Featured Philosopher: David Enoch

37 Replies to “Featured Philosopher: David Enoch”

1. Hi David,
   my question is born out of curiosity. As you say, most of your work has been in meta-ethics, where you defend robust and non-naturalistic moral realism. You have argued for this view, in part, by arguing that presupposing that there are moral facts is an indispensable precondition of practical reasoning, and by arguing against other views. Now, in the paper I commented on in Amsterdam, you argued that it makes sense to defer to moral experts sometimes, because we sometimes come to realize, after having first been reluctant to do so, that they were right all along. I am now excited to see that you are moving on to political philosophy. Given your attraction to realism about the moral, it seems in character, so to speak, that you would be “going after”, as you put it, Public Reason theories. I must confess – before moving on – that I have not read the papers on Waldron and Gaus, but I did just have a look at the abstract for the Arizona workshop, and it looks like exciting work indeed. OK, so now onto my question (again, which arises out of curiosity). I noticed that, in the Arizona abstract, you did not mention (or if you did, I missed it) what you wish to put in the place of the Public Reason theories. I am curious about what you would like to replace it with. And I am also curious about what your normative views and leanings are. Given that you are such a staunch and, as you put it, “shameless” defender of moral realism, it would be very interesting to hear a little about your normative views (viz. the normative or moral truths, as you see things). So, this is an invitation to put your meta-ethical and meta-political views into relation with the substantive or normative views they are meta-views about. I ask out of curiosity, as I say, because I take it when people defend moral or normative realism, and do so with as much passion as you do, then this is likely to come out of a strong passion about certain normative issues. The meta-ethical parts of Derek Parfit’s On What Matters, for example, give off the impression of being the meta-ethical outshoots of somebody who really takes agony to be something really, really bad, which we should work together to rid the world of. And when Parfit discusses well-being, he tends to say that “hedonism is a large part of the truth”. In his case, then, the moral realism seems, to a large extent, to be driven by a substantive concern with promoting happiness and preventing agony. So, out of curiosity, do you, in a similar way, have certain views in normative ethics that drive your meta-ethical and meta-poliitcal theorizing – or are you undecided in normative ethics and political philosophy, while yet being very decided in the meta-parts of the two?

2. Thanks, Sven.
   The relation between my metaethical views and some of my other views is not always entirely clear to me. As you note, there is at least a relation of temperament, and like you, I wouldn’t be surprised to see some correlations between people’s metaethical views and at least some of their normative views. Also, I wouldn’t be surprised if there are some interesting relations on the level of the motivation for views – indeed, my motivation for defending realism is normative (something about, well, taking morality seriously).
   Still, whether there are closer relations than this – entailment relations, perhaps, or perhaps something like coherence relations in one of the senses of this expression – remains to a large extent an open question, I think.
   How about Public Reason? Again, my official view is that there are no close relations to my metaethics here. I mean, you can sometimes hear the suggestion that we should go for a Public-Reason-ish political philosophy because [enter some less than fully realist metaethical view here]. But these claims are, it seems to me, especially confused, and the kind of view entered in the brackets is usually the kind of metaethical view that is not really a contender in the current metaethical literature (like some sophomoric relativism, or the kind of no-objective-truth-emotivism that it’s been a large part of the business of current expressivists to distance themselves from). Still, in critically evaluating Gaus’s book, I found myself referring to my metaethical stuff more than once. Perhaps this is not so much because of his conclusions as because of his arguments (that at times sure sound metaethical). Not sure.
   Ok then, what are my normative views? In many cases the answer will be “not sure”. But let me just say something quick about the political case. I think – but I’m not sure at all – that I’m a minimalist and a pluralist here. I’m a minimalist in the sense that I think we should reduce our expectations. Somewhat caricatureistically: I think that if a regime reduces the amount of blood spilt in the streets, it’s legitimate. Not much more, anyway, is needed for legitimacy. On the other hand, many thing may matter politically. That is, other things being equal, we should prefer a regime that satisfies them to one that doesn’t. In this way I can even accommodate something like the underlying Public Reason intuitions: I don’t think that anything about the acceptance of the governed is at all necessary for legitimacy. But it may very well be the case that other things being equal we should prefer a regime that enjoys some such acceptance to one that doesn’t. In this way, then, I’m a pluralist.
   But I don’t pretend that this is at all well worked-out. Still, you asked…

3. David,
   Let me start by admitting that this is somewhat self-serving, since (as you know) I’ve focused on your work a good deal in my own, and I’m currently working on a paper that continues this trend. I think, and certainly hope, though, that my question—or, at least, the broader issues it raises—will be of general enough interest to fit in here.
   I take the gist of your position in Taking Morality Seriously to be something like this: We shouldn’t pretend that Robust Realism (RR) doesn’t come with theoretical costs; it does. But there are good reasons not to be error theorists and no metanormative alternative can avoid error theory and also make good on various important features of normativity. For instance, you point out that, for quasi-realists, normative truth still somehow depend on us, and this seems to fly in the face of normative objectivity. In various places, you raise the just-too-different intuition against naturalism. And you just mentioned your work elsewhere against constitutivism.
   Now, as you yourself say, the indispensability argument you give for RR can be read as a kind of transcendental argument: Given our need for normative truth in deliberation, we are rationally committed to there being such truth. So, my first question is: Why is this kind of argument closed off to your competitors?
   For instance, quasi-realists like Gibbard seem to think that we are rationally committed to there being true and false normative claims. And as Mark Schroeder points out in Noncognitivism in Ethics, other non-cognitivists like Hare and (on a reading Mark alludes to) Korsgaard seem to think we are rationally committed to certain first-order normative truths. If any of those arguments can be made successful, it seems like they would provide a counter to the charge that dependence on us entails non-objectivity.
   (Similarly, I think that if naturalists could take your transcendental ball and run with it, they might be able to make good on just-too-differentness; but since there’s already a literature on the expressivism/constitutivism stuff, I won’t say more about that here, lest this get far too self-serving.)
   Of course, a simple response to all of this might just be that you think what Gibbard, Hare, Korsgaard, etc. have done fails. But that doesn’t show that, say, some other transcendental argument can’t work for some non-RR view. Thus, my second question: Given that you (a) already champion a transcendental argument of some kind and (b) seem willing to admit that if certain metanormative alternatives could make good on objectivity, just-too-differentness, etc., that would be best, since we wouldn’t have to bear the theoretical costs that attend RR—why isn’t a transcendental argument for non-RR normativity the avenue most worthy of pursuit?

