Some Of Our Books


« Paying People Not to Do Good: A Puzzle about Superogation | Main | 2009 Mad Meta: Program announced »

June 14, 2009


Feed You can follow this conversation by subscribing to the comment feed for this post.


Great topic. A few thoughts:
I worry that distinction #2 should be redescribed. As you have it:
'conditional' = 'valuable in relation to something else'
'unconditional' = 'valuable in all circumstances'

That sounds correct to me as a gloss on 'unconditional'. This seems to be Kant's usage when he contrasts the value of the good will with the value of happiness. The latter is good only on the condition that it is accompanied by good will, whereas good will is good regardless of whether it is accompanied by happiness (or any other good).

But 'conditional' ends up being nearly synonymous with 'extrinsic' (as you describe it): valuable because its value is acquired from something besides itself. Shouldn't 'conditional' be something more like 'valuable only in some circumstances'? I'm wondering if perhaps all unconditional values are intrinsically good, even if (as Langton's paper argues) not all intrinsic goods are unconditionally valuable.

Also, I share your thought that what does a lot of work in Kantian thinking about the value of humanity is that it is priceless. My reading of Kant's prohibition on suicide is that his argument rests on the inference that one's humanity is priceless because it is unconditionally valuable. But your remarks suggest (and I agree) there are problems with using this inference to support a total prohibition on suicide. It would justify the claim that humanity can't be traded for something of conditional value (happiness, say) but it does not justify the claim that it can't be traded for something also of unconditional value (humanity). Kant might well have been OK with that result though, given what he says in the 'casuistical questions' that follow the discussion of suicide in the Metaphysics of Morals.

I'm not sure the question of where the objects gets is value - from itself or from something else - is the best way of making the intrinsic/extrinsic distinction. A rare stamp does get its value from itself but it's not clear whether it has intrinsic value in Moore's sense (think of the isolation test). Perhaps a better way to distinguish intrinsic value is to ask whether the properties that give the object this kind of value are intrinsic properties of the object - properties which any dublicate of the object would have. The idea of getting value from itself is closer to the notion of final value. The rare stamp would have final, extrinsic value. Getting value from something else would then be called derivative value.

I agree with Michael's point about conditional value. One way to make that distinction would be to ask whether there are disabling/enabling conditions for the value in question.

I've also got worries about the way in which the priceable/priceless distinction is made. Note that on this definition, if we accept the consequentialist framework, then whatever has the most value would be priceless. But, I take it that this is not the idea. The distinction should cut between kinds of value rather than between amounts of value. And, even if I agree that spending time with a loved one is priceless, I don't see how it could not be rationally traded for anything (say, saving the world from being destroyed). I'm not sure what the better way to make the distinction is. Maybe values to be promoted and existing values to be protected would be better. Or, values which do not allow trades with objects of the same kind. Or, values not to be used as means.

Yes, regarding conditional value: I shouldn't have said "in virtue of." I meant the distinction to be symmetrical, so it's valuable in all circumstances or not. I think the suggestion about dis/enabling conditions is very helpful, thanks!

Jussi, can you give me an example of a what might have the most value, within a consequentialist framework? Would this be an accumulation of intrinsic goods, like population P's well-being? Wouldn't it or its parts always be at least in principle rationally tradeable for other intrinsic goods?

Jussi: Suppose we assume the Moorean framework for thinking about intrinsic value. On a particular occasion, there might be a maximally valuable outcome. It might be irrational for the agent who could actualize that outcome to trade it for any of the alternative outcomes which he might realize instead. But I think that Adrienne meant, when describing things that are priceless, to focus on outcomes or entities or relations or whatever that it would be irrational to trade for any other possible outcome, entitiy, or relation on any other possible occasion. Consequently, within a Moorean-style framework – provided that we make some typical Analytic philosophy assumptions about which situations are genuinely possible – nothing is priceless: it is always at least possible that there exist another outcome with greater intrinsic value. At least, this is so on all the axiologies that I have ever heard Mooreans suggest (according to which things like episodes of pleasure taken in friendship or the contemplation of beauty are intrinsically valuable, and the more such episodes that there are, the better).

Thanks for the post, Adrienne. I'm not following the explanation of the difference between intrinsic and extrinsic value. I would have thought that when something has value in virtue of being part of a larger valuable whole, it has value in virtue of its extrinsic properties. Which would make it extrinsically valuable. What am I missing?

