(This marks the fifth of eleven “meetings” of our virtual reading group on Derek Parfit’s Climbing the Mountain—see here for further details. Next week, we will discuss Chapter 7 of the June 7th version of the manuscript, which can be found here.)
In his Formula of Humanity, Kant tells us that “every rational being…must always be regarded as an end…and is an object of respect.” But, as Parfit points out, this requirement to respect all persons doesn’t tell us how we ought to act. In the narrow sense of ‘respect’ (where only acts such as those that are degrading, defamatory, or contemptuous count as being incompatible with respecting persons), we can act wrongly without treating people in ways that are incompatible with respect for them. “But Kant’s formula is intended to cover all wrong acts” (p. 117). So, Kant must have the in the wider sense of ‘respect’ in mind, where all wrong acts are necessarily incompatible with respect for persons. But, in this wider sense, Kant’s requirement to respect all persons tells us only that we must not act wrongly, which isn’t useful or informative.
Perhaps, though, Kant’s claim that rational beings must be treated as ends and that such beings have a kind of supreme value that Kant calls “dignity” is more informative. There are, Parfit claims, two kinds of value. “Some things have a kind of value that is to be promoted. Possible acts and other events [broadly construed to include whatever we can effect: acts, outcomes, processes, states of affairs, etc.] are in this way good when there are facts about them that give us reasons to make them actual” (p. 7). Other things, by contrast, have a kind of value that is to be respected, e.g., persons, our nation’s flag, and the oldest living tree. “Since these things are not events, we cannot want them to happen, or make them happen. But we can respond to them in other ways” (p. 121). We can, for instance, have reasons to treat them in respectful ways. Parfit agrees with Kant that persons have this second kind of value and thus are to be respected rather than promoted. This is what Kant means when he claims that persons have dignity rather than price. Parfit argues, though, that “we are not morally required to respect the value of anyone’s life in ways that conflict with this person’s well-being and autonomy” (p. 7), for it is not human life but the persons who live these lives that have the kind of value that is to be respected. Thus Parfit concludes that neither committing suicide nor assisting suicide need be the sort of act that fails to respect the value of persons.
Interestingly, although Parfit rejects the teleological conception of value (holding that it is not only events, but also certain things—e.g., persons—that are good in themselves), Parfit does endorse the teleological conception of reasons for action, for he endorses what he calls “the Actualist View,” according to which “We have a reason to act in some way if and only if, or just when, this act would in some way be good either as an end, or as a means to some good end” (p. 120). So although we have reasons to respond to things with the kind of value that is to be respected by acting in respectful ways toward them, the explanation for why we have these reasons to so act is teleological: such acts are themselves good ends, and the value of such acts is of the kind that is to be promoted/effected.
In the last section of the chapter, Parfit clarifies what Kant means by his claim that “humanity” has dignity. Although some of Kant’s remarks about humanity suggest that he is thereby claiming that non-moral rationality has supreme value, we should interpret Kant to be making the following, different, and much more plausible, claim: “all rational beings have a kind of value that is to be respected rather than promoted” (p. 127). This value is a kind of moral status or standing. Thus Kant believes, not that all rational beings are good, but that they all have the same moral standing. “But, for the idea of moral status to be theoretically useful, it needs to draw some distinction, by singling out, among the members of some wider group, those who meet some further condition” (p. 127). Kant, unfortunately, fails to draw any useful distinction, or so Parfit claims.
It seems, then, that, as with the last chapter, we haven’t made much progress in answering the question: “What ought we morally to do?” We’ve learned from this chapter that we ought to respect persons, but this, we found, isn’t of much help in figuring out what we morally ought to do. Ultimately, Parfit wants to know if we ever have most reason to do what’s wrong. But so far we haven’t made much headway in determining what we morally ought, or ought not, to do.
