Chisholm II

Well, since the comments on my last post were so interesting and helpful, let's see how things go with this, another apparent problem with Chisholm's views. Chisholm (in 1978) defined intrinsic goodness in terms of a generic account of intrinsic value states (e.g., either intrinsic goodness or intrinsic badness):

p is an intrinsic value state =df there is a world w such that p reflects all the good and evil that there is in w; and if p is not neutral, then every thing that reflects all the good and evil that there is in w either entails or is entailed by p.

This sure seems to spell out nicely the intuition that the goodness of an intrinsic good, for instance, doesn't require that there be some distinct other good which neither includes it nor is included within it.

However, Chisholm notes that this has the implication that disjunctive states of affairs can't be the bearers of intrinsic value. For instance:

A. Either 3 Canadians being happy or 2 Canadians being unhappy

His argument: This disjunctive state can't reflect all of the good and evil there is in any possible world, since in any world in which it obtains will also contain the good and evil that is in one or both of the disjuncts, and that good and evil will reflect all of the good and evil in that world.

As a result, we can't say, for instance, that A is intrinsically better or worse that any other state, not even the intrinsically neutral state

B: There are stones.

So here's the problem. Consider these disjunctive states:

C. Either 3 Canadians are happy or 4 Canadians are happy.
D. Either 2 Canadians are happy or 1 Canadian is happy.

Doesn't the disjunctive state C seem intrinsically better than D? But how can this be so if neither can be the bearers of intrinsic value? Indeed, they both must, since intrinsic goodness for Chisholm is defined as being better than neutral, and both C and D are that.

That is the problem. But how to fix the account of intrinsic value states, which it seems to me is basically right?

Another puzzling feature of this situation: If disjunctions can be the bearers of intrinsic value, then there are bearers of intrinsic value whose value is indeterminate. All that we can say is that it is better or worse that other states, but not how much better. C, for instance, is better than D somewhere in between 1 and 3 units, but we can't say exactly how much.

It could be that somewhere Chisholm altered his view on the value of disjunctive states, though I don't know where, if so. Or maybe this isn't a problem after all. Is it?

Comments

[Ben Bradley]

Disjunctions are an interesting problem, I think, and not just for Chisholm. It's not just disjunctions that are a problem either. What about: *there is at least one happy Canadian.* What's the value of that? Is it less or more than: *there are fewer than seven happy Canadians*?

Michael Zimmerman and Erik Carlson have written some good stuff on this topic. I think Zimmerman says that it doesn't matter what we say about the values of disjunctions and such, since they don't have *basic* intrinsic value. We might as well say they have no value. Carlson, I think, does something with ranges.

[Richard Chappell]

On first glance (which I'm afraid is all the glancing I've yet done!), it looks like the thing to say about disjunctions etc. is that their value is indeterminate between that of each basic disjunct. Then it's clear how C can be better than D. (And it's still indeterminate whether *there is at least one happy Canadian* is better than *there are fewer than seven happy Canadians* -- which seems the right result, right?)

[Robert Johnson]

Ben and Richard,
I'll have to look at what's beens said already. But there is a difference, between disjunction and the other cases you list. These other cases seem to me to have an indeterminate amount of goodness or badness, but that could be because of purely epistemic reasons. We know there's less that seven happy Canadians, but we just don't know how many. It's not that the number of happy Canadians, while less than 7, is itself indeterminate.

In the disjunction case, things are different. It does not seem to me, as Richard suggests, that we can reduce the question of the comparison of the disjunctions to comparisons of each disjunct. Over and above the betterness of 3 happy Canadians to 1 happy Canadian (and so on) there is the fact that the one disjuction is better than the other. It's not just that there is some determinate amount of betterness in the one over the other, and we just can't know about it. It seems that it really is indeterminate how much better C is than D, one that couldn't be remedied by further investigation, for instance or better epistemic gear.

[Richard]

Hi Robert,

I'm afraid I don't see the difference. I assume the State of Affairs *there are fewer than seven happy Canadians* just is (numerically identical to) the disjunctive SoA *there are six happy Canadians or five happy Canadians or four or [...]*

Now, I gather from your comments that you reject this equivalence. But could you explain for me how the two are meant to differ?

[Richard]

Another thing: "It seems that it really is indeterminate how much better C is than D, one that couldn't be remedied by further investigation, for instance or better epistemic gear."

I agree. I meant to speak of non-epistemic indeterminacy throughout. Now, if the value of C is indeterminately 3 or 4 utils, and D is indeterminately 1 or 2 utils, then it is objectively indeterminate how much better C is than D. To be precise: it is indeterminate whether C is better than D by 1, 2, or 3 utils. (But it is at least determinate that C is better than D, by supervaluation over all possible resolutions of the indeterminacies.)

[Ben Bradley]

FYI
Zimmerman, "Evaluatively Incomplete States of Affairs," Phil Studies 43 (1983)
Carlson, "The Intrinsic Value of Non-Basic States of Affairs," Phil Studies 85 (1997)
Anthony Anderson, "Chisholm and the Logic of Intrinsic Value," The Philosophy of Roderick Chisholm.

I think Richard is right that there's no need to bring in anything epistemic, even for quantified states of affairs that can't be identified with disjunctions.

[Robert Johnson]

Thanks for the citations!
I agree there's no need to bring in anything epistemic. It's that it can be brought in in the case of 'less than' and 'more than' judgments to explain the indeterminacy. I don't see how it could in the case of betterness comparisons among disjunctions. That is, if I am right that the betterness of C to D is more than the betterness of each disjunct of C to each disjunct of D.
Also, that, e.g., 'less than 7 Canadians are happy' may be identifiable with a set of disjunctions. But it is the betterness of one disjunction to another that I was commenting on.