4. Many thanks, David, for the reply!

5. Hi David (if I may), and thanks very much for doing this!
   My question has to do with your response to the epistemological challenge to non-naturalism.
   In Taking Morality Seriously, you defend your non-naturalism from epistemological objections with a so-called “pre-established harmony” explanation. Briefly, the moral beliefs that result from evolution are sufficiently correlated with the non-natural moral truths to explain how we get things right at least some of the time despite our not being in causal contact with those truths.
   One assurance we might want out of our moral epistemology is that there is some process available to us to improve our beliefs over time. Ideally, this process would do more than just render our previously held beliefs more coherent. So let’s suppose that your explanation does explain a greater than 50%-ish correlation between our moral beliefs and the moral facts. Does your explanation also explain how our moral beliefs can improve over time, and if so how?
   An initial response might be that my question isn’t one that your solution is supposed to answer, but a different question altogether. But the reason why your solution raises this worry for me is that since our beliefs aren’t directly responsive to the truths in question, even if evolution ensures that we are doing ok, it won’t (alone) ensure that we’ll get any better over time.

6. Hi David. Thanks, this is interesting.
   There is something dialectically surprising about your suggestion, I think – usually, when people smell a transcendental argument, they proceed to claim that it doesn’t support its conclusion, not that it can be the route to even more transcendental arguments… But of course, this is no response.
   Now, if the question is just “how do you know that we’re not going to find a transcendental argument supporting some alternative view?”, then the answer is “I don’t”. This is a just a particular instance of not knowing what argument in general (transcendental or otherwise) we’re going to find. We must remain at least this much open-minded – if new evidence comes in, we should reevaluate our beliefs, here and elsewhere. But from this it doesn’t follow that the mere epistemic possibility of future evidence coming in should somehow change how we respond to the evidence we currently have, no?
   You may be saying that I’m ignoring evidence that’s already there, because Korsgaard, or Gibbard, or … have already supplied such arguments. In that case I respond, as you anticipate, that I remain unconvinced (for reasons some of which I go through in the book and elsewhere).
   Or – you may after all be arguing against my own transcendental argument. You may be saying that some other views may in fact give us all that is needed for deliberation (say, a commitment to first-order normative truths). This would of course undermine my indispensability argument. But then we simply have to check the details and see to what extent this is so. Here I am happier with what I have to say against some views (e.g. constitutivism) than against others (naturalism, where pretty much all I have – except for going after specific naturalist views, like Mark Schroeder’s – is the “just too different intuition”). But I acknowledge, of course, that this is a necessary step in my indispensability argument.

7. Thanks, Preston.
   Not sure I thought about your version of the challenge (so this is helpful!), but at the end of the day it really doesn’t matter whether this is something my solution was intended to solve, right? What matters is whether it can make progress on this.
   Now, you nicely distinguish between getting closer to the truth by improving coherence, and otherwise, and then you (tentatively, with an “Ideally”) suggest that we want progress towards the truth not just by getting more coherent, right? I think I want to reject this further desideratum.
   I’m guessing the kind of model you have in mind is perception-like, right? Where – to put things in a somewhat oversimplified way – the picture is that regardless of one’s own other beliefs, the world intervenes, and this input tends (at least for a range of cases) to get our beliefs closer to the truth. Now, I don’t think this is exactly right even in the perception case, but I do think there’s something along these lines that makes sense (it’s not without reason that perceptually-formed beliefs are the example coherentists have that hardest time accommodating). But this model has limited scope. How can it be applied to a priori beliefs?
   Once again, then, I think we find ourselves in the following situation – moral epistemology is a non-special member of the epistemology-of-the-a-priori family; and here perhaps the best we can do with our starting points is to improve coherence (sufficiently broadly understood). Speculating again – there is some reason to believe that these internal, coherence-enhancing mechanisms are by-and-large okay, so there is some reason to believe that improving coherence will get us closer to the truths. But I think we’re going to have to settle for that. Disappointed?

8. Hi David,
   Since reading Faraci’s review of your Taking Morality Seriously, I’ve been interested to know how you harmonize your anti-constitutivist “schmagency” challenge with your avowedly transcendental argument in TMS, Chapter 3. In “Schmagency Revisited” you say: “noting that I do Φ is never a good answer to the question of whether I should Φ.” But in TMS, you say that the project of deliberation is “rationally non-optional” for us, and derive a form of Robust Realism from the indispensibility of normative truths for that non-optional project.
   If I understand you, you try to avoid your own “schmagency” challenge here by stressing that this non-optionality is normative, that the project of deliberation is rationally required of us. However, you try to underwrite the non-optionality of deliberation by claiming that “we cannot avoid asking ourselves what to do…opting out of the deliberative project as a whole…may not be an option for us.”
   Can you say why this line of argument is not vitiated by your arguments in “Shmagency Revisited?” If “noting that I do Φ is never a good answer to the question of whether I should Φ”, then how can the fact that “we are essentially deliberative creatures” underwrite a rational requirement to deliberate?

9. Thanks Nick,
   As I say in the book (P 62, and see footnote 33 there) my thought on this has evolved. I believe that earlier version of my argument were indeed vulnerable to this kind of criticism. Now, though, I don’t think that this is so, precisely for the reason you mention – I am relying on a normative premise about the rational non-optionality of the project of deliberation. This may make me vulnerable to other problems, but not to this one.
   Quite possibly, some of my wordings sometime reflect the older way I used to think about these things. If so, I’m sorry for misleading you.