Doug: Excellent summary, as usual. I just have a few comments. First, I’m not so sure about your “glass half empty” conclusion that we haven’t made much headway in determining what we morally ought or ought not do. Instead, what I think we’re finding more and more is that there’s just a single Kantian moral principle we need to determine what we ought or ought not do, viz. the consent principle. For instance, in this chapter Parfit agrees with Scanlon that we have reasons not to end someone’s life only if that person “‘has reason to go on living or wants to live'” (122), i.e., there are reasons having to do with his well-being or autonomy. This is because it’s the people living the lives, and not “human life” itself, that’s valuable, and so “we should respect this value” (123). But how in general should we do so? There are two ways cited, both of which seem awfully similar. There’s Scanlon’s way: “by treating people only in ways that could be justified to them” (123). And there’s Kant’s way: by treating them “only in ways to which they could rationally consent” (123). So I think what we’re getting in the last two chapters are specifications and clarifications of putative freestanding moral principles (the mere means principle and the respect of persons as ends principle) which are themselves simply parasitic on the consent principle. Now the consent principle is *not* the only moral principle — on that point Parfit’s been quite clear — but it does at least provide a sufficient criterion of wrongness.
Another thing I wanted to note was that there’s a very nice section in the chapter that you didn’t mention, Doug (for reasons of space, I’m sure), but that is worth mentioning, I think. It was on pp. 123-124, where Parfit distinguishes between Kant’s three kinds of ends: ends-to-be-effected (which includes aims/outcomes we could try to achieve), existent ends (e.g., rational beings/people), and ends-in-themselves (those things with dignity, having unconditional worth/supreme value). What Parfit does that’s interesting here is to undermine a standard interpretation of Kant that only some existent ends — rational beings — can have supreme value, insofar as that’s the only kind of value that can be respected (rather than promoted). But Parfit rightly points out that Kant himself discusses several ends-to-be-effected which have supreme value, and are thus things we ought to try to promote, including a good will, the Realm of Ends, the Greatest Good (the Realm of Ends plus everyone in it having the happiness they deserve by being virtuous), and, perhaps, the *existence* of rational beings.
Oh, one final remark: there’s some fun, dry humor sprinkled throughout this book that I’m really enjoying. One example from this chapter: when Parfit’s discussing Kant’s crazy statements about the wrongness of masturbation and lying (even to achieve a really good end), he remarks, “These are not the claims that make Kant the greatest moral philosopher since the ancient Greeks.” Indeed.
Dave,
I agree that we have made progress with respect to the question of what we ought morally to do, and I didn’t mean to suggest otherwise. The Consent Principle is genuine progress. The point was only that we haven’t made much progress with respect to this question since the chapter on rational consent. I should have said: “It seems, then, that, as with the last chapter, we haven’t, in this chapter, made much progress in answering the question: “What ought we morally to do?”
To be clear, we have made negative progress since the chapter on the Consent Principle. We’ve discovered that neither the Mere Means Principle nor the Formula of Humanity is going to be much help in determining what we ought to do, but we haven’t made much positive progress.
I should point out that I don’t mean to be disparaging in saying that we haven’t made much positive progress in answering the question: “What ought we morally to do?” For I have profited enormously from reading these last two chapters. There is lots of interesting stuff going on in these chapters.
One thing that I found to be most interesting was that we can admit both that it is not just events that can good in themselves and that persons have the kind of value that is to be respected rather than to be promoted, but that we shouldn’t draw any anti-teleological conclusions from such claims with regards to reasons for action, as Scanlon and others do.
Thanks for the other two points, which we’re definitely worth mentioning.
Thanks for the helpful summary, Doug. I was puzzled by many things Parfit says in this chapter. First, I fail to see why using the wider sense of ‘respect’ in the claim that all wrong acts treat people with disrespect is uninformative. I agree that if that’s all one were to say, it wouldn’t be too helpful. But those working in the respect-for-persons tradition construed broadly have done quite a bit to provide a specification of ‘respect’ in a way that covers all wrong acts but is still substantive. So maybe Parfit is right, but I thought much more work needed to be done to establish this point. Parfit seems to gesture in this direction when he says, at the end of sec. 19, that when we reject Kant’s silly claims about masturbation debasing humanity, we are not giving our strongest arguments. But while the gesture towards looking at substantive specifications of respect is the right step to make, I don’t follow Parfit’s point here, for those sometimes seem like very strong arguments to me (allowing, of course, that other times people make bad arguments when appealing to the Formula of Humanity). And much of the tradition Kant established seems like it’s going in a direction of refining, through stronger arguments in successive generations, what it means to debase or respect humanity. By my lights, for instance, Donagan’s substantive specification of respect for persons seems a bit better than Kant’s overall, and Hill’s better than Donagan’s.
