The Tribunal of Experience

A Question about Moral Disagreements

2004-06-30T20:21:00.000-04:00

Often, when reading discussions of moral disagreements and what we should think about them, I come across a claim like the following: Even if it turns out that quite a few moral disagreements result from disagreements about the non-moral facts, it's clear that some moral disagreements would remain even if people agreed about the non-moral facts.

Now, I assume that, though this is usually how the point is put, something more is being claimed here. It's not just that people would continue to disagree about moral issues while agreeing about all non-moral ones, but that they would continue disagree in these situations even in cases where the parties aren't making an errors in their reasoning. For these moral disagreements wouldn't be so troubling if they resulted from failing to fully understand the facts one know, to grasp their logical implications, etc. So we're also supposed to be imagining cases in which people make no errors in reasoning and fully appreciate the facts they know and so forth. And I suppose we should also rule out things like bias and the like here.

Furthermore, it also seems that the continued existence of such disagreement would be an interesting fact only if we assume that people also have some (or all?) of the information that would be relevant in coming to conclusions about the moral issues they disagree about. Even if two people have all the same beliefs about the non-moral facts, it wouldn't be too surprising that they reached different moral conclusions if their beliefs weren't the ones that were relevant to discovering the moral facts they disagreed about. So maybe we need to make the claim one about the continued disagreement between people who have all the non-moral facts or at least all the relevant non-moral facts. Let's restrict it to people who have all the relevant non-moral facts, as this is somewhat easier to conceive--and maybe we also have real cases of this. (I'm not sure we have any idea whether people would continue to disagree if they had absolutely all non-moral facts--and I'm pretty sure that no one has all those facts.)

I'm probably forgetting some additional qualifications here, but let's get to the point.

All right, so we're supposed to imagine people who agree about all the relevant non-moral facts, who don't make any errors in reasoning, who fully appreciate the relevant facts, and who aren't biased. Will such people still disagree about moral issues? It's supposed to be obvious that, at least in some cases, they will; but I've never found this all that obvious. I'm not sure that it's false--indeed, it has some prima facie plausibility--but it's far from obvious to me.

And I'm wondering what other people think about this. Is it obvious to you? And if so, why?

I can think of a couple arguments for the conclusion that such people would still disagree. I'm not sure what to think of either of them.

The first argument is pretty simple. It claims that we have some actual cases where people have all the relevant non-moral evidence and yet disagree. So this isn't something we have to imagine; it actually exists. Surely, that would be a good argument. But are there such cases? A case people often appeal to is one having to do with abortion. The idea is supposed to be that there are cases in which people with different views about the permissibility of a particular abortion know all there is to know about the development of the fetus and so forth. I don't know what to think about this sort of argument. But it's not obviously compelling. There are a couple of problems here. First, it's not really clear that cases like this are cases in which we have agreement about all the relevant facts. For disputes about abortion often turn on disputes about metaphysical and religious issues, and so it's not clear that there are all that many disagreements in which people agree about all the relevant facts. Now, of course, not all the moral disagreement about abortion reduces to this. There are, for instance, cases of atheists who disagree about abortion, and perhaps there are even such cases where the people agree about all the other relevant facts. Maybe. It's very hard to tell since it's hard to tell just what all the relevant facts are, and this brings us to the second problem here. How are we really supposed to determine whether we've got people who agree about all the relevant facts? This would seem to require appealing to some contentious issues in normative ethics, and so it's not altogether clear that we've actually got the sort of cases we're looking for. OK, enough of that.

The second argument is going to have to be more complex. Rather than pointing to cases in which people have all the relevant non-moral evidence, it appeals to more everyday cases of moral disagreement. The idea here is to argue that ordinary moral argument doesn't proceed in a way that suggests that agreement about all the non-moral facts will lead to agreement about the moral facts. So the main premise here is that moral positions are not responsive to additional non-moral facts in the way we would expect them to be if full non-moral information would lead to convergence of moral opinions. Is this true? I leave it up to the reader to decide.

I'm sure there are other arguments that some moral disagreement will persist even if people possess all the relevant non-moral information. But I'm not sure what they are. Can anybody give me some, or tell me how I've underestimated the plausibility of these other arguments?

Amazon List

2004-06-29T00:19:00.000-04:00

I finished up that amazon list I was talking about a while back. If anyone is interested in looking at it--and you know you've been waiting with bated breath--here's the link.

Thanks to Joe for a couple good suggestions for the list. I have, of course, included both the works he suggested.

If anyone notices any glaring omissions or mind-numbingly idiotic mistakes in my brief comments about the books (and those are always a possibility), please let me know.

Oh, and I'm thinking of writing up a list of anthologies with important papers in meta-ethics. I've come up with around twenty so far, and I could use some help. Naturally, then, I'd appreciate any suggestions anyone could give me.

Slow Posting

2004-06-29T00:07:00.000-04:00

Posting, I'm afraid, is probably going to be pretty slow here over the next couple of weeks. I can't keep up the herculean (for me) pace that I set for myself over those first four or five weeks. So I hereby apologize to (both of) my dedicated readers for not churning out third-rate philosophy at a pace of a couple thousand words a day over the next couple of weeks.

Why the slackening of my pace? Well, I'm trying to write up my dissertation prospectus. It looks like it's going to be on Mackie, if anyone out there was wondering. And if, for whatever reason, you're interested in looking at it, let me know and I'll send it your way here. (I expect to have something worth looking at in a week or so. I don't really expect anyone else to be interested, however.)

So this probably means that there's going to be a lot of discussion of Mackie on this blog. It's likely that whatever posts end up appearing on this blog in the near future will either be cribbed directly from my prospectus or will be questions that occur to me when writing it and that can be answered by someone smarter than me who happens to come across this blog.

Queerness and Reasons for Action

2004-06-25T18:54:00.000-04:00

In my previous post I tried to offer up a version of Mackie's argument from queerness that didn't depend on a lot of metaphysical assumptions. That argument depended on his argument that the existence of objective values would insure the truth of a form of motivational internalism. I'll now offer a second, different argument. This one focuses on what Mackie thinks would be the case with respect to moral reasons if there were objective moral values. It's more or less the same argument.

As I said above, the putative problem with objective moral values to which the argument from queerness appeals is closely connected to their prescriptivity. And it's clear from the argument of the first chapter of Ethics: Inventing Right and Wrong that Mackie thinks that prescriptivity involves moral facts providing people with categorical reasons for action. The argument I'm going to outline here is one that suggests there simply can't be categorical reasons for action, and so there can't be objective moral values. So this argument doesn't change the source of the putative queerness of objective moral values; it still resides in some aspect of their normativity. Furthermore, this is a problem for objective moral values that we can understand without relying on some form of metaphysical naturalism. Thus we might be able to understand the putative queerness of objective moral values without appealing to broad metaphysical theses.

The Argument

The argument here is quite simple. It's basic outline is this:

(1) If there were objective moral values, people would have categorical reasons for action.
(2) People do not have categorical reasons for action.
(3) There are no objective moral values. (from 1 and 2)

What are the details here? The truth of the premise (1) is something that Mackie thinks follows from the correct conceptual analysis of ordinary moral language and thought. Part of the normativity of objective moral values is supposed to be summed up in the thesis that if there were objective moral values, they would provide people with categorical reasons for action. The objective moral values would be necessarily reasons-providing, and the reasons they provide would be independent of people's desires.

The defense of premise (2) is where we have to supplement Mackie's explicit statements with some assumptions about his views. The assumption we need is that he thinks the only naturalistically respectable theory of reasons is an instrumentalist one. So he thinks that all of a person's reasons depend on her desires. Ignoring various complications, let's just assume that he thinks a person has a reason to do x just in case doing x would lead to the satisfaction of one of her desires. If this is true, there can't be any categorical reasons for actions.

So, in short, the argument is that the only plausible account of the nature of reasons for action rules out the possibility of people having the sorts of reasons for action that they would have if there were objective moral values. Hence we have good reason to think that there aren't objective moral values.

I don't have much else to say about this. What do you think?

A Possible Strand in Mackie's Argument from Queerness

2004-06-23T13:39:00.000-04:00

All this recent discussion of Mackie has brought me back to the main question I have about his work: What, exactly, is going on in the argument from queerness? It's pretty clear that it has something to do with the fact that objective moral values wouldn't fit very well into a naturalistic worldview, and it's also pretty clear that Mackie thinks this results from the fact that objective moral values, if they existed, would possess what he calls 'intrinsic and categorical prescriptivity.' But it's exceptionally difficult to determine just why Mackie thinks the intrinsic and categorical prescriptivity of objective moral values is supposed to be problematic.*

I've recently been reading Jean Hampton's paper "Naturalism and Moral Reasons",** and she makes this exceptionally clear there. She argues that Mackie must be assuming some form of substantive naturalism in his argument, and he points out that he never makes it clear just what his conception of the natural is. His argument seems to be: objective moral values, if they existed, wouldn't be natural; only natural things are ontologically respectable; objective moral values aren't ontologically respectable. Why wouldn't these things be natural? Because they would have to possess intrinsic and categorical prescriptivity. What Hampton does well is point out that, so far as one can tell by reading Mackie, this rests on some dubious intuitions about naturalness. For Mackie never spells out what criteria a putative entity must meet to be natural, and so it's not clear why things with intrinsic and categorical prescriptivity couldn't be natural. In other words, until we have a plausible account of what makes something natural, we can't really evaluate the plausibility of Mackie's argument.

I don't want to go into the details of this criticism of Mackie. I think there's something to it, and it gets at something that has long troubled me about the argument of that chapter. But rather than go into the details about the nature and plausibility of substantive naturalism, I want to consider whether we can find an interpretation of parts of the argument from queerness that don't appear to assume any broad metaphysical theses of this sort. I think that maybe we can. In an attempt to show how we might do this, I'm going to appeal to the stuff about Mackie's internalism that I was discussing in a previous post to draw out what might be one strand in the argument from queerness.

An Argument against Objective Moral Values

As I said above, the supposed problem with objective moral values is closely connected to their prescriptivity. And it seems clear that Mackie thinks that prescriptivity of objective moral values has something to do with the truth of a form of internalism. So, if we understand the problem as having to do with some form of internalism (as I'll be suggesting we can do), we won't be changing the source of the putative queerness of objective moral values. Now, we might be able to find a problem with internalism without relying on some form of substantive naturalism. Thus we might be able to understand the putative queerness of objective moral values without appealing to broad metaphysical theses.

So that's the strategy. What are the details here?

Here's the basic argument. Following my usage in the previous post, call the following thesis HI2:

Necessarily, if you correctly judge that is objectively wrong for you to x and your judgment is based on detecting or apprehending the facts that make it wrong for you to x, then you are motivated not to x.

The first premise is that HI2 would be true if there were objective moral values. So, if there were objective moral values, correct moral judgments based on detection or apprehension of the moral facts would be necessarily connected to motivation. For objective moral values are supposed to be such that detection or apprehension of them would be necessarily tied to motivation. The second premise is that we find any necessary connection between our actual moral judgments and motivation to act in certain ways. So far as we know, no moral judgment is necessarily tied to motivation in this way.

That's the basic argument. It's really quite simple. Basically, the two premises are that a form of motivational internalism would be true if there were objective moral values, and that the relevant form of motivational internalism is false. The first premise--namely that the existence of objective moral values would make HI2 true--is supposed to be based on Mackie's analysis of ordinary moral language and thought. That analysis is supposed to reveal that the truth of HI2 is part of the prescriptivity of objective moral values. The second premise, I suppose, should be based on empirical facts that suggest that it is never necessarily the case that a person who makes a moral judgment is motivated to act in a certain way.

