Expressivism vs. Nihilism

37 Replies to “Expressivism vs. Nihilism”

1. David, can you tell us whether nihilism, as you understand it, is a normative view? I can’t tell from what you say about your nihilist.
   Nietzsche’s nihilism is a normative view; Mackie’s is not. Both kinds of views can be called ‘nihilism’.

2. Hi David,
   I have to prep my teaching for today, but two quick thoughts off of the top of my head. If someone has “no substantive normative commitments,” then I think I’d be inclined to interpret ‘Murdering is not wrong’ as (c), but in any case, the “biforcated” interpretation would likely be (f). I don’t actually find anything wrong with directly expressing the lack of a mental state (though I know others do), but that aside, your question is how BEING For murder and not Being For murder are inconsistent mental states. So, second, would an expressivist hold that these mental states are inconsistent in the sense that it is impossible to be in both mental states at once? Or are you conceiving of an expressivist as needing some king of “stronger” inconsistency, such as the same mental state towards inconsistent contents?

3. Sorry, I meant “interpret … as (b).”

4. Jamie,
   Sorry if that wasn’t clear. I mean for nihilism to be a view that involves no substantive normative commitments. So it is (I think) closer to Mackie’s, though closer to his view about morality, not normativity in general, obviously.
   Dan,
   I do think that expressivists need a stronger form of inconsistency than the impossibility of being in both states. That was why I mentioned disagreement (which I tend to think of as the interpersonal analogue of inconsistency) in the original post—typically, lacking an attitude does not itself entail that you disagree with those who have it.
   Also, a quick comment about the direct expression stuff: Schroeder argues elsewhere (“Expression for Expressivists”) that expressivists should (maybe need to) accept a view on which expression is understood in terms of assertability conditions that necessarily involve some mental state. So, for instance, it might be that my having the belief that grass is green is part of the assertability conditions for “grass is green.” On the (I think plausible) view that the absence of an attitude is not a particular mental state, this is at least one reason to worry that expressivists can’t countenance direct expression of such absences.

5. So, when your nihilist says, “murdering is not wrong”, he is not saying anything normative, right? Then it seems to me he is not asserting the contradictory of what I say when I say, “murdering is wrong”. Normative assertions have normative contradictories; so do normative thoughts. (No?)
   So I’m not convinced there is any inconsistency problem here for expressivists to solve.

6. Jamie,
   When a nihilist about magic says “Jane is not cursed,” surely the nihilist contradicts someone who says “Jane is cursed.” Nevertheless, there is an important sense in which magic-nihilists have no substantive magic-commitments (e.g., they do not think that any magic-properties are ever instantiated), and thus an important sense in which their claim that Jane is not cursed is not a magic-claim. The normative case seems precisely analogous to me.
   So, no, I do not think that normative assertions/thoughts have normative contradictories any more than the contradictories of magic-claims are themselves magic-claims.

7. That seems wrong to me, David. A nihilist about magic does, I think, have very substantive commitments about magic. She thinks that each and every item is not cursed in exactly the same sense that a believer in magic thinks some items are not cursed; so if the latter’s judgment that my shoes aren’t cursed is a magic commitment, then it’s hard for me to see why the former’s are not. The magic nihilist, furthermore, behaves toward things that she thinks aren’t cursed in the same way that a believer in magic behaves toward things he thinks aren’t cursed. But Mackie definitely did not behave toward murder the way a believer in ethical properties behaves toward things she thinks are not wrong.
   I can see that this is a tricky issue; we probably don’t have a theory-neutral characterization of what it is for an assertion or thought to be normative. My point is that the things expressivists are forced to say about nihilist thinking are pretty much the things they would want to say in any case (that is, even if not forced).

8. I want to add something to Jamie Dreier’s question. Part of the problem of not being clear about what sort of nihilist we are considering here is whether in fact we do have to be able to entertain such a person, and in what sense. One could be a nihilist of either the Nietzschean *or* Mackiean/Joycean sort (or other sorts) and one would still have substantive normative commitments. Nietzsche and Mackie and Joyce all have substantive normative commitments, though all in their various ways attack morality or moral discourse (Nietzsche’s attack is much more interesting and fecund in my view, but let that pass). Obviously, the reason this is relevant to your post is that your argument seems to trade on the claim that normative nihilists don’t have any substantive commitments. But if and since all these ‘nihilists’ have normative commitments, then it is unclear what sort of nihilist you have in mind. Perhaps it is someone who explicitly disavows all normative commitments. It would be nice to have an example of such a person, but even if we had one, we should not infer from this claim that they in fact have no normative commitments. I think there are excellent reasons to think that no person that we would recognize as such is devoid of normative commitments, whether they claim to be or not. It’s part of how we operate in the world to have commitments that we treat as normative, even if we are confused or ignorant enough to deny this. Of course I can’t argue for this here, though others have (e.g., Korsgaard and Street, though my view differs from theirs, and is closer to Street’s). However, if in fact being a nihilist of whatever sort we think we have to entertain as responsible ethical theorists is not inconsistent with having all manner of normative commitments, then taking option (g) is available after all, and in fact seems like a fairly good option to take. You correctly note that BEING FOR should require normative commitments, but fail to see that having normative commitments is entirely consistent with denying that one has normative commitments, if for no other reason than that one does not appreciate the ubiquity and nature of such commitments. One may of course have other sorts of confusions and ignorances as well.

