Print This Post Print This Post

Dr. Neil Sinclair from University of Nottingham gave a talk this past week at the Oxford Moral Philosophy Seminar on “Presumptive arguments for moral realism.” An abstract of his talk is as follows:

Presumptive arguments for moral realism shape much of the landscape of meta-ethics and yet are seldom scrutinized. In this paper I hope to address this deficit. I distinguish two desiderata for meta-ethical theories. First, their ability to ‘save the appearances’ of moral discourse. Second, their consistency with our wider philosophical views. I then distinguish three species of presumptive argument: the first claim that only moral realism could save the appearances; the second that only moral realism has so far saved the appearances; the third that realism is the most natural or intuitive way to save the appearances. Dismissing arguments of the third kind I argue that the remaining arguments are prone to one of two mistakes. First, falsely assuming that some of the claims of moral realism infect the appearances; second, underestimating the resources available to the realists’ opponents. My tentative conclusion is that most extant presumptive arguments fail to support moral realism.

Neil’s paper can be found here, and he would welcome any comments/suggestions.


  1. 1. Posted by Don Loeb | February 7, 2008 9:31 pm

    I just published (whew) a paper on a similar topic in the most recent number of Ethical Theory and Moral Practice (the issue devoted to Mackie). The paper’s called “The Argument from Moral Experience,” and I’ll make it available as soon as I figure out how to send a link. Meantime, I’d be happy to send a copy on to anyone who is interested.


  2. 2. Posted by S. Matthew Liao | February 8, 2008 2:31 am

    Thanks, Don, for letting us know about your paper. Don’s paper can be found here. In case you are interested, for future reference, you can create a link by selecting a word, e.g. ‘here,’ click on the Link button, enter the link, and hit ok. I look forward to reading the paper :)

  3. 3. Posted by Don Loeb | February 8, 2008 6:45 pm



  4. 4. Posted by Daniel Star | February 12, 2008 2:40 am

    I thought I should mention that one of the best recent arguments for realism is provided by David Enoch in “An Outiline of an Argument for Robust Metanormative Realism”, Oxford Studies in Metaethics, Vol. 2, OUP, 2007, 21-50. This argument appeals to deliberative indispensability and is not discussed by Neil or Don.

    Also, a relevant paper by Stephen Finlay (“Four Faces of Moral Realism”) is presently being provided for free over at Philosophy Compass, and comments on this paper are being sought over at Thoughts Arguments and Rants.

  5. 5. Posted by Neil Sinclair | February 12, 2008 3:17 pm

    Dear Daniel

    Thanks very much for the references. Regarding the Finlay piece, I think it offers a classic case of what I call a type-2 strong presumptive argument. Regarding a the quasi-realist accommodation of moral truth and fact by adopting something like a ‘minimalist’ understanding of these notions as they appear in moral practice (an approach adopted by Timmons and perhaps Putnam), Finlay writes:

    “Insofar as Timmons’ and Putnam’s views remain antirealist their successful internal accommodation of morality’s objective pretensions can be challenged. Proponents of rival positions insist that ordinary practice is committed to moral TRUTH that exists even from a morally detached perspective, and moral FACTS that come to us straight from the WORLD. Particular attention has been directed towards the semantic antirealists’ ability to accommodate the appearance that moral claims stand in inferential relations with other (moral and nonmoral) claims. Expressivists have yet to substantiate their assurances that this can be done.” (§1)

    The obvious assumption here is that the truth assumed by everyday moralisers is realist truth, and that’s the assumption that I think is questionable. Ordinary moralisers assume that some moral judgements are true, others false, but it is unclear whether anything in their practice admits them to a robust realist understanding of truth. At least, that’s my suspicion. And at any rate, I don’t think simple observation of actual moralisers supports the claim that they assume a robust, realist notion of truth.

    The section from Finlay continues:

    “In any case, these antirealist maneuvers are at a significant disadvantage against semantic and ontological realism. Whereas realism can simply take logical relations and talk of truth, facts, properties, and descriptions in the moral domain to be continuous with those in other domains, according to our best semantic and metaphysical theories, antirealists must either distinguish distinct, moral equivalents for these, or defend radical revisions of our general theories.” (§1)

    Take the issue of logical relations. Whether expressivists must adopt a distinct ‘logic of the attitudes’ that is not continuous with ‘regular’ logic seems a controversial point. It all depends on what you take logic to be in the first place. Here is Blackburn:

    “[I]f anyone represented themselves as holding the combination of ‘p’ and ‘if p then q’ and ‘not-q’ we would not know what to make of them. Logical breakdown means failure of understanding…this result [is] secured, on my approach [for agents who deny moral modus ponens]…because the person represents themselves as tied to a tree of possible combinations of…attitude, but at the same time represents themselves as holding a combination that the tree excludes. So what is given in one moment is taken away at the next, and we can make no intelligible interpretation of them…Logic is our way of codifying and keeping track of intelligible combinations of commitments.” (Ruling Passions, OUP 1998, p.72)

    This is an account of logical relations that offers no discontinuity between the logical of the attitudes and the logic of other commitments (such as beliefs). Does this mean Blackburn is impaled on the second horn of Finlay’s dilemma? Has he radically revised our general theory of the nature of logical relations? He’s revised the realists’ theory, to be sure, but there is no reason to suppose that the realist theory was ‘our general theory’ to start with. I’m not sure I had a general theory of logical relations before doing philosophy, just that there is something wrong with some one who doesn’t follow modus ponens (for example).

