I’m interested in the equilibrium claim, not for the social issue of value pluralism, but because my hidden agenda is to think about a kind of balancing involved in eudaimonist thought, where as I see it the ancient theorists were trying to fit together both beliefs of various sorts and other attitudes (conative states of various sorts). It is, as you say, a matter of discovering (or establishing) a certain kind of coherence. I think there can be a kind of reconciliation of goods in a sense, though I don’t think that on the ancient views they really are goods in the sense that e.g. Berlin thinks they are (or that Williams assumes they must think they are). So I agree with you there. I also agree with you about the notion of there being an “equilibrium” to seek, if for no other reason than that that’s a metaphor I’d really like to be able to cash out. The “reflective” part I think (as indicated) we can cash out, but you raise some good points about the sense in which there is any equilibrium to be reached. I wish I had a clear story to tell about that too.

So let me change tack and go after the “equilibrium” in reflective equilibrium. For one thing, the back-and-forth, see-saw movement between practice and theory that it is meant to describe never really struck me as aiming for something that’s properly referred to as an “equilibrium”. For where’s the zero-sum dynamic? The point is to make things cohere, not to balance them against each other. But even accepting that some kind of balancing is involved, I would then complain that what we should be aiming for is the genuine reconciliation of goods, not merely (as with value pluralists) their accommodation. And this requires integrating not balancing, which seems to me to be what practical reason at its best should be all about. Otherwise put, we should think of it as more like “conversation” than “negotiation.” And real conversations never end, hence never reach anything that we might refer to as an equilibrium.

There’s a lot of interest in your comment, and I hope you’ll pull out some of the many threads and post them for discussion independently. I’ll confine myself to a defense of the metaphor you express reservations about in the last paragraph.

The interesting thing about “reflection,” considered as a process of thought, is that it is capable of taking itself as an object. This is true generally; it is true of political thinking, moral thinking, etc. And that, to me, is strongly evocative of the visual metaphor. In a looking glass you can see your own reflection — you can see yourself in the act of seeing, in the same way that in (metaphorical) reflection you can think of yourself in the process of thinking. I don’t know of any aural analog for that sort of reflectiveness, and in my view it is of the first importance in both moral and political thinking. That’s in part why I am unhappy about letting go of the very label you’d like to cut loose.

I’m also not sold on the contrast between dialogical and theoretical engagement with others, but that too is a subject for another day.

But I’m no coherentist myself. Oh I’m all in favour of being coherent, but not in the systematic way that “coherentism” requires. Mark raised the question of the motivation behind my posts and so I guess should I come right out and admit that I myself am an anti-theorist (albeit with good intentions! ). That’s why I’m more attracted to value pluralism than I am to what I – admittedly idiosyncratically – call “neutralism,” the form of political philosophy led by Rawls and Co. (I’m also very critical of value pluralism, but that’s for another post.)

So no, I don’t agree that value pluralism is compatible with theories of justice (I’m putting aside the issue of ideology for the moment, i.e., whether a given approach would have us be liberals, socialists, conservatives, or whatever). To me, it all comes down to the question: how should we respond to conflict? Value pluralists say that we should engage in a form of practical reasoning that is unguided by theory, namely, in good-faith negotiations, while Rawls and Co. would have us apply a theory of justice (and when it comes to those less fundamental issues where they welcome deliberation, well then they require that we deliberate in ways that respect the constraints provided by their theories of deliberative democracy). So even though Rawls may accept that his “well-ordered” vision excludes some real values, and so that there’s tragedy in justice, the point is that it is indeed a well-ordered, by which he means systematic, vision – something like a rulebook for a sport or a game that can be neutrally applied. Now while Berlin, like Rawls, is a liberal, his liberalism is not derived from such a theory; rather, it’s simply a rough ranking of values which, within a negotiation session, will call for giving more weight to the values that liberals favour. So there are neutralist liberals like Rawls and there are pluralist liberals like Berlin; the former are theorists, the latter are not.