Second, I think that the passage Dave refers us to, where Parfit says that the supreme value of dignity is had not only by rational beings (as the “mistaken” standard view has it) but also by good wills, the realm of ends, and the highest good (and possibly the continued existence of rational beings), is problematic. Parfit is right that they have supreme value in one sense: their value is unconditioned on anything else for that kind of value. But the standard view doesn’t deny that. The standard view holds, instead, that dignity is a unique subset of (unconditional) value, namely (unconditional) moral status, and only rational beings have unconditional moral status. It is not had by the good will (qua good will rather than qua humanity), realm of ends, or the highest good. (If anyone’s interested, I try to sort out the differences in some of these kinds of value (humanity and the good will) in a paper, coming out who-knows-when in the Journal of the History of Phil., called “Kant’s Conception of Humanity,” the penultimate version of which can be found on my Web site.) So Parfit is mistaken to say that these different unconditionally valued things all have dignity. Rather they all have unique kinds of unconditional value. Only humanity has dignity. This explains why, at the end of the chapter, Parfit is also mistaken to say that “Kant fails to distinguish between being supremely good and having the moral status that is compatible with being, like Hitler and Stalin, very bad.” Kant does make such a distinction: bad people have humanity (and therefore dignity or supreme moral status), but they have evil wills nonetheless.
Third, I found confusing the claim (five paragraphs from the end) that Kant ignores our non-moral reasons to be happy in saying that the principle of doing what would promote our own happiness applies to us only because we want to be happy (i.e., is a hypothetical imperative). Saying that it depends on our inclinations does not mean that the reason to want to be happy is not based on non-moral reasons. Indeed, hypothetical imperatives are by definition non-moral reasons for Kant, if only categorical imperatives supply moral reasons and if all imperatives are principles of reason (i.e., hypothetical imperatives are, in fact, reasons). Also, arguably Kant has a technical notion of happiness, that just means something like the satisfaction of one’s ends. If that’s happiness, it’s hard to see how the imperative to be happy wouldn’t be a hypothetical imperative.
Finally, the basic argument in this chapter that something close to the Formula of Humanity is not the best account of what we morally should do is unclear to me. On such a principle, at least as it’s commonly interpreted and put to work today, what makes right acts right is that they treat as an end, or respect, humanity (or some broader set of subjects). The consent principle and any other principle of what we morally should promote, etc., are on this account derivative from, or otherwise based on, this foundational principle, on the standard approach. I suppose that maybe the first few paragraphs of the chapter were supposed to dispatch this principle, but as I said I think that argument is too quick.
Josh,
You write, “the basic argument in this chapter that something close to the Formula of Humanity is not the best account of what we morally should do is unclear to me.” As I see it, the basic argument is not that the Formula of Humanity is not the best (or correct) account of what makes acts right or wrong, but that the claim that it is wrong to treat people in ways that are incompatible with respect for them “would seldom help us to decide, in difficult cases, whether some act would be wrong.” Thus, I think that Parfit is arguing that the Formula of Humanity isn’t a useful decision procedure, not that it isn’t the correct criterion of rightness. I think that this interpretation fits with his overall project. What it is important to know if we are to figure out whether we ever have most reason to act wrongly is not what makes act wrong, but which acts are wrong. Once we know which acts are wrong, we can ask whether we ever have most reason to perform them.
Ahh, finally a real-life Kantian to set us all straight!
Let me focus just on your second complaint, about Parfit’s treatment of dignity and supreme value. Your comments suggest you’re reading that section very differently from how I read it. I take Parfit simply to be denying the views of “some [unspecified] writers” who believe that Kant held that supreme value was something that could be had by only existent ends, the paradigm case of which are rational beings. All Parfit’s trying to show here, then, is that those writers are wrong, that Kant clearly held that supreme value could be had by ends-to-be-effected as well, consisting in the four examples I mentioned earlier. But this is perfectly compatible with these various examples having *unique* unconditional value, as you say.
