Missing Something? Crisp and Reasons-Realism

I’m currently reading Roger Crisp’s new book Reasons and the
Good. For the most part it is admirably clear and bound to create critical
reactions against many of the controversial views and arguments it introduces.
Occasionally I wonder though whether I’m just disagreeing too much or missing
something obvious. Just now I’m having big difficulties with how he introduced
debates about realism about normative reasons. Here’s the three sentences where
this is done:

(1) ‘[N]ormative reasons are properties of actions that
count, for the agent, in favour of their performance’.

 

That’s slightly odd view about what reasons are. Not
everyone would want to say that the favourers are properties of actions but let
that go…

 

(2) ‘These properties are ‘real’ in the sense that they are
not to be understood entirely in terms of, say, human projection or
expression’.

 

This is supposed to be the statement of realism about
reasons.

 

(3) ‘This kind of realism has in recent years been attacked
on two fronts – one Humean, the other Kantian’.

 

And, so the debate starts. But, wait. What would be
instances of properties of actions that would favour carrying out them? Crisp
believes that there is just one such property – being conducive to well-being.
Other people might be more liberal and accept that things like being keeping of
a promise, being kind, being enjoyable, being desire satisfying, being funny,
and so on can favour actions too. Whatever you think about this, I just can’t
see how the reason-realism debates would be about whether these properties are
real and independent of projection as (2) claims. Most people in the
reasons-realism debates accept that things like well-being, promise-keeping,
kindness, and so on, are very real properties and things. The debate is not
about that. Maybe the other skeptical debates concerning brains in the vats are
about the realness of these properties. However, the reasons-realism debate
must rather be about the counting in favour part in (1) and whether it is a metaphysically robust relation
in the world or a result of our projection or expression. Is Crisp being just
sloppy or am I misunderstanding him that badly? He goes on to say next very odd
things about the judgment-internalism debate too but that is for another posting.

19 Replies to “Missing Something? Crisp and Reasons-Realism

  1. Jussi
    You raise some interesting issues.
    Regarding #1 why can we not say that normative reasons as a propety are not attached to the action but to the relationship between the agent and the action? It seems to me that the same action could have a positive affect on one person and a negative affect on another. For example, the action of drinking milk as a source of beneficial nutrients, which would be conducive to the well-being of the agent, presupposes that the agent is not allergic to milk. So the propety ‘conducive to well-being’ cannot be attached to actions or the consequences would have to be positive or negative for all agents.
    Or maybe, I am confused.

  2. John,
    I don’t think you are. I think in (1) the condition ‘for the agent’ is supposed to take care of that. So, milk’s beneficial consequences are a reason to drink for the agents who are not allergic. Of course, for those who are allergic drinking milk won’t have such consequences and for them these consequences cannot be a reason. Also, in one actions I think refers to tokens not types of actions.

  3. Jussi,
    I don’t have the book but, assuming you are representing Crisp accurately, I think your criticism of him is on target. The realism debate would be over the favoring relation in (1), not the properties in (2).

  4. I’m starting to think that he does have the right thing in mind but cannot quite phrase it right. Here’s what he says in 30 pages later:
    ‘On my construal of intuitionism nothing more is required for the position from the metaphysical point of view than the claim that there are reason-providing properties, that is, that certain properties have the normative property of giving us reasons’.
    The first part of that sentence seems to commit the same ‘extensional fallacy’ than the earlier quotes (Thank you Jonas!). The debate is not about whether there are reason-providing properties. Everyone in the debate accept that those exist. The debate is more about the second part of the quote – the status of the favouring relation in the world. Even there, the anti-realists, including expressivists, do not deny that there are reasons and that they have the relational property of counting in favour of the actions (and facts…). It’s just not true that realism and nihilism are the only positions in the realism debates as Crisp claims. Furthermore, at least, even if it was shown that judgment-internalism was not true and Korsgaard’s view collapses to substantial realism, we haven’t even began the debate about the ontological status of the reason-relations…