Now, clearly, this argument doesn't lead directly to the conclusion that there are no objective moral values. If you accept the two premises, two different conclusions seem possible. One conclusion, of course, is that objective moral values don't exist. So the reason that we don't find any necessary tie between moral judgment and motivation is that there aren't any objective moral values to know about. This is Mackie's route.

The other conclusion is that we never make correct judgments about moral values on the basis of accurately detecting or apprehending them. So we can accept that moral values exist, but deny that we have any way to discover what they're like. Or we could claim that they exist we have a way to detect them, but that we never manage to come to correct moral judgments on the basis of detecting them. I don't know what to make of this latter response, and I'm not sure what Mackie would say about it. It's clear that it's something he needs to deal with, though. (And a similar response is possible when we're talking about the epistemological argument from queerness, as the mere fact that we'd need a faculty we lack to know about objective moral values doesn't show that they don't exist. Perhaps they exist, and we simply can't know anything about them.)

But, whichever of these two conclusions we draw, it seems to result in a pretty significant problem for the defender of objective moral values. We should conclude either that they don't exist or that they may exist but we can't know anything about them.

The Plausibility of This Argument

What can we say about the premises of this argument? The first premise is that if there were objective moral values, there would be a necessary connection between correct moral judgments and motivation; the second premise is that there is no such connection. Are these plausible? I'm not going to say anything about the first premise here, but I want to mention three reasons why Mackie might accept the second.***

So we want reasons for thinking that there is no necessary connection between correct moral judgments and motivation. The first way to argue for this would be to argue that we have empirical evidence of actual counterexamples. Perhaps the empirical evidence of the ties between moral judgments and actions doesn't suggest any such necessary tie between making correct moral judgments and being motivated to act in certain ways. We just don't see the close ties between moral judgments and actions that we would expect to find if (HI2) were true. When we look at the judgments we take to be most obviously correct, we don't find any close connection to action--or, at least, we don't find any connection close enough to suggest (HI2) is true.

Now, a second way to argue for this is to appeal to possible, though perhaps non-actual, counterexamples. Given whatever sense of possibility if supposed to be relevant to (HI), we can always describe possible counterexamples; we can always describe amoralists, people who make correct judgments but aren't motivated in this way.

The third way to argue that there is no necessary connection between correct moral judgments and motivation is to appeal to the Humean theory of motivation. Mackie is a cognitivist, and so he thinks moral judgments express beliefs (about objective moral values). If he accepts the Humean theory of motivation, then he thinks that beliefs alone are always motivationally inert. You need a conative state in addition to a belief in order to account for motivation, and beliefs and conative states are 'distinct existences'. So, given any particular belief, it's always possible to simply not care about the content of the belief and so not be motivated in virtue of having that belief. Hence, in general, there is no necessary connection between any belief and motivation to act in a certain way. Hence there is no necessary connection between (correct) moral beliefs and motivation to act in a certain way.

*. Now, some of Mackie's worries here are pretty clear. For instance, he thinks we'd need some special faculty of intuition to detect or apprehend objective moral values, and yet we don't seem to have any such faculty. That's clear enough--though it's not altogether clear why he thinks we'd need such a faculty to detect them.
**. This paper appeared in a supplementary volume #21 of the Canadian Journal of Philosophy.
***. In what follows, I'm going to ignore the complication that HI2 explicitly talks about correct moral judgments based on detection or apprehension of objective moral values. Instead, I'll talk about the connection between motivation and correct moral judgments in general. If it turns out that we can't find a necessary connection between motivation and any of our moral judgments, then it's clear that we have enough to establish the second premise.

Lee on the Demandingness of Consequentialism, Part II

2004-06-22T11:29:00.000-04:00

A couple posts ago I tried to sketch how someone might formulate the objection that consequentialism is too demanding and how they might do so in a way that gives the objection some prima facie plausibility. Now I'll say a bit about some ways in which a consequentialist might respond to this objection. The responses here are pretty schematic, but I'm really just trying to lay out the options. (You should also check out Christian Lee's detailed comments on the first post on this issue, as he discusses some of the same issues there.)

And away we go, with some possible consequentialist responses to this supposed problem. I'll talk about responses of two types. I'll begin with responses that go after this sort of reflective equilibrium argument, and then I'll briefly discuss a couple more arguments.

Against the Reflective Equilibrium Argument

The first move here is to admit that a reflective equilibrium argument will show that morality isn't as demanding as consequentialism implies, but to deny it's a problem. As I said, it's not likely that a reflective equilibrium argument of this sort is going to persuade someone who didn't think much of the consequentialism-is-too-demanding objection in the first place. The underlying reason seems to be that a person who doesn't buy this argument is likely to hold views about the issue of how revisionary a plausible moral theory can be that conflict with taking reflective equilibrium arguments very seriously. Now, the (and I hereby coin this word) revisionariness of a normative moral theory is a matter of how many of our considered judgments and mid-level principles we're required to give up in accepting the moral theory. Relying on reflective equilibrium seems to be based on a sort of methodological conservatism: we take our considered judgments quite seriously, and we don't abandon them lightly. So relying on arguments like this seems to be based on a conviction that a moral theory shouldn't be radically revisionary. If, on the other hand, you don't want to take these considered judgments or mid-level principles seriously--if you're happy with a revisionary moral theory--this sort of argument isn't likely to appeal to you. And that seems to be the source of the underlying debate here. So this is one way the defender of consequentialism can avoid the demandingness objection: she can refuse to accept the method of argumentation on which it appears to rely.

The second move is one I won't discuss in detail. It is to accept reflective equilibrium argumentation, at least for the sake of argument, but to deny the reflective equilibrium arguments shows that consequentialism is too demanding. Perhaps, if we adequately understand our considered judgments and mid-level principles, we'll find that it's not so clear that they conflict with a radically demanding moral theory. It may be that, once we fully understand them, we'll see that they do in fact imply that morality is very demanding; or it maybe that, once we fully understand them, we'll see that there are elements that suggest that morality is demanding and elements that suggest it isn't. I won't say much about the plausibility of this response, as doing so would take me too far afield. But I'll just point out that the arguments that morality is very demanding of which I'm aware--and I don't know much about this literature--involve some argument of this sort. In his "Famine, Affluence, and Morality" Peter Singer points to some fairly plausible mid-level principles, and he argues that they imply that we have very demanding moral duties. In Living High and Letting Die Peter Unger focuses more on our considered judgments about certain imaginary cases, and he argues that our considered judgments about such cases provide evidence for the view that we have very demanding moral duties in actual cases.

Against the Demandingness of Consequentialism

Both the sorts of arguments discussed above involve admitting that consequentialism, as defined above, does lead to the conclusion that our moral obligations are very demanding. Of course, one could also go the other way and deny that consequentialism leads to this conclusion. I'll now consider that possibility. One argument focuses in on (1) in the definition of consequentialism in the previous post; the other goes after (2b). I'll begin with the latter.

The way to go here is focus in on the relevant notion of goodness. What the consequentialist has to do is come up with a conception of goodness that gets around these worries about demandingness. I'm not going to claim that I know how to do this, since I don't. But I'll just sketch an example to show how it might work in a particular case that might seem to pose the demandingness problem.

Say you've got $1000, and you're trying to decide whether you should put in your child's college fund or send it to Oxfam. And suppose you take a fairly crude hedonistic conception of the good: what is good is the feeling of pleasure. With this conception of the good, it's pretty easy to figure out what you should do. You just need to ask yourself which course of action will lead to the greatest amount of happiness in the long run. Well, if this is the relevant issue, it seems clear that you could do more good by giving our money to Oxfam. For, though contributing a nice chunk of cash to your child's college fund might make her much happier (and it might make you happier as well), sending the same amount of money to Oxfam would make either a single person much happier (it might save her life and provide her with resources to live on for quite some time) or it would make a group of people happier. So more total pleasure is produced by giving to Oxfam. And it should be clear how this sort of argument can quickly lead to a very demanding conception of morality. For similar arguments are going to work for pretty much any amount of money you're planning on spending on yourself or the people close to you. Quickly, then, we seem to be led to the conclusion that you're obligated to give pretty much all your money away to Oxfam.

Can we avoid this? Sure, we only need to come up with a conception of the good that doesn't allow for this. So we need a conception of the good that tell us it's better to spend $1000 on your child than it is to send the money to Oxfam. Suppose we say that the fact that the state of affairs of your putting the money in the college fund is one involving a special relationship between people, and that this gives it more intrinsic value than the state of affairs in which you give the money to Oxfam. And if we can do something similar in other cases, we may not have to worry about demandingness.

A couple things about this sort of response. First, I don't even know that it can be done. I don't know that we can come up with a conception of the good that will avoid all problems of this sort. Maybe it can be done, and maybe not. Second, there's the obvious problem that the conception of the good suggested above can seem pretty counterintuitive. The money sent to Oxfam could lead to a state of affairs that saves the life of a young child, or that feeds a hungry child for several weeks or months. Yet, according to this conception of the good, it's really better from an impersonal point of view to send your child to college than it would be to send the money to save the life or feed a child you don't know. I'm not sure how plausible a conception of impersonal good this is, and it seems to undermine what's supposed to make consequentialism plausible in the first place. What seems to make consequentialism a plausible moral theory is that it allows us to understand the nature and purpose of morality by appealing to some intuitively plausible conception of the good. If, for instance, we're considering utilitarianism, we're able to understand the nature of happiness, why it's a good thing, and how morality can be understood as a system for promoting happiness. However, if we're going to develop the sort of conception of the good that is needed to avoid this objection, it's not clear that we're going to be able to appreciate its goodness. (Furthermore, you might wonder whether a conception of the good that is tailored to allow consequentialism to be consistent with our ordinary moral thought is really doesn't turn into a moral conception of the good.)

Another possible way to argue that consequentialism doesn't require too much of us could be based on a focusing in on (1) and what a person can do. The move is pretty simple: we try to argue that (many of) these extremely demanding obligations get thrown out as things we can't do. So, for example, we have to argue that giving your money to Oxfam rather than putting it in your child's college fund is something you can't do.

But, of course, there is a sense of 'can' in which you can do this. What sense is that? Well, it would seem to be that, given your actual psychology, it is in your power to act this way. It may be difficult to do so--it may require that you give up things you value most--but it is still an option for you. And this is why it seems that your obligations are demanding--they are can require to give up those things you find most valuable in order to do the right thing.

Now, the way to avoid this objection would be to understand the sense of 'can' in (1) in such a way that it doesn't turn out that you can often ignore those things that you find most valuable. If we could do this, then it couldn't turn out that consequentialism tells us that doing what we ought to do requires a sort of heroic detachment from what we actually care about. Is there any such sense of 'can'? Maybe. It does seem true that there's an ordinary sense of 'can' in which it's true to say that most normal people in normal situations can't give up their most central projects and ignore obligations resulting from their closest relationships in order to do create more good for others. Given their motivational set and how much they care about the things, they can't do this (at least not in the relevant sense of 'can'). And if they come to think their moral obligations require this of then, then you're just going to think, "so much the worse for my being moral" and they're going to act accordingly. It's not just that people would find these obligations too demanding when compared to what they usually think of their moral obligations, but that, in some way, such obligations would be too demanding for them to be able to do it.

Can this be plausibly argued? Maybe, but I'm not sure how to do it. The first and most obvious problem is that there is empirical evidence that some people can do this sort of thing. Now, that in itself isn't a problem for the defending of this sort of response. They could always admit this but argue that most ordinary people can't do this sort of thing, and so the response still works in most cases. But they've got to assume that there's some substantial psychological difference between the people who do give up these things in order to meet such perceived obligations and ordinary people who they're going to claim couldn't do this. Is there such a difference? I don't know. Call in the psychologists.