Doug, this is really a great summary. Thank you for the hard work!
One thing. As I understood things, the main question Parfit wants to answer is whether we often have most reason to act wrongly. (I mistakenly wrote ‘ever’ myself once or twice.) However, he has since clarified for us that he agrees with you and Mike A. What he really wants to know is whether we often have sufficient reason to act rightly.
Thanks again for the great summary.
Doug,
thanks for the summary from me as well, it was very helpful! I’ve quite enjoyed following this reading group.
I’m sure I haven’t fully appreciated what’s going on with the Actualist View of goodness of events and the teleological view of reasons, but I wonder whether we go astray if we say that acts of “treating with respect”-kind have value of the to-be-promoted kind.
I have two clarificatory questions about what Parfit says about this:
– I wonder why in formulating his Actualist View, Parfit writes “intrinsic properties and features”, when in the footnote 136 that follows, he admits, following Korsgaard, that an act or event may be “extrinsically good as an end”. Is the definition meant to be only rough and ready?
I suppose it shouldn’t matter whether the feature “expresses respect towards X” turns out to be intrinsic or extrinsic/relational?
[“the Actualist View: Possible acts and other events are good as ends
when they have intrinsic properties or features which give us
reasons to want them to be actual, or to happen, and to make
them actual if we can. …”(120)]
– On p.122 Parfit writes,
“Though this kind of value is not a kind of goodness, and is not to be
promoted, when we can respond to the value of such things by treating
them in respectful ways, these acts may be good as ends, having the
kind of value that is to be promoted.”(122)
I wonder why it says “may be”? Are there cases where “treating respectfully” is not good (in the reason-giving sense) as an end? And if there are, don’t we in these cases have any reason to treat the thing in question with respect?
– apart from these, I have to say I quite like the claim (on p.124) that worms have dignity (so I take it, “supreme value”)
This is very embryonic, but I wonder if Alto’s penultimate point might not be expanded to express a worry about a possible Paradox of Deontology. That is, if acts like “treating X in respectful ways” may themselves be good as ends, and are thus to be promoted, then one wonders if there might not be cases in which in order to promote (maximize?) such acts one had to treat a number of specific respect-worthy things with disrespect. Of course, the “may be” in the clause Alto cites might be intended to head off such a worry, but I’d need to hear more about the details.
Dave,
Yeah, I definitely agree that Parfit intends to show that more than just rational beings have value that is supreme in the sense of unconditioned. But, first, he says more than that, because he says that other things have not only supreme value, but also dignity, which seems like a stretch. (The highest good has no dignity in Kant’s sense of supreme moral status, even if it is the highest good.) The logic behind this stretch is unclear to me, but it seems like he’s equating “dignity” with “unconditioned value,” without distinguishing different kinds of value, such as moral goodness and moral status; I think this is the source of the problem. So that part of his claim is, I would argue, false. And, second, I don’t know who those “some writers” are anymore, since, as far as I know, no one–or at least none of the usual suspects he’s been citing–denies that, say, the goodness of the good will is an unconditional/”supreme” kind of value. After all, that’s right up front in the Groundwork. So I guess it’s also starting to look a bit like a straw person argument.
But, right, none of this negatively impacts his point that different kinds of things/states-of-affairs have unconditioned value. It just means that it’s unclear who he’s disagreeing with in making this point, and that he overstated his point by saying that different kinds of things have dignity.
Very nice summary indeed.
One thing that was bothering me about ch. 5 is still bothering me though. I can’t quite see what Parfit is up to. His argument that the humanity formulation can’t tell what we ought to do I think is based on the assumption that respecting a person’s humanity or treating it as having ‘incalculable worth’, must require us to do something directly with or to that person. This, for instance, explains the jab he makes at the well-known example of Kant’s rediculous views on masturbation. The debasement of humanity, even here, isn’t found in the person’s mishandling of something in himself directly (stop sniggering!). It is in acting contrary to principles (Kant wrongly thought) would be accepted by each of us under the right circumstances. Although Parfit nods toward a difference between treating some person in a certain way and treating her humanity with respect, he apparently thinks that the latter will have to be cashed out in some way in terms of the former. But this isn’t required. Treating a person’s humanity with dignity is just conforming your behavior to what they would agree to as principles governing conduct were they better placed. (More or less, something like the consent principle.) So, again, he’s right that we get nothing further out of the idea of respect for persons; but this doesn’t seem to be news. Surely it can’t be news to the Kantians he discusses; they all know well Kant’s claim that all the formulations of the CI are in some sense equivalent. So I guess I’m a bit baffled now by the trajetory his thinking has taken here.
