Here’s a problem with Roderick Chisholm’s well-known account of intrinsic goodness I’d like solved, and I bet some of you will know how to solve it.
Here’s a problem with Roderick Chisholm’s well-known account of intrinsic goodness I’d like solved, and I bet some of you will know how to solve it. I favor such an account myself, primarily because, as Chisholm noted, it provides a nice way of reducing concepts of value to those of deontology. Also, it gives a nice version of the isolation test for intrinsic value, a version in which contemplation focuses on p and q as such.
Chisholm defines intrinsic goodness in terms of intrinsic betterness, and intrinsic betterness in terms of required attitudes. More precisely, at the bottom of the analysis is this principle, which yeilds an interpretation of the isolation test of intrinsic value:
p is intrinsically better than q only if it is appropriate for one to prefer p to q and the contemplation of just p and q requires one to prefer p to q.
Problem: Assume that we can’t get ‘ought’ from ‘is’. Now note the second conjunct in the consequent. It appears to get an ‘ought’ our of an ‘is’. In particular, the second conjunct says that either one is required to prefer p to q because of the natures of p and q themselves, or else because of the contemplation of p and q.
If it is the former, then either p and q themselves contain values that demand the preference ordering or not. If so, then we haven’t gotten anywhere with this definition of intrinsic goodness, and certainly not to a reduction of axiological concepts to deontological ones. But if not, then we’ve gotten an ‘ought’ out of an ‘is’.
The latter is little better. Why should the contemplation of two states of affairs result in a requirement to prefer one to the other? Again, either ‘contemplation’ is somehow itself value-suffused or not (that is, either Chisholm thinks of contemplation as in some sense ‘valuing’ or not). If it is, we again appear not to have gotten very far, or at least not all the way to a reduction of value concepts to deontological concepts; if not, then again, we get an ‘ought’ from an ‘is’.
My guess is that Chisholm is assuming in the consequent some general normative principle which says something to the effect that it is ‘appropriate’ to prefer p to q and, when contemplating things such as p & q as such one is required to prefer p to q. But that is just a guess at this point.
Robert, you write,
“If it is the former, then either p and q themselves contain values that demand the preference ordering or not. If so, then we haven’t gotten anywhere with this definition of intrinsic goodness, and certainly not to a reduction of axiological concepts to deontological ones”
It’s difficult to see how p and q do not “contain value”. But of course value is analyzed dispositionally here in terms of appropriate responses. The second conjunct of the consequent differs from the first conjunct (of the consequent) in specifying an *exceptionless* disposition to respond in a certain way under certain conditions. That is, I take *a required response* to be *an exceptionless disposition to respond* in a certain way. So I guess I don’t see where an ‘ought’ is derived from an ‘is’. Maybe I’m missing something.
Hi Robert,
I think I need to hear more about what the problem is supposed to be. You ask, “Why should the contemplation of two states of affairs result in a requirement to prefer one to the other?” If Chisholm wants requirement to be fundamental, why should he think this question has an answer?
Also, on not getting an ought from an is. Suppose the second conjunct should be interpreted as saying: if (a) one were to contemplate p and q, then (b) one would be required to prefer p to q. Your worry is that if (b) is supposed to follow from (a), then we’re getting an ought from an is. Is that right? I think Chisholm isn’t saying that (b) follows *logically* from (a). Maybe he just thinks that it’s necessarily true of certain pairs of propositions that one is required to favor one over the other if one contemplates them. Maybe this would involve a sort of necessary connection that you don’t like: a necessarily true conditional with a non-evaluative antecedent and an evaluative consequent. I think Chisholm would have been totally OK with it. So I suspect the problem you see for Chisholm comes from your ought-from-is principle, which he would reject.
Mike,
Chisholm himself is explicit in regarding ‘required’ deontologically. So it is, in his mind at least, an ‘ought’. But your suggestion might be worth pursuing instead of Chisholm’s. Still, I want to keep the ‘ought’ reading of ‘required’ and see how to defend that.
Ben,
I was hoping you’d comment. I don’t know if he thinks it fundamental or not. It seems to me that it’s not, and, in any case, getting a requirement out of contemplating something seems to me to be bizarre just in itself, aside from the derivation of the ‘ought’ from the ‘is’. One way to make sense of it is as a response, as Mike suggests; but then how to keep the ‘ought’ understanding of ‘required’?
