Coming Soon:

Now Available: Volumes I, II, III, and IV of the Collected Published and Unpublished Papers.



Total Pageviews

Wednesday, February 23, 2011


Having completed more than two hours of discussion of Michael Oakeshott's splendid essays, "Rationalism in Politics" and "Rational Conduct," I turn this afternoon in my UNC graduate Public Policy seminar to the famous theories of John Rawls. Even though I have written a book on those theories [UNDERSTANDING RAWLS], I yesterday quickly re-read the seventy-five pages I have assigned to the doctoral students in Public Policy who are taking the seminar. As I skimmed the familiar pages, I was reminded of a curious characteristic of Rawls' prose, one that struck me very powerfully when I read A THEORY OF JUSTICE shortly after it appeared in 1977. I actually included a brief discussion of this curiosity in the original manuscript of my book, but my then wife Cynthia Griffin Wolff and my good friend Robert J. Ackerman persuaded me to remove the offending passage. Now, more than thirty years have passed, Jack Rawls has gone on to his reward, and I think it is not entirely insensitive of me to offer my observation, for whatever interest it may hold.

As most of you will be aware, Rawls first sketched the theory that would later become the subject of his book in an essay entitled "Justice as Fairness." In that essay, he offered two principles of justice, the first an equal liberty principle and the second a principle of distributive justice usually referred to as "the difference principle." Rawls claimed, in the essay, that individuals exhibiting what Hume had described as the circumstances under which the idea of justice arises and committed to a bargaining procedure designed to arrive at unanimously agreed upon principles for regulating their social interactions and institutions, would necessarily [as a Theorem of Bargaining Theory] coordinate on those two principles. In point of fact, this claim was wrong, a fact that I demonstrated in an essay published in THE JOURNAL OF PHILOSOPHY. In a second essay, entitled "Distributive Justice", Rawls, who appears to have recognized the invalidity of the theorem quite independently of my criticisms, introduced the famous device of the "veil of ignorance" and revised the two principles, this time rewriting the difference principle so that, for the first time, it made reference to the "least advantaged."

There is nothing surprising or noteworthy about the fact that Rawls' revised his principles. He tried a first formulation, discovered that it would not fly, and dramatically altered his theory so that he might plausibly claim to be offering a Theorem in Bargaining Theory. But in A THEORY OF JUSTICE, Rawls does something that I find genuinely weird. He puts forward the first formulation of his two principles, correctly observes that it cannot be the solution to the bargaining game he has imagined, and then asks the question, "What other interpretation of these two principles might we find that is more plausible and defensible?"

Now, this question simply makes no sense. Why on earth suppose that the words must be preserved, but that they need a new interpretation inasmuch as the interpretation readers might be led to put on them -- the interpretation Rawls himself placed on them in his first article -- cannot be defended? Let me put it this way: With respect to what other text do readers and interpreters regularly engage in this sort of inquiry? The answer leaps immediately to mind -- Divinely Revealed Texts.

When devout readers of the Bible find a text that rational persons simply cannot take literally, they have one of two possible responses. Since the text is the Revealed Word of God, they can either deny reason and accept the words as written -- which is to say, they can become Fundamentalists, Inerrantists. Or they can employ the techniques of textual interpretation originally devised for such occasions and find an interpretation that allows faith to lie down with reason.

What I find so strange about Rawls' mode of discourse is that he appears to view the two principles as revealed texts that cannot be altered -- as though they were of divine origin -- but also cannot be assigned what seems to be their most obvious meaning. Since Rawls is a rationalist, not a fundamentalist, he opts in Chapter Two of A THEORY OF JUSTICE for what can only be described as scriptural interpretation, coming up eventually with the final, mature version of the two principles. [These too are, as it turns out, not a plausible solution to his bargaining game, but that is neither here nor there.]

Well, I must gather up my things and go off to class. I just thought some of you might find this interesting.


Greg said...

TJ was published in 1971, not 1977.

Also, in answer to the 'weird' bit you point out: reflective equilibrium.

Murfmensch said...

I think Greg is right about reflective equilibrium being a more charitable take on what Rawls is doing.

We believe a few plausible claims. When they aren't "working" we have to decide which one to revise. We tinker until they work again.

He seeks to conserve beliefs for reasons like those Pierce gave when he laid out the way we make our ideas clear.

On another note: There are times when I think Rawls demands too much of himself when he seeks to make his two principles the product of a rational-choice-theretical model.

Amato said...

While I trust your right professor--that Rawls "correctly observes that it cannot be the solution to the bargaining game he has imagined"--I never picked that up in my own reading of A Theory of Justice. Where is that located?

As I understood it, Rawls thought the equal liberty principle and the maximin of the difference principle were the natural outcome of the bargaining game (original position) under the constraints he set up--rational, without jealously, self-interested, and under the veil of ignorance. Whether or not he was right is a whole other matter.

Amato said...
This comment has been removed by the author.
Amato said...

Oh professor, this is completely off topic for this post, but I thought you might be interested in these charts related to income inequality in the U.S.

Eggs Maledict said...

Ok, I'm really confused. How does reflective equilibrium hep Rawls, there? He's still got a position which is incorrect. Reflecting on his intuitions isn't going to make it any less so.

On a different note, I find Rawls' constructivist Kantianism reeeeeaaaally strange (read: incoherent) and would greatly appreciate someone explaining how it can possibly work.

Robert Paul Wolff said...

Amato, it would take a long time to respond fully [see my book on Rawls for the whole story.] A few quick points:

1. Once the veil of ignorance is introduced, the choice situation ceases to be a bargaining game, because the participants are identical, and therefore could not possibly disagree. ZSo it becoems a rational choice problem -- very different.

