Rawls and ‘Liberalism’

by John Holbo on February 7, 2008

It is often suggested that what distinguishes Rawls’ Political Liberalism from his earlier A Theory of Justice is the ‘political’ bit. This second book is a ‘political’ interpretation of the first one. But I just noticed something. The word ‘liberalism’ does not appear in the index of Theory, and occurs in the text (thank you, Amazon search inside) only three times; none of the three is a self-reference to features of his own theory. What about ‘liberal’? It has no entry in the index either (one entry is for ‘liberal equality’). It occurs 18 times, which is still pretty light. Again, none of the occurrences has a clear ‘mine is a liberal theory’ character. There are several references to works by others with the word ‘liberal’ in the title. The one bit that even makes it into the index is a brief, ‘liberal’ interpretation of equality that is, however, rejected in favor of the ‘democratic’ conception encoded in the so-called ‘difference principle’.

I don’t really have any point to make. But I’m curious. When did Rawls become a ‘liberal’ – when did justice-as-fairness become a theory of ideal ‘liberalism’?



Neil 02.07.08 at 10:41 am

In ToJ, Rawls presents a theory that all rational agents are committed to, whether they know it or not. Nothing liberal about that; that is, nothing that presupposes liberal views on the part of the contractors. In PL, he presents a theory (well, the same theory) now pitched as something to which citizens of liberal democracies are committed, whether they know it or not.


John Holbo 02.07.08 at 11:04 am

“In ToJ, Rawls presents a theory that all rational agents are committed to, whether they know it or not. Nothing liberal about that;”

Well, there is something liberal about it if it turns out that all rational agents are committed to liberalism, whether they know it or not.

Seriously, what I am wondering about is, specifically, when did the term ‘liberalism’ become associated so strongly with Rawls. Because, at a certain point, it became quite common to argue that, since ToJ was flawed, liberalism was flawed. And yet he doesn’t use the word in ToJ. I don’t think this is philosophically crucial – the label game – but I’m investigating the history of the use of the term, and this seems like a potentially interesting data point.


Neil 02.07.08 at 11:12 am

Perhaps this is tied up with the history of the term ‘liberal’ in the US. When did this become a term of abuse? It is strange for an outsider, especially one from a country in which the conservative party is called the Liberals.


abb1 02.07.08 at 11:57 am

Here in Switzerland ‘liberal’ means center-right, associated with big business, mostly banks and pharmaceuticals. Sorta like the US, when you think about it.


aaron_m 02.07.08 at 12:25 pm

To 1, 2 & 3

Rawls argument in a ToJ is NOT that it is rational to accept his principles of justice. What Rawls does is use a thought experiment where agents act in a self-interested and rational way in a certain kind of choice situation. But the reasons why actual people should be moved to accept the principles chosen in the thought experiment is because they see that the construction of the original position is a good representation of what it would mean to base society on a set of rules that treats each individual as a moral equal and in an impartial way.

What Rawls’ thought experiment basically does is force us to question whether any move we make away from his principles of justice are founded in a commitment to treating individuals as moral equals or founded in our own partial interests because we do not think we would personally benefit from what is chosen behind the veil of ignorance. If you agree that in the original position people would choose the difference principle but do not think that society should adopt this view of distributive justice you must ask yourself if you are simply objecting to the difference principle because you are allowing your own partial interests to dictate what are supposed to be impartial standards of justice. Thus Rawls’ entire approach is based on appealing to our moral motivation NOT to give our own rational and partial interests too much weight in considering issues of social justice. Another way to put it is that we must rely on individuals’ moral motivation for them to accept what is chosen in the Ralwsian thought experiment. Rawls has a theory about what is reasonable not what is rational and that is a liberal approach. Hobbes’ theory is an example of a rationalistic view on social justice that should not be considered liberal.

For more look at Barry’s Theories of Justice. TM Scanlon’s theory should also, I would argue, be understood as essentially adopting the same kind of argument as Rawls but avoiding the use of a rationalistic thought experiment that cause a great deal of confusion. Instead he explains the theory in terms of reasonableness directly.