9. Jamie,
   You may be right that there’s no theory-neutral way to say what counts as a (susbtantive) normative commitment. Surely, though, we can agree that the magic nihilist and the magic believer differ in that only the latter has any commitments that involve instantiations of magical properties/correct application of magical predicates. Hopefully you do agree. So let’s just call this class of commitments “positive commitments” or PCs.
   Now, it seems to me that the biforcated attitude semantics (BAS) analogue of having a normative PC is BEING FOR something. After all, that’s what makes BEING FOR different from other attitudes. BEING FOR blaming for murdering is, according to BAS, the same as judging that murdering is wrong. ‘Being curious about blaming for murdering’ or ‘being happy at the thought of blaming for murdering’ is not. So BEING FOR must be what has the positive normative content, so to speak. If it has that content in virtue of the kind of attitude that it is (and how else could it?), then BEING FOR anything is analogous to having a normative PC of some kind.
   If that’s right, then nihilists cannot BE FOR anything, because whether or not they have “normative commitments” of some kind, they don’t have any normative PCs. But, as I’ve been arguing, expressivists can only interpret nihilists who say “murdering is not wrong” such that they disagree with those who think that murdering is wrong by saying that nihilists are FOR something. And I think that’s a problem.
   It might be useful to point out that I think the problem isn’t really with the expressivist interpretation of nihilistic instances of “murdering is not wrong,” but with all of them (because, again, I agree with you that nihilists and believers express the same thing when they say murdering isn’t wrong). The problem is that I think (in cognitivist terms) that when anyone says “mudering is not wrong,” they do not express a PC. Rather, they express a belief merely that it is not the case that murdering is wrong. Anything beyond that, such as that murder should be tolerated because it’s not wrong and only wrong things are blameworthy, is implicature (just as, if in magic discourse, non-cursed things should be hugged, we might interpret the believer, but not the nihilist, as implying that your shoes should be hugged when they each say that the shoes aren’t cursed). But expressivists build this further PC (tolerance or, in BAS terms, BEING FOR not blaming for murdering) into the semantics, and that’s what leads to the tension with nihilism.

10. Eric,
    I agree with you that the kind of nihilist I’m talking about has to be utterly devoid of normative commitments (or PCs in keeping with the comments I just made to Jamie). I also agree that the most famous historical “nihilists” have not been nihilists of this kind. But I do not think that that’s because such nihilists are impossible, but rather because it is only recently that philosophers have started to step back from concerns about morality in particular and look at normativity more generally. And there are, I think, current nihilists who are nihilists about normativity more generally (I might be wrong, but I think that Stan Husi and Bart Streumer are two examples).
    Now, of course, as you point out it might be that such “full-blown” normative nihilists are in trouble because they just can’t help but make normative judgements. For instance, some people have claimed that because belief has a normative component, you can’t coherently believe normative nihilism. (Matt Evans and Nishi Shah have a paper forthcoming in Oxford Studies in Metaethics on this.)
    Let’s set aside for the moment the question of whether this line of argument works. Even if it does, all it shows is that nihilism isn’t something we are capable of consistently believing. It doesn’t show that nihilism is false. And it certainly doesn’t show that nihlism is not a view we can entertain. But expressivism seems to me incompatible even with this minimal kind of coherence. It’s one thing to say that nihilists have some deeply buried normative committment somewhere that makes them inconsistent. We can still understand how they can sensibly entertain the (still real) possibility that there is no normative truth. It’s another thing entirely to hold that, as a matter of semantics, each and every negated normative claim one makes is itself an expression of a normative commitment (PC). In that case, nihilism isn’t even a sensible view, just given what our normative terms mean. And as I said in the post, especially given the kind of “weak objectivism” many of us accept about normative thought and language, this result seems wholly unacceptable.

11. David, I don’t have an intuitive grasp of what it is to have a positive normative commitment. That’s not a term of ordinary language, and I wonder whether it has any theoretically neutral characterization. Will it help if we just stipulate that your nihilist IS not FOR anything? We don’t have to try to imagine such a thing (we probably can’t, because there is no such attitude as BEING FOR – Mark just uses the term as a kind of wild card).
    So then an Biforcated Attitude expressivist should say that your nihilist is not speaking correctly when he says “murder is not wrong”. He is not using a sentence that has as its linguistic function to express the state of mind he is actually in. Maybe he is confused about the language, possibly because of some bad metaethical theorizing. As Eric Campbell says, your nihilist doesn’t sound like Richard Joyce or John Mackie, and I’m happy to add Bart Streumer; they ARE FOR lots of things. And they plan to do many things, and they rule out planning to do others, and so on.

12. Jamie,
    Let’s set philosophers aside for a moment. It seems to me that, as a completely philosophically unsophisticated teengager, I was able to entertain the thought that nothing I or anyone else does or thinks or says matters, that nothing is really right or wrong, good or bad, and that life is meaningless. And it seems to me that in such a nihilistic moment, I might have asserted that “murdering is not wrong.”
    I take it the expressivist could say one of two things about me: First, she could say that I was expressing my BEING FOR not blaming for murdering. Given my nihilistic mindset, and that such an attitude is just as incompatible with that mindset as BEING FOR blaming for murdering, that sounds wildly implausible to me.
    Second, she can say that I’m using the language wrong, just as she might say that a Platonist is using language wrong when he says “murdering has the Platonic property of wrongness.” But I don’t think that’s even mildly plausible. I’m not infected with any weird metanormative views (as the Platonist might be); I just have the very natural thought that (a) there might have been some truth about what matters, what I ought to do, etc. but (b) there isn’t. That’s the kind of weak objectivity I think almost everyone can entertain pre-philosophically. And it’s the kind of objectivity that, I’m arguing, is straightforwardly ruled out by a view on which all normative assertions—positive and negative alike—are expressions of some non-cogntive attitude.