10. No, thanks David, that’s very helpful.

11. Hi David,
    In your book you distinguish between “general supervenience” question and a “specific supervenience” question. The later is something like: “why do THESE specific moral properties superveniene on THOSE specific natural properties?” You give a quick answer that can be summarized as “because the content of the true moral principles specifies these moral properties as obtaining with those natural properties.” E.g. the moral principle says “causing pain is wrong.”
    But, it is important to see that the only way that moral principles could explain the distribution of moral properties, is if moral principles make it the case that the property instances obtain where and when they obtain. For an analogy, take the Titius-Bode Law, which describes the distribution of planets in our solar system. “If the radius of the Earth’s orbit is normalized to 10, then the radii of all planetary orbits can be given by rn = 4 + 3 × 2n, where n is -∞ for Mercury, and 0, 1, 2, … for succeeding planets.” But this “law” is only a description of the distribution, and cannot be offered as an explanation of that distribution. Why not? Because this “law” has nothing to do with what makes it the case that there is the distribution that there is. It tells us nothing about why the planets are distributed as they are. To co-opt a point from Lewis, calling something explanatory does not make it so.
    So, in order for moral principles to explain supervenience, they must do so by explaining the distribution of property instances. And they can only do this if they are somehow responsible for the distribution. This raises some questions that seem to me to be fairly serious:
    First, how is it that moral principles are able to “attach” moral properties to natural properties (or to agents, actions or events with certain natural properties)? That is, how do moral principles make the particular distribution of moral properties come about?
    Second, what must moral principles and moral properties be like in order for this to be possible?
    It seems to me that without giving answers to these two questions, you has not given a complete explanation of supervenience. All that you have done as done is claim that a further fact (the content of moral principles) is relevant, but you has not explained how that fact is relevant or the (metaphysical) mechanism through which it operates.
    I have my own thought about how the non-naturalist should fill out your explanation, but I was curious about your take.
    Thanks!

12. Hi David,
    I have what I think is a rather simple question, but it has bugged me for a while. It’s about how the indispensability argument of Ch.3 of TMS is actually supposed to support RR (Robust Realism, or the thesis that there are irreducibly normative truths). The argumentation seems geared towards supporting the conclusion that we are epistemically justified in believing RR. And this is in fact how you yourself state your conclusion at the end of ch.3. But I was expecting an argument for RR, not an argument for the claim that we’re epistemically justified in believing RR.
    (I guess this is just a version of a usual sort of worry about transcendental arguments.)
    Thanks in advance! And sorry if I’m missing something obvious about the argument.

13. Thanks for your response, David. That’s helpful.
    I don’t think that the relation between us and the moral properties needs to be perception-like, though of course that’s one way of solving the problem. I think you’re right that we can look at general epistemologies of the a priori for help with our moral epistemology. But some plausible accounts of other a priori disciplines won’t adequately extend to moral epistemology (e.g. Balaguer’s (1998) Plenitudinous Platonism, analytic accounts of a priori knowledge). So I think we should be careful.
    All of that being said, since you reject the further condition I proposed in my first comment, this might all be moot for you.

14. I agree, Preston, that Balaguer’s Plenitudinous Platonism doesn’t have an obvious parallel for morality (or the normative). But I think it in anyway doesn’t give us anything like the objectiivty we intuitively want, not even in mathematics. (Which is a part of what’s nice about it – many people think you can have objects without objectivity; he shows that you can have objects without objectivity!)
    I also agree that analytic account are implausible for the epistmeology of the normative. Then again, I also think that they’re implausible for most other (interesting) cases of a priori knowledge and justification.
    So I think that the companions-in-(what-mistakenly-looks-like)-guilt strategy here is still on the table.

15. Thanks, Hille.
    You’re right that the immediate conclusion of the indispensability argument is that we’re justified in believing RR, not RR itself. You’re also right that there’s something weird about that, and that what’s weird about it is weird about transcendental arguments in general (to the extent that I understand what those are, at least).
    But I think you’re wrong to imply that I don’t argue for RR. The way I see things, I do argue for RR, via arguing for our being (epistemically) justified in believing in RR.
    After all, if we are justified in believing p, doesn’t it follow that we’re justified in believing p? We’re not talking about deductive arguments, right? So what more is needed in order to, well, justify our belief that p?

16. Thanks, Aaron. Very interesting indeed.
    Let me start by saying that I am indeed worried about my story on supervenience (I’ve been pushed on this by Tristram McPherson and David Plunkett – I hope to have more to say about this soon).
    And another preliminary – do share your thoughts on how to explain supervenience from a non-naturalist perspective. (If you don’t feel like arch-sharing here, on line, email is also an option).
    Anyway, substance:
    I think that in your comment you draw together two things that should be kept apart. One is the (right) claim that mere generalizations do not explain in general, nor (in particular) do they explain supervenience. The second is the (false, I think) claim that generalizations only explain if they are, roughly speaking, the relevant truth-makers. I don’t know enough about these things, but this requirement seems to me much too strong. It’s just that explanation is a wider category than this, no?
    So – I agree that if moral principles were mere generalizations, my explanation would not amount to much by way of explanation. But I think they aren’t mere generalization, so that’s fine. I don’t want to commit to the principles being the truth-makers, or the things that give moral properties their nature or any such thing. But such a claim, it seems to me, is not needed for my explanation to go through.
    If I’m right in all this, then your two specific questions rely on false presuppositions, I think.
    What do you think?

17. Dear David,
    This is partly self-serving in a self-promoting kind of way, and partly self-serving in a curiosity-driven sort of way: could you say a bit more about how you intend to attack public reason? I think the recent criticisms by Steve Wall and Bohman & Richardson are damning, and I don’t think even Gaus’ The Order of Public Reason manages to save the day, despite the shift from politics to wider social morality. I admit I haven’t studied that book though.
    The self-promoting bit is this. I’ve a forthcoming paper against public reason, in which I think I make similar points to the ones you mention in your Arizona abstract:
    http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2232789
    Crudely, it seems to me that public reason wants to have it both ways: it uses voluntaristic language (‘acceptability’ etc.), but then idealises the consenting/accepting citizenry in ways that make the theory effectively driven by substantive considerations rather than consent/acceptability/accessibility relations. I’ve also got an older paper on modus vivendi in which I argue that non-idealised public reason is a non-starter — so mine is a sort of pincher movement, a bit like what you hint at in your paper, perhaps. I’d be thrilled if you could reveal more of the detail of your argument.