Sam C 02.07.08 at 12:38 pm

Just to add a little to Aaron_M’s very clear post: one might say that the difference between A Theory of Justice and Political Liberalism is that in the first, Rawls takes it that anyone, in principle, could recognise the reasonableness of his principles of justice. In the second, he retreats to the position that only people in societies with liberal and democratic traditions are likely to do so, and helps himself to many more of the specifics of those traditions. The move is towards Michael Walzer’s version of cultural relativism, and away from A Theory of Justice‘s ambition to see things sub specie aeternitatis. The two books both talk about what we can agree on, but give different answers to ‘who’s we, paleface?’.


Jacob T. Levy 02.07.08 at 12:48 pm

Brian Barry’s immediate book-length commentary on Rawls (published 1973, so written 1972) is titled “The Liberal Theory of Justice: A Critical Examination of the Principal Doctrines in “A Theory of Justice” by John Rawls.”

And CB MacPherson’s review of the Barry book
says “Barry’s title is well chosen. *The* liberal theory of justice is indeed what Rawls’ theory is now well on its way to becoming”– Can J Pol Sci 1974. That it is *a* liberal theory of justice can presumably be taken for granted.

(MacPherson predicted both that Barry’s would be the primary work of criticism of Rawls for a long time– just as Anarchy, State, and Utopia was being published– and that Rawls’ theory might well not “survive” Barry’s critique, whatever that might mean.)

I think the answer to John’s question is: immediately. That doesn’t tell us when Rawls himself became concerned about the word or concept of “liberal,” but it was received as a liberal view right away.


John Emerson 02.07.08 at 2:04 pm

Perhaps this is tied up with the history of the term ‘liberal’ in the US. When did this become a term of abuse? It is strange for an outsider, especially one from a country in which the conservative party is called the Liberals.

Liberal became a term of abuse in the U.S. as the result of a decades-long media campaign. Liberalism became identified with illicit sex and drugs, military defeat, government largesse to unworthy, poor recipients, and amoral, lax and indulgent childraising, education, and criminal justice. The accuracy of these accusations is somewhat beside the point by now, since a considerable majority of the population accepts them to some degree.

This was a rejection of the liberal Democratic tradition running from FDR to LBJ (1932-1968). While liberals never dominated the Democratic Party, they were very influential during this period.

Why the term “liberal”? My theory is that FDR realized that Americans would never accept socialism, so that he shoehorned a weak kind of socialism (i.e. fascism, thank you Jonah) into liberalism by calling it a freedom: “Freedom from Want” in the Four Freedoms. This was not a freedom recognized by earlier liberals.

As a footnote, around 1930 Minnesota’s governor Floyd B. Olson of the Farmer-Labor Party declared himself to be a socialist while stating his willingness to work with liberals. But a high proportion of Minnesotans at that time were first or second generation German, Scandinavian, or Slavic immigrants who did not feel the American antipathy to Socialism.

Today’s right traces back (through William F. Buckley and others) to people who never accepted Roosevelt’s liberalism, and as we have seen with the Social Security “reforms”, they really want to turn the clock back 77 years or more. (George Will wants to go back even further, to the per-Teddy Roosevelt era).


Matt 02.07.08 at 2:20 pm

By looking at the index of Rawls’s _Collected Papers_ it looks like the first time he really starts talking about liberalism very much is in “Justice as Fairness: Political not Metaphysical”. That makes sense in that it was around this time the he was moving to emphasis the ways that liberalism was (or can be) independent of comprehensive moral views. Before this point it wasn’t as important to make the emphasis be on liberalism as such, I’d think.