13. Hi David,
    just about this:
    “But that won’t work, because BEING FOR something means having a substantive normative committment, and he doesn’t have any of those”.
    I don’t think this is the right way to read Schoeder. This is because, on his view, having any belief whatsoever consists of being for things. As Mark puts it,
    “Believing that p involves being for proceeding as if p, true. But it also involves being for not proceeding as if ¬p. (p. 98)”
    This is a general view of what is being expressed by statements like ‘grass is green’. If you say that, you express your attitude of being for proceeding as if p were true and also being for not proceeding as if ¬p were true. But, even if this were the case, asserting that grass is green hardly counts as undertaking a normative commitment. So, on this view, something further than being for is required for a normative commitment. My proposal would be that this something would be being for reactive attitudes. And, neither (f) nor (g) constitute such commitments. So, it seems to me that the nihilist can be expressing either of those attitudes.

14. David, think of the overwrought student of existentialism Hare describes in “Nothing Matters”. I take it you find Hare’s diagnosis wildly implausible, not even mildly plausible, and so on. But it actually strikes me as pretty reasonable. Similarly, I think it’s incoherent for a modal nihilist to suspect that although some things might have been possible and some impossible, in fact nothing is either, even if it seems to the modal nihilist that what he’s thinking is perfectly natural.
    It looks like the role played in teenaged David’s reasoning by the thought you associate with the sentence “murder is not wrong” is very different from the role that thoughts that things aren’t wrong play in most people’s reasoning. Maybe it’s a quite different thought that you express with the same words.

15. I’m generally sympathetic to Eric and Jamie’s objections. But I’m wondering whether you need to appeal to the extreme example of a full-blown nihilist. I take it that the central point you want to make is that (c) and (g) are different. Can’t you establish that by appeal to more mundane examples?
    For example, take the following two possible truths:
    (i) I think murder is not wrong
    (ii) I think nuclear decay is not wrong
    Assuming that I think the latter because I think that nuclear decay is not deontically assessable at all, it seems to me that (i) and (ii) are really quite different. But both receive a similar analysis on the Schroeder style view. That seems problematic, and indeed, looks like the problem you are trying to raise, but on weaker assumptions. Is that right? (I could say a little more, but I’ll hold off in case someone has a simple reason why I’m off track here.)

16. Alex,
    just wanted to say that you are not off the track. I’m vary of speaking for Schroeder but it seems to me that he recognises the need to distinguish between disbelief (and in general disacceptance) and believing that not-p. I take it that (ii) is an instance of disbelieving (not even believing so to say) and (i) is belief that not-p.
    There’s an easy solution to this in the case of non-normative beliefs with the framework’s biforcated account of beliefs. And, Schroeder does explain what the solution to this problem would need to look like along the same lines for the normative beliefs. But, the solution here would not be as straight forward for reasons he gives. This is the new new negation problem of the section 8.4 (see also 167 in the lessons).

17. Alex Gregory, that’s interesting.
    I’ve generally resisted the idea that there is one more slot for negation than appears in the syntax, but it’s hard to deny that your (i) and (ii) display a semantic difference. It would be out of the question, I take it, to rephrase (ii) as

    (ii*) I think nuclear decay is permissible.

    (Or, ‘right’, ‘acceptable’, etc.)
    But I’m a little skeptical that this is particularly a problem for Mark Schroeder’s view, or for expressivism.
    Does it seem at all plausible that (ii) uses metalinguistic negation? Notice that (at least as it seems to me) it would be more natural to put that thought like this:

    (ii**) I don’t think nuclear decay is permissible.

    Compare,

    (iii) She doesn’t believe he’s a jerk, she knows he’s a jerk.

    which is very plausibly metalinguistic.

18. Jussi,
    I agree with Schroeder that expressivists are going to have to understand belief in terms of BEING FOR. But I think that what follows is that expressivists are going to have to accept some version of the view that belief has a normative component. Again, unless BEING FOR X involves, because of the kind of attitude it is, some kind of normative commitment, I don’t see how it can be what we’re expressing when we use normative language. You might think, for instance, that BEING FOR X rationally commits you to including X in your future plans and having certain negative attitudes towards failure to X, and that that’s why BEING FOR blaming, rather than ‘being pleased at the prospect of blaming’ is an appropriate way to understanding wrongness judgements. But if that’s right, then BEING FOR proceeding as if P will, likewise, involve including proceeding as if P in my future plans and having certain negative attitudes towards failure to include proceeding as if P in one’s future plans, and that seems the same as saying that belief has a normative component (again, if having those attitudes towards blaming is sufficient for judgements of wrongness to be normative judgements).