18. David,
    You’re right that my point is dialectically unusual; that’s one reason I raised it here, since I’m taking for granted that you buy into the promise of transcendental moves in metaethics!
    As you suspect, part of my question was just whether you had reason to think transcendental moves would be unavailable for non-RR theorists. And, as I suspected, you don’t. That’s not surprising, or even a problem in itself; I just wanted to give you a chance to voice one if you had it!
    Interestingly, though, I think your response to Nick in this thread might point to a reason for thinking that at least some non-RR views can’t make your transcendental move after all. You say that you take your indispensability argument to avoid shmagency (or shnormativity) objections because you are relying on a normative premise—that belief in normative truth is rationally non-optional. Extending this, you might claim that all transcendental arguments rely on claims of rational non-optionality, and thus that all transcendental arguments rely on normative premises. With that in hand, it’s not hard to see why you might think such arguments are most friendly to RR, since non-RR views typically ground normativity in something non-normative, and so it seems it will always be the argument for that grounding—not the transcendental one—that’s doing the real work.
    I don’t think this argument will work against everyone, though (not that you did; I made the argument, not you!). Suppose Bill is a Blackburn-style expressivist. Bill argues that we are rationally committed to accepting that all substantive normative claims (e.g., ‘lying is wrong’) are either true of false. Then he reads your book. First, you tell Bill that his view has to be wrong, because it entails that normative truth somehow depends on us and our attitudes, and that this is inconsistent with normative objectivity. Second, you argue that belief in normative truth is rationally non-optional.
    Bill is surprised: You’ve both argued for the rational non-optionality of normative truth. The only difference seems to be that his normative truth is deflationary, while yours is robustly real. If he’s to take your arguments to favor your view, it seems that must be because his deflated truths can’t assist in deliberation, perhaps because they aren’t objective.
    Now, here’s where things get fuzzy for me. On the one hand, I feel the force of the intuition that somehow the normative truth still depends on us for the expressivist. Really, though, that seems only to be the case in the sense that we wouldn’t have Bill’s transcendental argument if it weren’t for us and our attitudes. But can’t the same thing be said on your account? We’re only rationally committed to the existence of normative truth because of our need for deliberation.
    What’s more, the only argument (that’s not mere intuition-pumping) you’ve offered regarding objectivity—the stuff about standing your ground in the face of disagreement—doesn’t seem to hit home. If Bill is right and we’re rationally committed to there being normative truth (even in a deflated sense) then surely we’re warranted in standing our ground. (And I don’t seem any reasons beyond that why non-deflated truth is necessary for deliberation.)
    Ok, so that’s the kind of thing I’m thinking about (it’s related both to some of what I said in my review and, if I’m remembering correctly, to some concerns Kate had): It seems like there might be transcendental moves that expressivists (for one) can make that would seemingly serve to undermine your arguments against them and thus to undermine the indispensability of RR normative truth (though maybe not of normative truth period).
    (It could be that you just don’t buy the expressivist argument for deflated truth. If you’ve addressed this directly somewhere, and I missed it, sorry.)
    Finally, a bit about your comment about epistemic possibility:

    [I]t doesn’t follow that the mere epistemic possibility of future evidence coming in should somehow change how we respond to the evidence we currently have, no?

    I agree, for the most part. But I do think that such epistemic possibility can create strong theoretical pressure. Think about the history of using God to explain things. As science progresses, God seems required to explain less and less. There are still things science can’t explain, and it might even be that current evidence supports a theistic explanation (or a non-natural one, or no explanation at all) but it still may be that, given a choice between defending theism and pursuing naturalistic explanations, we should choose the latter.
    There are (at least) two reasons for this: One is inductive and may or may not apply in the normative case. (I suppose it depends on whether you take non-RR theorists to have satisfied other concerns you’ve had to give you reason to think they’ll be able to satisfy, say, the just-too-different intuition.)
    But the other reason is theoretical: The theistic explanation isn’t as useful as a scientific explanation would be, so we should pursue the scientific one rather than accept the theistic. I’m tempted to say something similar here: Your argument for RR only goes through if either (a) there can’t be non-RR normative truth or (b) non-RR normative truth wouldn’t help us deliberate (these might amount to the same thing).
    Generally, then, the claim is that we should pursue RR only as a last resort, and I’m not sure you’ve shown we’re at the end of the line. More specifically, what I’m trying to highlight here is that your own use of transcendental arguments seems to me to illuminate ways for your opponents to resist (a) and (b)—e.g., Bill might argue that, for him, it is commitment to acceptance of normative truth that depends on us and our attitudes, not the truths themselves and that, in this way, his position may not be so different from your own.