But, even if we had empirical evidence that there were such psychological differences, there could still be problems with this response. The first, of course, is that it's not clear how much help such a response would be. It might only rule out extremely demanding things--say, giving up your entire life and going off to help the indigent--and not giving the money to Oxfam rather than putting it in your child's college fund. Second, it might seem that taking this route is going to lead to serious problems making sense of moral obligations that we think people actually have. For, even though we think that consequentialism can be too demanding, we still think that morality can be pretty demanding, that it can sometimes require people to give up things they really want and value. But if we're employing some notion of what people can do that closely ties people's abilities to their actual desires, values, aims, etc., we might have trouble making sense of this. Furthermore, a consequentialist who offers this response might have trouble making room for the fact that people with very strange desires, values, aims, etc. can have ordinary moral obligations. We certainly don't want to say that the serial killer isn't under any moral obligation not to kill people just because he really wants to kill them (and so can't, in this sense, not kill them). But if we tie what a person's alternatives are to what they actually want in the way that this response seems to require, it's not clear we can avoid this sort of problem.

(Nearly) Shameless Self-Promotion

2004-06-20T23:40:00.000-04:00

I haven't yet mentioned this here, and so I'll do so now: I've written some reviews and compiled some lists for Amazon. Honestly, most of this stuff isn't very good. But I thought it was worth mentioning, and I think I'll put a link to my stuff on my sidebar.

If you're interested in checking out what I've done, you can look here.

Why am I posting about this? Well, first and most obviously, I'd like to have some people go and look at what I've written. And if you feel that some of my reviews are helpful, then you can give me votes. (I suppose you can also vote if you don't find them very helpful.)

Second, it might be nice to get some constructive criticism about this stuff, and especially about the reviews. My aim in writing the reviews is to get some practice in developing the sorts of skills that I'll need to write academic book reviews in the future. Writing a review also forces me to get straight on the content of the book and to find a clear and brief way to express some of its main ideas. Furthermore, preparing to write a review often provides some impetus to get through the whole of a book. (I can use some help sometimes!)

Now, the format there isn't exactly the format of an academic review. For instance, they only give you approximately 1,000 words to work with; and that usually requires me to do quite a bit of cutting. Consequently, I tend to avoid trying to do much philosophical criticism in the reviews--there's just not enough space for detailed and fair criticism--and instead I focus on letting the reader know what the book is about and giving some of my general impressions about its strengths and weaknesses. So, in some ways, these are quite different from good academic book reviews. But I do find that writing them has been helpful to me, and I think it would be nice to know what people who have some knowledge of these matters think of them. And, of course, it would also be nice to know if I've made some glaring and egregious errors in my descriptions of the views of other philosophers.

Third, you, dear reader, might be able to help me. I'm thinking about compiling another list about contemporary books in meta-ethics that I think people ought to read, and I wouldn't mind getting some suggestions from people. I've already got two lists (this one and this one), and I'm willing to listen to suggestions for the third. Plus, you might suggest something I don't yet know about. (I don't claim to have anything approaching an encyclopedic knowledge of the literature here.) So, if you know of some great but unduly obscure treatises in contemporary meta-ethics, let me know.

Thanks.

Lee on the Demandingness of Consequentialism, Part I

2004-06-20T13:54:00.000-04:00

I'd like to say something about the objection to consequentialism that Christian Lee discusses in this post. I'm going to break up my discussion into two post. The first sets out the problem as I see it; the second briefly discusses some possible consequentialist responses.

The basic objection is that maximizing consequentialism is false since it leads to an account of the nature and extent of our moral obligations on which those obligations are clearly too demanding.

Here's Lee's account of the objection:

Consequentialism: An act A is obligatory for S iff (1) A is an act from amongst S's alternatives that S can perform and (2) If S performed A, then the world would better off than it would have been had S not performed A. In case of ties between alternatives for best act, some act from the tied acts is obligatory.

A consequence of this view of obligations entails that it is possible that someone is obligated to perform a momentous act, e.g. moving to Tanzania to help Aids victims.

But...so the objection goes, that is too demanding! Any moral theory which could require us to put our lives in an upheavel and give up what we want to do with our lives, sacrificing our autonomy to choose our own careers, is false. Dead stop.

I find this objection wanting. To put it crudely, who ever said that morality and its requirements was easy. I would think it is quite the opposite. Morality is very, very hard. This objection presupposes something that just seems wrong to me, that obligations are easy to meet and I can't see why someone should think this.

I'm pretty sympathetic to what he says here. But I have a few things to say about this.

First, a quibble. I think that Lee's being somewhat unsympathetic to this sort of objection when he says that is must rest on an assumption that it's easy to fulfill our obligations. I don't think a defender of this objection needs to say that. She might still be able to say that some obligations can be hard to fulfill. What it seems she needs to say is not that all obligations are easily fulfilled, but that some obligation the consequentialist says we have is too hard to fulfill. In other words, she needs to argue that our actual moral obligations aren't that demanding, though they may well be quite demanding.

Second, a little on Lee's definition of consequentialism. Here it is:

"An act A is obligatory for S iff (1) A is an act from amongst S's alternatives that S can perform and (2) If S performed A, then the world would better off than it would have been had S not performed A."

When I read this, all (2) seems to be saying that A is obligatory if doing A makes the world better than not doing A. But the problem is that there are lots of different ways of not doing A. And A may be such that for some alternatives, doing A would make the world better; for others, doing A wouldn't. So it's not clear how to apply (2) in all cases.

Let me explain what I'm getting at by considering an example. Suppose I have $1000 in disposable income, and I'm trying to decide what to do with it. Well, it seems that one thing I could do is give $50, and only $50, of it to Oxfam. What does consequentialism, defined as it is above, tell me about whether or not I'm obligated to give $50, and only $50, of it to Oxfam? Does this action meet condition (1)? Sure, giving $50, and only $50, is one of the actions that's available to me. Does it meet condition (2)? It seems it's going to depend. If I don't send the $50 to Oxfam, I might just put it in the bank. If that's what I do instead of donating the money, it seems it would have made the world a better place to give it to Oxfam. But if I donated $100 to Oxfam instead of donating $50, then it seems not donating $50, and only $50, would make the world a better place.

So it doesn't look like (1) and (2) together tell us whether it is obligatory for me to donate $50, and only $50, to Oxfam. It depends on what would have happened had I not done so.

To me, then, it seems (2) needs to be something like this:

(2b) If S performed A, then the world would been better than it would have been had S done anything else from the alternatives she could perform.

This, I think, makes clear that we're supposed to compare the results of doing A with the results of all other possible actions open to S. Maybe Lee's original formulation implied this, as it seems clear that he was assuming something of this sort. If it did imply this, though, I must be misreading it.

OK, enough of this preliminary stuff. Is there any way to make this objection to consequentialism seem more plausible?

I suppose the first way to back up the argument that this sort of consequentialism is too demanding is to give a sort of reflective equilibrium argument. We have lots of considered judgments about our particular obligations (e.g. it was wrong of Oswald to kill Kennedy) and lots of mid-level principles (e.g. murder is wrong), and these are evidence for and against moral theories. Some of these considered judgments and mid-level principles may be rejected in the course of developing a moral theory, but we have good reason to be suspicious of any moral theory that requires us to give up lots and lots of them. And, the argument goes, this sort of consequentialism does require us to give up lots and lots of them. It requires us to give up considered judgments and mid-level principles about the stringency of moral obligations: that normal people in normal circumstances don't have moral obligations that require them to give up their most central projects, that normal people in normal circumstances don't have moral obligations that require them to make themselves miserable in order to help others, etc. It also seems to require that we give up considered judgments and mid-level principles about the existence of other obligations: that normal people in normal circumstances have stringent obligations to devote themselves to helping people with whom they have special relationships, that normal people in normal circumstances have stringent obligations not to steal from people or break promises or the like in order to help others, etc. And so on.

Now, admittedly, this is really just a more complicated version of the original objection that our moral obligations simply aren't this demanding. So people who didn't find the original version of the argument very persuasive may be unlikely to find this more elaborate version more persuasive--but at least it gives the defending of the objection more ammunition than a mere assertion that our obligations aren't as stringent as consequentialism says. And this could be very helpful, especially if the defender of this objection can point to some specific considered judgments and mid-level principles that we find very plausible and that seem to conflict with this form of consequentialism. That, certainly, would give her a more to work with than a mere sense that our obligations aren't as demanding as consequentialism tells us they are.

Mackie Question

2004-06-19T13:24:00.000-04:00

I have a question about Mackie's analysis of ordinary moral language and thought in the first chapter of his Ethics: Inventing Right and Wrong. It seems clear that Mackie thinks that, if there were objective moral values, then some version of motivational internalism would be true. But what I'm not sure of is whether he thinks that some version of motivational internalism is in fact true, even though there are no objective moral values.

So far as I can tell, what Mackie explicitly claims about moral motivation is that objective moral values are such that if a person apprehended or detected them, then she would be motivated to act in a certain way. Let's simplify this by focusing on objective wrongness. Mackie, it seems, is committed to the following thesis about moral motivation:

(MM) Necessarily, if you apprehend or detect that it is objectively wrong to x, then you are motivated not to x.

If we suppose that apprehending or detecting that something is objectively wrong involves judging that it is objectively wrong, then (MM) appears to imply what David Brink calls 'hybrid internalism' in the third chapter of his Moral Realism and the Foundations of Ethics. Take hybrid internalism to imply the following thesis about judgments of moral wrongness:

(HI) Necessarily, if you correctly judge that is objectively wrong for you to x, then you are motivated not to x.

And it is important that this thesis is about correct judgments about objective moral wrongness.* This doesn't tell us whether there will or won't be motivation if a person makes these judgments when there are no correct judgments about objective wrongness. So, if this is the sort of motivational judgment internalism that Mackie accepts, he needn't say that people are in fact motivated when they make judgments about objective rightness and wrongness. Accepting (HI) would, of course, commit him to saying that people are sometimes motivated when they make moral judgments if he thought there were cases in which people made true judgments of this sort. But, since he thinks there are no objective moral values to make correct judgments about, accepting (HI) doesn't require him to say that people are ever motivated when they make moral judgments.

So, if (HI) is all Mackie's analysis of ordinary moral language and thought commits him to, he isn't committed to any thesis about actual moral motivation. But is (HI) all he's committed to? This is what I'm not sure of.

Consider the thesis of motivational internalism: that is, the thesis that there is some necessary connection between moral judgments and moral motivation. However we formulate motivational internalism, it seems it's going to imply the following these about judgments of moral wrongness:

(MI) Necessarily, if you judge that it is objectively wrong for you to x, then you are motivated not to x.

Notice that this thesis doesn't say anything about whether the relevant judgment is correct or incorrect. So (MI) tells us that there is a connection between moral judgment and motivation, whether or not there are objective moral facts for people to make judgments about. And since Mackie's analysis of ordinary moral language and thought implies that people do actually make judgments of objective rightness and wrongness, then he is committed to making claims about the actual connection between (at least some) moral judgments and motivation if he accepts (MI). Thus, if Mackie accepts (MI), he is committed to the view that people are actually motivated to act in certain ways when they make judgments about objective moral wrongness. Even those their judgments about these things are never correct, their judgments are still motivationally efficacious.

So does Mackie accept (MI) or only something like (HI)? Does he only make a claim about what moral motivation would be like if there were objective values, or is he claiming something about the connection between motivation and moral judgments in worlds where there are no objective values?