  5. Jussi,
    I think you are right — Crisp was a little sloppy.
    But I write now with ulterior motives. I’m reading Crisp’s book, too. I had a few things about the book I wanted to discuss with you (or anyone willing to listen), but I didn’t want to start a whole new post to do it. So is it ok if I do it here?
    If so, here’s one issue.
    The point of Chapter 1 is to argue for the claim that “morality in itself provides no [ultimate] reasons” (2). So, to take Crisp’s example from p. 14, the fact that it would be wrong for me to kill Jack does not provide me an ultimate reason not to kill Jack.
    My question is, What does ‘wrong’ mean here? Does it mean prohibited by the positive morality of my society? Or does it mean REALLY morally wrong?
    After introducing the case of killing Jack, Crisp writes, “But I now want to suggest that positive morality is like law in providing only non-ultimate or derivative reasons” (14). This suggests the first, positive-morality reading.
    But then a few pages later, Crisp brings up what he calls “‘natural’ or ideal morality” (16). He asks, “What extra reason against, say, killing in certain circumstances is provided by that killing’s being prohibited by some set of special principles?” (16).
    So I guess I’m inclined to conclude that Crisp is targeting both positive morality and REAL morality. Do you think this is right?
    Here’s another question: Against whom is Crisp arguing? The view that positive morality provides ultimate reasons is, in my view, about as silly as the view that the law provides ultimate reasons. Only the view that REAL morality never provides ultimate reasons is, in my view, really worth discussing.
    But have there been any philosophers who have explicitly endorsed the view Crisp is attacking? I’m inclined to think that most moral philosophers have never discussed it. Moreover, I suspect that very few philosophers have commitments in conflict with Crisp’s thesis.
    You might be inclined to think that deontologists (such as, e.g., Ross) who deny that the good is prior to the right are committed to rejecting Crisp’s thesis. But (as Crisp himself points out (26)), this isn’t so. Ross can say that, when some act would be wrong because it is a promise-breaking, the ultimate reason not to perform that act is provided not by the fact that the act would be morally wrong, but rather by the fact that it would be a promise-breaking. I’m inclined to think not only that Ross could say this, but that he would say this.
    Maybe Crisp’s thesis could be shown to conflict with Kant’s view about the value of acting from duty rather than from inclination. Perhaps Kant’s view here is plausible only if duty itself provides ultimate reasons.

  6. Chris,
    I would be happy to talk more about his book even though I’m only half way through. I’m also worried that ‘People’ are reading this and I get into trouble as usual… So I better be careful (shhh).
    I think you are right about the course of the argument. He does start from the positive morality and argues that it is not reason-providing. I wasn’t quite sure whether what was not here thought to be reason-providing was supposed to be the beliefs of the community or the moral properties as projected to the world by the community. I agree that not many people have held the kind of views that would be attacked by these arguments. Maybe some relativists who think that the real morality collapses into positive morality and some contractarians (Harman?).
    The move then, as I see it, is to argue that similar arguments apply to all moral properties – none of them can be reason-providing. I’m not sure why he thought of the real morality as a set of special principles. Anyway, the arguments against that view are not clear at all. I don’t see how the Euthyphro could get started against the moral realist. I don’t see why wrongness could not be additionally reason-providing to the wrong-making features (like Scanlon thinks), let alone why cruelty, honesty, and so on could not be, and why moral reasons, whatever the grounds, could not be identified by their role in our practical life. The debunking view of evolutionary explanation for our moral beliefs is also pretty suspicious. I’m not sure there is an evolutionary explanation for our moral beliefs and even if there were I don’t see its debunking force (ask Simon Blackburn about this).
    But, even if these arguments against the reason-providingness of moral properties would succeed I don’t think he would get his conclusion. His claim is that there are no moral reasons at all because in descriping what ultimate reasons we have we do not need to use moral concepts to state them. If this principle would be right, then there could not be non-eliminative reductions anywhere. Take the case of water. In describing the wet stuff around me, I do not need to talk about water at all. I can state my claims just as well in more fundamental H2O terms. If Crisp’s principle was right, this would imply that there is no water. But, no-one thinks that. There still is water, full stop, even if you don’t need that concept to describe anything. So, if you can have non-eliminativist reductions, moral reasons are safe from all the arguments Crisp gives. Maybe our talk about moral reasons is reducible to reasons some more basic properties provide and we could stop using the moral terms as we could stop using water. We don’t need to use moral terms to pick these reasons but just as well we can. Still, the moral reasons exist as water does.

  7. Jussi,
    Thanks for that. Two comments and a question:
    1. I suppose I feel some of the force of Crisp’s rhetorical question, “What extra reason against, say, killing in certain circumstances is provided by that killing’s being prohibited by some set of special principles?” But I wish he’d say more than he does.
    2. Regarding your last line of thought, I don’t think Crisp denies that there are moral reasons. He denies that there are no ultimate moral reasons. So he can agree that the fact that some act would be morally wrong can provide a non-derivative reason not to do it. (E.g., it might be that whenever you do something wrong, you feel guilty. Since you have reason not to do what makes you feel guilty (since feeling guilty feels bad), an act’s being wrong always (though derivatively) gives you reason not to do it). In this sense, he can say that there really are moral reasons.
    Analogously, he’s say, sure, there’s water, but there is no “ultimate water.” Water exists in a derivative way. It’s not a basic thing. But, still, there really is water.
    3. Aside from any intrinsic interest a person might have in the question of whether there are any ultimate moral reasons, does Crisp’s thesis in Chapter 1 matter? What effect will our view about this thesis have on whatever other views in ethics we hold?