19. Hi David,
    I’ve learned a tremendous amount from your contributions in meta-ethics, and I’m pleased to see that you’re making a foray into political philosophy—particularly on the topic of public reason. I have been thinking on Gauss’s approach a lot, and I’m quite sympathetic to it. I haven’t yet read the whole of your forthcoming critique of Gauss, but I’m really glad you shared it.
    With regard to Gauss, I have a question about your claim that “it isn’t clear that morality is even the kind of thing that can have a function” (p. 11 of Gauss critique). The idea that morality has a function is pretty ubiquitous in moral philosophy. And, from what I’ve read, it’s all the rage among social scientists currently writing on the evolution of morality. The tradition in moral philosophy includes Hobbes and Hume as classical figures, but runs through Warnock and Gauthier in the 20th century, and in recent meta-ethics notable proponents include David Copp and Philip Kitcher. The social scientists I have in mind include Christopher Boehm, Jonathan Haidt, and Dennis Krebs.
    Among all these thinkers, the function ascribed to morality is that of either limiting selfishness or promoting mutually beneficial social cooperation (or both). In what sense is this the function of morality? The same sense in which the function of a watch is to indicate time, and the function of a heart is circulating blood. The theoretical literature on function is complex, but a pretty uncontroversial formula is that that function of a thing is the effect it was designed or selected to have. Moral systems are either designed (by intentional agents) or selected (by natural selection) to have the effect of limiting selfishness and promoting social cooperation. To me it’s clear that Gauss is thinking of the function of morality along these lines in The Order of Public Reason, given his discussions of the evolution of cooperation (chapter 7) and the process by which a society coordinates on a morality (chapters 19 – 20).
    I would like to know what you think is so odd/obscure about the idea that morality has a function. Maybe you think that theorizing about the evolution of morality only gives us the function of a social phenomenon (analogy: the study of physics) rather than the function of Morality Itself (analogy: Physics Itself). But moral naturalists like Copp and Kitcher think we can derive moral objectivity and reasons to act morally from the premise that Morality Itself is a social phenomenon. For his part, Gauss suggests that we can derive a notion of practical rationality from an understanding of social morality’s role in upholding cooperative social orders (see especially OPR p. 131). So, again, why can’t Morality Itself have a function?
    Thanks, and sorry for the long post…

20. Thanks, Enzo. I am looking forward to reading your forthcoming paper.
    Yes, it sounds like at least one of my points is rather similar to yours. I put things in somewhat more general a context, perhaps, by criticizing the idealization Public Reason account incorporate in a way that draws on my Why Idealize? – in that paper I make a more general claim against all idealized-response-dependence views of normative concepts, the move is rather similar to the one you describe. Basically, it’s the claim that for an idealization to be legitimate, it must be consistent with the underlying motivations for going response-dependence in the first place. Otherwise – if the idealization is really just meant to avoid counterexamples – it’s objectionably ad hoc. And this, I believe, is the case with Public Reason account.
    I also have a few other points there – something about self-defeat that follows some of Wall’s work, and also some “larger” things – like what is the way to be appropriately modest in the face of disagreement, and how to accommodate the underlying Public Reason intuitions without falling victim to this kind of stuff.

21. David,
    I am curious to know if you have an analogous example of some entity or phenomenon that explains some other entity or phenomenon, such that the first one does not at least a partially ground the second (causally or otherwise). Without that, I’m not sure you can say that moral principles can explain the distribution of moral properties without the principles grounding that distribution. It seems to me that X is only an explanation of Y if X is some how responsible for Y.
    It’s true that there are other things that we call explanations, but these aren’t really explanations. What immediately comes to mind are explanations of action. Why did Al mow the lawn at 6 in the morning? So he would wake up his neighbor. But the waking up of the neighbor isn’t really explanatory related to the mowing, since it comes after (or might not happen at all). What really explains the mowing is Al’s desires and beliefs. When we ask why does this property supervene on that property, I think we are looking for something more like this explanation.
    I’ll email you the paper I’m working on regarding this issue.

22. Again thanks, David (Faraci),
    Small correction – the premise I am relying on is not that belief in a normative truth is rationally non-optional, but rather that the deliberative project is.
    I should say that I would be very cautious with generalizations about all transcendental arguments. I think the category of all things that are called by this name is very broad and heterogenous, and it won’t help to distinguish between those that really are and those that aren’t worthy of the name. The name, after all, is a technical term, not some natural language term we all have intuitions about. In fact, I am not even sure that the term “transcendental argument” captures a philosophical kind.
    Having said this:
    I certainly agree that there is a sense in which the availability of the argument (for me or for Bill) depends on us. What I don’t accept is that this is the only sense in which for the expressivist normative truths are ultimately about us.
    So the real issue becomes whether my argument about IMPARTIALITY etc. works. I don’t think that you’ve refuted it here. The mere fact that there are truths doesn’t suffice. Consider a disagreement about what is ugly, and assume a response-dependence meta-ugliness view. If this is the right thing to say of ugliness, and if we need to coordinate, and all other thigns are equal, then it seems to me clear that both of us should go impartial. Similarly for moral cases, if a response-dependence metaethical theory is true. Similarly, then, for expressivism, if (and I know this is a big “if”) it is sufficiently like response-dependence views in the relevant ways.
    Re epistemic possibility of future evidence: Note that what I said was just that the mere possibility of future evidence shouldn’t change etc. Of course it may make a difference together with other things, like perhaps evidence that futire evidence will go this way rather than that, etc. I think this is true of the case you describe.
    Is RR last resort? In some respects, it should be the default position, because it’s so intuitively plausible. Or in other words, I think it’s less useful to argue over the place different philosophical positions should occupy in line – you just evaluate all the arguments and let chips fall where they may, no?

23. Again thanks, Aaron. I’m looking forward to reading your paper.
    These are not things I know much about, but here’s my attempt:
    The thought that there is no explanatory connection without some grounding relation is an interesting one. Not sure what I think about it, exactly. But I think I can take in on board, seeing that you’re willing to be (I think) rather liberal on the kind of the relevant grounding. You say – causally or otherwise. Well, how about moral grounding? Surely, the fact that the action caused pain morally grounds its being wrong. Isn’t this enough? You may think that moral grounding is somehow not good enough for your purposes, but I don’t see how you can possibly reject it without begging the question here. I mean – causation is ok, constitution is ok, presumably some other relations that are at the bottom of mathematical or other a priori explanations are ok, but moral grounding is not? Why? (And what about other normative examples? Like the inconsistency which ground the epistemic irrationality, etc.)

24. Thanks for your kind words, Andres.
    All I officially commit myself to in the critique of Gaus is, as you note, that it’s not clear that morality has a function. Now, I also think that it doesn’t, indeed that it’s a category mistake to think that it does. But I don’t think I need this in this context.
    What do I say of that tradition? Well, that they’re wrong – to the extent, that is, that your characterization of their views is accurate. I’m happy to take your word for it. I don’t do history. (But I suspect that here as almost everywhere else, attributing this view to a great dead philosopher will not be easy or uncontroversial – one good reason not to do history).
    And yes, I do want to emphasized the distinction between a practice or even a body of beliefs having a function, and the content of those beliefs having a function. So don’t get me started on philosophically-sounding social scientists. This is a distinction I bet it’s going to be awfully hard to even make them see, let alone respect. Am I wrong about this?
    But again – though I am happy to commit myself to this, I don’t think anything in my critique of Gaus depends on this. At the very least, if he wants to rely on this highly controversial metaethical thesis, he should defend it, or at least explicitly note that the whole Public Reason project (as he understands it) depends on it.