19. Jamie,
    As you guess, I think that Hare’s response is absurd; I’m with Parfit on this. Also, the modal realism case seems different to me. Isn’t thinking that something “might have been” a certain way just to think that it’s possible? So this view seems straightforwardly incoherent in a way that normative nihilism doesn’t.
    We may just be at loggerheads here, but let me see if I can diagnose the disagreement and see if that moves things forward at all. In our discussion of magic-nihilism above, you said that you think that the magic nihilist “thinks that each and every item is not cursed in exactly the same sense that a believer in magic thinks some items are not cursed.” But it sounds like you think normative nihilism is different, when you write: “It looks like the role played in teenaged David’s reasoning by the thought you associate with the sentence ‘murder is not wrong’ is very different from the role that thoughts that things aren’t wrong play in most people’s reasoning.”
    Now, as I said above, I think magic nihilism and normative nihilism are the same. I think that when anyone says “murdering is not wrong,” the semantic content of their claim is merely that it is not the case the murdering is wrong. I just think that for most people, when they say that “murdering is not wrong” they express (via implicature) further normative attitudes (e.g., the belief that it would be inappropriate to blame for murdering). And my complaint about expressivism is, in essence, that it can’t help but make those further normative attitudes part of the semantic content of “murdering is not wrong.”
    I’m guessing, though, that you see the two nihilisms as different because you think that such further attitudes are indeed part of the semantic content of normative claims, and that’s why normative nihilism is either incoherent in much the way that modal nihilism is, or else normative nihilists are just using the language differently. Is that right?
    If so, let me push on this a bit with a case that I alluded to earlier. Suppose that it’s part of magic-discourse that when you think something is not cursed, you have to perform some sort of spell to make sure it stays that way. And suppose this is common enough that in general, when people say that things aren’t cursed, they are understood as proposing performing the anti-cursing spell. Now, the analogy to my view would be to say that this is implicature and that when the magic nihilist says that something isn’t cursed, but with absolutely no intention of suggesting performing an anti-cursing spell, his claim still has the same semantic content as a believer’s. The analogy to your view (as I’m reading you) is that magic-nihilism is either incoherent or the magic nihilist is using the same words as believers but means something different. I’m sure you don’t think magic-nihilism is incoherent, so you presumably would either agree with the implicature view or hold that magic-nihilists are using the same words to mean something different from believers. Now, the latter sounds totally implausible to me. If it sounds right to you, I’m not sure what else to say. But if you agree that it’s implausible, why is it plausible with respect to normative nihilism?

20. Alex (and Jussi and Jamie),
    What a cool suggestion! In my comments to Jamie above, I was trying to say something about why I think this problem arises—because the case of the nihilist brings out the fact that, for expressivists, negated normative claims express further normative attitudes (that I think people only express implicature). Your example brings this out nicely, perhaps even better than the nihilist does.
    One place we might disagree: I don’t think that (i) and (ii) differ semantically. I just think that (i) is most naturally read with an implicature that (ii) lacks, because competent speakers will hear (ii) and recognize the category error. But notice that I could have a strange view on which certain kinds of actions aren’t deontically assessable either, and I could thus say “murdering is not wrong” in order to say something analogous to how we naturally read (ii). In that case, I don’t think the meaning of my assertion changes, just what I’m trying to express.
    Certainly, I think this is part of where you and I are disagreeing, Jamie, as I say above, and as I think your final comments indicate.
    Also thanks, Jussi, for pointing out the relevant bits in Schroeder.

21. A couple of little things:

    Isn’t thinking that something “might have been” a certain way just to think that it’s possible? So this view seems straightforwardly incoherent in a way that normative nihilism doesn’t.

    Indeed, but I thought we were ignoring that feature, since otherwise your teenage self is also straightforwardly incoherent (in that he is flouting supervenience). Or did you mean it’s perfectly coherent for your teenage self to flout supervenience?

    I think that when anyone says “murdering is not wrong,” the semantic content of their claim is merely that it is not the case the murdering is wrong.

    This doesn’t seem to me to be a very distinctive claim on your part, unless your point is about semantic ascent. (Some people think “it is not the case that” just means “it is false that”.) Who denies what you’re claiming here?
    I wouldn’t say that the further normative attitudes are part of the semantic content of moral terms; more plausibly things like “it makes sense to blame someone for x” is part of the content of “x is wrong”; probably this is what you meant. So, yes, I think that’s pretty plausible. But, importantly, it is not built into expressivism. It’s just a popular view that’s incorporated into some versions. E.g., it’s built into Gibbard’s view about moral terms, but certainly not into Hare’s, and not, I believe, into Blackburn’s.
    I like your Magical Practices story. But that’s not implicature, is it? It’s just common background beliefs or knowledge that you’re talking about. So I guess that is what I’d say about the magic talk. But it’s very different from normative talk, I think.
    It’s at this point that I wonder what non-naturalists like you and Parfit are thinking, quite honestly. Suppose there really are some non-natural normative properties, and suppose some community has predicates that denote them. Do you think that is sufficient for the predicates to be normative predicates? If, e.g., they never, under any circumstances, use their thoughts about these properties to decide what to do?
    If you do, then I can see why you would think the magical discourse is so closely analogous (indeed, normative discourse would in a real sense be magical discourse!). It’s also why Parfit’s metaethical view would be the last one I would consider adopting.

22. Jamie,
    How is my teenage self flouting supervenience? I didn’t think he was. He wouldn’t deny that normative properties supervene if they are instantiated at all, just that any are instantiated.

    I wouldn’t say that the further normative attitudes are part of the semantic content of moral terms; more plausibly things like “it makes sense to blame someone for x” is part of the content of “x is wrong”; probably this is what you meant. So, yes, I think that’s pretty plausible. But, importantly, it is not built into expressivism.

