral judgments are fact-dependent: “it is, for example, bewildering to try to say what principles we would affirm for beings who were otherwise like us as we are in our adult state but whose normal life spans occupied only twenty-four hours…” (p.246) Yes, that is bewildering. I suppose that if pressed, I might concede that we don’t fully understand the concept of justice until we know what it would require for beings such as that – and for all other conceivable beings in all conceivable circumstances. But there are still plenty of things that we – or at least I – don’t understand about justice for beings like us in circumstances like those that we face. And I’m quite confident that our prospects for improving our understanding of justice are greater when we make certain factual assumptions, even if doing so postpones the question of justice in its full generality.