*. I realize I'm ignoring a possible difficulty here. (MM) only talks about judgments one makes based on apprehension or detection, whereas (HI) talks about correct judgments. And it seems this might be a problem. For one can imagine a person arriving at a correct moral judgment in some way other than through apprehending or detecting objective moral values, and perhaps Mackie wouldn't way to say that correct moral judgments arrived at in (all) these other ways would be accompanied by moral motivation. So maybe the version of (HI) that make would accept would have to be somewhat more complex. Perhaps he'd only accept something like this:

(HI2) Necessarily, if you correctly judge that is objectively wrong for you to x and your judgment is based on detecting or apprehending the objective moral facts that make it wrong for you to x, then you are motivated not to x.

I'll ignore this complication here.

Philosophy and the Movies

2004-06-17T16:34:00.000-04:00

There's recently been quite a bit of discussion in the philosophy-related blogosphere about philosophy and the movies. (See here and here.) All of this has been prompted by a list of recent English-language movies with philosophical themes that has been compiled by Jason Brennan.

With one notable exception, though, no one weighing in on this issue seems to know much of anything about movies. It may be that Brennan does know more than his list lets on, though. He claims he wants a list focused on recent movies in English, since he takes it that those films are more likely to keep students interested. That may be true, but that doesn't strike me as a good reason to focus on those films as opposed to older, harder, better ones. I don't think I'd be too worried about challenging them with a film that might not be the sort of thing they'd go out and see on their own. After all, I doubt many students are really all that interested in reading philosophy, and yet we're going to force philosophy on them. So why not force some good movies on them too?

I'm going to focus on films that raise philosophically interesting moral and political issues. But, before I go ahead and give my little list, I'd like to point out that I find little serious overlap between philosophy and the cinema. Film, it seems to me, isn't a medium that readily lends itself to the sort of abstraction that's required in philosophical thought. What a good film can do is raise some important philosophical issues in a vivid manner; it's not so good if you're looking for serious investigation of these issues.

That said, there are some films that I find philosophically interesting. First, I've always thought that the films of Fritz Lang resemble analytic philosophy more than any other films of which I am aware. His best films, I think, are models of how one can both raise and investigate intellectual issues within traditional narrative cinema. He pushes for a sort of abstraction by giving his films stripped-down, simplified stories, and consequently his films often seem like good philosophical thought experiments. Moreover, since Lang almost totally avoids sentimentality and usually keeps a healthy distance from the characters, his films call for dispassionate analysis of those characters and the situations in which they find themselves.

Still, these are largely formal matters--they are ways in which his films read like analytic philosophy--and they're unlikely to be things that beginning students of either philosophy or the cinema pick up on. But some of his films also raise philosophical issues. And the best of his films, M, is also the most philosophically interesting. It raises a large number of interesting philosophical issues, including what we should think of the moral status of the death penalty, how we should think various forms of insanity affect moral and legal responsibility, and, most interestingly, how we should weigh individual rights against the interests of the larger community. In M, the focus is largely on the larger community and the ways in which it can be affected by a single destructive individual. On the other hand, in his later Fury, the focus is on the single individual and the ways in which an individual can be unjustly treated by the larger community. In addition, a few more of his films might be interesting to watch in conjunction with discussion of certain philosophical issues: Beyond a Reasonable Doubt raises questions about the death penalty, and both Rancho Notorious and Die Nibelungen have things to say about revenge.

And believe it or not, there are some explicitly philosophical narrative films out there. During his later, world historian period Roberto Rossellini made films about Pascal, Descartes, Socrates, and Augustine. (These films are not documentaries.) I haven't seen any of these films, as they are very hard to find. But I believe they do concern these figures' ideas, and not just the details of their lives.

Of all religious films, I think there's no question that Dreyer's are those that should be most interesting to philosophers. The two that would be most worth seeing and discussing are Ordet and Day of Wrath. The latter raises a number of interesting issues: the importance of religious tolerance, the nature and existence of evil, the different ways in which believers and non-believers see the world. Ordet has something to say about the first and third of those issues along with several others, viz. different types of religious faith, different conceptions of God and His relation to the world, and the nature and possibility of miracles. (Dreyer's Gertrud would also be interesting to watch in conjunction with discussions of what makes for a good human life. But it's an exceptionally demanding film, and one that few students are likely to appreciate or understand.)

Of course, there are many other interesting films with religious themes. Here are a few (with the relevant themes listed in parentheses): Tarkovsky's Stalker (faith vs. reason/science); Bunuel's Nazarin (problem of evil); Bresson's Diary of a Country Priest (value of faith, conflict between religious and secular worldviews) and Au hasard balthazar (sin, the problem of evil).

If you can track them down, any number of Frederick Wiseman's documentaries would be valuable. Primate and Meat would be worth seeing in a class discussing the moral status of non-human animals. Near Death would be great to watch when discussing euthanasia. Welfare and Public Housing should be of interest in classes discussing social justice and income redistribution.

There are several interesting films that might be worth watching in connection with discussions of feminism. Any number of Mizoguchi's masterpieces would be relevant here; The Life of Oharu, Sisters of the Gion, and Osaka Elegy would be especially useful. Other films that might be interesting here: Akerman's Jeanne dielman, 23 quai du commerce, 1080 bruxelles; Cassavetes's A Woman under the Influence; Romero's The Season of the Witch (aka Jack's Wife); Pabst's Joyless Street; von Sternberg's The Devil Is a Woman; and a number of Godard's films, including Two or Three Things I Know about Her, Numero Deux, and The Married Woman.

Fassbinder directed several films that raise important issues about sexuality. Some of the most useful here are: Fox and His Friends, In a Year of Thirteen Moons, and The Bitter Tears of Petra von Kant. His early, and unduly obscure, The Niklashausen Journey is also an interesting film about revolution and its possible consequences.

OK, I think I've written enough about this. I'll just list the rest of the films that fall into various categories. In parentheses I'll list additional themes in the movie that might be of philosophical interest.

Immoralism
Bresson's Pickpocket (redemption)
Hitchcock's Rope

Explicit Marxism
Dudov's Kuhle Wampe
Eisenstein's Strike and October
Godard's Tout va bien
Ivens's New Earth
Pudovkin's The End of St. Petersburg

Non-Marxist (or Not Explicitly Marxist) Criticism of Capitalism
Bresson's L'argent
Chaplin's Monsieur Verdoux (immoralism, deontology vs. consequentialism)
Clair's A nous la liberte
Pasolini's Porcile
Polonsky's Force of Evil
Renoir's The Crime of Monsieur Lang
Romero's Dawn of the Dead
von Stroheim's Greed (greed as a vice)

Suicide
Bresson's The Devil, Probably and Mouchette

Death Penalty
The fifth episode of Kieslowski's Decalogue or his A Short Film about Killing
Oshima's Death by Hanging

Moral Status of Animals
Franju's La sang des betes

Issues of Race and Ethnicity
Cassavetes's Shadows
Fassbinder's Ali: Fear Eats the Soul and Katzelmacher
Romero's Night of the Living Dead (watch carefully!)

Search Engine Help

2004-06-14T22:37:00.000-04:00

Does anyone know where I can find a good, free search engine for my blog? I've been looking around, and I haven't yet found anything.

If it's possible, I'd prefer to have something very simple. I'm going for a minimalist aesthetic here, as should be obvious by now.

Thanks.

A Second Point about Putnam's Ethics without Ontology

2004-06-14T14:48:00.000-04:00

One more point about Putnam's Ethics without Ontology.

Moral and Non-Moral Value

In a previous post I discussed a concern about Kantian ethics that Matthew Yglesias brought up a long time ago. I didn't buy his particular worry--namely, that Kant can't account for the badness of actions that don't result from free human actions (e.g. a comet's killing a large number of people--but I did think he might be getting at something. What I thought he might be getting at was that Kant can't really account for the importance of non-moral value, of non-moral goodness and badness. In particular, it seems that, for Kant, considerations of moral value always trump considerations of non-moral value, and this it something one might find implausible. Now, I took this interpretation of the problem because I agreed with some of the commenters on Yglesias's original post that Kant was only attempting to provide an account of moral value, and that moral value applies only to the actions of rational beings and the consequences of those actions. The assumption here is that moral issues arise only when we're talking about the actions of rational beings. If we're talking about, say, a comet's killing people, moral issues don't even arise. This may be a bad thing, but it's not a morally bad thing.

But, in his discussion of the diversity of ethical evaluations, Putnam appears to deny that the sphere of moral issues is restricted in this way.

There are many different kinds of ethical judgments. For example, there are ethical judgments which involve praise and blame and ethical judgments which have nothing to do with praise and blame (an example of the latter, which is of historical importance, is the judgment that the Lisbon earthquake of 1755 was a very bad thing; this is also a counterexample to the idea that all ethical judgments have the function of "prescribing" conduct); there are ethical judgments which imply "oughts" and ethical judgments which do not imply "oughts"; and there are a host of ethical judgments which are not happily formulated using the moral philosopher's favorite words, ought, must, mustn't, good, bad, right, wrong, duty, obligation--the idea that all ethical issues can be expressed in this meager vocabulary is a form of philosophical blindness. (pp. 72-3)

So Putnam appears to have a wider conception of the breadth of moral value. The judgment that the Lisbon earthquake was a bad thing is an ethical judgment. Hence ethical badness isn't limited to actions of people; we can make sense of moral value being present in cases where we don't have free human actions. If this is right, then perhaps Yglesias's original objection to Kant goes through; perhaps there is some aspect of moral value that Kant doesn't account for.

But what, exactly, is supposed to make these judgments moral judgments? It's not obvious.* If ethical judgments aren't limited to judgments about free actions of rational beings, what do they concern? Why think that a judgment about the Lisbon earthquake should count as an ethical judgment?

Putnam gives an account of what he means by talking about ethics in the first of the four lectures in the first part of the book. He doesn't offer a clear and precise way of distinguishing ethical from non-ethical considerations, and this appears to be intentional. For he thinks ethical thought involves taking seriously a group of different, though interrelated, concerns that can't be brought under any single system. He main considerations he mentions are considerations concerning our obligations to alleviate the suffering of others; considerations concerning Aristotelian ideas of human flourishing and the relation between being virtuous and leading a good and admirable human life; and considerations concerning universality/ethical equality that lead us to thinking that everyone matters ethically. But none of these considerations, he claims, is the central element of ethical thought. They all have a role to play in ethical thought, and none of them plays the central role. So it looks like Putnam's conception of ethics combines elements of consequentialism, deontology, and virtue theory. None of these is types of consideration is supposed to be more basic than the others; there's no reducing all these concerns to any single type. The attempt to reduce all ethical concerns to any of these types, he thinks, is the result of a sort of "moral blindness."

Now, before I go on to discuss the implications of this for the issue I discussed above, I want to consider just what Putnam is claiming about ethics here. When he provides this account of what he means when he talks about ethics, he notes that there are possible "ethics" that take considerations about courage or "manly prowess" to be the chief considerations. But, as he uses the term 'ethics', considerations of this sort don't fall within ethics. The question I want to ask about this but don't know how to answer is: Is Putnam claiming that these systems flout some logical or conceptual requirement on being ethical theories, or is he claiming that they flout some substantive moral requirement? In other words, is the person defending an ethic of manly prowess not thinking ethically at all, or is he simply thinking poorly about ethics?

This, I think, brings up a question about Putnam's account of ethical considerations. Is this account supposed to be making a metaethical point or an ethical one? The metaethical point would be that, if you fail to appreciate Putnam's favored ethical considerations, then, as a logical or conceptual matter, you're no longer thinking ethically. The ethical point would be that, if you fail to appreciate these considerations, then you're thinking ethically but displaying a moral failing; then you're making a moral mistake rather than a logical or conceptual one.