    I think that’s what I meant, yes. But while I agree that that particular content isn’t built into expressivism, surely some distinctively normative content must be. According to BAS, judging that murdering is wrong means BEING FOR blaming for murdering. Now, BEING FOR is a just a placeholder; but it’s a placeholder for something particular. It couldn’t turn out, for example, that BEING FOR is just ‘being pleased at the thought of’ because obviously one can be pleased at the thought of something without making a normative judgement. In order for BEING FOR blaming to be the same as judging wrong, BEING FOR is going to have to be an attitude that has distinctively normative features—perhaps it rationally commits one to doing what one is FOR, or something like that.
    Let’s suppose that’s right, and that normative judgement entails such rational commitment. We get the following pictures:
    Cognitivist:
    Judgement that murdering is wrong = belief that murdering is wrong. Belief that murdering is wrong entails rational commitment to blaming for murdering.
    Judgement that murdering is not wrong = belief that murdering is not wrong. Belief that murdering is not wrong entails nothing further. Both believers and nihilists alike can make this judgement.
    BAS Expressivist:
    Judgement that murdering is wrong = BEING FOR blaming for murdering. BEING FOR blaming for murdering entails rational commitment to blaming for murdering.
    Judgement that murdering is not wrong = BEING FOR not blaming for murdering. BEING FOR not blaming for murdering entails rational commitmment to not blaming for murdering. Only believers can make this judgement.
    That seems problematic to me because I think that nihilists and believers mean the same thing when they say “murdering is not wrong.” Any additional attitude the believer has (such as thinking that it makes sense only to blame for wrong things, and thus that he is rationally committed to not blaming for murdering) is not part of the semantic content of the claim. (This is what I was talking about when I said that expressivists don’t think that negated normative judgements are merely judgements that it’s not the case that the positive judgements are correct. But I see why this was an unhelpful way of trying to make that point.)

    It’s at this point that I wonder what non-naturalists like you and Parfit are thinking, quite honestly.

    I should have been clearer about my agreement with Parfit. I agree that there is an important (normative) sense of “matters” that Hare’s analysis is missing. What I’m trying to argue, in essence, is that it’s not just Hare that’s missing it, but expressivism by its very nature. But I don’t accept the rest of Parfit’s metanormative view. I am not a non-naturalist. And I agree with you that it is utterly implausible that denoting non-natural normative properties is sufficient to render a predicate a normative one.
    Is that, perhaps, where the disanalogy with the magic case is supposed to come from? That normative claims, unlike magic ones, involve more than just the denotation of the relevant properties? I agree that that’s a disanalogy, but I don’t see why it’s a relevant one.

23. Jussi and Alex,
    I just took a look at the relevant Schroeder stuff. He offers two suggestions. The first, which he doesn’t like, is to try to come up with minor attitudes for normative predicates. The second is to bite the bullet and say that there really is no such thing as disaccepting a normative claim (other than accepting its negation). Maybe one way to read what I’m doing here is as an argument that this is too large a bullet to bite, but I’m not sure about this.
    I do think, though, that focusing on the nihilist (and on Alex’s case, too, perhaps) explains why the first move won’t work. No matter what that minor attitude you propose, it will involve BEING FOR something, so nihilists can’t disaccept anymore than they can negate, and that seems problematic.
    Alex’s case is different. Since the person who says that nuclear decay isn’t wrong isn’t a nihilist, there is no barrier to seeing him as BEING FOR something. But it’s not clear that there’s anything plausible that we could interpret him as BEING FOR. Schroeder, for instance, suggests that we might understand judging wrong as ‘BEING FOR blaming for, BEING FOR disliking’. That would mean that disacceptance would be BEING FOR not disliking. But surely that’s not what Alex’s guy is expressing.

24. Hi David,
    I’m still worried about this:
    “No matter what that minor attitude you propose, it will involve BEING FOR something, so nihilists can’t disaccept anymore than they can negate, and that seems problematic.”
    To me, this seems to misunderstand what role the attitude of Being For is supposed to play in Schroeder’s theory. This is best explained in his response to Skorupski which you can find from his website (called “Skorupski on Being For”).
    Basically, the idea is that Being For plays no semantic role at all when we give an account of the content of normative sentences. According to the semantic account, normative words like ‘wrong’ stand for a relation of blaming to things. As Mark puts it, the semantic value of ‘stealing is wrong’ is the property of blaming for stealing. Thus, the attitude of being for plays no role whatsoever in the explanation of the meaning of normative sentences. On this view, I assume, the semantic value of ‘stealing is not wrong’ is the property of not blaming for stealing.
    In contrast, the attitude of being for is supposed to be a theory of what it is to believe a sentence, or more properly what it is to believe believe one of the properties mentioned above. Now, I assume that whether believing a property constitutes undertaking a normative commitment depends on the property in question. If you believe the property of blaming for stealing (ie., if you are for blaming for stealing), then presumably you do undertake a normative commitment. In contrast, it does not seem to me that if you believe the property of not blaming for stealing (ie., if you are for not blaming for stealing), you do so. This seems to me to be more akin to believing that grass is green (ie, being for proceeding as if grass were green and being for not proceeding as if not-p).
    So, I guess I want to deny that nihilists who lacks normative commitments could not be for things. This is because being for plays no semantic role in the meaning of normative concepts. That’s exhausted by attitudes such as blaming which the nihilist presumably lacks and which would constitute normative commitments.

25. Jamie: On meta-linguistic negation, that seems like it might be right, at least as the sentence would normally be used. I suppose I’m inclined to think that I could *also* use that same sentence to straightforwardly say that nuclear decay is not wrong, in just the same way that I’m inclined to think that your (iii) could *also* be used to say that she doesn’t believe he’s a jerk but only knows it. But I’m not sure that I know how to justify this inclination of mine. (There’s probably a better way to express what I just said, but hopefully you get the idea.)
    David: I’m also sympathetic to the idea that (i) and (ii) are semantically similar, and that the difference is purely pragmatic. But I was working on the assumption that this is not what the expressivist would want to say, and so it seems fairer to proceed on the reverse assumption.
    I’d forgotten that Schroeder had talked about some of this. I’ll try and take a look later, if I get a chance.