Mark, I agree with you that not all theories are codes or systems like Rawls’. I distinguish between two kinds of holism, the “systematic” and the “organic”. The former involves wholes made up of interlocked elements or modules, things that can potentially be grasped independent from the whole; while the latter’s parts are always more-or-less integrated together – they cannot be separated from the whole since that whole is literally present in every part. Theories are wholes and ancient theorists, e.g. Plato and Aristotle, almost always aim for organic ones, while most modern theorists (Hegel being the major exception), shoot instead for systematicity. And as I see it, “reflective equilibrium” is perhaps the most sophisticated account going of how we should produce systematic theories (not that, again, I think we should). But even if we choose to give it a wider application, I’m uncomfortable with using it for non-theoretical approaches. “Reflection” makes me think of theoretical reason because, at least metaphorically speaking, both have ocular connotations (theoria, as you know, means “contemplation” or “viewing”). Practical reason unguided by theory is an aural rather than ocular endeavour since it is thoroughly dialogical and dialogue, of course, requires that one listen to one’s interlocutor. But perhaps I’m being, not pedantic now, but anal. Waiter, I think I’ll have that glass of wine after all.

Certainly there shouldn’t be any inconsistency between an endorsement of value pluralism and the very endeavour of having a liberal theory of justice. Rawls comes very close to endorsing Berlin’s value pluralism in Political Liberalism, when he agrees that there is no social world without loss (p. 197; I take it that Rawls does not mean the trite sense of no world without loss that merely notes the contingent limitations on our abilities — I take it he means that in principle not all goods are compossible). He also seems to draw back from endorsing Nagel on the fragmentation of value for reasons to do with the constraints of political liberalism’s philosophical commitments (p. 57).

One claim about RE that I would want to push is that insofar as it is a form of coherentism, it does not necessarily imply greater systemisation in our array of moral intuitions, beliefs, principles, commitments, etc. Sure, we try to build support for each point where it can be found, and we must try to massage away inconsistencies, but an overly neat or codifiable [thanks Mark] outcome is, I would say, less coherent rather than more. To use a bad analogy (and I’m allowed to use bad analogies, since we’re only blogging here) a conspiracy theorist has a very tight web of connections between his various beliefs about the assassination of JFK, the Freemasons, the Roswell UFO case, credit card banking, and the music of Nana Mouskouri. [He has, so we say, a theory for everything.] But that tightness and systemisation comes at the cost of a great deal of coherence with our recognition that sometimes things just do kinda sorta just happen.

But that doesn’t mean we shouldn’t see Rawls’ own view as a conception of what form reflective equilibrium, in a broader sense, might take. Rawls didn’t invent the term either; he boosted it from Goodman, who was advocating it as a test of inference rules. The idea is a powerful one precisely because it is such a ready image — trying to fit a body of attitudes (beliefs among them) in a way that they cohere and we can make sense out of the whole (and perhaps equip ourselves to project the sorts of new or future cases that would also cohere.) Beliefs do cohere or fail to cohere not only with each other but with other judgment-sensitive attitudes (to borrow Scanlon’s term), and the same sorts of balancing and weighing are involved in the process of reconciling and making sense of the broader range of attitudes. We could give that process another name if we needed to, but I don’t see why we need to.

Now, one strain in your thought seems to be a resistance to codification, and therefore a resistance to anything that proposes to produce “theory” — which is, presumably, precisely a codification. And here too I grant that that is one form a theory make take, but it’s hardly the only one. It seems to me we have ready models of efforts at understanding that are valuable to us despite an unsuitability for codification (in fact, I’m inclined to think this is true, in the end, for most interesting forms of understanding). You can tie theory to codification, and tie theory to systematization, but I don’t see why we must or should. Again, granted, we need to observe the ways that other thinkers have done so, but why we should follow suit I simply don’t see.