(I’d also insist that Sam C’s account, though a commonly accepted one, is a bit misleading as to Rawls’s development here- his later view surely has some similarities with Walzer but they are more on the surface than deep. Walzer’s view is a metaphysical view in _Spheres of Justice_ is a metaphysical view about value, that it depends on culture in important ways, and so really is relativist in that sense. That’s not what Rawls supports in his later works at all. He never gives up the claim that some version of the view presented in _TJ_ is _the right_ view of justice, just the claim that any reasonable and rational person should be able to come to accept it. So, to the extent there is any relativism at all in Rawls’s later work (to my mind putting it like that obscures more than illuminates) it’s at the level of politics, not deep claims about the nature of value as found in Walzer and other communitarians. (Mulhall and Swift’s chapter in the _Cambridge Companion to Rawls_ is good on Rawls’s relationship to communitarian thought.))


aaron_m 02.07.08 at 3:14 pm

To defend Sam C’s interpretation, Rawls’ project of political liberalism seems to be a bigger move towards Walzer than Matt allows for. This is because Rawls appears to be working more in the spirit of interpreting what he takes to be standards of social justice in a certain political culture and backs away from universalistic interpretations.

This is not always obvious in PL, but in The Law of People Rawls is explicitly defending a diversity of political cultures in a way that does appear to be communitarian. He argues that decent hierarchal peoples (i.e. hierarchical political cultures) and liberal peoples are to be viewed in a law of peoples as self-authenticating sources of claims (i.e. ends in themselves). Individuals in decent hierarchical societies are “not regarded as free and equal citizens, nor as separate individuals deserving equal representation,” while they “can recognize when their moral duties and obligations accord with the people’s common good idea of justice” (LoP, p. 71). The upshot is that Rawls is advancing a theory that justifies the interests of communities to have their ends, at least to some extent, take priority over the ends of individuals. This is not a liberal standard of justice. Still, what dictates the standards of social justice individuals have access to in their respective political communities is dependent on what kind of political culture the society happens to have, with some limits for a fairly basic notion of human rights (i.e. what is to count as a decent hierarchical society against a non-decent one). Thus, if we think of what the argument in the LoP attempts to justify from the perspective of individuals in hierarchical societies the move towards communitarianism is made apparent.

On the above interpretation one might suspect another reason for Rawls’ increasingly using the term liberal in his latter work. In the ToJ stuff there was no need to use the label because that is how the theory was clearly going to be interpreted, while in his later work he advances a theory that seems to challenge key liberal tenants and thus there is more need to use the name. Well maybe this is part of the reason, but it is just a wild guess.

Of course some argue that Rawls has from the beginning been advancing an interpretive theory for liberal democratic states in his domestic theories of justice.


Sam C 02.07.08 at 3:15 pm

Matt makes some good points, but I stand by the claim I made about Rawls’s increasing relativism. What he and Walzer come to share is a view about justification: the kinds of reasons one can give for adopting a political regime are drawn from shared culture, not from rationality or from shared human nature (again, who exactly is the we appealed to in ‘we agree…’?). I didn’t make it at all clear that this was what I was getting at above, though (and I agree about Mulhall and Swift).


Matt 02.07.08 at 3:19 pm

Aaron_M- I’d argue that while your reading of _The Law of Peoples_ is also a common one (found in many good works by Tan and others) that it’s also not right. Rawls thinks that “decent hierarchical societies” are _unjust_ and never says otherwise. The mistake in the approach you favor, I think, is seeing _LP_ as presenting a theory of _global justice_ while what it means to present is a theory for the international policy of a liberal society. So, the question is not, “are decent heirarchical societies just?” (Rawls’s answer to that is clearly “no”.) but rather, “Should we be willing to cooperate with DHS on equal terms”? To that Rawls says yes. Now, that migth be wrong (though I think not) but it’s not a move that involves relativism in any deep way at all. (To my mind the best essays helping bring this point out are the two on _LP_ in Samuel Freeman’s _Justice and the Social Contract_.)