2. To make the rational choice problem even imaginable, Rawls needs a whole lot of ad hoc assumptions -- Life Plans, Primary Goods, etc -- and one that is simply incoherent, namely the assumption that one can define a unidimensional Index of Primary Goods. WEven then, he cannot get started until he gives an implausible argument for maximin as a rational rule of choice. See my book for a detailed Game Theoretic analysis of what that assumption implies about where in history the rational chooser is located.

Greg said...

Amato & EM,

Those are very good questions with fairly complex answers. Rawls is an enormously important thinker in contemporary political philosophy, worth understanding on his own terms even if we don't end up agreeing with him. So for a more sympathetic (read: accurate) reading than Wolff's you might consider Samuel Freeman's truly excellent book "Rawls", which offers a remarkably comprehensive but accessible discussion of Rawls's work. On reflective equilibrium in particular, see Tom Scanlon's chapter in the Cambridge Companion to Rawls, and (for a highly technical but truly full defence) see Norman Daniels's book on the subject.

Greg said...

*Tim Scanlon

Marinus said...

Greg, we should say that one of the many, many things Rawls does in that book is that he repeatedly claims that it is something of a theorem. Prof Wolff's book is the only book on Rawls I've read which takes that claim at face value (though I am by no means a scholar in this field). So when you say 'take Rawls on his own terms', you can only mean 'taking Rawls charitably in such a way to ignore the claims he makes which are insupportable'. Which, between the speculative psychology and handwaving economics, means ignoring 2/3rds of the book.

Of course, what happens in the other 1/3rd is why people still read that book, and read it carefully. But Rawls does present his results as a theorem, and that theorem does not fly.

Daniel said...

As I recall, the problem Rawls has with the first formulation of the two principles in TJ is that they're vague. That version of the difference principle says something about everyone's advantage rather than the greatest benefit to the least advanced, for example, and it's not clear whether `everyone's advantage' should be understood in a maximin way or a Pareto efficiency way (or even a utilitarian way). My Intro to Philosophy students usually think it should be understood in a communitarian way -- `everyone' as a singular, not collective, noun.

If my recollection's right, then Rawls' worry here is prior to any worries about whether the two principles show up in a theorem of either bargaining theory or collective choice theory. We can't try (and fail) to prove claims about them until they make sufficiently precise claims. He's not shifting things around to save some cherished doctrine from refutation; he's removing the ambiguities so that claims about the two principles are susceptible at all to either proof or refutation.

I will agree, though, that shifting about to save a cherished doctrine is what Rawls was doing in the years leading up to the writing and publication of TJ. I just don't remember seeing him do something like this in the text of TJ itself.

Robert Paul Wolff said...

I think that is wrong, Daniel, but it would take me too long to explain why. I have gone into this in great detail in my book on Rawls [UNDERSTANDING RAWLS]. The problem is not that the first formulation is vague. There is as perfectly straightfoward interpretation of it. The problem is that the first formulation is not plausibly the solution of the barganing game Rawls defines in his first stab -- i.e., "Justice as Fairness."

By the way, Rawls does indeed claim in that first article that he is asserting a theorem in Bargaining Theory, and he echoes and repeats that claim in his book. Honest.

Greg said...


"So when you say 'take Rawls on his own terms', you can only mean 'taking Rawls charitably in such a way to ignore the claims he makes which are insupportable'. Which, between the speculative psychology and handwaving economics, means ignoring 2/3rds of the book."

No. What I can (and did!) mean is that we should take Rawls in the interpretive light which he himself practised/advocated. Texts are to be read and understood first on their own terms, but also on the strongest terms available in an interpretive exercise. In this sense Wolff's assertion that TJ is incoherent or yours that 2/3 of the book is insupportable are altogether too cavalier and convenient. To point to a 30yr old book as though it's the final word on a major point of contention in Rawls's work is just lazy. It CAN fly, and to insist otherwise is to not take the material very seriously at all.

Marinus said...

'Theory of Justice' is more than a text: it's the presentation of a philosophical position. So, when you are trying to make sense of it, you have the usual interpretive exercises, but you also must draw out the substantive position articulated in the book. Rawls throws his net very wide indeed, and makes an enormous raft of claims, struggling to get all of it to cohere. Prof Wolff claims that all the claims are not coherent with each other, and I suspect he is right. Rawls does a lot of things of great interest -- something which hasn't been mentioned in the blog posts or the comments yet is the extended and profound discussion of why utilitarianism isn't likely to be chosen in the bargaining game -- but it's a long book, and not all of it is equally good.

Greg said...

Enough with the vague broadsides. Which parts of Rawls do you find 'insupportable'? Be specific.

Charles Richard Booher II said...

"...we should take Rawls in the interpretive light which he himself practised/advocated. Texts are to be read and understood first on their own terms, but also on the strongest terms available in an interpretive exercise."

Rawls was an awful interpreter of philosophical texts. His constructivist reading of Kant has had a disastrous effect on Kant scholarship in the English-speaking world. I don't believe that there is any good reason to follow his example of reading a text 'in its own terms' (which, for him, meant ignoring the context in which it was written and pretending it was addressing only the problems in which Rawls was intersted - hence, this in fact meant on Rawls' terms, not on the terms that the author of the text accepted or recognized).

Further, I see no reason why one should take the principle of charity as far as you suggest one should. Most philosophers were wrong. It is good philosophical interpretation to show how and why they were wrong (or right), not to try and save their views through interpretive and/or technical tricks.

Rawls' influence is real, but his greatness as a philosopher is severely overrated. Instead of Freeman's hagiography, Raymond Geuss presents a strong case for dismissing Rawls rather than engaging him on his own terms. I think that Wolff's book is incredibly charitable in its attempt to take Rawls' pretensions to use Bargaining Theory and Rational Choice Theory.