    Similar points apply to the other features. Whether the expressivist offers an account of moral ‘facts’, for example, that is discontinuous with our understanding of nonmoral ‘facts’ depends on what that understanding was in the first place. The thrust of my argument is that the pre-philosophical understanding of moral facts is meta-theoretically neutral, so that the expressivist accommodation of moral ‘fact’ doesn’t generate any discontinuity. Nor does it revise our general notion of what facts can be, since there is no reason to suppose that our general understanding of facts was ever the realist one.

  6. 6. Posted by J. Edward Hackett | February 15, 2008 11:20 am

    It is surprising to me that all this talk of how moral experience seems doesn’t wish to constrain those appearances that such appeals to see what is actually there. A methodological procedure to “save the appearances” is useless without first looking to appearances in a way where we do not presuppose the content of what we want to “see” in those appearances. As such, a Husserlian procedure at looking what people often call moral phenomenology through the phenomenological reduction could aid in this endeavor.

  7. 7. Posted by Stephen Finlay | February 23, 2008 5:26 am

    Neil, thanks for the discussion of my Philosophy Compass piece. I’m afraid I’m late to the conversation here.

    I disagree with you less than you might think. Because this is a survey paper, I’m largely sketching arguments rather than advancing them. (Hence I start by saying that the antirealists’ moves CAN BE challenged, and I attribute the challenge to others). My official position at that stage of the paper is that it is for the reader to decide whether these are fatal problems for expressivism.) But I may be being too defensive on this point, since you merely say that I ‘offer’ the argument, which indeed I do.

    In the next passage you quote, I do take more ownership of the objection. To take, as you do, the case of logical relations, my point here is not that expressivism must adopt a logic that is discontinuous with ‘regular logic’, if by that you mean the logic underwriting ordinary inferences, but that it requires a logic that is at odds with our ‘general theories’ (of course, I also say ‘best theories’, which is more vulnerable to your point). You question this, too, but I think I’m on stronger ground here. There’s plenty of evidence, it seems to me, that our standing theories of logic have been nonexpressivist ones, and indeed there are serious questions about whether an expressivist logic can even be made to work. (My colleague Mark Schroeder, in his forthcoming book BEING FOR, does a terrific job of arguing that it can. But he also finds that success here comes with significant costs in the revision of our understanding of nonnormative logic.) None of this entails, of course, that our general and long-standing theories are correct. But the metaethical theory that would have us revise everything else has a greater burden of proof than one that is compatible with what we already thought about everything else.

  8. 8. Posted by Neil Sinclair | February 25, 2008 4:01 pm

    Dear Stephen

    Thanks for the comments and reference, and apologies if I got a little over-excited about what was only intended as a survey piece.

    I am very interested in your claim that “There’s plenty of evidence, it seems to me, that our standing theories of logic have been nonexpressivist ones”.

    First of all, I’m not totally sure what a theory of logic is, but I guess in my earlier post I was assuming that it is an account of why certain inferential moves (e.g. modus ponens) are privileged and certain others disprivileged. If that’s all a theory of logic is, I’m not sure that our prephilosophical theories of logic are non-expressivist (or non-realist for that matter) – they are simply a set of views on which inferences are appropriate and which inappropriate.

    Perhaps what you’re thinking of is a our standard philosophical view of logic, which involves truth-preservingness, truth-tables and so on. I think that this is our standing philosophical view of logic, but (and I don’t think I mentioned this previously) I’m not sure this is incompatible with Blackburn’s account, once we understand that account as earning us the right to systematise our inferences using truth-conditional semantics. On this view understanding logic through truth-tables and truth-preservingness is always an artefact of understanding logical relations in other ways first. For realists, this is done through notions of the mutiple realisability of various states of affairs (witness Guttenplan’s pictures of palm trees!); for expressivists through notions of intelligibility of combinations of committments. So if there’s a worry that expressivism will force us to revise our predominantly truth-conditional approach to logical relations, I think it’s unfounded

    Of course, that’s no more than a sketch of the approach and the devil may be in the details – but in any case I’d be interested in hearing further about our ‘standing theories of logic’, and in particular about the evidence for their non-expressivist nature.

  9. 9. Posted by Stephen Finlay | February 27, 2008 12:04 am

    Hi Neil,

    I had in mind our philosophical theories, not our prephilosophical theories. (I think it’s generally the case that we have the naive philosophical theories that we do (e.g. the correspondence theory of truth) because they capture what it’s most natural to think when we first start reflecting. That’s not to deny that they can be wrong, of course, but I do think it gives them a presumption of truth).

    In regard to expressivistic logic, I should say that this is one area of metaethics that I generally prefer to stay out of. So I’m happy to be corrected, but I’ve been convinced by the people whose work on expressivistic logic I respect the most (Jamie Dreier and Mark Schroeder, for example), that expressivists still have a lot of work to do. Dreier points out that expressivists have a problem with negation, and Schroeder generalizes this for other logical relations. I can’t do these concerns justice, but you should definitely look up their work if you’re not familiar with it.

    Dreier is optimistic that the problems can be solved, and Schroeder argues that they can be, but at the cost of having to make considerable revisions elsewhere. But these are difficulties that an ordinary truth-functional logic doesn’t face. It looks like we should prefer our traditional theories of logic, unless we’re forced to give them up.

Post a comment

Name: (required)

Email Address: (required) (will not be published)