A simpler picture would have been something like this:
p is intrinsically preferrable to q only if, taken as such, p ought to be preferred to q.
That gets the reduction Chisholm alludes to without the added trip through contemplation.
I guess I don’t see what’s bizarre about getting a requirement out of contemplating, in the sense Chisholm is talking about. How is it so different from a utilitarian saying that if (a) you’re in a situation where you can bring about some pleasure and no pain, or else bring about some pain and no pleasure, then (b) you ought to bring about the pleasure? Here we’re getting a requirement out of just being in a certain situation. The only difference is the sort of situation you find yourself in – in one case having certain options, in the other contemplating certain things.
I’m not sure the trip through contemplation is necessary, I’m just not seeing the problem.
Following up on the first half of Ben’s first reply: If P is an intrinsic value, then how could it be analyzed into some other value? That would suggest that P is valuable because of some other value to which P bears some extrinsic relation. Now, Chisholm isn’t analyzing intrinsic value but but intrinsic betterness, but the same point would apply. To my eyes, it looks like Chisholm is butting up against a common dialectici: He’s trying to analyze intrinsic value without making judgments thereof mysterious, appealing to self-evidence and the like. But any analysis of intrinsic value that doesn’t morph it into extrinsic value will have to rest content with dispositional analyses like the two right-hand conjuncts.
Ben,
I think you will agree with my initial idea, won’t you, that there must be some normative principle in the background here. You can’t get ‘you ought to prefer r to s’ out of ‘r has more pleasure than s’ without assuming background principles, e.g., ‘you ought to prefer what has more pleasure’ or, better, ‘you ought to prefer what is best’ and ‘what has more pleasure is best’. Likewise, you can’t get ‘you ought to prefer p to q’ out of ‘you contemplate p and q’ without some similar background principles.
It’s not clear how we should proceed in the Chisholm case. But here is a way that occurs to me: The fact that it is appropriate to prefer p to q can’t require you to prefer p to q (excising the strange introduction of contemplation for the moment) unless it is inappropriate to prefer q to p and inappropriate to be neutral. So perhaps he should have said just that.
p is intrinsically preferable to q only if it is inappropriate to prefer q to p and inappropriate to be netrual.
BTW, this reference to contemplation may just be a hangover from his thinking about these things in terms of Brentano’s theories. That’s just wild speculation.
Robert,
I am not clear on how your suggested version differs. What do you mean by ‘taken as such’ in ‘p is intrinsically preferrable to q only if, taken as such, p ought to be preferred to q’? In particular, I am not sure how it differs from contemplating just p and q.
Also, what does Chisholm mean by ‘appropriate’? If he allows that attitude/state given reasons can make a response locally appropriate, then he might have the trip through contemplation there to cut out strike of the demon type cases that would generate false positives on a pure appropriateness view.
Brad,
I wasn’t thinking of ‘taken’ as an intentional act or state. Rather, its just ‘p isolated ought to be preferred to q isolated’. The ‘contemplation’ bit is Chisholm’s way of deploying the idea of isolation. But that is just exactly what I’m interested in, how to get an ought out of the isolation test, as it were, so explained — explained as a focus of contemplation.
Mine’s different because I’m directly defining betterness in terms of required preferences. It’s sort of a bridge principle.
I don’t know that Chisholm here defines ‘appropriate’, but something such as ‘ok’ or ‘permissible’ would do.
Robert,
Ok. But I am not clear how your version is an advance. You say you want to see “how to get an ought out of the isolation test” as Chisholm glosses it. Do you then aim to be giving an alternative explanation of how to get this our of an isolation test?
If so, I can’t see how just defining betterness in terms of required preference does that. Aren’t you just asserting that the isolation test gives you an ought?
Or are you rejecting the need for an explantion? Is the idea that Chisholm gets into trouble by trying to hint at an explantion by talking about contemplation?
Say I isolate Bill marrying Kate and Bill marrying Jack. Which ought I prefer? Following Chisholm’s advice I contemplate. I want to ask, as you put it, “why should the contemplation of two states of affairs result in a *requirement* to prefer one to the other?” But how does your definition help me understand this any better (how does it serve as a bridge)?