I'm also not following, and have never really understood, the difference between conditional and unconditional value, even granting Michael's revision. Kant says there is this sort of value that the good will has, such that it has that kind of value in all circumstances. So it has that kind of value unconditionally. The question is: what kind of value is it? Maybe the claim is that the good will retains its intrinsic value in all circumstances. Fine; but then the same is true of other alleged intrinsic goods, such as states of affairs consisting of an innocent person getting some pleasure from some beautiful artwork. Anything that is intrinsically good and has its intrinsic properties necessarily will have its intrinsic value unconditionally. So it's not clear what sort of value can be plausibly attributed to the good will alone. Maybe the claim is hat the good will retains its overall value, its value derived from all its properties whether intrinsic or extrinsic, in all circumstances. But Kant never makes any attempt to justify this claim, and since it is prima facie absurd, he probably didn't mean it. So what sort of value is supposed to be had unconditionally by a good will? (There are papers by Fred Feldman in the Journal of Ethics and Erik Wielenberg in the Journal of Value Inquiry on this sort of stuff, probably making similar points, but I don't have them handy to see what they say. They would be worth looking at.)

If the question is whether there is conditional but priceless value, then what comes to mind are theses to the effect that the essence of human happiness is being in a certain relationship. The classical, supernaturalist versions involve a relationship with God, but you could have a naturalist version that said that happiness was all about loving relationships. Then you could say that happiness was valuable conditionally (because of the dependence on relationships with others) but priceless (because there is nothing worth trading happiness for).

Adrienne and Jason,

I was thinking of infinities. Imagine that I could invent a pill that would give someone an infinitely long, self-contained happy life of an oyster. Doing this would thus have an infinite amount of value. Doing other things would thus have less value and it's not clear why it would be rational to trade this outcome to anything else. It's true that you might suggest that at least the parts of this are tradeable but so would I think of the parts of spending time with loved ones or other persons.


I wonder about this:

"Fine; but then the same is true of other alleged intrinsic goods, such as states of affairs consisting of an innocent person getting some pleasure from some beautiful artwork. Anything that is intrinsically good and has its intrinsic properties necessarily will have its intrinsic value unconditionally"

I'm not sure this is right. What if the artwork the person is looking at depicts an aesthetically pleasing instance of torturing innocent people - something very evil. In that case, it seems to me, the state of affairs of an innocent person getting some pleasure from a beautiful artwork no longer has value. Looking at other people being tortured is not something one should enjoy. It's still true though that the state of affairs has in other situations intrinsic value and has its intrinsic properties necessarily. The point is that, for any state of affairs, there can be external considerations that disable the abilities of the state of affairs to make that state of affairs valuable.

You can avoid this by saying that the value-bearers are states of affairs that contain the whole world (and their histories and futures). But, I'm not sure why we should say that. It would collapse the intrinsic/extrinsic distinction for one.


I think what you are really suggesting is that the intrinsic value of a state of affairs does not depend solely on its intrinsic properties. Is that right? If so, you're disagreeing with Adrienne's definition of intrinsic value.

It seems overblown to say that for any state of affairs, "there can be external considerations that disable the abilities of the state of affairs to make that state of affairs valuable." Suppose the pleasure is taken in something innocent. Wouldn't that be enough? Why think the bearers of basic value would have to be whole worlds?

Hi Adrienne -

I posted this yesterday, or thought I did, but apparently it didn't take so here goes again. Really, I just have a quick question. I find your claim that "consequentialism embodies the idea that intrinsic value can be priceable" (I'm paraphrasing) very interesting. Could you say more about what you mean here?

Thanks, everyone, for all of the interesting ideas and questions! A few thoughts:


I had in mind Jason's point: if we focus on outcomes or entities or relations or whatever that it would be irrational to trade for any other possible outcome, entitiy, or relation on any other possible occasion. Consequently, within a Moorean-style framework – provided that we make sometypical Analytic philosophy assumptions about which situations are genuinely possible – nothing is priceless: it is always at least possible that there exist another outcome with greater intrinsic value."

More generally, I was thinking of hedonistic and preference-satisfaction utilitarians, who attribute intrinsic value to and only to pleasure/preference-satisfaction, and seem to infer from this that they are the basic currency in a values economy. The whole point about humanity, for a Kantian, is that it can't be put into such an economy.


You write: "Kant says there is this sort of value that the good will has, such that it has that kind of value in all circumstances. So it has that kind of value unconditionally. The question is: what kind of value is it? Maybe the claim is that the good will retains its intrinsic value in all circumstances."

This is almost my point. But I'm saying that both labels--"intrinsic" and "unconditional"--indicate ways of *having* value. I still incline toward talking about the "source" of a thing's value, in elaborating what it is to be of intrinsic value, but I'll indicate some hesitation about this below.

However, perhaps talking about value in some versus all circumstances is not the best way to elaborate the conditional-unconditional distinction. I say this because I take it Rae Langton is right when she says self-conferred value is intrinsic (you get if from yourself) but conditional (you're valuable only on the condition that you value yourself). And this would be true even if one necessarily values oneself, a la Korsgaard's view--i.e. even if the condition is always fulfilled, one is still of intrinsic but conditional value. I like Jussi's suggestion that the question of conditionality is whether there are any enabling conditions on a thing's value.