  8. Chris,
    good questions and ones I’ve been thinking about myself. I feel the force of the rhetorical question when it is put in terms of special principles. I don’t think the principles themselves carry the normative force. But, take the property of wrongness. I’d need to fill in the details but I think it is possible to imagine two euthanasia cases where the killing has equal consequences for the well-being of all agents concerned but in one case the killed has consented and in the other she does not want to die. Crisp’s view implies that in these cases there is just as much reason to kill both patients – only well-being can provide reasons. But, it is somewhat plausible to think that the consent makes a difference to the wrongness of the killing in the cases. Killing the unwilling patient is wrong but killing the willing patient not necessarily in this case. My intuition in this case is that wrongness of the killing provides a strong reason against killing that is independent of well-being.
    Regarding (2) you are right, he accepts derivative moral reasons (but wants to ignore them because they do not hold independent normative weight). But, accepting derivative reasons is a different matter than reducing moral reasons to well-being based ones. You can see that when you think of the water case. It’s different to think that water is h2o than to think that there is water that gets its existentance derivatively from the existance of water. Identity is different from making relation.
    Regarding (3), in a sense it doesn’t as he thinks all moral theories can be phrased in a form that only attributes non-moral reasons. But, I think he thinks that nothing is lost if we stopped using moral language and that the moral beliefs we have are intrinsically not to be trusted is suspicious. I think if changed our ways of thinking to Crispian one it would have practical effects to our way of life that would make it worse. And, I agree with Williams that thick moral terms are to which our moral knowledge is ingrained and ignoring them would be morally harmful.

  9. Chris (and everyone else),
    I have an intuition question about the book. Here’s Crisp’s basic principle:
    The Self-Interest Principle (SI). Any agent at time t who has (a) a life that can go better or worse for her and (b) a range of alternative actions available to her at t which will affect whether that life does go better or worse overall for her has a reason to act at t in a way that makes her life go better overall, the strength of such a reason varying directly in proportion to the degree of promotion of her well-being.
    Crisp says that anyone who has a rational intuition and masters the relevant concepts grasps the truth of SI and that grasp provides the ultimate justification for believing it.
    Any such invariable normative principle awakens my internal Dancy and starts to play the counter-example game. Crisp’s reply to this challenge is ‘I cannot imagine a case in which SI would not be true’. I’m not sure about this. A good case I can conceive is the basic Hitler test. Say Hitler has two available options – order the extermination of the Jews or not to do so. The consequence of the first act is tens of millions of dead Jews and (counterfactually) a moderately happy Evil Dictator. Consequence of the second action is at least that the Jews stay alive and live more or less happy lives but we will have a moderately frustatrated Evil Dictator. SI implies that Hitler has a reason to order the extermination of the Jews and the happier he will be as a result the stronger this reason is. My intuition goes completely against. My bedrock is that there cannot be any reason to exterminate the Jews – nothing at all favours that sort of action (and more happiness and well-being cannot give more reason to do so). That would imply that SI is not self-evident as it is not even true. I’m sure not everyone will accept my intuition but surely some intelligent, competent people do. Could this mean that as a result SI is not self-evident or does it mean that I lack the rational intuition that reveals the true normative order of the world (or that I lack the relevant concepts)?

  10. Jussi,
    as regards your latest comment. The natural answer is to say: Hitler had no moral reason to exterminate the Jews, but if his well-being is really promoted to some extent by exterminating the Jews (something we might argue about), he had some non-moral, self-interested reason to do that. There may be self-interested reasons to act in morally horrific ways. It is also for this reason that it would not be plausible to respond by adding a “unless the action is wrong” clause to Crisp’s principle. By and large, we shouldn’t moralize reasons of self-interest, no matter where they lead us (morally).
    Also, a point about strength: the happier Hitler is, the stronger his self-interested reason will be as a self-interested reason. This may not imply that the happier Hitler is, the stronger the overall case for exterminating the Jews. A very strong SI reason may be comparatively very weak, as is in this case. I agree though that Crisp is too swift claiming he can’t imagine a potential counterexample to the principle.