I don't know how to interpret Putnam here. But let's get back to the main issue.

Clearly, Putnam's broad conception of the ethical has some bearing on the Kant issues about moral vs. non-moral value. Putnam's views suggests that there may be another sort of moral value here--the moral value of situations not involving human agency in any direct way. How does this apply to the case of the Lisbon earthquake? Putnam doesn't really explain, but it seems like we can infer something like the following from what he says about ethics. He could argue that considerations of human suffering are ethical concerns even when free human agency doesn't play a role in leading to the suffering. And, clearly, the Lisbon earthquake resulted in considerable human suffering, even though it wasn't caused by free human actions. This, then, is why we can make an ethical judgment that the Lisbon Earthquake and the suffering it caused were bad things.

So the problem for Kant's moral theory is that it doesn't allow us to make sense of this judgment as a moral judgment. Kant's moral theory

It seems, then, that Putnam has a conception of the ethical that will allow him to argue something like I was arguing against Kant. I framed it as Kant not being about to take moral value seriously enough, while Putnam will claim that Kant isn't actually accounting for all types of moral value. But, however we put it, the main point is the same Kant's moral theory only takes a small class of the genuine concerns under consideration, and so he has a one-sided ethical theory. We want to say that the badness of the Lisbon earthquake matters morally, and it's unclear that Kant can say that.

Could Kant save his theory by arguing that he was only aiming to account for a certain sort of ethical consideration that Putnam mentions? Importantly, Putnam talks about ethics, and, with Kant, we're talking about morality. So maybe we can draw a distinction between the moral and the ethical. The moral is one--but only one--part of the ethical; moral considerations are one--but only one--kind of ethical consideration. Kant wants to account for the moral, not for the ethical in general. So here we take Kant to have a Williams-style contrast between the ethical and the moral in mind. Ethical considerations are those that have a broad relevance to what people ought to do, to both what is moral and what is rational, to what sorts of actions will lead to a good--and not just morally good--human life. And while moral considerations are a particular sort of ethical consideration, they aren't the only sort of ethical consideration. All Kant wants to do is offer an account of moral considerations, and not of the ethical in general. So there's really no conflict between Kantian moral theory and Putnam's pluralism about genuine ethical considerations.

Two responses to this:

(1) Putnam doesn't seem like he'd be very sympathetic to a move of this sort, as it doesn't seem to be consistent with the nature of his pluralism about moral considerations. He seems to think these considerations are connected, and that they need to be backed up by one another in important ways. They aren't individual parts of the ethical that can be pulled out and understood in isolation from one another.** But this response on Kant's behalf appears to assume that moral considerations can be understood in this way.

(2) Even if we suppose that we can make sense of the various types of ethical considerations in isolation from one another, I think Putnam still has room to complain about Kant's moral theory. The problem for Kant is that he's still going to have to downplay the other ethical considerations in favor of those that count as moral considerations. The moral considerations will trump the others, and I think it's clear Putnam doesn't want to allow this. For allowing moral considerations to trump other ethical considerations in all cases seems to suggest that ethics is really a system with a single central component consisting of morality and some additional considerations that supplement moral ones. It's pretty clear, though, that this isn't what Putnam has in mind. He thinks there's a genuine plurality of ethical considerations here, and that none is the ruling set of considerations. So it doesn't seem that he'd the moral always trumps the non-moral. Thus the ethical considerations in one's moral thought that arise from the recognition of some natural evil like the Lisbon earthquake may sometimes override one's narrowly moral considerations. In some cases it may be more important to alleviate suffering of this sort than it is to ensure one possesses a good will or acts on principles that can be universally willed by other rational beings. Kant, it seems, can't allow this.

*. Putnam isn't understanding ethical judgments as covering anything evaluative, as he thinks there are evaluative epistemological and methodological judgments that don't count as ethical.
**. Putnam also claims that he accepts the Hegelian criticism that Kant's ethical theory is formal and empty. I'm not sure what his reasons are for this, as he doesn't explain them. But the interrelation between the various ethical considerations could be part of what he thinks is behind this problem. What Putnam thinks Kant's ethics focuses in on is the universality of ethical considerations and the importance of acting on principle. And he may think that, while these are important elements of ethical considerations, focusing on them doesn't get you to a robust moral theory unless you bring in the other ethical considerations on the ground level.

A Point about Putnam's Ethics without Ontology

2004-06-14T11:40:00.000-04:00

I've been reading the first part of Putnam's Ethics without Ontology, and, while I don't have a whole lot to say about the general issues there, Putnam does say a couple minor things that I wanted to mention and briefly discuss. [I'll discuss one point here and another in a later post. And I may, in an event later post, come back to the general issues raised in the book--but, then again, I may not.]

Abortion and Personhood

In discussing the persistence of moral disagreement as an argument that morality isn't subjective, he considers the case of disagreement about abortion. And he says:

Disagreements about the morality of abortion are usually also disagreements about the question of just when a fetus becomes a person--sometimes this is put in metaphysical terms, as "When does the fetus acquire a soul?" To assume that the irresolvability--if it is irresolvable--of the question of the legitimacy of abortion is simply an example of the "irresolvability" of ethical disputes, and not, for example, an example of the irresolvability of metaphysical disputes is, on the face of it, unwarranted. (pp 75-6)

I think this passage makes a pretty good, if somewhat well-worn, point about moral disagreements--namely that they often depend, at least in part, on what are contentious non-moral issues. Nevertheless, I'm not so sure that the point applies in this particular case. The issue here is what the relevant sort of personhood is when we're concerned with the abortion debate, and it's not clear to me that we're concerned with a metaphysical and non-moral sort of personhood. If the question is, as one of Putnam's formulations suggests, one about the fetus and whether or not it has a soul, then it does seem to be a straightforwardly metaphysical question. But this question seems too narrow to be what he's concerned with.

There seem to be two types of persons (or senses of "person") that might be relevant here: a moral type (or sense) and a non-moral metaphysical type (or sense). The moral type of person is, I think, a being with the moral status that we accord to normal adult human beings.* The non-moral metaphysical type of person is what we're trying to accurately describe in offering a theory of personal identity through time. And what the passage above suggests is that Putnam thinks it's the latter, metaphysical type of personhood that matters when we're talking about the abortion debate. However, I usually take the relevant type of personhood to be the moral type. What we want to know about abortion is whether we ought to accord to the same moral status to fetuses that we accord to ordinary adult human beings, and not whether fetuses are metaphysical persons.

Of course, one might think that there isn't any clear distinction between these two types of personhood. Perhaps metaphysical persons, and only metaphysical persons, are the ones with the status of moral persons. If you offer some psychological continuity account of metaphysical personhood, this strikes me as having some prima facie plausibility. Indeed, the following then stikes me as prima facie plausible: a being has moral personhood if and only if it is a metaphysical person. Why does this seem plausible? Because it seems that the psychological characteristics that are going to go into making someone a metaphysical person seem to be closely connected with the psychological characteristics that go into making someone a being with the moral status of a normal adult human being.

What if you have a physical continuity view of metaphysical persons? Well, whatever physical characteristics it takes to be a person, they don't look to me like they're going to be a necessary condition for moral personhood. (Admittedly, if the physical criteria have to do with the continued existence and activity of physical processes underlying a being's mental life, this might not be the case.) We can imagine physically quite different beings we'd want to say have moral personhood. But having those physical characteristics might be sufficient for moral personhood.

Does any of this seem right to people?

*. Of course, it needn't be limited to normal adult human beings. But normal adult human beings are those beings who most people would agree are persons.

Some More on Hedonism

2004-06-13T15:00:00.000-04:00

What I was after in the previous post was the claim that a hedonist can't recognize qualitative distinctions in pleasures and pains; here I'm concerned with another, not unconnected, objection to hedonism. I'm after the objection that hedonists have to count any pleasure, however disgustingly obtained, as an intrinsic good of some kind. I'm not hedonist, but that's not the reason why. It's not that I think all pleasures are intrinsically good; I don't. It is, rather, that I don't think the hedonist has to claim this.

Hedonism, as I'll be understanding it here, is the thesis that pleasure, and pleasure alone, is intrinsically good, and that pain, and pain alone, is intrinsically bad.

In the previous post I argued that the hedonist can make qualitative distinctions in kinds of pleasures. The move I suggested there was that the hedonist should be careful not to identify intrinsic goodness with pleasantness: that is, not to identity the property of being intrinsically good with the property of being pleasant. And I argued that, if the hedonist is careful not to do this, it seems she can argue that the cause of a pleasure (or pain) partially determines how intrinsically good (or bad) that pleasure (or pain) is. So pleasures (or pains) that feel the same to the people experiencing them can have a different amount of intrinsic goodness (or badness) depending on how they're caused; one can be better (or worse) than the other even if the persons experiencing them couldn't tell the difference simply through introspection. This, I think, gives the hedonist a way to have qualitatively differences of the sort she needs to claim that, for instance, we can distinguish pleasures obtained from playing pushpin from those obtained from reading poetry.

First Problem: Non-Good (and Maybe Bad) Pleasures

But this alone isn't enough to get us around the objection that the hedonist still has to count all pleasures as intrinsic good experiences and all pains as intrinsically bad ones. Does she still have to say that all pleasures are good in some way? If so, this may seem counterintuitive. Consider, for example, a case in which someone gets pleasure from torturing a small child just for the hell of it. What is supposed to be counterintuitive here is that the hedonist has to say there is some intrinsic good here, at least insofar as the torturer gets pleasure from it. It may be very slight since the cause is torture, but there's still some intrinsic good here. But this looks preposterous, since the torturer's pleasure is worthless. Indeed, one might want to say that this is a case in which the pleasure is intrinsically bad; it is a case in which the situation is worse by the presence of the torturer's pleasure, inasmuch as the pleasure derives from someone enjoying torturing another human being. So there's not some, perhaps minor, good in this situation that derives from the torturer's pleasure. That pleasure, as a matter of fact, makes the entire situation worse than it would otherwise be. And since a hedonist can't account for this--for either pleasure not being intrinsically good at all or for its being intrinsically bad--hedonism is false.

Now, admittedly, not everyone has these intuitions. Some people have the intuition that there is in fact some intrinsic goodness to the torturer's pleasure, but that it is overridden by the pain caused to the child and other factors. I don't want to get into this issue here. All I want to do is suppose we have the intuition that the torturer's pleasure isn't intrinsically good at all and see whether the hedonist can make sense of this.

I don't see why the hedonist can't. If, as I have suggested, the hedonist doesn't identify being pleasurable with being good, then I think there's really no problem here. For she could then argue that there are pleasures that are caused in such a way that they lack the property of intrinsic goodness. So it's not just that different causes of pleasures can affect the degree to which they are intrinsically good, but pleasures caused in certain ways need not be good at all. Still, though, pleasures and pains are the only things that have intrinsic value. It's just that, depending on their causes, pleasures can be intrinsically bad and pains can be intrinsically good.* Thus the hedonist can account for the intuition that the pleasure of the torturer isn't any good at all. Moreover, it appears she can account for the intuition that the torturer's pleasure is actually intrinsically bad.

Interlude: A Worry about This Response

But does this take too much away from the intuitive idea behind hedonism, the idea that pleasure is good and that pain is bad? This, it seems, is supposed to give hedonism whatever intuitive plausibility it has: that pleasure definitely feels good and that pain definitely feels bad, and so maybe those feelings determine what is good and what bad. Furthermore, it's clear that people want pleasure, and they want to avoid pain. This doesn't reduce the case for hedonism to psychological hedonism, but it does appeal to the common-sense idea that, in general, people like pleasure and dislike pain. These sorts of considerations are what gives hedonism whatever initial plausibility it has, and it should be clear why they might lead us to want to identify the property of feeling pleasurable with the property of being intrinsically good.