I think constitutive value may be another case of intrinsic but conditional value. Going back to the example I take from Mill: virtue has only conditional value--it is valuable on the condition that it is a constitutive element of happiness. But when it is valuable, it is valuable because of what it is. Now, maybe this is a mistake, since the relevant property is 'being a part of happiness,' which is not an intrinsic property.

So maybe I'm eliding two distinctions. Go with the Moorean way of elaborating intrinsic value in terms of intrinsic properties. Then the distinction I was trying to capture with the terms "intrinsic" versus "extrinsic" needs a pretty label to indicate the difference between internal and external sources of value. Do you get your value from yourself, or are you valuable only because of a relation you bear to something else of value? Virtue gets its value from itself, when it is a part of happiness--happiness isn't *something else* of value; happiness is, in part, virtue.

Yeesh, sorry to go on so long. I'll stop now.



you might be right. What I want to distinguish between is in virtue of what something has value and what is relevant for the value. I think the talk about being dependant is ambiguous between the two. So, I would want to say define intrinsic value in terms of the good-makers being intrinsic properties of the value-bearer and leave room for the idea that despite this, there can be external factors that affect whether the normally good-making features make the object intrinsically valuable in the given circumstances.

You are also right about the second point. All I would actually need is that for *some* state of affairs there are disabling considerations. That would be enough to show that it's not true that 'Anything that is intrinsically good and has its intrinsic properties necessarily will have its intrinsic value unconditionally'. That the value-bearer's would be whole worlds would rule out this possibility. If value-bearer's are not whole worlds, then it's not clear why some of the disabling conditions could not be external to the value-bearing state of affairs.

OK Adrienne, that helps. So your answer to my question was, I think, that the kind of value something has unconditionally is just plain value. This is a generic property that is had by things in different ways. It can be had in virtue of intrinsic or extrinsic properties, it can be had conditionally or not, etc. Happiness and the good will both have this property, as do lots of other things. I think Chisholm called this "generic value." Generic value is the fundamental value property in virtue of which the other value properties - intrinsic, extrinsic, conditional, unconditional- are defined. Is that the idea?

I just want to point out that there is a different way of thinking about these things. Intrinsic value is the fundamental value property; things that are extrinsically valuable do not have the fundamental value property at all. Maybe there is a disjunctive value property, being intrinsically or extrinsically valuable, that is shared by intrinsic and extrinsic goods; but it is not an interesting property and certainly not fundamental. I don't know if I stated this very clearly- I find this a very hard issue.

Hi Ben. I'm not sure about this notion of generic value. For one, I'm inclined to think that saying something has priceless value is not a way of saying it has generic value in a certain way. But perhaps I need a notion like generic value to make sense of this "source" talk. Couldn't there be a plurality of kinds of value that can be had in various ways? E.g. aesthetic, moral, priceable, priceless (not that these are all on the same dimension). I'll have to think about it some more.

Dale: Just to be clear, I don't think that consequentialism embodies the idea that intrinsic value can be priceable (in Adrienne's sense). "Consequentialism" is supposed to describe a certain class of theories about the nature of morally right action. However, certain accounts of intrinsic value (of its nature and its bearers) that have been favored by consequentialists do seem to imply that intrinsic value can be priceable. These accounts typically have several features in common: they hold that intrinsic value attaches to certain episodes or happenings and that intrinsic value is additive, so that (other things being equal) the greater number of such episodes or happenings, the more intrinsic value there is. In principle, one could be a consequentialist and an intrinsic value aficionado but adopt some offbeat theory on which, e.g., only one entity (e.g., God) was i-good, or value is not additive, etc. Then, there might be entities or situations or relations that were priceless in Adrienne's sense.

Jussi points out that, if a life full of oyster-grade pleasure could be infinitely long, it would contain an infinite amount of pleasure. Consequently, he says, it could be priceless. I'm not so sure. If this is possible, would it not also be possible for there to be a life full of higher-grade (I don't mean higher quality, necessarily, I just mean more intense) pleasure that was also infinitely long? If so, this life might be rationally preferable to the life full of oyster-grade pleasure. To countenance such nomological impossibilities as immortality, we are going to need something beyond high school math, which is (sadly) all that I have.