Matthew Kuzma 02.07.08 at 3:26 pm

When it became clear that conservatives have no interest in justice as fairness and greatly prefer the personal morality narrative that stems from justice as individual accountablity?


engels 02.07.08 at 3:37 pm

But is Justice as Fairness a liberal fascist conception?


aaron_m 02.07.08 at 3:46 pm


In LoP Rawls in clearly not advancing only a pragmatic argument about what is justifiable in international relations in an imperfect world. In LoP he clearly compares the idea from PL of accepting a plurality of individuals’ views on the good with the idea of accepting plurality of views on the good political culture. Thus as Nagel notes, Rawls is advancing a principled defence of hierarchal peoples (i.e. pluralism on the good political culture) in the same way he advances a principled not pragmatic defence of individual pluralism of the good in his domestic theory (i.e. a principled defence individuals illiberal conceptions of the good). The problem is of course that one cannot universalistically see individuals as self-authenticating sources of claims and then claim that communities can also be self-authentication sources of claims in the same way. This is because some political cultures reject the liberal principled defence of individual pluralism on the good. Thus the choice is to either give up on a universal principled defence of individual pluralism on the good or to give up on a principled defence of pluralism on what counts as a good political order. In LoP Rawls defends thus does not give up on the later, which leaves only giving up on the former. Rawls of course does not say this and he may not intend it, but the choice remains.


SeanD 02.07.08 at 3:46 pm


It seems right that Rawls’s theory can’t be a theory of what a fully just world would look like, since it allows for (admittedly) unjust decent societies. But I don’t think we need move all the way to the ‘foreign policy only’ interpretation to admit this point. One interpretively plausible middle-ground position would be to take LP to be a theory of justice for the global basic structure; where that last is (normatively) understood primarily in terms of relations among peoples. It doesn’t seem crazy to suppose that a global basic structure of this kind could be just even though not all of the societies participating in it are just; witness domestic basic structures, which can , I’d argue Rawls’s theory implies, be just even if they contain unjust associations (church hiearchies, schools, families to some extents, etc…)

There may also be an argument to be made that a theory of just foreign policy presupposes a theory of justice. How, after all, would you know that the foreign policy was just in the global context, if you didn’t know what justice was for that context?


Matt 02.07.08 at 3:59 pm

Aaron_M- I doubt we’ll settle this in blog comments and we’re now pretty far from John’s post but I still don’t think that’s quite right. (I don’t think Nagel is the best interpriter of Rawls, either. I’d be pleased if Jon M, who is an excellent interpriter of Rawls, would pipe in, though.) But still, Rawls’s defense of DHS’s is only in the matter of international cooperation- his point is that we should be willing to cooperate with such societies on equal terms. I don’t think there’s more to it than that. There’s no claim that societies count as sources of claims “in the same sense” as people do- that’s just a misreading. Again, Rawls’s view is that DHS are _unjust_- it’s an imperfect order that contains them. But, given that they do exist, should we be willing to cooperate with them on equal terms? That’s the question of this part of LP and I don’t at all see that it implies what you (and Nagel, I guess) think it does.


Paul 02.07.08 at 4:27 pm

Regarding the initial question, Matt is right. The focus on “liberalism” gains prominence in “Political Not Metaphysical.” There Rawls is concerned to explain how a society can avoid “autocratic” uses of political power, by which he means (admittedly still cryptically) uses of political power that can’t be justified in terms that it is “reasonable to expect all citizens reasonably to endorse.”

In “Domain of the Political and Overlapping Consensus”, Rawls attempts to explain how what he there calls a “political liberalism” can avoid using citizens’ corporate political power “autocratically.” Here the notion of “liberalism” is at the center of the discussion, and I think a lot of this has to do with Rawls’s appreciation of many of the points made in Waldron’s “Theoretical Foundations of Liberalism” (Rawls cites Waldron in a footnote in “Domain”). The aim of addressing justifications of the use of political power to each citizens “common human reason” is associated with liberalism by Waldron. Since Rawls is concerned with the best way so to address each citizen, he appears comfortable appropriating the idea of liberalism for this project.


