Skip to main content.
April 17th, 2006

Counterexamples to Lewis on Value

In “Dispositional Theories of Value”, Lewis endorses the following two claims.

Here are a couple of counterexamples to this pair of theses. I don’t know whether these are at all original; I’m not very familiar with this literature.

Some people have many thwarted desires; others don’t. I value being one of the ones who does not. Or at least I think it is valuable to not have many thwarted desires, so if Lewis’s first thesis is right then I would value this under ideal circumstances.

But I don’t desire to desire this. To be sure, I do desire to not have thwarted desires. But I don’t regard this status of mine, desiring to not have thwarted desires, as something I have pro-attitudes towards. It seems to me constitutive of having desires that one desire to not have many thwarted desires, since I’m essentially a thing that has desires. So necessarily I desire to not have thwarted desires, so if I desired that I desire to not have thwarted desires, I’d be desiring something that I recognise as a necessary truth. And this seems like a very odd attitude to have. At any rate, I don’t have this attitude.

So this is a value, or at least something valuable, that I don’t desire to desire, and that I wouldn’t desire to desire if my circumstances were more ideal.

Perhaps there is a gap in that argument. I said it is essential to me that I desire not to have many thwarted desires. But I only have that property if I exist, and I might not exist. (Indeed, barring a dramatic medical revolution I won’t exist one of these centuries.) Maybe my desire to exist is a desire to desire that I not have many thwarted desires. I don’t really think it is. When I introspect I don’t see any second-order desire to desire to not have thwarted desires, but maybe I’m just not looking closely enough.

Still, considerations of existence and non-existence suggest a second counterexample to Lewis’s theory. Poor Billy is slowly and painfully dying. He belives (rightly or wrongly) that this protracted death is an affront to his dignity, and because he so values his dignity he wishes he were already dead.

Does Billy desire to desire dignity? No. He does desire dignity, but he wishes that he didn’t desire it, because he wishes that he had no desires at all. So Billy values something he doesn’t desire to desire.

Note that I’m not saying that anyone who desires not to exist thereby cannot reasonably desire anything that entails existence. That would be a most implausible claim about desire. (Or so I say; there are some who deny this, or something slightly weaker than it.) Rather, I’m just making it a condition of the case that Billy’s state is so deplorable by his own lights that as a matter of fact he does not desire anything that entails living, such as desiring dignity. That seems to me compatible with valuing dignity, so the second-order desire analysis of valuing fails.

Posted by Brian Weatherson in Workbench

15 Comments »

This entry was posted on Monday, April 17th, 2006 at 12:17 am and is filed under Workbench. You can follow any responses to this entry through the comments RSS 2.0 feed. Both comments and pings are currently closed.

15 Responses to “Counterexamples to Lewis on Value”

  1. Ralph Wedgwood says:

    These are very interesting counterexamples! Surely, though, they’re first and foremost counterexamples to the second of the two theses that you list — i.e. to the equation of the attitude of valuing with desiring-to-desire, not to the equation of being a value with being something that we would value under ideal conditions.

    Anyway, your second counterexample seems to me more persuasive than your first. Your first rests on the claim that it is “constitutive of having desires that one desire to not have many thwarted desires”. But surely there are creatures (small children, etc.) who have desires, but lacking the concept of desire, have no desires about desires at all?

    Your two counterexamples differ in that the first attacks the sufficiency of desiring-to-desire for valuing, while the second attacks its necessity. The stock objection to sufficiency just focuses on higher-order desires that are viewed by the individual subject herself as irrational fetishes or the like. (Examples of that sort have been made fairly familiar by Gary Watson’s criticism of Harry Frankfurt’s invocation of higher-order desires.) I can’t off the top of my head recall any attempts to construct counterexamples to the necessity of desiring-to-desire for valuing; so your second counterexample seems to me more original as well as more persuasive.

  2. Ralph Wedgwood says:

    Oops! On second thoughts, I see that both your cases are counterexamples to sufficiency: neither is a counterexample to necessity. What was I thinking?

    As I said, I can’t offhand remember any other attempts to give counterexamples to sufficiency. But perhaps Michael Smith or someone influenced by him has objected to sufficiency by insisting on the possibility of a kind of akrasia — in particular, the possibility of a kind of akratic agent who not only fails to desire what he values, but even fails to desire to desire what he values?

  3. Brian Weatherson says:

    Thanks for those comments Ralph.

    I agree with the point about small children. It was too strong a claim to say that desiring not to have many thwarted desires is constitutive of being a desiring agents. I still sort of think that I value not having many thwarted desires without desiring to desire this, but the argument I gave for that doesn’t work.

    I should look up some of the akrasia literature to see whether examples like me ‘death with dignity’ example exist. I’m tempted to say that my example is of someone who isn’t at all irrational, but saying that depends on hard questions about the debate about euthanasia. Perhaps euthanasia is somehow practically irrational. I doubt it, but I don’t want to take a stand here.