26. 1. Supervenience:
    Your teenaged self thinks that in a world naturalistically like this one, some acts are wrong. He thinks that in the actual world nothing is wrong. So, according to him, there are acts that share all natural properties with different normative properties.
    2. I don’t really understand what it is for an attitude to have “distinctively normative features”. Do you just mean, it has certain normative properties? Which are the distinctive ones?
    3.

    Is that, perhaps, where the disanalogy with the magic case is supposed to come from? That normative claims, unlike magic ones, involve more than just the denotation of the relevant properties? I agree that that’s a disanalogy, but I don’t see why it’s a relevant one.

    It’s a relevant one because the question was supposed to be, why is it implausible to say that the magic nihilist and the magician are using the same words to mean something different, but not implausible to say the parallel thing about the normative nihilist and the normician? And the answer is that “the same meaning” for the magic case is just denoting the same property, whereas in the norm case it is plainly (to sensible people like us) something more.

27. Alex, if (i) can be used a the contradictory of “nuclear decay is wrong”, then I think it could in that use be paraphrased as (ii*).
    Hm, also, I just noticed that I overnegated my (ii**).

28. Ah, I see: you think it’s either metalinguistic or else a claim about permissibility. I’m inclined to think that there is a third option – a rejection of the idea that a certain property is present -, but I can see that I’d have to say more to justify that claim. I suspect that part of what’s going on here is that I prefer to think in terms of reasons rather than all-out right/wrong/permissible, and it seems doubtful to me that thinking [X is not a reason for phi-ing] involves a positive commitment to anything at all (and it clearly isn’t metalinguistic negation either). But this is possibly an issue for another time.

29. Jamie,
    I’m not sure why teenage David is thus committed. Why couldn’t he just think (implicitly) that normative properties are epistemically but not metaphysically possible?
    Jamie and Jussi,
    Thanks, Jussi, for the clarification on the view and the pointer to the Skorupski article, which was really helpful. I think that perhaps I haven’t been thinking about the Being For view exactly as Schroeder wants, but I also think that my worry can withstand the adjustment.
    Suppose that the semantic content of “murdering is wrong” is ‘blaming for murdering’. (By the way, does this mean that the semantic content of “grass is green” is ‘proceeding as if grass is green’? Weird.) Now, there are many different attitudes I might bear towards blaming for murdering. I might be curious about it, be excited about it, care about it, be worried about it, and on and on and on. Clearly, most of these attitudes can’t fill in the BEING FOR placeholder, because it’s not at all plausible that when I judge that murdering is wrong, that just means that I am curious about, excited about, care about, or am worried about blaming for murdering.
    Now, Schroeder himself says that “someone who believes that stealing is wrong is committed to blaming for stealing” (emphasis mine). This gives us some hint at what BEING FOR is going to have to be like in order for it to plausibly be the attitude that we have when we judge that things are right, wrong, good, bad, etc.—it involves a kind of commitment. What kind of commitment? Well, a natural suggestion might be that ‘commit to’ is the same as (say) ‘plan to’, ‘intend to’, or ‘be disposed to’. But none of these will work, because planning, intending, and being disposed to blame for murdering are not plausible understandings of what it is to judge that murdering is wrong. (Some or all might be entailed by such judgement; but they do not exhaust such judgement). The upshot is that whatever kind of attitude BEING FOR is, it must, by its very nature, capture whatever this distinctive kind of commitment is that comes along with the judgement that something has a normative property (as well as anything else that is distinctive of such judgements). Call that kind of commitment n-commitment.
    Hopefully, this helps clarify what I have been trying to get at in my comments to you, Jamie, when I talk about BEING FOR having distinctive normative features, or about BEING FOR involving a substantive normative commitment. If I’m right that BEING FOR is a placeholder for whatever attitude involves n-commitment, then it doesn’t seem appropriate to say that nihilists are FOR anything. After all, being n-committed is (part of) what’s distinctive about judging that something has a normative property, and nihilists don’t ever judge anything to have a normative property.
    Of course, you could resist this by claiming that while a judgement that something has a normative property is sufficient for n-commitment, it is not necessary, and thus one can be n-committed in virtue of making judgements of other kinds. If that’s right, then normative nihilists can plausibly be understood as having n-commitments. But that doesn’t sound right to me. It seems to me that the sense in which I commit to X-ing when I judge that I ought to X is different from any other way in which I might commit to X-ing, and that no other judgements (that did not also involve something’s having a normative property) could lead me to commit in the relevant way. Indeed, that’s why normative judgement is so special. Maybe I’m wrong about this, but let’s suppose I’m right for the moment.
    The reason this would be a problem for the expressivist and not the cognitivist is that, for the cognitivst, negated normative judgements don’t have to involve n-commitments, because negated normative judgements aren’t necessarily judgements that anything has a normative property. For instance, suppose that judging X wrong n-commits one to blaming for X. For the cognitivist, it doesn’t necessarily follow that judging X not wrong n-commits one to not blaming for X. But for the expressivist, it does, because if the nature of BEING FOR ensures that BEING FOR blaming for X n-commits one to blaming for X, then it likewise ensures that BEING FOR not blaming for X n-commits one to not blaming for X.
    This is also why I don’t think the difference between magic and normativity is relevant, because while judging that murdering is wrong n-commits one in a way that judging that ones’s shoes are cursed presumably does not, judging that murdering is not wrong involves no n-commitments, and is thus no different (in this way) from judging that one’s shoes are not cursed. So, again, the nihilist and the believer alike—whether about magic or normativity—should be able to assert “murdering is not wrong” or “my shoes are not cursed” without thereby n-committing themselves in a way that is in tension with their nihilism.