Hi all-

Dale's been having trouble posting, and so emailed me directly. He gave me permission to post his comment, and my response, so that I can keep this discussion all in one place, for future reference:

Hi Adrienne - I think I see the point better now. But I wonder if it's true (insofar as this is not really what you were interested in discussing, you're perfectly free to tell me to go suck a lemon). It might, of course, be true of certain *forms* of hedonist utilitarians, for instance. But let's say that you have a certain axiology where there is a certain value, x, that's lexically dominant on all other values. Let's say you're a certain kind of bastardized Millian who believes that there are three kinds of pleasures: lower, higher, and highest. Higher pleasures are superior to lower pleasures (but not lexically), but the highest pleasure (of which there is only one, i.e., the contemplation of God, or the great spaghetti monster, or something like that) is lexically dominant on all other pleasures. Couldn't one still be a consequentialist and hold that the highest pleasure is, on your understanding, priceless? You might respond by saying that the consequentialist still has to hold that the highest pleasure is not priceless: it can be rationally traded for more of the highest pleasure (perhaps we trade-off one person contemplating God for millions of people contemplating God). But this just seems to me to be an axiological claim on which consequentialism remains neutral. One might simply say that it always makes a state of affairs worse to trade-off the highest pleasure for any amount of anything else. Or one might make some other claim, viz., that there really is, in this universe, only one quantum of the highest pleasure to be had--it literally couldn't be traded-off. Under either of these assumptions the highest pleasure would be priceless (at least as I understand "priceless"). But even under any of these assumptions, one could still be a consequentialist, couldn't one?

Thanks, Dale. I think you're right. As I wrote my response to you, I was
thinking that, really, there is nothing in principle preventing a
consequentialist from embracing priceless value--it's just that I don't know of
any who do so.

Hi Adrienne,

Great post! I find the distinctions you draw extremely helpful and I’m still trying to puzzle over them. In particular, I’m wondering about the case of constitutive value. You say that you think constitutive value might be another case of intrinsic but conditional value because virtue has value intrinsically (it is valuable because it is virtue) but it’s being valuable is conditional on it being constitutive of happiness. [Though as you allow perhaps being constitutive of happiness is not what makes virtue virtuous and so not an intrinsic property of it.]

I was wondering about the relationship between constitutive relationships of this sort and the priceless/priceable distinction. It seems that the priceability of an A which is constitutive of some intrinsically valuable priceless B [and gets its value because it is constitutive of B] depends on whether there are other ways to constitute B. If there are other ways to constitute B, then A is priceable [tradeable], but if there aren’t other ways of constituting B, then A is priceless.

Let’s suppose, for example, that being healthy has intrinsic priceless value. If being active is the only way (or a necessary part of the only way) to constitute health, then being active has intrinsic priceless valuable, but if there are competing ways of constituting health that don’t involve being active, then being active is priceable… and still intrinsically valuable? I’m not sure about that last part. I’m unclear about how to think of the property of the non-uniqueness of a constitutive part of something valuable; it seems to affect its priceability but why does it (or does not) affect its intrinsic/extrinsic value? For example, in the case of virtue being constitutive of happiness, if being virtuous were to only be one of many ways of constituting happiness, wouldn’t that weaken the claim that virtue’s value is intrinsic?

Hi Jen!

Really interesting line of thought. I'm definitely torn about whether to say a constitutive element in an intrinsic good has intrinsic value. I hadn't really separated out these two ways of characterizing intrinsic value--in terms of intrinsic properties versus in terms of sources of value. But supposing it does make sense to attribute intrinsic value to a constitutive element of an intrinsic priceless good, your point that the pricelessness of the element depends on whether it's a necessary element seems right on.

I guess I don't see why the intrinsic value of the constitutive element would depend on this, however. It seems like the element has the intrinsic value when it is in fact an element of an intrinsic good, and lacks intrinsic value when it is not. So virtue would have intrinsic value when it is a part of happiness, even though you (or some individual) can get happiness without virtue.

I also need to think through the notion of final value, which a couple of people have brought up, and consider how it fits into this scheme.

Great topic!

In your original post, Adrienne, you suggest that all extrinsic value is also priceable. While this might be true for things that are extrinsically valuable in virtue of being instrumentally valuable, there are other ways things can have extrinsic value, and it seems less clear that priceability follows.

The property of uniqueness, for example, is an extrinsic property, and it seems that things can be valuable in virtue of being unique. If something is valuable in virtue of being unique, it doesn't follow necessarily that it could be traded for something of equal value. Historical artifacts might be examples of such a case.

Thanks, Gwen! Yes, it looks like again the difference between value had in virtue of in/extrinsic properties and value gotten from in/external sources may be causing me some trouble. Uniqueness is, as you say, an extrinsic property. But does a unique object get its value from external sources? From the nonexistence of other objects of its kind? That sounds strange to me.

If we go ahead and say the value had in virtue of uniqueness is extrinsic value, then it does seem plausible that something might have extrinsic yet priceless value. I was thinking that if an object got value (only) from an external source, then it would always be rational to trade that object for the source of its value.

Is there anything to this notion of sources of value, or is it just a manner of speaking? I'm not so sure, at this point.

The comments to this entry are closed.

Ethics at PEA Soup

PPE at PEA Soup

Like PEA Soup

Search PEA Soup


  • Unless otherwise indicated, the views expressed in any given post reflect the opinion of only that individual who posted the particular entry or comment.