  4. Mike Almeida says:

    Does Billy desire to desire dignity? No. He does desire dignity, but he wishes that he didn’t desire it, because he wishes that he had no desires at all

    This seems to depend closing desire/preference under implication. Something like: If Billy desires to die, and if dying entails (in some sense) having no desires, then Billy desires to have no desires.

    But I doubt that’s true, even if Billy knows the relevant entailment (i.e., that dying entails having no desires). After all Billy might desire that he die and also desire that death does not terminate of all of his desires. Certainly he can desire what he knows won’t/can’t happen. But in that case he doesn’t desire to have no desires.

  5. Brian Weatherson says:

    Well Billy could be like that, but I stipulated at the end that he wasn’t. Now maybe that’s an incoherent situation, or maybe it makes it the case that he doesn’t really value dignity, but I think it’s fair game to make such stipulations.

    Perhaps more explicitly, the because there is a causal because. Billy wishes to die. This wish has caused him to cease wishing for most things incompatible with his being dead, including the existence of his own desires. That seems plausible. Certainly if I wish for p, and p entails q, I can be caused to wish that q. That’s what I’m stipulating for Billy.

  6. Gideon says:

    Billy’s first choice is immediate death. But how does he rank the following possibilities: (a) No death yet plus the loss of the desire for dignity; (b) no death yet plus the retention of the desire for dignity? If he’s indifferent, then it seems plausible that he does not really value dignity. If he prefers (b) to (a), then he both values it and desires to desire it.

    But even if this is a fair response to the original counterexample, it does raise a question. (I don’t remember whether Lewis discusses this phenomenon.) We can easily imgaine that Billy values dignity above all else. It’s his supreme value. If valuing is desiring to desire, how is intensity of valuation to be understood? One natural thought is this: If X values Q above all else, then he has a a desire of unsurpassed strength to have a desire (of unsurpassed strength) for Q. But this case is a counterexample to that proposal. If the desire for the desire for dignity really were of unsurpassed strength, then Billy would prefer retaining that desire to dying. But he doesn’t.

  7. Mike Almeida says:

    Here’s a small reductio. Suppose Billy desires his own dignity. Now provide Billy with the following options: (1) die with dignity or (2) no longer desire dignity. Suppose for reductio that Billy chooses (2). In that case Billy most prefers to go on living (however undignified that life might be) provided that he is not pestered by this inconvenient desire, and that is inconsistent with the example. Suppose then that Billy chooses (1). In that case Billy most wants to die, but that is consistent with the fact that he still desires to desire his dignity. Just stipulating that he does not happen to have this desire to desire his dignity seems close to question begging, no?

  8. Brian Weatherson says:

    While I’m happy to take on board Gideon’s modification to the argument, I’m not convinced that his argument is right that Billy does desire to desire dignity. (I think Mike is making the same point, though his case is harder to parse. Given the choice he’s offering, Billy would say “Both!”.)

    To simplify matters, I was assuming that Billy does prefer (b) to (a). What I’m denying is that Billy desires to desire dignity. To see why I might deny this, consider the following case.

    It’s possible that I’ll get a $1000 pay raise for next year. (I.e. get an extra $1000 for doing the sasme amount of work.) It’s also possible that I could take on a huge amount of extra work for an extra $1000. My preference ranking is as follows.

    (x) Get a $1000 pay raise.
    (y) Get no pay raise and not take on the extra work.
    (z) Get no pay raise, take on the extra work, and hence get paid $1000 more.

    Given that preference ordering, do I desire that my pay be $1000 higher? (By ‘my pay being $1000 higher’, I mean something that could happen either by getting the pay raise or by taking on the extra duties.) The simple answer is “Of course I do! Who doesn’t want more money?” And that’s consistent with preferring (y) to (z). So desiring that p is compatible with preferring that ~p given that my preferred way of having p come about is taken off the table.

    A more complex answer is that whether I desire my pay to go up depends on how plausible it is that I will get the $1000 pay raise. If the pay raise is a mere fantasy, something that isn’t a real possibility, then I think it is natural to say that I don’t really desire my pay to go up. I’d much rather stay on lower pay and lower duties. But if the pay raise is really a live possibility, then it feels more natural to say that I would desire my pay to go up by $1000.

    I’m using fairly heavily here Frank Jackson’s idea that we should understand desires counterfactually. I desire that p if I prefer the nearest possible world in which p to the nearest possible world in which ~p. (That needs extra clauses to deal with ties about the nearest world etc, but that’s the core idea.) If the world where I get the $1000 as a straight raise becomes more salient, then it makes more sense to describe me as desiring that my pay go up by $1000.