Elias 02.07.08 at 4:59 pm

Oh, it’s not only that. All sorts of people think that Locke’s Second Treatise was written by an English dude. But I just did a text scan for “English dude” and it never appears in the text. You know what else doesn’t appear there? “Lockean.” Mind-blowing.


aaron_m 02.07.08 at 5:12 pm



Too little time to write my own answer

I have included some long quotes from GLOBAL JUSTICE AFTER THE FALL: Christian Realism and the “Law of Peoples” by Edmund N. Santurri that may help explain my interpretation of Rawls.

And I promise stop now and not include such long quotes or posts in the future, but I just could not let it go this time.

The problem

“Rawls admits that from the vantage point of reasonable liberal peoples and, it would seem, from the perspective of justice ideally considered, decent hierarchical societies are less than fully just inasmuch as they lack the full array of basic democratic institutions that would be affirmed by self-interested rational contractors in the original position set out in the first stage of the argument described earlier in this article (Rawls 1999a, 62, 78, 83).
This admission of Rawls’s, naturally, raises the question whether allowing representatives of unjust, albeit decent, societies as parties in deliberation about principles of international justice might not necessitate a compromise of the ideal in the eventual principles chosen. And if the perspective of “outlaw societies” is judged irrelevant to the determination of international principles of justice in ideal theory, why is the perspective of unjust, albeit decent, societies permitted in the ideal formulation of those principles of international justice especially given reason to believe that such an allowance would affect the determination of principles in a theoretically significant way? For example, since a decent hierarchical society typically will privilege, in the Rawlsian account, particular religions or “comprehensive doctrines” through the establishment of religious qualifications for political office, representatives of such a society would not assent to a principle of international justice that identified discrimination of this sort as in violation of a fundamental human right to equal liberty of conscience. Accordingly, Rawls indicates that such a principle would not be included in the Law of Peoples. Yet Rawls also contends that in the first stage of “justice as fairness” hypothetical individual contractors who are fairly deprived of specific knowledge about their religious convictions would indeed settle on equal liberty of conscience as a fundamental principle of justice, and such would seem to suggest that the principle ought to be recognized generally as a standard of just international association, given the attractiveness of the standard under fully impartial conditions of deliberation (Rawls 1971, 205–11)” (Santurri, p. 788)

Rawls approach to explaining the influence of DHSs in ideal theory is:

“decent peoples are said to warrant our moral respect because they acknowledge and support a range of human rights, accord their citizens a consultation role in public policy, and “allow a right of dissent” matched with procedures of “respectful reply” by governmental and judicial officials (Rawls 1999a, 61). Though decent hierarchical peoples are not fully just and though liberals rightfully hope that these societies will reform in liberal directions, the moral respect decent societies should elicit argues against their being subjected to political, economic, or military sanctions designed to advance liberalization. On the contrary, moral respect for decent hierarchical peoples argues for their “self-determination,” affirmation of which follows particularly from “fair equality and a due respect for other peoples” (Rawls 1999a, 61–2, also 85, 111–2). Thus, toleration of nonliberality, given this line of argument, is a matter of moral principle, where considerations of “mutual respect” trump considerations of justice (Rawls 1999a, 62)” (Santurri, p. 798).

Saying that DHS are ‘unjust’ but that ‘morality trumps justice,’ seems to simply be an offside semantic manoeuvre that does not change the fact that Rawls appears to be offering a principled defence of political culture taking priority over the interests of individuals.

“It is one thing to say that liberal foreign policy must set aside certain demands of justice, given the exigencies of international political reality. It is another thing to say that in contractarian determinations of international principles of justice for ideal theory, the principles chosen must be acceptable to representatives of unjust, albeit decent, peoples even if the content of the principles chosen will be compromised by admitting the perspectives of those representatives” (Santurri, p. 799).

Doing the later is a whole lot different than the pragmatic interpretation you are advancing of Rawls.