25. David,
    Correction noted; I was playing a bit fast and loose there. I also agree about transcendental arguments. I was really just using this (here and in the work I’m doing) to capture a kind of argument regarding rational non-optionality in metaethics that I think you, Korsgaard, and some expressivists share. Obviously, if that’s the case, I can’t go around saying that “transcendental arguments” more generally always involve rational non-optionality!
    It sounds to me like the sticking point against Bill is going to be whether or not deflationary normative truth suffices for deliberation. I have to admit I find what you say here rather unsatisfying; it seems to me to be relying extremely heavily on an intuitive sense of what it means for something to be “response-dependent.” I don’t know if Bill should say that his view isn’t response-dependent, or that it is but not in a way that supports impartiality. But it’s not clear to me that enough has been said about response-dependence to close off either avenue. For instance, suppose Bill argues that certain particular substantive normative truths are rationally non-optional. Those truths might still be response-dependent in some sense, but it’s not at all clear to me that if someone disagreed with me about one, I should go impartial. Surely, I should try to convince them to be more rational!
    You say you think evidence of future evidence is present in the theistic case I describe. Does that mean you think it is not present in the metaethical case?
    Finally, you say:

    I think it’s less useful to argue over the place different philosophical positions should occupy in line – you just evaluate all the arguments and let chips fall where they may, no?

    I agree, insofar as we are talking about making decisions based on current evidence. But my concern here is more prospective (and ad hominem): You admit that you have less to say against competitors than you’d like (e.g., re: just-too-differentness). You are also depending on an argument for RR—indispensability—that explicitly relies on the unavailability of alternatives (thus, in a way, making itself the last resort!). So, the thought is that your present evidentiary position is uncertain enough to warrant searching for further evidence. If so, and you can choose what kind of evidence to look for (as surely we can, especially as philosophers), isn’t evidence for (e.g., arguments for) non-RR the right thing to pursue? Of course, you might just have no idea how to do that. And that’s part of what this post is about: To my mind, what is particularly striking about your work (especially in contrast with other non-naturalists, like Parfit) is that you yourself are working with one (I think the best, maybe the only) type of argument—rational non-optionality—that might vindicate non-RR.

26. Thanks again, David.
    I’m not sure what you mean by “deflationary truth”. I think the sticking point is about response-dependence, not about truth (much less about the kind of truth, to the extent that truth has kinds). And I guess you’re right that in a way I’m relying here on intuitions about response-dependence, though I don’t think this is a helpful way of putting things. “Response-dependence” is a term of art, and we don’t really have intuitions about it. The intuitions are about things like whether at the end of the day it’s all about me. It seems to me clear that on expressivism it is. Now, I acknowledge that showing in detail that this is so turns out to be more complicated that we may have thought. This is because quasi-realists are really quite clever in their efforts to make it hard for us to say this truth. But they don’t, it seems to me, do much by way of showing that it’s not a truth after all. I hope to write more on this soon.
    Yes, my thought was that there was a difference between your point in the metaethical case and the theistic case. In the former, it was just about the possibility of future evidence, and in the latter about the likelihood of evidence in one direction, or some such. No?
    I don’t know what the right priorities should be in allocating research resources – putting more work into refuting RR, supporting it, or perhaps forgetting about the whole debate and getting into more sensible endeavors (like business ethics? modal ontology? Dynamic semantics? The semantics of epistemic modals? Or – I fear we must mention the possibility – not philosophy at all?). I guess we should use methodological considerations here – perhaps we should devote our resources to trying to refute RR, simply because it’s the leading option out there. (You know I’m kidding, right? The “business ethics” example was supposed to convey this message clearly.) But I don’t think there’s a deep disagreement here between us.

27. Thanks very much for the response, David. I see what you’re saying. I think my discomfort with the style of argument might be partly methodological. If my conclusion is that p, I tend to want to focus my argumentation more directly on what makes p true, or why p must be true, or whether p is the best explanation for something, etc., than on whether we (who?) should or must think that p. (Unless the latter sort of thing, about what someone should or must think, is explicitly the thing I want to argue for for some reason.)
    I can see how there is some overlap here, though: sometimes an argument that p turns on evidence that p, and of course any evaluation whether the evidence really supports p, or makes p likely, must be made by an epistemic agent. Still, it seems like the conclusion should be that p, not that we’re somehow bound to think that p. If I’m trying to argue for p and end up somewhere short of p, then I tend to feel like I haven’t yet understood something.
    Well, maybe this gets too metaphilosophical to carry on with here. Anyway, thanks again for the response.

28. Thanks for your reply, David.
    I look forward to the first issue of OSPP (and of course I’d be delighted to read any draft you’d be willing to share — I’m especially interested in the ‘modesty in front of disagreement’ discussion you mention). I must also re-read Why Idealize in the light you describe. Somewhere in my PhD (2008!) I had given a rudimentary counterpart-based account of what happens with PR’s bad idealization, but can’t remember whether that made into the above piece or elsewhere. Anyway I’m not sure I agree with myself on that any more (pun almost not intended).

29. David,
    I just meant that the question will be whether the “truths” Bill champions (presumably by accepting a deflationary account of truth) make good on the relevant intuitions. I think we’re roughly on the same page here. Indeed, for what it’s worth, I agree with you about expressivism (I have a paper that’s sort of on this, too; if you’re interested you can read it here.) But I do think the kind of transcendental argument I’ve been talking about gives them a good shot, at least inasmuch as when I think about what success along those lines would mean, the “it’s all about me” intuition starts to flag.
    As I said, I think whether there’s a difference between the metaethical and thesistic cases depends partly on whether you think non-RR theorists have been making strides that do point in one direction. Obviously, this itself will be a point of disagreement (between you and them, that is, regardless of whether between you and me).
    Ultimately, I think you’re right that there’s not much real disagreement here between us, except perhaps about the strength of certain intuitions and maybe some methodological stuff. Indeed, as someone who leans away from RR, I appreciate that your work gives me gives me something to latch onto regarding what it would take to move one way or the other. In my view, far too many of your non-naturalist brethren are content to, as it were, stand at the top of Default Mountain, hurling objects down at other climbers.