30. Alex,
    As you might be able to tell from what I just said to Jamie and Jussi, I think we agree that claims that a normative property is not instantiated don’t commit one in same the way that claims that a normative property is instantiated do. That’s part of why I’m not comfortable granting that (i) and (ii) differ semantically, even for the sake of argument. After all, why would they differ (on the non-metalingusitic reading of (ii)) unless (i) did involve some such additional commitment?

31. David,
    Jussi has anticipated everything I would want to say. I think the right way of thinking about BAS – and, I would suggest, about expressivism in general – is as offering a surprising view about the nature of belief and propositions. It can’t be surprising that belief involves a kind of commitment, though, because we do take people to be committed to things in virtue of their beliefs. If the nihilist wasn’t committed to her nihilism, then we couldn’t infer what she would be committed to saying about P, if she believed that P is normative. So involving commitment can’t be what makes belief a normative attitude. So I still don’t see what the problem is supposed to be.

32. David,
    Sorry, I now see that I misspoke in my initial comment. The point should have been that even if the expressivist interpretation of (i) is mistaken, it’s nonetheless difficult to make that case directly. So strategically it might be better to focus on (ii) for which the expressivist interpretation may be less easy to defend, even if you ultimately think it’s mistaken in the very same way.
    (Though Jamie’s point about metalinguistic negation seems to show that this strategy won’t work anyway, at least without some further support.)

33. Mark (et al.),
    Thanks, that’s helpful. I think perhaps the disagreement here is about whether the kind of commitment involved in belief is the same as the commitment involved in judging that a normative property is instantiated. I don’t think that it is. I agree, of course, that there’s a sense in which if I believe nihilism, I’m committed to nihilism. But I don’t think that that commitment alone can commit me to doing anything (relevantly, I don’t think nihilists are committed to proceeding as if nihilism is true). That’s (part of) what’s distinctive about judgements that normative properties are instantiated—only such judgements commit me to action (or inaction). So my problem here with BAS isn’t with how it understands the semantic value of phrases like “murdering is wrong” but with how it understands the nature of judgement, because on BAS, all judgements apparently involve a commitment to action (or most, I guess, since there might be some Ps such that no particular action counts as proceeding as if P). And I think that what’s problematic about this is revealed by the fact that, according to BAS, when someone—even a nihilist—judges that murdering is not wrong, he commits himself to acting (or not acting) in certain ways—i.e., to not blaming for murdering.
    Alex,
    I see. That’s useful. I’ll have to think about that some more.

34. I’m worried about your setup given your aims here. Some disagreements (especially philosophical disagreements) can only be understood as combinations of various attitudes which fully partition logical space. The nihilist seems like a case in point.
    First up, here’s the complete list of possibilities, with the expressivist-friendly construals in parenths:
    (a) You think that murdering is wrong (You are FOR blaming for murdering)
    (b) You do not think that murdering is wrong (You are not FOR blaming for murdering)
    (c) You think that murdering is not wrong (You are FOR not blaming for murdering)
    (d) You think that not murdering is wrong (You are FOR blaming for not murdering)
    (e) You do not think that not murdering is wrong (You are not FOR blaming for not murdering)
    (f) You think that not murdering is not wrong (You are FOR not blaming for not murdering)
    Someone who has no substantive normative commitments – a skeptical nihilist, if you like, or maybe just a plain skeptic – is someone for whom BOTH (b) and (e) hold. She also refuses to assert/express the relevant claims in (a), (c), (d), or (f). (She is analogous to someone who firmly thinks a certain proposition, p, is neither true nor false.) So the skeptical nihilist disagrees with me, the moralist, who holds a very different combination of commitments – i.e., both (a) and (f), as well as (e). True, the skeptical nihilist doesn’t disagree by negating a proposition I assert, but that’s not the only way to disagree. Denial does the job too. (Someone who contends that p is neither true nor false disagrees with me, the proponent of p, when they deny what I say. Notice how we couldn’t do justice to this disagreement without talking about the combination of attitudes the person holds.)
    An interlocutor who disagrees with me might also (as here) reject as wrong-headed a whole discourse that I engage in – they will be such that (b) and (e) (alone) hold where ‘murdering’ is replaced by the name of any arbitrary act type. This distinguishes a nihilist from someone who just doesn’t think that some particular type of behavior is deontically assessable (like dreaming, say).
    As for the residual idea that there’s a problem with expressing an absence of an attitude, I’d have to hear more, but I don’t see why that would be off the bat.
    Another kind of nihilist – someone who asserts that “God is dead, everything is permitted” – is someone of whom both (c) and (f) hold – and (b) and (e) too. The same is true when ‘murdering’ is replaced by the name of any arbitrary act type. I think that’s the guy who is more likely to go around asserting “Murdering is not wrong,” although the skeptical nihilist might also say something along those lines to pragmatically implicate (b).
    But the important thing is that the fact that someone says “murdering is not wrong” isn’t enough, when taken alone, to differentiate a host of possible positions that they may hold. We need to hear more about their overall position, vis-à-vis the above list, to know where they stand and hence to know enough to disagree. So, without including all of the beliefs/attitudes on the above list, I don’t think you’re going to be able to represent a nihilist’s position adequately. But, once you have done that, it’s not clear to me why there is a problem for the expressivist in theorizing the relevant attitudes and disagreements here.