Matt 02.07.08 at 5:31 pm

Okay- one quick last reply. You can see that this reading is mistaken from this bit here, _”It is another thing to say that in contractarian determinations of international principles of justice for ideal theory, the principles chosen must be acceptable to representatives of unjust…_”

Why is it mistaken? Because DHS’s are not part of (fully, at least) ideal theory. They are already in a part of _LP_ that is extended beyond ideal theory. I think you’ve got an unreliable guide here and I suggest finding a better one (Freeman is, as noted before, probably the best place to start.)


John Holbo 02.07.08 at 5:45 pm

Thanks for the assistance – Jacob, others. And feel free to indulge in erudite, tangential Rawlian bickering, aaron, others. I think I have my specific answer now.


Jon Mandle 02.07.08 at 6:17 pm

Flattery will get you everywhere, Matt…
On The Law of Peoples, very briefly, I think Matt is right that Rawls never says or intends to say that decent hierarchical societies are just. Rawls notes that such a society has a “common good” conception of justice. What this means, exactly, is obscure. But he is clear about one thing, at least: “To repeat, I am not saying that a decent hierarchical society is as reasonable and just as a liberal society.” (p.83) My own view is that the common good conception of justice is tied to the issue of legitimacy, and decent hierarchical societies, although not just, do have a legitimate political order. And this grounds the attitude of tolerance toward them. (I discuss this in a paper called “Tolerating Injustice” in Harry and Gillian Brock’s collection, The Political Philosophy of Cosmopolitanism.)

John’s observation is interesting and I’ll think about it some more. But off the top of my head, I’d say that it’s important to remember that when Rawls wrote TJ, TJ hadn’t yet been published. My guess (I’m happy to be corrected on this) is that the term “liberalism” wasn’t commonly used to designate a position in political philosophy, but rather an orientation in politics or public policy. I don’t think Rawls would have wanted to have his view tied too closely to this orientation – it might not even have occurred to him – for two reasons. First, he didn’t want the theory tied too closely to any particular set of policies. But second, to the extent that “liberalism” was associated with meritocracy, welfare state capitalism, and the rejection of socialism – not to mention Vietnam – that label just didn’t fit. He was to the left of such “liberalism”. When TJ came to be recognized as a definitive statement of liberal political philosophy – immediately, as Jacob points out – this shifted how the term was used, at least in philosophical contexts.


loren 02.07.08 at 6:47 pm

“Again, none of the occurrences has a clear ‘mine is a liberal theory’ character.”

I guess, but by the second (substantive) page of the book Rawls has asserted the primacy of justice over considerations of order and efficiency; he’s linked justice to the inviolability of persons; and he’s insisted that the rights secured by justice be removed from the vagaries of political bargaining or a “calculus of social interests.”

If it walks like a duck …


trane 02.07.08 at 7:13 pm

Don’t worry about the length of comments and quotes. My theory of your comments is that all reasonable (but Rawls-ignorant) people) such as myself, learn something. Thank you.


abb1 02.07.08 at 7:39 pm

I have learned that this doctrine is as full of contradictions and open to interpretation as any other, or more.

“…decent hierarchical societies, although not just, do have a legitimate political order” – a bit heavy on semantics, isn’t it.


aaron_m 02.07.08 at 7:46 pm


Well Rawls brings in DHS in what he calls “The Second Part of Ideal Theory,” and he uses this reasoning to reject the ideal theory advanced by cosmopolitans, e.g. global distributive justice, because it would limit the primary interests of peoples to self-determination on their own conceptions of the good political order.