Doug,
Thanks for the clarification that we’re looking for a helpful decision procedure, rather than a determinate criterion of rightness. That still, though, leaves me with a concern for Parfit: over the years, the standard interpretation is that the Formula of Humanity is to be used as a decision procedure in light of a certain substantive specification of what it means to treat humanity as an end in itself or treat subjects with respect. Kant offered such a specification, as have Donagan and Hill, to name a couple of the bright lights. So before dispatching with the Formula of Humanity as an uninformative decision procedure, it really must be considered in light of the candidate substantive specifications of what it is to respect people that have been proposed over the years. Here’s an analogy: consequentialists can defend non-substantive consequentialist decision procedures (e.g., do whatever you subjectively determine to maximize the good) without a substantive account of the good, in which case it might seem kind of uninformative. But before we dismiss it as uninformative, we should consider how the procedure fares when supplemented with various candidate substantive accounts of the good.
On a separate note, Doug writes, “What it is important to know if we are to figure out whether we ever have most reason to act wrongly is not what makes act wrong, but which acts are wrong. Once we know which acts are wrong, we can ask whether we ever have most reason to perform them.” This raises an important question that I kind of wish Parfit were more clear about: we need to know which acts are wrong, i.e., we need a decision procedure, but that decision procedure must then track the right acts as determined by the correct moral criterion, so I wonder whether the moral criterion and decision procedure might not dovetail here? Maybe not: there are some famous kinds of consequentialism where they do not dovetail. And on some readings of Kant’s ethics, for example Timmons’, the universalizability principle is the decision procedure, while the Formula of Humanity is the moral criterion (Herman suggests something similar). But I’d like more information about how the criterion and the procedure are related, because ultimately, I take it, we want to know not just how to decide whether we have most reason to act wrongly, but, more basically, whether we do have most reason to act wrongly. And this, in turn, requires an account not just of how to know what the wrong act is (as determined by the decision procedure), but, more basically, of what the wrong act is (as determined by the moral criterion). Or maybe I’m reading too much into Parfit’s project?
Josh,
in your first message you wrote you “find confusing the claim (five paragraphs from the end) that Kant ignores our non-moral reasons to be happy” and you end the point by “if that’s happiness, it’s hard to see how the imperative to be happy wouldn’t be a hypothetical imperative.”
I thought Parfit thinks that Kant ignores reasons (“imperatives”) to want to be happy, or reasons to have happiness as a goal, and that Kant thinks there are merely reasons to do what subserves the goal of happiness, if we happen to have that goal (as we in Kant’s view inevitably do – for no reason perhaps). So I think in Parfit’s reading, Kant does not give us “an imperative to be happy” and in that sense ignores our non-moral reasons. (No doubt such an imperative or reason would be non-moral in Parfit’s view).
Arto,
You write, “I wonder whether we go astray if we say that acts of “treating with respect”-kind have value of the to-be-promoted kind.” What if we say instead that such acts have value of the to-be-effected kind? Acts, unlike persons, are things that we can effect. Acts are events, and so we can have reasons to want them to occur, and to make them occur. Persons are not events, so we cannot have reasons to want them to occur, or to make them occur.
You also wonder: “why in formulating his Actualist View, Parfit writes “intrinsic properties and features”, when in the footnote 136 that follows, he admits, following Korsgaard, that an act or event may be “extrinsically good as an end”. Good question. I don’t know the answer. I’m with you I don’t see why it should matter whether the features are intrinsic or extrinsic.
You also ask about why he says “may be” in the quote from p. 122. I don’t know the answer to that either. But these are good questions. Perhaps, you should pass them along to Parfit—he is looking for comments.