But if we expand our conception of what hedonists can and cannot say about pleasure and pains in the ways I've been suggesting, do we have to give up these sorts of considerations that might give hedonism its plausibility? That is, do we have to give up the case for hedonism?

I don't know what to say about this at the moment.

Second Problem: Overall States of Affairs and Not Pleasures and Pains?

The second worry is that this way of responding to the worries about qualitative distinctions and about bad pleasures is really a way of abandoning the view that pleasures and pains have intrinsic values. If the hedonist accepts this sort of view, she has instead taken to attributing intrinsic value to states of affairs that aren't simply states of affairs of people having pleasures and pains. She's attributing intrinsic value to the entire state of affairs that includes both the person's having a pleasurable or painful experience and the events that led up to that pleasurable or painful experiences. What's intrinsically bad in torturer case, then, isn't the torturer's pleasure, but the overall state of affairs including both the torturer's pleasure and the activity of torturing a small child that leads to that pleasure. And that isn't hedonism, for no longer can the hedonist say that pleasures and pains, and pleasures and pains alone, have intrinsic goodness and badness.

I'm not sure that this is the right way to describe what the view is. I'm not sure that accepting this amounts to attributing intrinsic value to something other than pleasures and pains.

Perhaps an analogy would help draw out the intuitions here. Consider two killings, both of which result from a person being shot. The number of shots is the same in both cases, the person killed is shot in the same place in both cases, the person suffers the same amount of pain in both cases, the killing leads to the same amount of grief among the killed person's family and friends in both cases, etc. However, the one killing was an intentional murder by a contract killer and the other was an accidental shooting.

Now, it strikes me that the former case is worse--much worse, in fact--than the latter. But it seems to me that there are two ways to describe what's worse here, and I'm not sure which is the correct description. One way to describe the difference would be to say that the killing is worse in the former case, since, in that case, the killing was the result of a murder, whereas the first killing was only an accident. So here the difference is in the killings, and it's based on their causes. The other way to describe the difference would be to say that the killings, qua killings, are equally bad, but that the overall situation in which the person is killed in a murder is worse than overall situation in which the person is killed in an accident. So here the difference isn't one of the badness of the killings, but one of the badness of the overall situations of which the killings are a part.

I hope it's clear how this is analogous to the issue concerning hedonism. Saying that the one killing is worse than the other is like saying that the pleasures and pains have different intrinsic value, whereas saying that the one overall situation is worse than the other is like saying that overall states of affairs that include pleasures and pains have different intrinsic values. So, if we knew which was the right way to describe the differences in the killing case, we might know which was the right way to describe the differences in the case of pleasures and pains.

But, unfortunately, I don't know which is the right way to describe things in either case. So let's just suppose we go with the overall situation/overall state of affairs version. Is this a problem for hedonism?

Well, the hedonist then can't say that pleasures and pains, and only pleasures and pains, have intrinsic value. But perhaps the hedonist could alter her view to account for this way of looking at things. It's not the pleasures and pains themselves that are good and bad, but the overall state of affairs including both the subjective mental state of feeling pleasure or pain and the series of events that leads up to it. What has intrinsic value (intrinsic goodness and badness), then, is an overall state of affairs, and not merely the subjective mental state. And the intrinsic value of these states of affairs will depend on both the subjective mental state involved and the other events included in this state of affairs. But neither the mental state alone nor the other events have any intrinsic value on their own. These are Moorean organic wholes; they have intrinsic value, though none of their parts have intrinsic value in isolation from the others.**

Is this still hedonism, though? Why think it wouldn't be? I guess because it gives up the key hedonist thesis that pleasure and pains, and pleasures and pains alone, have intrinsic value. (Indeed, this is how I defined hedonism above.) For now we deny this and instead claim that intrinsic goodness and badness are possessed by these overall states of affairs that include both these mental states and the events leading up to them.

Still, though, intrinsic goodness and badness is restricted to states of affairs in which sentient beings experience pleasure or pain. Pleasure and pain are still requisite for intrinsic value of any sort. And this seems closely related to hedonism. But is it hedonism? I don't know deciding whether to call this hedonism or not is anything more than a terminological matter, but it seems to me that this is close enough to what I originally defined hedonism to be that I'd say this is a form of hedonism.

*. Of course, accepting my move of refusing to identify pleasurableness and goodness doesn't require the hedonist to say this. She can still say that all pleasures are good to some extent, and that all pains are bad to some extent. She isn't required to claim that some pleasures are bad and some pains good.
**. And, of course, this account of intrinsic goodness allows us to avoid a bad pleasure-style objection. For a hedonist of this stripe doesn't have to say that all states of affairs involving the feeling of pleasure are good. Whether a particular state of affairs involving a feeling of pleasure is good will depend not just on the subjective mental states involved, but also on the other parts of the overall state of affairs. And this leaves the hedonist will room to argue that some overall states of affairs involving pleasure have no intrinsic goodness, and that some such states of affairs even have intrinsic badness.

Ewing on Hedonism

2004-06-12T23:37:00.000-04:00

[I hereby warn the reader that what I'm not sure of the truth of anything that I say in this post. Actually, I'm not sure of what I say in any of the posts here. So if you're one of those types who can't stand to read something false (and perhaps obviously false), proceed no further. I doubt anyone who's come across this site is of that type, though. It is a blog likely to appeal to readers of philosophy, after all.]

I've been reading through A. C. Ewing's short book Ethics, and I've come across something there that I want to talk about. In the third chapter Ewing presents several objections to hedonistic utilitarianism; I want to talk about one of them. The objection is that the hedonistic utilitarian can't account for the fact that some pleasures are of greater value than others. For the hedonistic utilitarian, the only thing that is good in itself is pleasure, and all pleasures have to be considered to be of equal (qualitative) value. Like Bentham, the hedonistic utilitarian par excellence, we can discriminate between pleasures based on their intensity, duration, proximity, certainty, etc., but we can't make qualitative distinctions between pleasures.

And Ewing takes Mill to task for trying to deny this and yet remain a utilitarian.

Mill tried indeed to reconcile his utilitarianism with the admission that a lesser pleasure might rationally be preferred to a greater on the ground of the superior quality of the former, but it is generally, and I think rightly, agreed among philosophers that he failed to escape inconsistency. To say that pleasure is the only good and yet admit that a lesser pleasure may be preferable to a greater is like saying that money is the only thing which counts and then adding that money earned by public work is better than the same amount of money earned by business. If pleasure is the only good, the more pleasure always the better.

The analogy there, I admit, is pretty clever. But I don't buy the underlying argument, and it's an argument that a lot of people make. For instance, in his Morality: An Introduction to Ethics, Bernard Williams quips that Mill honorably contradicted himself by drawing qualitative distinctions in his account of human happiness, even though this was clearly inconsistent with his hedonism. Honestly, I don't think there are good reasons to think there is some contradiction here. In the latter part of this post I'll consider some reasons to think that there is one, and I'll reject them. But, first, I want to suggest why it seems to me that there's no obvious reason to think there's a contradiction here and why I think we can appeal to Ewing's own example to show this.

Pleasures and Quality

Suppose we respond to Ewing that he's misunderstood the lesson he should draw from his own analogy. The hedonist who thinks there are qualitative differences between pleasure could say that Ewing's way of thinking is like saying that money is the only valuable thing and then thinking that whoever has the greatest number of bills has the most of most value. That is, it's like thinking that if you have 100 $1 bills and I have ten $100 bills, you have money that is of more value than I the money I have. For, if you don't know how money gets its value, you might confuse having a lesser number of bills with having money that is of less value. And once you take into consideration how money gets its value, you'll realize that sometimes a group of fewer bills is more valuable than a group of more bills.

Similarly, the hedonist could argue, if you don't know how pleasures get their value, you could confuse having more pleasure (i.e. pleasures that are of greater intensity, that are of longer duration, that are more pure, etc.), with having pleasure that is of greater value. And once you take into consideration how pleasures get their value, you'll realize that sometimes a less intense, long-lasting, pure, etc. pleasure can be of more value than a more intense, long-lasting, pure, etc. one.

So the problem with Ewing's objection is that he's got an overly crude conception of the value of pleasures. He's taking their value to be wholly determined by their quantitative characteristics: that is, by their intensity, duration, purity, etc. And while it's the case that the value of pleasures is partially determined by intensity, duration, purity, etc., their value is not wholly so determined. For, the hedonist could argue, the quality of your pleasures also partially determines their value. Again, compare this with the case of the money. The value of the money you have is partially determined by the number of bills you have, but it's not wholly so determined. The type of bills that you have also matters.

The value of a pleasure, then, is a function of its quantity and its quality, and sometimes a pleasure whose quantity is less than that of another can still be of more value due to its greater quality. So, in this way, it can be rational to prefer a lesser pleasure to a greater one. Pleasures and pains are still the only things with intrinsic value, but some types of pleasures are more valuable than others. And it's not clear to me why this is inconsistent with the spirit of hedonism.

Two Putative Problems for Qualitative Distinctions

Why might someone think that drawing qualitative distinctions between the values of pleasures is inconsistent with the spirit of hedonism?

I can see two reasons one might think this. The first is pretty straightforward; it aims to explain away the thought that pleasures can have qualitative differences as based on something that is clearly inconsistent with hedonism. Consider Bentham's famous example of the pleasure one gets from pushpin and the pleasure one gets from poetry. One might claim that the reason why we might mistakenly think that the pleasure of reading poetry is qualitatively superior to the pleasure of playing pushpin is that we think reading poetry is superior to playing pushpin. In other words, the reason that the pleasure we get from appreciating art is of greater value than the pleasure we get from playing pushpin is that appreciating art is an activity of greater intrinsic value than the activity of playing a game of pushpin. The problem, of course, is that this is clearly inconsistent with hedonism, since this attributes intrinsic value to something other than pleasure (viz. the actions of playing pushpin and reading poetry). So the hedonist can't distinguish the intrinsic value of pleasures by relying on the intrinsic value of their causes, since, according the hedonist, those causes don't have intrinsic value.

But I don't see why the hedonist has to make this mistake. The hedonist can claim that pleasure, and pleasure alone, has intrinsic value, while holding that the intrinsic value of particular pleasures can depend on more than their quantitative characteristics. She can attribute the qualitative differences between pleasures to differences in their causes, but she needn't attribute the qualitative differences to differences in the intrinsic value of their causes. So she can argue that two pleasures with identical quantitative qualities, one caused by reading poetry and one caused by playing pushpin, have different intrinsic value. The pleasure from reading poetry is of greater intrinsic value, because it is pleasure caused by reading poetry as opposed to playing pushing. But it's not the case that the activity of reading poetry itself has any intrinsic value.

So the first objection to qualitative distinctions between pleasures makes clear one thing the hedonist can't argue, but it seems it doesn't rule out any possible explanation she can give of the qualitative differences between pleasures. So why think she can't make distinctions of quality? Here's what I take to be the intuitive argument against making such distinctions. Pleasure is a particular subjective mental state, and what is of value in pleasure are the subjective qualities of this mental state. In other words, what is of intrinsic value is how pleasure feels. And if there were any qualitative differences in the values of the pleasures, they would have to be qualitative differences revealed by how the pleasures feel. If there were qualitative differences between pleasures, they would have to be felt differences. But there are no qualitative differences in pleasures that one can discover by introspection. To the person who gets pleasure from playing pushpin, the pleasure may feel exactly the same as the pleasure of the person who gets pleasure from reading poetry. There needn't be any distinction in how the two pleasures feel to the people who have them, and so there can't be any qualitative differences between them.