SeanD 02.07.08 at 7:54 pm

The following TJ passage might be relevant to the original question; as it seems to entail, at least, the Rawls-of-TJ doesn’t take being liberal and being just to be the same thing:

A conception of social justice, then, is to be regarded as providing in the first instance a standard whereby the distributive aspects of the basic structure of society are to be assessed. This standard, however, is not to be confused with the principles defining the other virtues, for the basic structure, and social arrangements generally, may be efficient or inefficient, liberal or illiberal , and many other things, as well as just or unjust. (TJ 9)


Jon Mandle 02.07.08 at 7:55 pm

I didn’t mean to rely heavily on semantics. Justice and legitimacy are not the same, although I believe they are related. To get the sense of the difference, consider the possibility that I believe candidate A will promote policies that are more just than candidate B. But suppose in a fair election, candidate B wins. I might say that justice would have been served better if A had won, but that the outcome is legitimate.


abb1 02.07.08 at 8:21 pm

“Legitimate”, as I understand, could mean “lawful” or it could mean “just”.


abb1 02.07.08 at 8:30 pm

And what’s ‘decent’ – unjust but with their clothes on?


John Emerson 02.07.08 at 8:53 pm

When it became clear that conservatives have no interest in justice as fairness and greatly prefer the personal morality narrative that stems from justice as individual accountablity?

At the dressed-up, philosophical, justification level “accountability” is the buzzword, but at the gut level I think that American conservativism at least only needs the establishment of hierarchies by some sort of comepetition, and you don’t have these hierarchies if the losers and sinners and failures aren’t punished. The people punished might be lazy, criminal, immoral, genetically or racially inferior, untalented, or just unlucky, but their existence is part of the happiness of the successful winner, and their goal is not only to accept their lot, but also to unquestioningly accept that they deserve their suffering.

Meritocratic and moralistic versions of this kind of schadenfreude are easiest to justify, but Calvinistic notions of God’s mysterious grace stress the arbitrariness of God’s judgement, and one major GOP faction — the high-stakes gambler types — live to see someone’s fate decided by the turn of a card or the flip of a coin. That’s how completely profane whoremongers manage to fit so neatly into a Christianist political party.


ben saunders 02.07.08 at 8:54 pm

I read that quotation as simply saying that justice and liberalism aren’t the same thing, so it isn’t an a priori truth that the just basic structure will be liberal; though it might turn out (and indeed does) that the substantive requirements of justice have liberal features.


ben saunders 02.07.08 at 8:55 pm

Sorry, #33 is referring to TJ p.9 quoted in #28.


SeanD 02.07.08 at 10:58 pm


That seems right; I didn’t mean to present it as evidence that Rawls takes his theory of justice to be a non-liberal one. It might suggest, though that at the outset of TJ, unlike the outset of his later work, he doesn’t see what he’s doing as fundamentally a project of developing a liberal conception of justice; since if that’s what he’s doing, one wouldn’t think that he’d have explicitly disavowed the project of saying what a liberal basic structure would look like.


Dr Zen 02.08.08 at 12:53 am

He probably didn’t mention it because he felt it went without saying. What else would a liberal believe? I am a liberal *because* I believe in justice as fairness. As a pragmatic matter, I think equity is more achievable than equality, and arguably more desirable, and Rawls appeals on that basis.


James Bourke 02.08.08 at 4:28 am

What many of the comments have seemed to ignore, and one aspect of ToJ that seems to me to contribute to a liberal interpretation of the theory, is that for Rawls the first principle is lexically prior to the second, and when he clarifies that in response to Hart the liberties of the moderns take priority over the political liberties, if memory serves (granted the response to Hart comes much later, but it’s to clarify the aim of ToJ). One of R’s motivating concerns in ToJ was to defend the priority of rights against utilitarian conceptions that threatened not to take seriously the distinction between persons. As he stresses here, justice as fairness is a rights-based conception. It seems to me that that counts for something in chalking the early R up as a liberal.


zdenek v 02.08.08 at 1:11 pm

There is similarity between the problem Rawls faces in Political Liberalism ( ie in modern society people may hold different philosophical and mtaphysical doctrines which are all ‘reasonable’ and that in the face of this fact we cannot justify political policy on metaphysical basis since such grounds are not acceptable to all ) and what confronts Kant in the opening argument of the 3rd section of the Groundwork when he is looking at the foundation of the categorical imperative : Kant thinks that there is a kind of contradiction here since the will has to be autonomous but it must also be based on a principle but such law / principle cannot be imposed from outside. So he thinks the will must adopt a principle for itself ( but this is a problem because it looks like the free will by imposing a principle upon itself is forced to restrict its own freedom. ).