    Similarly with Billy. If dying is not really an option, and he prefers (b) to (a), then I agree he does desire desiring dignity. But if dying is a sailent option, perhaps the most realistic option in which he ceases desiring dignity, then I think it’s natural to say that he doesn’t desire desiring dignity.

  9. David Sobel says:

    One could say, Billy does not value dying according to Lewis because Billy does not want a situation in which he wants to die, rather he wants a situation in which he has no wants. But that seems wrong.

    But I don’t think that is telling against Lewis, although I think other examples are. Billy would not be content to lose the desire to die (or to have dignity) unless he lost all desires. Thus so long as he is capable of having values, he values dying (and dignity).

    Several people have pointed out that Lewis’s view misses cases like this: you are imprisoned with very little chance of escape. Being free is of value to one and suppose one counts as valuing it. Does it follow that you want yourself to want freedom? No, that would just be frustrating.

    Perhaps if Brian’s examples are seen as a problem for Lewis, they could be overcome by fixing the above problem. Valuing X is your informed self wanting your actual self to get X. Then we could at least see how one could value death.

  10. Mike Almeida says:

    (1) and (2) are true about Billy, right?

    (1) Billy prefers the nearest world w in which he dies to the nearest world in which he does not. [Billy desires to die]

    (2) Billy prefers the nearest world in which he retains his dignity to the nearest world in which he does not. [Billy desires his dignity]

    In world w, Billy no longer desires his dignity or desires to desire his dignity. Trivially so, since Billy does not exist in w. But it is also true in w that Billy does not retain his dignity. Again trivially so, since Billy does not exist in w, he does not retain his dignity or anything else there.

    So if Billy satisfies his desire to die, he frustrates his desire for dignity: he actualizes a world in which he no longer exists so does not retain his dignity. But you say, Poor Billy … because he so values his dignity he wishes he were already dead. But if Billy finds a protracted death an affront to his dignity, he cannot satisfy his desire for dignity by taking his life. He can only frustrate it that way.

  11. Brian Weatherson says:

    I don’t really know why you think Billy’s dying means that he loses his dignity. I wouldn’t have thought that this was a property that required being alive.

    If you want to use ‘dignity’ for a property that only living people have, choose some other property Billy to value that is consistent with being dead, like not being undignified. I don’t think this is a good use of terms, and I certainly don’t think it is English, but it will make the example work.

  12. Dan López de Sa says:

    According to Brian, Billy values something he doesn’t desire to desire, so it would be a counterexample to the necessity of desiring to desire for valuing. So I side with the first Ralph and not the second Ralph about it.

    As another candidate counterexample to the necessity of desiring to desire for valuing, Gilbert Harman (‘Desired Desires’) offers the case of someone who values listening to Mozart, but is indifferent with respect to her first-order desire to listen to Mozart, and thus lacks the desire to desire it.

    As to the sufficiency of desiring to desire for valuing, it seems to be put in jeopardy by Harry Frankfurt’s (‘Freedom of the Will…’) physician that, trying to understand his addict patients better, desires to have a desire for the drug, without valuing it.

    In my own view, the upshot of the whole discussion was that the Lewisian proposal should be understood, or otherwise amended, along the following lines: valuing is desired desiring, or perhaps merely desiring one does not desire against.

    But Brian’s Billy would be a counterexample also to this, as Billy is supposed to wish not to desire dignity, as opposed to lacking the desire to desire it, is this right?

  13. Mike Almeida says:

    I guess I was thinking that Billy would have to exist to have any property truely predicated of him. ‘Billy’ doesn’t refer to any being that exists in w. So ‘Billy retains his dignity’ seems false, since ~(Ex)(x retains his dignity). But maybe that’s inconsistent with metaphysical views you find more plausible. Maybe you want to say that Billy exists in w, but he happens to be dead there.

  14. Brian Weatherson says:

    Mike,

    That’s right, I do think that a dead person has properties. For one thing, they have the property of being dead. In any case, I’m an eternalist, so I think that in one good sense things never go out of existence.

    Dan,

    I was wondering about whether we could amend Lewis’s theory along those lines, but the point you raised seemed to tell strongly against it. Billy doesn’t just lack the second order desire, he has a conflicting second order desire.

    I thought about trying to say that S values p iff they desire p and don’t have any conflicting desires at any level, but that rules out a priori that people could have conflicting values, and that seems wrong. (Lewis discusses this point when he’s talking about the value of activity and the value of tranquility.)

  15. kyangadac says:

    A problem with Billy’s conflicted desires for dignity and death as posited in this thread is grammatical. Death is being used as a verb and as a noun willy-nilly. Dignity as a possession of a living human (or a just dead corpse) can be seen as valued (desired to be desired etc.) But death is an event that Billy would hasten or otherwise change the circumstances of. In saying that death is equivalent “to wishing he had no desires at all” i.e. wishing he were dead, are you not conflating the event that Billy wishes for with the state that he will be in after the event?