30. I see your point, Hille, of course. However:
    First, you should remember that the relevant justification (in the interim conclusion that we are justified in believing p) is epistemic. So this makes things better, right? I mean, it’s not as if the interim conclusion is that it would be useful for us to believe that p or any such thing.
    Second, if I’m right, the case of inference to the best explanation (which you cite) is also of the exact same nature. In that case too, it seems to me, your justification for the conclusion (“there are electrons”) proceeds via an indispensability argument, that is, the indispensability of the believe (that there are electrons) for our explanatory project. Once again, then, there’s parity between my kind of argument and IBE.
    Third, you may think about the whole story of going via the interim conclusion (that we’re justified in believing p) not as a part of the official arugment for p, but as the background conditions in which p is directly supported by the evidence. Perhaps this is how we usually think of IBE.
    Finally, note that you’re still talking as if by saying that my conclusion is that we’re justified in believing p, it is ruled out that the conclusion is that p. But I take both to be conclusions of mine. It’s just that I conclude that p via the interim conclusion that we’re justified in believing p. Some oddity remains, no doubt. But it’s not exactly as if p is left out.

31. Thanks, David, that helps. (Esp. the remarks about IBE.)

32. David,
    Hmmm… I replied earlier but it seems to have vanished. If it comes back, just assume whichever one sounds better is what I really mean.
    All I meant is that it seems like the sticking point will be whether the “truths” Bill champions (presumably by endorsing a deflationary account of truth) can stand up to the relevant intuitions. So I think we’re on roughly the same page here. Indeed, for what it’s worth, I agree with you about expressivism (I have a paper that’s partly about this, too; if you’re interested you can read it here). But I have to admit that when I think about what the rational non-optionality move might do to the expressivist picture (along the lines I suggested for Bill in my last post), my “it’s all about me” intuitions start to flag.
    As I said earlier, I think whether the metaethical case is different from the theistic depends on whether non-RR theorists have been making progress. Obviously, this will be a point of contention (between you and them, at least, regardless of whether between you and me).
    I also agree with you that we don’t have much substantive disagreement here, except perhaps with respect to the strength of certain intuitions and perhaps some methodological stuff. Indeed, as someone who leans away from RR, I really appreciate the ways in which your work helps illuminate paths in both directions^*; and I think the transcendental path—which I’m fond of—might ultimately lead either way.
    ^*In contrast, it seem to me that some of your non-naturalist brethren are far too comfortable standing atop Default Mountain, hurling objects down at the climbers.

33. Hi David!
    Glad to hear that you’re developing the position from the Gaus review. Have you read Nagel’s “Moral Conflict and Political Legitimacy”? Might be worth a look, if not.
    I have a question about your view regarding the “role of the political philosopher.” In the review of Gaus, you say:
    “The deeper point, though, is that there is nothing modest at all about public reason, and not just about Gaus’s version thereof. The thought that the political philosopher, ensconced in his study, can elevate himself above the normal controversies of political life, look at all of it from above, as it were, without taking sides; and then offer, as an impartial arbiter, the theory that justifies social morality or political authority to all those fighting in the arena – this thought is both arrogant and patronistic. It is arrogant, because it substitutes attributing substantive mistakes to some of the views in the political arena with attributing a category mistake (mistaking the private and the public) to all of the views there. And the philosopher putting forward such a view is patronizing others, in that he treats those in the arena as if they were children, and himself as the responsible adult who doesn’t participate in their childish game but serves as an arbiter.”
    I would have thought the political philosopher “looking down” on the first-order substantive political disagreements both (a) substantively disagrees with one side, takes a side, participates in the “childish game,” attributes substantive mistakes to her opponents, and argues with those people (perhaps while wearing a different hat); and (b) is trying to think about what can and should be done in the face of this disagreement, given that it doesn’t seem to be going anywhere, despite all of her efforts in (a).
    In particular, one has to think about what kinds of political and social institutions (if any) to have, what those institutions should do, how they should be limited, how they should be populated and structured, what justifies them, and so on–all in the face of this disagreement and heterogeneity.
    Now, I’m with you against the Public Reason approach, if it suggests that the (b) task is somehow value neutral, or that it should be seen in that way, or that it could be undertaken without (possibly controversial, substantive) normative commitments, or in a way that all sides of the disagreement could embrace. That might be the dream, but I don’t see how it could be accomplished in reality (lots of familiar arguments here, with that Nagel paper being an early instance). But I’m against you if you think that task is somehow inappropriate, or requires taking a patronizing or condescending attitude.
    There’s a way in which this (b) task is “just” a practical problem, in that it is a series of problems that arise only contingently, because of actual disagreement and the possibility of bad things happening as a result of that disagreement. (In a society with no disagreement, we might not get to these issues at all. Although one might worry about the possibility of domination, if one were to change one’s view, so that there would still be need to think about the (b) issues.)
    But there is a point to the (b) tasks even if the point is not to evade normative commitments or resolve normative disagreement uncontroversially.
    For one thing, the disagreements may be different with respect to the (a) problems (abortion, affirmative action, taxes, education, etc.) and the (b) problems (what political equality requires in terms of institutions, what the constitution is for and how it should be understood and protected, whether we should use representatives and how representatives should conceive of their role, when coercive use of state force is permissible, etc.), so that we might come to live stably and peaceably amongst each other, even if we routinely disagree about the (a) issues.
    For another, it seems that the arguments that might be offered in the (a) domain might often have no or little bearing on the (b) domain, assuming that they can’t just make the (b) domain unnecessary. And we need to address the (b) tasks (or so we might assume) in some way or other to keep bad things from happening.
    One question I have is whether I’m thinking of a first- and second-order political question/disagreement distinction that you think collapses or should be collapsed (e.g. 1st order question: what should the tax rates be; 2nd order question: how should we as a polity determine what the tax rates should be).
    On a side note, I’m struck by how your position regarding political philosophy appears to be similar to Dworkin’s position regarding metaethics. By doing it, you are just doing first-order theorizing! This leads me to think that I’m misunderstanding your view…
    At any rate, I look forward to reading the fuller exposition of it as you develop it.