In other words Kant’s and Rawls’ problems are similar because in both cases we are looking for principles to ground out action/policies but the very set up of the situation forbids us to choose any *specific* principles.

What is interesting is that Rawls approach to this problem ( PL 89-125 ) is also similar to Kant’s approach : constructivism ( does it work ? ).


josh 02.09.08 at 5:30 am

On the question of whether Rawls thought of his position as ‘liberal’ during the writing (and re-writing, and re-re-writing …) of ToJ, I don’t think its quite right to say that ‘liberal’ would have been thought of as referring to a political, not a philosophical, position. There were quite a few people applying the terms ‘liberal’ ‘liberalism’ and ‘the liberal tradition’ to positions in political philosophy. And Rawls was certainly aware of this. As we know from Thomas Pogge’s book on Rawls, Rawls’s thinking was influenced by a seminar in which he participated at Oxford in 1953, on ‘The Moral Presuppositions of Liberalism’; the seminar was convened by Isaiah Berlin and Stuart Hampshire, and Herbert Hart participated prominently in it. If memory serves, Hart used ‘liberal’ to refer to a political-philosophical position in print, and Berlin certainly did; I suspect Hampshire did as well, though I can’t think of any instances before the publication of ToJ.
Now, I do think that, given how ‘liberal’ was used by others, Rawls would have likely considered himself as writing from a broadly liberal position; but he may also have thought that characterising his theory as liberal would make it appear more opposed to socialism than it actually was. But that’s sheer speculation on my part (though I think it fits well with the distinction between ‘liberal’ and ‘democratic’ equality mentioned by John in the original post — which also suggests that Rawls conceived of liberalism as a political-philosophical position, no?)


Jon Mandle 02.10.08 at 12:56 am

I think you’re right, Josh. A quick search for entries in the 1960s in philosophy in JSTOR found many references to “the liberal tradition”. There was a consensus that Mill and Locke were part of it, but there was some controversy about Kant. My sense – again, speculative and subject to correction – is that what made a philosophical position “liberal” wasn’t entirely clear, and the dominant use of the term came from political science, policy, and maybe even economics. Barry’s 1965 Political Argument has an explicit discussion where he is trying to pin down a properly philosophical definition of liberalism.


josh 02.12.08 at 3:54 am

Thanks for the endorsement, Jon! I’ve been doing a bit of work on definitions of the ‘liberal tradition’ — in relation to Berlin, rather than Rawls — and it did vary pretty widely (the one point on which everyone agreed being, as you note, that J.S. Mill was a liberal). The seminar that Rawls attended at Oxford looked at Condorcet, Kant, Mill, and G.E. Moore (!) Berlin variously identified Mill, Constant, de Stael, Paine, Bentham, Tocqueville, Kant, Locke, Sismondi, Montaigne, Montesquieu (with qualifications), the young Schiller and Fichte, Humboldt, Erasmus, and even Ockham as part of, or at least contributing to, the ‘liberal tradition’; Plamenatz’s Readings from Liberal Writers included Milton, Locke, Montesquieu, Voltaire, Diderot, Rousseau, Hume, Burke, Bentham, Constant, Mill, Tocqueville, Acton, and Tawney.
All of which does not really inspire confident belief in the existence of a coherent ‘liberal tradition’; there certainly doesn’t seem to have been much philosophical coherence to it. I tend to think that the idea of a ‘liberal tradition’ and the use of ‘liberal’ associated with it derived from historical works which projected late 19th and early-20th century liberalism back to Locke — though there were certainly political scientists and economists (notably Hayek) who contributed to this, and you’re probably right that they tended to use the term more than philosophers.

Comments on this entry are closed.