Josh, where does Parfit say that these other ends-to-be-effected (such as the good will and the greatest good) have dignity? He doesn’t, as far as I can tell. Instead, he says they have, according to Kant, supreme value, contra some authors who think only existent ends can have such value. But to have supreme value doesn’t entail the having of dignity. Dignity is defined as “absolute, unconditional and incomparable value or worth.” And it’s only “incomparable” value that, he asserts, means “supreme value.” So just because rational beings have dignity, and dignity has supreme value, that doesn’t mean everything with supreme value has dignity.
Doug,
thanks; you ask: what if we say instead that such acts have value of the to-be-effected kind?
I think that’s ok (and indeed we want to have some such story of how the value of persons/things and value of acts hang together), but it now seems to me we have to drop the claim about teleological explanation of reasons, for one of two reasons. I’m not sure of this, but here’s a go.
“To be effected” sounds like “to be done”, and that does not sound teleological. Actions which we we have reasons to do because they are deontically fitting responses to situations, (or ones for which there are “expressive” reasons), are also “to be done”. But it seems that such acts are good as ends only because (or in the sense that) they are to be done.
Whether we say “because” or “in the sense that” is a matter of whether we accept a buck-passing account of value of events or not.
If we do (with Parfit?), then to be of value is just to have features that give us reasons.
Isn’t it then so that there’s no room for the remark from your summary: “So although we have reasons to respond to things with the kind of value that is to be respected by acting in respectful ways toward them, the explanation for why we have these reasons to so act is teleological: such acts are themselves good ends, and the value of such acts is of the kind that is to be promoted/effected.”
That doing X is of value cannot really explain why we have reasons to do X because it is just the same fact? I’m not sure now if Parfit’s view of goodness in the reason-involving sense is committed to these being the same fact – I should check.
If we don’t accept the buck-passing view, we can distinguish between two cases: some reasons for action are “teleological” or value-based (say, doing something because it is enjoyable in itself) and some may derive from the “fittingness” of the action as a response to a situation (say, expressing consolations to someone in grief). [There are complications: the latter action is (in a secondary sense) good because fitting, or reasons-responsive. The former action is (in a secondary sense) reasons-responsive because responsive to value – and thus there are two good-making features involved: the enjoyment, and the fittingness of effecting enjoyment.] It seems that “reasons of respect” are of the second kind, good because fitting, and not teleological. But in any case, Parfit seems to go with the buck-passing option.
Arto,
You raise some interesting worries, and I think that I might have a response on behalf of teleology, but it won’t be something quick. Right now, I have to pack for my trip to San Diego. Let me think about it and try to get back to you tomorrow morning. If not early tomorrow morning, then it will have to be Saturday, as we’ll be on the road all of Friday. But I do hope that you’ll tune in then, because I would really like to continue this discussion.
Sure; I’ll actually be travelling a bit as well (from Jyväskylä to Frankfurt), couple of soccer matches to catch: Brazil vs France (not in the stadium though), and, no joking, Kantians vs Hegelians 🙂
Thanks, Dave, I see where we have an interpretive dispute going on here. I take Parfit to be associating the supreme value of the ends-to-be-effected with the dignity of rational beings when he says “According to some writers, Kant believes that such supreme value is had only by some existent ends, such as rational beings, whose value is of the kind that is to be respected rather than promoted.” I’ve been taking the “such supreme value” to be the supreme value of existent ends, namely dignity. But I see how it could be taken in the more austere sense that it’s simply unconditioned value (or incomparable value?), rather than the same unconditioned value (I guess “such” is ambiguous in that way). But if it’s that austere sense he intends, then his argument heads even further in the straw person direction, since it’s without question that (almost?) no one denies what he says “some writers” deny, namely that the highest good is highest (!) or that the good will’s goodness is unconditioned (!). Those would be strange claims indeed, but more to the point, everyone (I think!) agrees that the good will (say) has supreme moral goodness.