35. Kate,
    I think that the nihilist I’m concerned with here is more akin to your “God is dead” nihilist than to the skeptic (though I don’t think God’s being dead is the only reason—or even a reason—to be such a nihilist). The nihilist thinks that murdering is not wrong, and by this they mean that it is false that murdering is wrong. So I’m not sure what your suggestion is with respect to this nihilist: If I’m right (pace various other commentors), this nihilist can’t BE FOR anything. And so the BAS expressivist can’t understand him in line with (c) and (f), as I take it you agree they should.
    As to your skeptic: If I’m understanding you correctly, the skeptic thinks that normative discourse involves a kind of presupposition error. So when he says “murdering is not wrong,” he doesn’t mean that it’s false that murdering is wrong. Personally, I think that the move from not true to false is pragmatics, given the possibility of just this kind of presupposition error, and so I think that even the skeptic should be understood as in the cognitivist versions of (c) and (f).
    Let’s suppose, though, that you’re right that the skeptic can’t be understood this way. Your suggestion, I take it, is that it is true of this skeptic that (b) she does not think that murdering is wrong or (e) that not murdering is wrong. First, I don’t see how this can distinguish her position, since presumably someone who thinks that murdering is permissible (i.e., ought be permitted) also fits (b) and (e). Second, and related, simply not thinking that murdering and not murdering are wrong does not mean that you disagree with people who think that one of them is wrong. Someone who simply hasn’t considered the normative status of murdering might fit both (b) and (e), but that person does not disagree with the moralist. So holding that the skeptic is expressing (b) and (e) when she says “murdering is not wrong” doesn’t solve the disagreement part of the negation problem. To do that, I think you’d have to say that the skeptic is expressing something that’s not on the list—presumably some second-order claim about the presupposition error in question. (And again, that’s only if you deny that she can be understood as in (c) and (f), which I don’t.)
    As to expressing the absence of an attitude: Personally, I don’t think that anyone should think that you can directly express the absence of an attitude. But let me just offer one reason for thinking that expressivists have to take this position: In “Expression for Expressivists,” Schroeder argues that expressivists should accept a view according to which declarative statements have “assertability conditions”—conditions under which it is appropriate to make the claim in question. Those conditions, he argues, necessarily involve mental states that the speaker is (purportedly) in. But while it’s not hard to believe that an attitude is a particular mental state (or set thereof) it seems doubtful that the absence of an attitude is any particular mental state (or set thereof). So it’s doubtful that merely lacking an attitude can be a determinate condition under which it’s proper to assert something, and that means that the mere lack of an attitude isn’t something that can be directly expressed.

36. “So I’m not sure what your suggestion is with respect to this nihilist: If I’m right (pace various other commentors), this nihilist can’t BE FOR anything. And so the BAS expressivist can’t understand him in line with (c) and (f), as I take it you agree they should.”
    Yeah, I agree with the other commenters that the ‘everything is permissible’ nihilist can be for lots of things, once that idea is understood aright. I don’t really have anything to add to that part of the discussion, which Jussi and Mark covered. My thought was that, if you insist on holding that the nihilist has no positive commitments vis-a-vis blaming, then he should be understood as the skeptical nihilist above. I think you may also be equivocating between these two characters in the above discussion somewhat.
    “I don’t see how this can distinguish her position, since presumably someone who thinks that murdering is permissible (i.e., ought be permitted) also fits (b) and (e).”
    Yes, but someone who thinks that murdering is permissible ALSO fits (c) and (f), as I say above. The skeptic is such that only (b) and (e) hold for her.
    “Second, and related, simply not thinking that murdering and not murdering are wrong does not mean that you disagree with people who think that one of them is wrong.”
    But this is just the general issue of how to think about cases where the person hasn’t entertained the proposition in question. This is a problem for everyone, isn’t it? You might add to the above list (distinguishing ‘positively not thinking p’ from ‘having no position whatsoever about, having never entertained p’), and I see no reason (as yet) why the expressivist can’t make the analogous moves. OR you can say that (b) and (e) hold of the person who hasn’t considered the moral status of murdering, but for different and strictly provisional reasons, to do with lack of reflection on the agent’s part. And, plausibly, I can only disagree felicitously with you on matters where my position is somewhat stable/informed by reflection. Whatever the thing to say here, exactly, I can’t see that the expressivist will have a special problem in saying it.
    Hope that helps!

37. Kate,

    You might add to the above list (distinguishing ‘positively not thinking p’ from ‘having no position whatsoever about, having never entertained p’)

    I’m not sure I understand what ‘positively not thinking p’ is such that it’s different from both thinking not-p and “merely” not thinking p (like the person who hasn’t considered whether p). Do you mean that the person has actively decided not to take a position on p, rather than simply lacking one? If you do, perhaps the idea is that this person thinks something like “there is no good reason to think p (or not-p)?” If so, the disagreement isn’t about p, it’s about whether there’s reason to think p. I have no idea what the BAS analogue of this would be, though of course that doesn’t mean there isn’t one.
    In any case, as I said I’m more worried about the nihilist and his capacity to BE FOR things. But that’s just to reopen that Mark/Jussi discussion.