Now, you originally proposed a phronetic-type activity as a way of comparing values that are incommensurable — and like others I am sympathetic with that idea, taking the incommensurability point to be precisely one about some metric that would allow for a codified method for adjudicating conflicts. If you think the results of an aspiration to reach reflective equilbrium must be capable of systematic expression or codifiability, then I could see why you would reject the connection. But there is no pretense that the method itself is codifiable, and I am not sure why we must think that, if we engage in it, such results can or must be possible. Rawls may think that, but why should we?

Setting aside the point about the reading of NE VI, reading any of this back into Aristotle seems unwarranted. He clearly thinks we can reason about ends, but if he thinks that the result of that reasoning amounts to a “system” of any very robust sort he manifestly fails to produce it. There are things we can know about our final end — and we can arrive at these through social and individual reflection — but casting that as a matter of “theory” independent the very sorts of processes that go into determining what to do seems quite alien to the conception of practical philosophy that runs throughout both the NE and the EE. In fact, he castigates Plato for doing just this in his (Plato’s) conception of the good in Republic. My sense in general is that this is to foist a modern and unwarranted conception of what moral (and other) theorizing must look like on an ancient approach that would have little of it (not just Aristotle, but Plato before him and the hellenists after him). (A really fine recent treatment of this issue, if you haven’t seen it, is Henry Richardson’s Practical Reasoning about Final Ends.)

Finally, I think you can conceive of liberalism as “neutral” either in its practices or in its grounding and legitimating ideals. These might, but maybe in some views might not, be related. I was thinking of it in the second sense. I don’t find the first sort (which I take you to be characterizing) attractive either.

Regarding “reflective equilibrium,” I guess I’m just uncomfortable using the term to refer to a kind of practical reasoning that doesn’t aim for systematic theory. I mean, it’s Rawls’ term, and only those who do his kind of political philosophy seem to favour it. Also, to my knowledge, none of the value pluralists have ever used it to describe their approach. Then again, maybe I’m just being pedantic.

As for phronesis, I would translate NE VI.9 as asserting that good practical reasoning “is conducive to” or “brings success at attaining” or “promotes” some end. Practical reason doesn’t “apprehend” ends in the sense of recognizing or becoming clear about their nature since that’s the job of theoretical reason (as regards ethics). Of course the debate about what Aristotle really meant when he said that we deliberate not about ends but about what promotes ends (NE 1112b) has been going on for a few centuries now…

Consider the individual case first. Rather than insisting that the target equilibrium of such reflection be a systematic theory, for example, why not think of the aim as a set of attitudes — a combination of cognitive, affective, and conative states, of just the sort that the ancients (at least Aristotle and Plato, at points) thought constituted the mental economy of the virtuous agent? And there need be no appeal to justice, either to principles or intuitions, as occupying some special place.

Neither do I think you are right about the “theory-givenness” of the target of phronesis, though theorizing might certainly be part of our reflection about it. Aristotle doesn’t deny that we can reason about ends; he denies that we can deliberate (the process giving rise to boulesis or choice — the “efficient cause” of action (NE VI.2), about ends, but even this point he qualifies by insisting that we can deliberate “about what sorts of thing conduce to the good life in general” (NE VI.5). Practical wisdom, not theoria, is the “true apprehension” of the end (NEVI.9).

The social case is tougher, obviously, because there’s no obvious social analog to this apprehension that can occur in individual agents. But neither, I’d think, would there be any obvious reason why some social function couldn’t be proposed as such an analog (though, so construed, that would seem to me to involve a perfectionist sort of liberalism as opposed to the neutrality sort that Berlin at least seems to favor.) The idea in any event is not simply to appropriate the model of reflective equilibrium from Rawls, nor of phronesis from the ancients, but to identify a new model of the former based on elements of the latter. Anyway, that’s the way I was thinking of it.

I say ‘phronesis-like’ rather than ‘phronesis’ because Berlin, Hampshire, Williams, etc., reject the theory-based target for which Aristotelian practical reason always aims. To Aristotle, one cannot reason (practically) about the ends, only about the means-towards-the-ends; the ends (i.e., the hierarchy of virtues that he believes is necessary for realising human well-being) are arrived at through theoretical reason (theoria). To value pluralists, by contrast, we must engage in practical reason/negotiation as regards both ends and means.