Parfit’s book is so full of counter-examples against many things that one influence it has had, for good or bad, on me has been that it’s made me look for one’s. So, what you think about this one. He defines good ends as:
“the Actualist View: Possible acts and other events are good as ends
when they have intrinsic properties or features which give us
reasons to want them to be actual, or to happen, and to make
them actual if we can. …”
I think there’s something odd about this. It seems too loose at least when ‘possible’ is not further qualified. So, there is a possible event of sudden immediate curing of all disease in the world. Some weird cosmic radiation perhaps could bring it about or mere chance even though the odds are minimal. As an end it satisfies all Parfit’s conditions for a good end: global healthiness has features that give us reasons to want it to be the case and there would be reasons for us to bring it about if we could. Given his criteria then it is a good end. But it sounds like a mad end to me. If someone said she had that end, I would hesitate to call it good. I wonder if good ends would need some sort of contstraint in relation to what the agents can bring about or at least can with some justification believe they can bring about.
Arto,
Let me start by explaining that the view that I have in mind is the “purely teleological conception of reasons, according to which, since any rational action must aim at some result, reasons that bear on whether to perform an action must appeal to the desirability [i.e., value] or undesirability [i.e., disvalue] of having the result occur, taking into account also the intrinsic value of the action itself” (Scanlon 1998, 84). Scanlon rejects this view. Take his tennis example, for instance. He thinks that, given that I’ve decided to play to win, the fact that my making a killer shot would have some bad result (namely, causing my opponent embarrassment) doesn’t provide me with the least reason not to make the shot. Scanlon thinks that some reasons bracket off other considerations such that these other value considerations don’t count as reasons. These reasons are not, Scanlon claims, teleological or value-based at all.
Now you worry that if we accept the buck-passing view (BPV), then to say that some act is good as an end cannot explain (teleologically) why the agent should perform that act, for, on BPV, the fact that some act is good is just the fact that it has the purely formal, higher-order property of having other properties that provide reasons to perform that act. But we can accept BPV and think that different types of things call for different responses. Maybe, on BPV1, to say that some non-event, N, (e.g., a person) is good is to say that N has the purely formal, higher-order property of having other properties that provide us with reasons to act in respectful ways toward N. But, maybe, on BPV2, to say that some event, E, (e.g., an act) is good is to say that E has the purely formal, higher-order property of having other properties that provide us with reasons to desire that E occurs. In this case, the fact that some act X is good is not, as you claim, just the same fact that there are reasons to do X. So, it seems to me that we can say that the fact that some act is a good as an end can explain why one has reason to perform it. The fact that it is good as an end means that one has reason to desire that it occurs. On this view, reasons for action are explained in terms of reasons to desire. We have reasons to act so as to produce a certain possible outcome (construe outcomes broadly so as to include the acts themselves) just when we have reasons to desire that that possible outcome be actual.
Jussi,
You write,
When Parfit talks about some event being good as an end, he is not committed to it being a good end to have, in sense of being one’s goal or aim, where a goal or aim is something one intends to put effort into. Whether something is a good goal to have does depend on whether it is a feasible goal. Parfit, though, is talking about what is good as an end in contrast with what is good instrumentally or good as a means. So if you want, you could substitute “Possible acts and other events have final/intrinsic goodness” for Parfit’s “Possible acts and other events are good as ends.”
Doug,
thanks. I know that’s something he has in mind but the formulation doesn’t quite get there. The substitution too is problematic. If we did that then we would face the Dancy/Väyrynen objection. Having intrinsic value would require providing reasons for wanting the events to happen or to make them happen. But, there seems to be plenty of cases where something has intrinsic value without in fact in the circumstances providing reasons to anyone as there is no one around to do anything. The talk of ends solves this problem as it is already assumed that someone is in a position to do something about it. But that should somehow be spelt out in the formulation.
Josh: I’m glad you agree that the good will has supreme value. I thought I’d heard you recently muttering that its value was only “so-so.”
Seriously, though, you emphasize a point I’ve been worrying about more and more as I make my way through the book, viz., a tendency Parfit has to devastate views held by “some writers,” who nevertheless remain uncited. This is especially odd, given Parfit’s obvious lack of reservations about launching powerhouse objections as well to people he names with regularity, including Wood, Scanlon, Korsgaard, and Herman. Perhaps these “some writers” are just lesser lights he doesn’t want to embarrass? But it’s difficult to assess the objections he raises if we’re unable to determine what the original position actually was. I for one hope he includes citations to these “some writers” in the final version of the manuscript.