  11. (cont.)
    Or maybe your thought is that distinctions such as moral/non-moral are too “sophisticated” or reflective to be part of a principle that claims self-evidence for itself? Aren’t you perhaps working with an overly simplistic notion of self-evident as “immediately compelling” which Crisp himself (I believe) would reject?

  12. Francesco,
    I don’t want to overly moralise reasons and accept that there can be often be self-interested reasons to act in immoral ways. But, the something you say we might want to argue about is something I really want to argue about. I just cannot accept that there was something that *really, ultimately, normatively favoured* ordering the killing of millions of Jews. I’m not certain what could even possibly do that.
    If by self-interested reasons we mean something else that actual favourers then I’m ok probably. Otherwise, I don’t see any temptation for saying that Hitler had good, outweighed reasons for the order besides trying to save the universal principle. I’m happy to say it is false. I accept that most of the time well-being and happiness give reasons but in some contexts this force is disabled. No, I’m not thinking of self-evident as obvious. I’ve thought about the principles and cases a lot.
    Funnily enough here is what Crisp says of me:
    ‘Any dissenting thinker we can imagine must fall into one or other of these categories’ which are ‘misunderstanding the principle or commitment to doubtful theory’. I’m not sure I’ve committed to any theory (I’m not a particularist yet as far as I know), let alone any doubtful ones. So, I’m by necessity taken to misunderstand the principle. That hurts. I would have thought he’d think better of me.

  13. In these very interesting reflections, Jussi
    writes that
    ‘My bedrock is that there cannot be any reason to exterminate the Jews – nothing at all favours that sort of action (and more
    happiness and well-being cannot give more reason to do so).’
    But, it is in one respect funny that Jussi should say that since he seems to accept the Dancyian view that we should be careful in making categorical claims about reasons since there are always potential counter-instances lurking in the background.
    Suppose that, unless we exterminate ten million Jews, some evil dictator will exterminate all the world’s sentient beings after first having had them tortured. In this case I don’t think that it is clear that we could not possibly be said to have any reason at all to exterminate ten million Jews.
    We might next argue as follows:
    So, it seems that, contrary to what Jussi claims, there can be reasons to exterminate the Jews. If there can be such reasons, then we no longer have a principled reason to deny that we could ever have self-interested reasons for doing this. But, as before, these reasons would be strongly outweighed by our other reasons to refrain from acting in such a horrible way.

  14. It certainly appears that “Hitler has a reason to phi” implies “There is a reason to phi”. But perhaps there’s a subtle equivocation here. That is, maybe there’s a sense in which saying that Hitler has a reason to have the Jews and other undesirables killed is not to say that there is a reason to have the Jews killed, nothing that favours that course of action. If that were the case, we would most likely be expressing with the former that given Hitler’s twisted perspective, there was something that made his action intelligible.
    Speaking of Dancy, I think this is an interesting case for his apparatus of favourers and various kind of enablers. Maybe we could say that while Hitler’s happiness would in suitable circumstances ground reasons for him (or even us) to do something, there is an enabling condition for self-interested reasons that is lacking in this case, namely “x does not involve monstrous suffering of others”. This wouldn’t be to moralize reasons (favourers) themselves, but the circumstances in which they are fit to play the favouring role. So the very thing that would be a reason in a different context would not be such in the context in which it would favour mass murder. This wouldn’t just go for Hitler’s self-interest but other things, like saving resources (millions of fewer people would reduce consumption of the Earth’s resources), improving the economy (not that eliminating the Jews actually did that, but counterfactually it could), and so on – in the absence of the disabling condition, these very considerations would favour the course of action, but as it is, none of them speaks even a little bit in its favour. This would save Jussi’s intuition that nothing could justify the killing.
    Come to think of it, this reminds me of what McDowell says about ‘silencing’ reasons – the virtuous person does not think that the money to be had by stealing and selling a friend’s car provides even the slightest reason to do so. On the sort of story I sketch above, the virtuous person can think this way while fully acknowledging that in suitable circumstances, making money is a great reason to do things.