Take two values X and Y, that are contingently incompossible (and not reducible to some more basic value Z) such that a choice of Society 1 is bad terms of X and such that a choice of Society 2 is bad in terms of Y.

The first question is: Can there be any rational grounds to prefer 1 over 2?
The second question is: Can there be any grounds to think 1 morally superior to 2?

I think the answer to both questions is yes. Just because there is no common metric value Z, does not mean there are no other rational or moral resources at our disposal. Indeed, the pluralist will say there is an abundance of such resources. Dime a dozen. Comparability comes quite cheap, and is not really particularly mysterious.

For instance, we can appeal to the values of M, N, O, P, Q, R, S, and T to support the choice of society 1 or to explain why Y is a particularly important value; and appeal to the evils of A, B, C, and D as reasons not to choose society 2. All of these will be rational grounds, and some of them will be moral grounds. (In particular, some of these rational and moral resources will involve value claims specifically about how other values figure in our lives.)

Then the question becomes: what rational reason is there to suppose choice of 1, with its values of M, N, O, P, Q, R, S, T, Y, ~A, ~B, ~C, and ~D, is preferable to the choice of 2, with its value of X. Doesn’t the question just get pushed back to this further level?

Well, yes and no. Yes, in the sense that there aren’t going to be conclusive reasons to demonstrate the superiority of 1 against all possible responses. But I don’t think we should be looking for such an absolute answer here. So no, the question is not problematically pushed back, because 1. some good rational and moral reasons have been given so far, and 2. what’s left is a question of coherent reflective equilibrium (perhaps this is what Charles meant by phronesis). Note, I think this is going to be the same question of coherence that any moral theory is inevitably going to end up with, whether it incorporates value pluralism or value monism. I don’t see why an inconclusive but very coherent liberal pluralist political theory is any more philosophically troubling than an inconclusive but very coherent utilitarian political theory.

The other way to go (which I quite like) is Charles’ point. It matters a lot what is built into the idea of ‘incommensurability.’ If it is just a lack of a common metric, which if we had it would enable us to use a “more is better” heuristic, that still doesn’t preclude the possibility of other bases for rational determination as to how to resolve competition. But my impression (again, no great expertise here) is that sometimes that further possibility is just what the incommensurability claim is meant to rule out: that there neither is nor can be any rational basis for choosing between the claims of competing values. If that is the view, then I don’t see how an appeal to phronesis could help, because phronesis is nothing if not a response to reasons (albeit perhaps not in codifiable form)! I’d think if phronesis is seen as the right response (and I myself am attracted to this way of proceeding) then we have to think there is some rational purchase to be had, even if it is not by means of a common metric. That is, we’ve got to be taking the incommensurability claim in the first (weaker) way. Then, of course, the only remaining task is the teeny one of making out what that rational basis is (but of course one can safely leave that exercise to the reader!).

Thanks, that’s much more clear now, and I see what you mean. There is a lot more work to do, here. I think, however, that Berlin perhaps tacks a lot more on to the Incommensurability Thesis here than it strictly requires. It may be that there is no “common currency” with which to resolve conflicts between values, but this doesn’t mean that their realization must necessarily negatively covary, which seems to be a logically seperate issue. You’re right though, insofar as Berlin adopts this second thesis, nothing about the nature of values necessarily leads us to the conclusion that more of them ought to be realized.

At the very least, it seems that in order to render Williams’ claim consistent with VP, one must supply a good deal of additional content in the form of premises tacit in Williams’ remark. Perhaps details could be filled in which render the claim consistent with VP, perhaps not. But this much is clear: the Williams remark should not be appealed to as an argument *in itself*. But this is how it is often employed in the VP literature: it’s simply quoted as if it were an *argument* for “more is better.” “More is better” is then used as the basis for the entailment from VP to liberalism. So if “more is better” is, to say the least, questionable as thesis to which the VP could endorse, then the entailment it is supposed to secure is jeopardized.