Now as to this specific passage we’ve been discussing, Josh, it’s not just “some writers” who hold the view that it’s only existent ends that can have supreme value; it is also the case that “Kant sometimes makes such remarks.” But then in light of that fact Parfit’s rebuttal has more bite.
Jussi,
Could you say in greater detail what the Dancy/Väyrynen objection is. And references would be great too. Thanks, Doug
It’s the objection Dancy makes against the buck-passing view in his:
‘Should We Pass the Buck?’ in Anthony O’Head (ed.): Philosophy, the True, the Good, and the Beatiful (Cambridge Univ. Press 2000).
I have a reply to it in my:
‘Reasons and Value – A Defence of the Buck-Passing Account’ Ethical Theory and Moral Practice 2004 paper.
Pekka brings this up in his Oxford Studies in Metaethics paper but has an unpublished paper where he takes the objection further and tries to argue that my reply doesn’t work.
I guess one good way of putting the objection is to take any object we take to be intrinsically valuable, say a great work of art or an astonishing scenery and place it in a possible world in which there are no beings with judgment-sensitive attitudes and thus the objects are not providing any reasons that are relations to such attitudes. If to have value then is to provide reasons, then, by definition, those things cannot have any value whilst in those worlds. But, they were assumed to have intrinsic value. Thus, Dancy claims that value and reasons have different logical structure, the latter are relational in the way value is not. Therefore he thinks, or maybe thought, that value is unaccountable for in terms of reasons.
I argued that there is a way for at least the buck-passing view suitably formulated in terms of having features that have dispositions for providing reasons to avoid the problem.
Doug,
good, sorry for the delay. I can see the difference between an event having features that give us reasons to desire that it occur, and an event having features that give us reasons to make it occur.
The features are the same, I suppose, but they play two roles, or give reasons for two kinds of responses: desires and actions.
If I get it right the suggestion is that the fact that the features play one role could explain the fact that they play the other role? If so, then good, they are indeed not the same fact.
A small point to the suggestion that non-events might be “good” when they have features which give reasons to respect – Parfit might want to say that in those cases they “have value”, and the value in question is not a form of goodness,
(as Hitler and Stalin are not good but to be respected).
Just a follow-up. Earlier in this thread, some of us (particularly Doug and me) seemed to converge on the idea that Parfit’s overall question is whether Kant’s principle(s) give us a useful decision procedure, rather than a moral criterion. In Chapter 8, Parfit says something that suggests this is a mis-reading of him. He says (p. 152): “our main aim…is to find out whether a Kantian moral theory can help us to decide which acts are wrong, and help to explain why these acts are wrong.” In other words, Parfit is seeking both a decision procedure and a criterion of rightness.
This might re-open some questions that were closed above. It also might call into question the more general way we’ve been framing Parfit’s claims and understanding his overall aims.
I noticed this too, Josh, and was going to raise the issue again when the ch. 8 discussion opens up on Thursday.
Dave and Josh,
Josh writes:
I don’t see how the fact that Parfit is seeking both a decision procedure and a criterion of rightness shows that we were (or, at least, that I was) misreading him in thinking either (1) that finding a decision procedure is crucial to finding out whether we ever have most reason to act wrongly or (2) that the basic argument of chapter 6 is not that the Formula of Humanity is not the best (or correct) account of what makes acts right or wrong, but that the claim that it is wrong to treat people in ways that are incompatible with respect for them ‘would seldom help us to decide, in difficult cases, whether some act would be wrong’.
Doug,
I guess I was thinking about your (2) (maybe Dave was also?). The chapter 8 comment suggests, to me anyway, that both are the “basic” concerns. Each time a new principle is considered, it’s to see whether (a) it explains rightness, i.e., it provides the moral criterion, and (b) it provides a useful decision procedure to allow us to ultimately know whether we ever have most reason to act wrongly.
This is very late in the game, but, Arto, thanks for your (much!) eariler reply on the Kant-ignoring-our-reason-to-be-happy issue. Somehow I just missed it back when it was originally posted. In any event, I think that’s a plausible reading of Parfit’s point.