Could it be that he was trying to formulate a buck-passing account of value but was not quite there yet? Maybe it’s only my buck-passer lenses speaking. But, it does seem appropriate (sorry…). Take the analysis:
‘p is intrinsically better than q only if it is appropriate for one to prefer p to q and the contemplation of just p and q requires one to prefer p to q.’
The fact that it is appropriate to prefer p over q could be explained with the fact that p provides stronger reasons than q. These reasons make it appropriate to prefer p in the first place. The same stronger reasons which p provides also explain why the contemplation of p and q (and their reason-providing features) require preferring p to q. In addition, because we ground the appropriateness of preferring p over q and the grounds for the requirement to prefer p upon contemplation on reasons, the account is not circular. No value concepts need to be used in explaining the reasons – just the normal reason-providing features. And, we don’t get ought from is but from reasons. Just a thought.
First point: Chisholm didn’t mean there to be two distinct clauses in that analysis, I don’t think. See p. 52 of Brentano and Intrinsic Value where he defines appropriateness in terms of requirement. Appropriateness drops out of the official analysis, which is:
A is intrinsically preferable to B =df. A and B are necessarily such that, for any x, the contemplation of just A and B by x requires that x prefer A to B.
The definiens is just Chisholm’s way of cashing out the claim that it is appropriate to prefer A to B.
I confess I’m still not getting the problem. Suppose hedonism is true, and suppose that A contains more pleasure than B (and neither contains any pain). Then, on Chisholm’s view, if we contemplate A and B, we are required to favor A. The requirement to favor doesn’t come from the mere contemplation all by itself. It comes from the contemplation, plus hedonism, plus the hedonic facts about A and B.
I guess you might still wonder why contemplation comes into it at all. That’s reasonable enough. One way to see why it might come in is like this. Suppose we take your suggestion, and say that A is intrinsically preferable to B iff A ought to be preferred to B. Then we might ask: ought to be preferred *by whom*? One answer: by everyone. But that won’t work. There are some states of affairs that are too complicated for me to contemplate. Plus I might be busy doing more important things, so it might be wrong for me to do any favoring. So maybe a better answer is: by anyone who contemplates A and B.
Sorry for the long comment.
Ben,
If what you say is right, viz., the assumption of hedonism (all and only pleasure is intrinsically good) is what requires us to favor p over q when we contemplate them, then why does he say that his formula shows us how to reduce value concepts to deontological ones? He’s supposed to be explaining betterness in terms of our being required to prefer it, not our being required to prefer it in terms of what is better.
Or are you saying something else?
We could think of hedonism as a substantive theory about what we are required to favor. It would say roughly that we are required to favor A over B iff A contains a greater balance of pleasure over pain. From this, and his analysis of intrinsic betterness, we could derive the claim that A is better than B iff A contains a greater balance of pleasure over pain. So value concepts are still being reduced to deontological ones. Couldn’t that work?
Yes, that must be it. I was assuming that in contemplating p and q, it just couldn’t be their values that generated the requirement. But if what you say is right, the story could just be that goodness is whatever we are required to prefer to neutrality, but we are assuming a theory in which we are required to prefer pleasure to neutrality.
It is, however, confusing, since Chisholm tells us initially that he is assuming hedonism as a theory of value. But it turns out that this is a trickier assumption that one might think.
You mean we figured something out? Sweeeet.
Maybe so!
Two other thoughts.
One: Maybe we’re supposed to read the consequent as saying, roughly, p is intrinsically preferable to q only if x ought to prefer p to q, where x is understood as being aware of p and q. That is, he doesn’t want to have requirements on preferences in absence of someone’s being aware of the things that are to be preferred. You could have a different theory, a more ‘objectivist’ view of requirement, in which you ought to prefer p to q even if you aren’t aware of them.
The other: This now strikes me as classic Chisholm-type argument. He gives a very careful analysis of the relationships beteween concepts bottoming out in something like this required preference claim. But then If we ask where the requirement comes from, we really get no story (or at least no obvious story). Its like getting a very nice conceptual analysis of the relationship of epistemic justification to evidence and belief, but when you want to know what justification itself is, well, no story.
Well, the story has to end somewhere, right? But Chisholm doesn’t merely give conceptual analyses. He also gives (in Ch. 6-7 of B&IV) a substantive theory of value (which of course is at bottom a theory of requirement). The story about where the requirement comes from is told by that theory.
You mean that ‘fittingness’ theory? Hmmph.