Does this argument work? I don't think so. The key assumption is that if there were any qualitative differences in the values of the pleasures, they would have to felt differences. And I don't think that the hedonist needs to accept this assumption. But to see why someone might think they do accept this assumption we need to look at how one might interpret Ewing's claim that the hedonist thinks "pleasure is the only good."

I think it only looks like she does have to accept the assumption I mentioned above if you understand the hedonist to be making the following claim in claiming that pleasure is the only good: that the property of being pleasurable is identical to the property of being intrinsically good. Now, presumably, you can discover everything there is to discover about the pleasurableness of some experience through introspection. So, if the intrinsic goodness of an experience were identical to its pleasurableness, you would be able to discover everything there is to discover about the intrinsic goodness of an experience through introspection. Thus, if there were qualitative differences in the intrinsic goodness of experiences, you would be able to discover them through introspection, i.e., through the way the experiences felt.

This, it seems to me, is what's ultimately behind this objection: the assumption that the hedonist identifies intrinsic goodness with pleasure. And I don't think that the hedonist needs to argue for such an identification. So we need to be careful about what we take the hedonist to mean by claiming that pleasure is the only good.

What the hedonist should argue is that in claiming pleasure is the only good she is claiming that being intrinsically good is a property that pleasurable mental states, and pleasurable mental states alone, have, but that she is not claiming that the property of being intrinsically good is identical to the property of being pleasurable. Pleasure, then, is not intrinsic goodness--though all, and only, pleasures have the property of being intrinsically good. How, then, do we get qualitative differences in value? Different pleasurable mental states have the property of being intrinsically valuable in different ways. While all pleasures have the property of being intrinsically good (and so being valuable), some have it more than others (and so are more valuable than others). And the extent to which a particular pleasure has this property depends on both its quantitative characteristics (i.e. its intensity, duration, purity, etc.) and on the causes of the pleasure.

Furthermore, the degree to which a particular pleasure has the property of intrinsic goodness is not something that one obviously should be able to know by introspection. (Perhaps Mill could say that his competent judges are better able to determine which pleasures have this property than non-competent judges are. I don't want to defend Mill's views here, but that might help him out.) And so there is no reason to find it problematic that there need not be a qualitative difference in the experiences of people experiencing pleasures of different levels of intrinsic goodness. There is, therefore, no reason to find it problematic that the person who gets pleasure from playing pushpin may have a feeling that is qualitatively indistinguishable from the feeling of the person who gets pleasure from reading poetry.

Ethics Group Blog?

2004-06-12T16:44:00.000-04:00

I posted about this a while ago, but I thought I should do so again since I now have (a few) readers.

I think it would be interesting to have a graduate student group blog on topics in ethics and cognate areas of philosophy. What areas am I thinking of? I guess normative ethics, applied ethics, meta-ethics, and political philosophy. (I realize meta-ethics may not fit in there all that well.) And if I didn't mention something you think belongs, go ahead and let me know.

When I originally announced that I'd like to do this, I was unaware of any group blogs devoted a single area of philosophy. But, with the arrival of summer and people finally getting some free time, several have popped up and I'm sure even more will appear in the near future. (Who knows, maybe there are already plans for a group ethics blog. I really wouldn't know.) So why not one on ethics and cognate areas?

In any case, if you're a philosophy graduate student and you'd like to contribute to a group blog on ethics and related areas, contact me and we'll see if we can get to work on something. (You can find an email address in my user profile.) I'm willing to listen to pretty much any suggestion about how the blog should be organized, where it should be hosted, what program we should use to write it, what it should be named, and so on.

Cullison on the Doctrine of Double Effect

2004-06-11T16:07:00.000-04:00

In this post over at the Rochester Group Blog, Andrew Cullison raises a problem for the Doctrine of Double Effect (hereafter "DDE"). DDE says something like the following:

One may never intend to do evil, but one may, in some cases, intend to do something that one knows to have evil effects.

Basically, Cullison denies that it's clear the proponent of DDE can have the distinction between intended effects of an action and known but unintended effects of an action. Here's the case he thinks results in a problem for this distinction.

Suppose someone told you to pick up a the first piece of glass you see. You see a pint glass. You run over and pick up the mug. The glass also happened to have Chris's beer in it. Suppose you knew that the glass contained Chris's beer. Did you intentionally pick up Chris's beer?

(This isn't a trick)

The answer is 'yes'.

And if we say yes here, it seems we have to say yes in the following analogous case that Cullison discusses. A bomber pilot intends to blow up a munitions plant. He knows that there are going to be at least ten civilians killed in the bombing. Does he also intend to kill the civilians? If we say you intended to pick up Chris's beer in the previous example, it seems we have to say that the bomber pilot intended to kill the civilians in this example.

But proponents of DDE usually want to say that their principle can help us to understand why we think the pilot may, in some cases, kill the innocent civilians in the bombing raid. With the distinction between intending to do evil and intending to do something with known but unintended evil effects, she could argue that, in the bombing case, the bomber may not intend to kill the civilians (as this is a great evil), but he may drop the bomb(s) with the intention of blowing up the munitions plant while knowing it will have the evil effect of killing ten civilians. For the bomber wouldn't be intending to do evil here; he'd simply be doing something, namely blowing up the munitions plant, that he knew to have these unintended evil effects. The problem Cullison thinks DDE faces is that his other case appears to undermine the needed distinction between intending evil and doing something with known but unintended evil effects.

Let's go back to his example in which you pick up the beer mug. What he asks is whether you intentionally pick up Chris's beer, and he says yes. And I'm tempted to agree, but I think this question is somewhat ill-formed. I think I'm tempted to agree to this question because I'm thinking of it like this: Do you intentionally do something that can be correctly described as picking up Chris's beer? I'm tempted to say yes to this, and, similarly, I'd be tempted to say that the bomber intentionally does something that can be correctly described as killing civilians. But I'm not similarly tempted to say yes to the following question: Do you intend to pick up Chris's beer? Similarly, I'm not sure I'd say that the bomber intended to kill the civilians.

But let's put these intuitions to one side.

Here's what I'm tempted to say in order to draw a distinction between what one intends in one's action and what is a known but unintended effect of one's action. It's an explanatory matter. Intention is closely connected to what explains one's action. What explains your going and picking up the mug is that it's the first piece of glass you see, and not that it contains Chris's beer. What explains the bomber's dropping the bombs is that it will blow up the munitions plant, and not that it will kill ten civilians. This is why it's your intention to pick up the mug and why it's the bomber's intention to blow up the munitions plant, and it's why it's not your intention to pick up Chris's beer and why it's not the bomber's intention to kill ten civilians.

How do we draw the distinction? How do we know what does the explaining and what doesn't? I'd say that part of what we need to do is look at counterfactual situations. If your intention was to x, then you wouldn't do the action if you couldn't x by doing it. This is part of why appealing to your intention to x does explanatory work here. If you would have done the action whether doing the action would be doing x or not, then the fact that your action was an action of x-ing doesn't seem to explain your doing that action.

This, perhaps, is somewhat unclear. Let's formalize what I'm getting at here:

(I) If you intended to do x in doing action a, then in the closest possible worlds in which you know that doing a doesn't lead to x, you don't do a.

(I) gives us a necessary condition for your intending to do x in doing a. I'm not going to claim it's sufficient; I doubt it is.

And, of course, (I) gives us a test for what you intend in doing some action. If the putative intention fails to meet the condition set out in (I), then it's not an intention. Let's apply our test in these cases. If you intended to pick up Chris's beer in picking up the mug, then in the closest possible worlds in which you know that picking up the mug isn't picking up Chris's beer, you don't pick up the mug. This what we get in applying (I) to this case. So what do you do in these possible worlds? Do you still go and pick up the mug? Yep. It's still the first glass you see, and you were still told to pick up the first glass you see. So it can't have been your intention to pick up Chris's beer; otherwise, in the possible worlds where you know that picking up the mug isn't picking up Chris's beer, you don't go and pick up the mug.

Now, go back to the killing of civilians case. What is the intention here? If the bomber pilot intended to kill ten civilians in dropping the bombs, then in the closest possible worlds in which the bomber pilot knows that dropping the bombs doesn't kill the civilians, the bomber pilot doesn't drop the bombs. This is what we get in applying (I). What does the bomber pilot do in these possible worlds? Does he still drop the bombs? Yes, he does. He still wants to blow up the munitions plant, and he still knows that dropping the bombs is the way to do this. So the bomber pilot didn't intend to kill the civilians in the bombing. If he had intended to do so, then he wouldn't have dropped the bombs in counterfactual situations where he knew the bombing wouldn't lead to civilian deaths. But he still drops the bombs in those situations.

And notice that the intention we want to say that the bomber had, namely the intention to blow up the munitions plant, does pass our test. For, in the nearest possible worlds in which the bomber pilots knows that dropping the bombs won't blow up the munitions plant, he doesn't drop the bombs.

So any intention needs to meet the test we've given here, but a known but unintended effect doesn't need to meet it. This allows us to draw the distinction the proponent of DDE needs in the cases above. Picking up Chris's beer and killing civilians both fail to pass the test, and so neither is an intention in the relevant actions. [I may consider some problems with this solution in a later post.]

DCT and Moral Epistemology, Part I: The Problem

2004-06-11T10:49:00.000-04:00

[I've decided to break these large posts up into a few smaller ones. I think this will make things easier to read, and it'll make it look like I'm posting more than I am. So it's all to the good.]

Here's a problem that DCT may run into with respect to moral epistemology. DCT tells us that our moral obligations depend on the commands and prohibitions of God, and this, I think, may result in a problem for moral epistemology.

We think we can know our moral obligations in at least many ordinary cases, and any adequate moral theory should allow us to understand how this be the case. The possible problem is that, if DCT is true, it's not clear that we can know our moral obligations in most cases. (The argument I'll be giving here is largely based on an argument Mark Johnston develops against what he calls the "bare-locus view" of personal identity in his "Human Beings." It's a very interesting paper, and I hereby recommend it.)

The First Argument

The general form of this style of argument is this: We know that p; if theory T were true, we wouldn't know that p; therefore, T isn't true. In this case, the argument is: we know moral proposition p; if DCT were true, we wouldn't know that p; therefore, DCT isn't true.

Let's consider a real-world scenario in which it seems clear that you know some moral proposition. Take a case that gets brought up in the God-could-command-awful-things objection to DCT: the case of someone torturing a small child for fun. Suppose you come across some person doing this, and it's clear from the information you possess that this is what's going on. You see the person torturing the child in an especially painful way, you know this person had a history of abusing children, you have no reason to think that someone's putting on an act to trick you into thinking this is what's happening, etc. It seems that you know that the torturer is doing something wrong if you come to the situation with this information and end up believing that the torturer is doing something wrong. Add whatever details of this sort you feel you need to add to be sure that you know.

So here's our argument: We know that the torturer is doing something wrong; if DCT were true, we wouldn't know that the torturer is doing something wrong; therefore, DCT is false. But why wouldn't you know that this if DCT were true? The simplest argument is that you wouldn't know DCT because you'd have to be able to know about God's commands, and you simply can't know about them.* Why can't we know about God's commands? Some possible reasons: (i) the idea of God and of His commanding something is straightforwardly unintelligible; (ii) we can make sense of this, but we have no way of acquiring evidence about what God has and has not commanded; (iii) we have a way to acquire evidence (e.g. revelation), but the evidence is never sufficient to provide us with adequate warrant for knowledge; (iv) we sometimes have adequate warrant for knowledge here, but not in enough cases to account for a large enough number of the moral claims we think we know (several conflicting claims about what God's commands are here, and none of them is clearly better than the others); etc. If any of these reasons is a good one, it seems to preclude our having moral knowledge in this case if DCT is true. Yet it seems clear we do have such knowledge.