34. Good to hear from you, Alex, but really, is there a need for this kind of language? Reminding you of Dworkin on metaethics? 🙂
    I can see that there’s a characterization of what’s going that will lead to this way of putting things. But this means, of course, that it’s not a very enlightening way of characterizing what’s going on. Perhaps we’re both arguing against some distinction, but that’s about as close as we get. I argue (in a way) against some strong distinction within the normative domain – even Public Reason people accept that both “private” morality and political justice are in the normative domain. Dworkin argues (I guess) that meta-normative discourse just is a part of normative discourse. Huge difference, it seems to me. But this is just me succumbing to the temptation of addressing this side note of yours. So – substance:
    I think we are almost entirely in agreement. What’s arrogant and patronizing is precisely the attempt to treat (b) issues in a way that is neutral on (a) issues. Indeed, there was already something problematic about the way you put the (a) condition (the problem is not in your way of putting things, but in the things you’re trying to put this way): How should we understand the different-hat metaphor? This suggest that what’s going on is something like: We’re going to share ice cream, just one flavor; wearing my private hat, I go for vanilla; wearing my public hat, I suggest we flip a coin. But this is precisely my point, of course: Treating our deeply held moral views like mere preferences is a deep, terrible mistake, and treating others’ views in this way is indeed condescending.
    I also agree, of course, that there are also very good type-b questions (how should we go about our political business, given that we don’t agree about a-issues, and indeed also b-issues). In addressing them we should consult (what we take to be) the true moral principles, and see what they have to say about such cases, cases in which there’s no consensus about the true moral principles. So am I collapsing b-issues to a-issues? Well, I’m okay with drawing a distinction if you want, and we can do so (vaguely) based on the content of the relevant questions. But what I do claim is that b-issues are not that special compared to a-issues, certainly not in the way PR-theorists think. And I think you agree, no?
    In the fuller version – which I am also looking forward to seeing… – I will also have a section accommodating the underlying PR-intuitions. I actually think that there is something good about treating people in ways that they can (or rather do) accept, etc. The way to accommodate this intuition without falling prey to all the problems facing PR-theory is simple, somewhat boring, and as far as I know totally neglected in the literature (let me know if I’m wrong – I’m a newcomer to this literature): The value of interacting with people on terms they accept (or some such) is just another value, to be included among all the other desiderata for political arrangements. It is in no way a necessary condition for legitimacy, or any such thing. Notice that this way of thinking about things avoids self-defeat worries, worries about excluding the unreasonable, worries about idealization, and so on. On the other hand, it delivers a rather weak result. This, it seems to me, is as far as we can go towards PR-intuitions.
    What do you say?

35. Hi David,
    Knowing you, I knew that would be a low blow. So fair enough.
    I agree we mostly agree. I didn’t mean the “hat” thing to be controversial, just acknowledging that we might say different things and talk in a different way depending on the context. I don’t think this requires treating our deeply held moral views like mere preferences, and I think we agree that the (a)/(b) distinction doesn’t require this.
    I think your way of accommodating the underlying PR intuition will not make the PR people happy, but you probably know that already. Personally, I think there is a necessary condition in the vicinity: a political system is legitimate with respect to a person living under the system, P, only if P could reasonably agree to live under the system—-while in possession of information about the system and P’s relative situation under the system—-and not just because P’s or others’ alternatives to the system are all particularly bad. Of course, it all turns on how we understand “could reasonably agree,” and for that I am quite happy to make that assessment in a full-throated, normative, objective way. There’s no suggestion that everyone (or all reasonable people) will or should accept the system, or would accept it under these idealized or hypothetical conditions. From what you say, it seems that you wouldn’t agree with this, but I’m not sure.
    As for the literature, I think you’re right that the option you describe is neglected, perhaps because the PR people will think it misses the point (many will agree about the importance of, say, domestic peace and “stability for the right reasons”). One person you might look at, although not exactly on point, is Richard Arneson, particularly his “Democratic rights at national and workplace levels” and his “Defending the Purely Instrumental Account of Democratic Legitimacy.”
    Looking forward to the paper(s).

36. Of course my suggestion won’t make PR-theorists happy. It’s not them we’re trying to accommodate – rather, it’s whatever force the underlying intuitions on which they’re relying. Normal people do not have intuitions about it being objectionably authoritative to treat people according to true principles that they can’t reasonably be expected to accept, or any such thing. All of this is far too theory-laden. But we do have intuitions about it being a good idea to be able to explain yourself to others in a way that they can understand, etc. So I’m trying to accommodate this benign intuition, not the PR-dogmas the inculcation of which is the sad story of political philosophy (in the analytic tradition, I guess) at least since Rawls’ Political Liberalism.
    Your own suggestion sounds interesting – I need to hear more, though (do you have something you can share, or not yet?). My initial suspicion is that in this vicinity there isn’t going to be any condition of this thought that is both true and non-trivial. I mean, with your non-PR understanding of “could reasonably agree” the danger is that anyone can reasonably agree to anything that’s right and just – so that we’re getting close to saying that a system is legitimate if it is legitimate. But this is all awfully quick, of course. So again, do send me what you’ve got.
    Thanks, Alex!

37. Hi David,
    The condition I suggest is just a necessary condition, and it’s supposed to be very minimal (accommodation through deflation, one might say). So the full story about legitimacy might be much grander and more controversial (probably more in line with the kind of account you think is required). Looking forward to further discussion, and I’ll send you an early draft…