  15. Antti,
    it seems to me that ‘Hitler has a reason to do x’ implies at least that ‘there is reason for Hitler to do x’. If the reason is provided by some fact about how Hitler’s doing x would promote his well-being, then insofar as his having reason to do x implies anything about what we have reason to do it would seem that there is reason for us to do x for Hitler’s sake.
    Those who accept desire-based or aim-based theories of reasons would say that truths about what Hitler has reason to do does not necessarily imply anything about what we have reasons to do since our aims and desires might be very different from those of Hitler. But, on most substantive or value-based theories, except for on some kinds of Egoism, if Hitler has reason to act so as to promote something, then we would also have reason to act so as to promote this thing.
    It is, it seems, a separate question, just as Antti suggests, whether these substantive or value-based views would do best to claim that, in some cases, reasons to promote our own well-being are silenced by reasons against doing what these things that would promote our well-being or happiness.
    Maybe Jussi could defend his view by arguing that, if Hitler had reason to exterminate the Jews, then we would also have reason to do this since Hitler’s having reason to act so as to promote something, S, would, on most plausible views, imply that we all have some reason to do this thing which would promote S. In other words, if Hitler had reason to extermine the Jews—since this would promote his happiness—then we would also have reason to exterminate the Jews, since reasons, as Korsgaard claims, are not private entities, but things we share. However, Jussi could continue, it is very implausible to say that we all have reason to exterminate the Jews. Therefore, we might conclude, Hitler has no reason at all to do this.

  16. Thanks for the great comments Sven and Antti. Couple of quick points.
    Sven,
    regarding your first post, I agree with you that I shouldn’t have said that there could be no reason to exterminate the Jews. Maybe there are conceivable thought-experiments where there is at least some reason to do so. But, acknowledging this does not make any more appealing to say that Hitler’s happiness has a force to do so.
    Antti,
    I agree actually that there might be a sense of instrumental reasons under which we might say that Hitler had a reason for doing what he did. I would want to add that this sense is devoid of any normativity and only is able to make the agent’s actions understandable in a very weak sense given the aims he has. I had Dancy’s disablerers in my mind and mentioned them already in the previous post. Scanlon uses cases like this to introduce his take on reasons holism in a way that very much resembles what you say about McDowell. I do agree that the disablerers/absense of enablerers would take away the normative force of most considerations in this context (save cases like Sven’s). It is odd that Crisp finds all this inconceivable.
    Sven,
    I like the argument you give. Thanks. Maybe the hard egoist would deny the step that we have a reason to help others what they have reason to do.
    Here’s my objection of the day against Crisp now that I got going. I think he gives an argument in the first chapter of his book against his own view welfarism about reasons later on. He argues against moral reason by saying that because we do not need moral terms to state the ultimate practical reasons we have there are no such reasons. He then argues (well, intuits) for welfarism about reasons – Any ultimate reason must be grounded in well-being. After this he argues for hedonism about well-being – well-being just is (is of identity) the balance of enjoyment over suffering.
    Now, if the latter claim is true, then you can replace ‘well-being’ in welfarism by ‘greatest balance of enjoyment over suffering’. For this reason, you don’t need the concept of well-being to state the ultimate reasons there are. If that is true, and the need for using the concept in stating the ultimate reasons is the criterion for the existence of such and such reasons (as in the case of moral reasons), then there are no ultimate reasons based on well-being and welfarism is false due to Crisp’s own arguments. Voila.

  17. Jussi,
    what does Crisp exactly mean by “there are no ultimate moral reasons”? That there are no ultimate reasons that can only be formulated by (thin) moral concepts–something I’d tend to agree to–or that there are no ultimate reasons given by wrong-or right-making features, but which need not be formulated by citing wrongness or rightness? (such as promise-keeping, (no) harm to others etc.) The latter seems much less plausible, but this depends on issues such as monism and pluralism, not on which concepts are used, it seems to me.

  18. Francesco,
    He means that there are no ultimate reasons that can best be stated by using ANY – THICK OR THIN – moral concepts. So, the idea is that when we use moral concepts and say that something is right or wrong, cruel or kind we assume that we refer to properties that are reason-giving. On a closer look this turns out to be an illusion. There is a debunking evolutionary explanation for our use of these terms and the thoughts that we are referring to reason-providing things. Happily, in many occasions the objects we thought had these properties possess certain other, more basic features that provide ultimate reasons. These reasons are not however moral reasons at all by his criterion. However, they can ground the illusion of moral reasons by creating derivative reasons to act in moral ways. At this point, I do worry about the second horn of Prichard’s dilemma as described by Scanlon. Anyway, Crisp is furthermore a welfarist about reasons so the ultimate ground for reasons must always be well-being. If somewhere where we thought there were moral reasons for acting in morally good ways no well-being improvements are in the offering, then there just is no reason to act in the given way at all. So, any normative umph thin moral terms might have is buck-passed to well-being and not the wrong-making features as pluralists like Philip would think. My head hurts at the moment so I might be getting this completely wrong.

Comments are closed.