Of course, the plausibility of this particular argument depends on the plausibility of these reasons for thinking we don't know enough about God's commands. But someone might deny that these reasons are goods one. Suppose someone does. Can we formulate a different argument of the same sort, an argument that is immune to this particular response?

The Second, and Better, Argument

I think we may be able to. (And there is where I'm drawing pretty heavily on Johnston's paper.) Let's try an alternative type of argument: We know that p without relying on evidence for q; if theory T were true, we wouldn't be able to know that p without relying on evidence for q; therefore, T isn't true. And our particular argument: We know that the torturer is doing something wrong without relying on evidence that God commands that people not torture small children for fun; if DCT were true, we wouldn't know that the torturer is doing something wrong without relying on evidence that God commands that people not torture small children for fun; therefore, DCT is false.

So the problem here isn't that we can't know about God's commands. For, even if we can know about them, the problem here is that it doesn't seem we need to in order to know about our moral obligations. Yet, if DCT were true, then we'd have to know these sorts of things about God in order to know what our moral obligations are.** Now, we just need to argue that you don't need to know anything of this sort to know that the torturer is doing something wrong. And this strikes me as plausible. Even if you've never given a thought to God and His commands, you can know that what the torturer is doing is wrong. Provided you see what the torturer is doing and know enough about the circumstances, you know this is wrong--regardless of whether you know anything about God and His commands. If DCT were true, this wouldn't be enough to know the torturer is doing something wrong. So DCT is false.

We might make this clearer by appealing to what one might think is a particular necessary condition for knowing something: that one rule out the relevant alternatives. (Here, again, I'm following Johnston's argument pretty closely.) If one thinks the moral status of some action depends on God's commands, then the relevant alternatives one has to rule out are the possibilities that God has commanded something else in this situation. If DCT is true, then, when you come upon the torturer doing his thing, you'd have to rule out the alternative that God willed that the young child be tortured (and thus made this action right) in order to know that the torturer was doing something wrong. This would be a relevant alternative, and it would be one that your evidence would need to rule out.

But it doesn't seem that your evidence needs to rule this out. All the evidence you need is that the torturer is torturing a small child for fun, and that evidence doesn't appear to rule out the alternative that God had commanded the torturer to do so. Hence you know that what the torturer is doing is wrong, and you know this without ruling out that God commanded the torturer to do so. Hence you don't need to rule out this alternative to know that what the torturer is doing is wrong. Hence DCT is false. And, of course, this particular case is supposed to make the general point that you can know the moral facts without ruling out alternatives that you'd have to rule out if DCT were true. Hence DCT gives us an inaccurate picture of moral epistemology. Hence DCT is false.

Motivational Internalism and Tolerance, Part III: Some Final Issues

2004-06-10T17:37:00.000-04:00

In two previous posts (found here and here), I discussed a possible tension between accepting motivational internalism and accepting tolerance of people's moral opinions. I want to return to this issue now and finish up my discussion of it. (I'm not so certain that there's a real issue here anymore, but I leave it up to the reader to determine whether I've simply been wasting my time thinking about this.)

In the second post I considered responses to this possible tension that one could make by denying that reasonable people should tolerate moral opinions. Here I want to consider the second way in which one might argue that there is no tension here--by arguing that one could argue allow that we should, or at least some reasonable people could, think that we should tolerate people's moral opinions, and that this is perfectly consistent with accepting motivational internalism. Here's how I see this strategy being used. We show that our reasons--or the reasons any reasonable people would have--for accepting such a form of toleration simply aren't threatened by the truth of motivational internalism. The justification for tolerance needn't rest on there being a disconnect between moral judgment and action of the sort that the motivational internalist denies, and so there is no need for the internalist to worry about justifying intolerance of moral opinions.

Strategy II

Argument A. The first possible reason to be tolerant in this way is that not being tolerant of people's moral opinions would place a great imposition on people. It would require an unacceptable amount of interference in people's lives to monitor whether people have unacceptable moral opinions and to punish all those who turn out to have unacceptable ones. So whatever bad we could avoid by not allowing people to have the wrong moral opinions is outweighed by the bad we would cause by not being tolerant of people having these opinions. This, clearly, is a consequentialist defense of toleration: it claims that the consequences of intolerance would be worse than those of tolerance, and so we ought to be tolerant.

Argument B. We can also offer a rights-based defense of this sort of tolerance. Even if we think that the consequences of being intolerant of such opinions would be better than the consequences of being tolerant of them, we could argue that people nevertheless have a right to hold whatever moral opinions they hold. And, indeed, we could argue that the right to hold one's moral opinions, even objectionable ones, trumps the bad that is likely to come from tolerating such opinions. There would be two ways to argue this. First, one could argue that, while some bad is going to come of tolerating such opinions, it won't be enough bad to justify interfering with this right. Second, one could argue in a more absolutist vein that, irrespective of how much bad comes form tolerating objectionable moral opinions, a person's moral opinions deserve toleration.

I'm sure there are other particular moral justifications for tolerance for which someone could argue, and I don't want to go into all of them here. But I hope this gives one a sense of how the argument might go. Instead, I want to focus on a consideration that might be used to supplement any of these particular moral arguments (and its broader implications for this issue).

Tolerance, Moral Motivation, and Action

The possible tension here is supposed to arise because the motivational internalist thinks there is a very close connection between moral opinion and moral action. In particular, she thinks that it's a necessary truth that if a person believes he ought to do something, then he is motivated to do it. And this suggests that the motivational internalist thinks there is a close connection between having objectionable moral opinions and acting in objectionable ways. (Consider the racist example I discussed in the previous posts.) But this seems to be a problem, as one way of justifying tolerance of people's opinions is to point to the fact that they're only beliefs. Sure, we might think, this person thinks some terrible things, but that doesn't mean she's going to act on them. And until she acts on them, it's none of our business. It's surely unfortunate that she's so bigoted or cruel or unsympathetic, but we shouldn't interfere with her unless there's a good reason to think her being this way is likely to result in her injuring someone else. The problem for the motivational internalist is supposed to be that, because she holds that there's an especially close connection between opinion and motivation, it's not clear she can defend our thinking in this way.

But here we're assuming there's a close connection between motivation and action, and the motivational internalist may deny this. What she affirms is that necessarily, if a person believes she ought to do something, then she has some motivation to do that thing. And, importantly, this needn't be overriding motivation--in fact, it needn't even be very strong motivation. It may be, then, that this special moral motivation is, in many cases, likely to be overridden by other motivations a person may have. And if this is the case, there need be no close connection between moral opinion and action. The point here is that a close connection between moral opinion and action, which is what the internalist defends, isn't obviously the same as a close connection between moral opinion and action. But the tension arises only if the internalist affirms that this latter connection exist. She needn't affirm this, and so she needn't face such a tension.

So, what is the connection between moral motivation and action? And what do internalists (and externalists) think the connection is? I think this is where the interesting action is here. And, as a matter of fact, this is where the interesting action is in the other strategy for responding to this problem. The move in that strategy was to argue that an externalist should also think there is a close connection between moral opinion and action, and so this isn't a special problem for the internalists. The move, that is, was to point to a close connection between moral opinion and action, which is supposed to be a special problem for the internalist, and to claim that it's a datum with which any plausible meta-ethical theory must deal. And if that's true, whatever problem there is for toleration of moral opinions is a problem for any plausible meta-ethical theory. The move here is in the opposite direction: it is to deny that there is any very close connection between moral opinion and action, since the motivation that necessarily accompanies a moral opinion needn't be very strong.

Let's look at this second move first. The move is to sever the close connection between moral opinions and action while allowing a necessary connection between having a moral opinion and having some motivation to act in a certain way, and this is supposed to allow the internalist to avoid some tension with allowing for tolerance of other people's moral opinions. I think there are a couple of problems with this response. First, it appears to turn motivational internalism into an uninteresting claim. Motivational internalism, it seems, isn't an interesting thesis unless it's the thesis that moral judgment often involves a sort of motivation that has a pretty close connection to how people in fact act.* Second, it appears that this response raises issues about why we should believe motivational internalism is true in the first place. Suppose you accept that there is isn't any very close connection between moral opinion and action, since the motivation that necessarily accompanies a moral opinion needn't be very strong. What, if you accept this, is your evidence that's leading you to internalism as opposed to externalism? If the motivation that is present isn't one that has much effect in many cases, why think it's even there in those cases? It seems the usual evidence for motivational internalism is taken to be the close connection between a person's sincere moral opinions and her actions. If someone really believes it's wrong to do X, then it's unlikely that they're going to do X. And if they think it's their duty to do Y, then they're likely to do Y. It is this putative empirical data about the connection between moral opinion and action that is supposed to support motivational internalism. But if the motivational internalist ends up saying that there's only some, perhaps not very strong, motivation that accompanies moral opinions in many cases, aren't they admitting that the evidence of a close connection between moral opinion and action isn't all that strong? And if the internalist admits this, why do they still think there is a necessary connection between moral opinions and motivation at all?

How about the first move? The move is arguing that there is a close connection between moral opinion and action, which is supposed to be a special problem for the internalist, and that it's a datum with which any plausible meta-ethical theory, whether it's internalist or externalist, must deal with. Consequently, there's no special problem for internalism when it comes to tolerance and the connection between moral opinions and actions. Both internalists and externalists have to allow that there is a close connection of this sort, and so the implications for tolerance of moral opinions, if there are any, are going to be implications for both groups.

But is this really true? Doesn't the internalist want to claim there is a closer connection between moral opinion and action than the externalist? If there were no real difference, what's the debate about? And if there is a real difference, then internalists may not allow as broad a tolerance as we want. They are likely to find a stronger connection between morality and action than non-internalists, and this is likely to lead them to argue that we should restrict tolerance more than reasonable people might want to. Or, at least, it's likely to lead them to reject certain arguments a person might give for intolerance--arguments appealing to the fact that moral opinions needn't be closely connected to action.

Again, though, all of this is going to depend on the details of the views being defended. The real differences between externalists and internalists about this issue may be quite minor. For the plausibility of motivational internalism turns on whether having a moral opinion without being motivated is a possibility, not whether it is an actuality. So the debate may be about merely hypothetical cases, and internalists and externalists may agree about the presence of motivation in all actual cases. Furthermore, whether someone is an internalist or externalist doesn't say much of anything about the strength of motivation that they think accompanies moral opinions in actual cases, and so it would be possible for an externalist to think that having a moral opinion usually is accompanied by a stronger motivation than an internalist thinks.**

Minimally, all an externalist has to allow is that there are some possible cases in which a person makes a moral judgment but has no motivation to act on it. They needn't allow that this is usual the case, or indeed that it is ever the case in the actual world. So the question we need to ask is: Just how odd are these cases supposed to be? Is it just a rare, and perhaps imaginary, sort of person who makes moral judgments and is not motivated? And does this person have an exceptionally odd psychology? If the answer to this last question is yes, then it's not clear that this alone is going to make any practical difference (and we may think we have no good reason to tolerate these sorts of immoralists at all--perhaps they're psychopaths). However, if the externalist thinks that, in some cases, any normal person (or quite a few normal people) can be this way, it will make some practical difference. That is, if she thinks that even ordinary people can sometimes make moral judgments and yet have no motivation to act on them, then there might be some practical implications here. Accepting this amounts to moving towards a view that severs action from moral opinion, and this allows an externalist of this sort of accept a certain reason to tolerate the moral opinions of others: that having those opinions needn't be closely connected to motivation to act. Of course, it is exactly this sort of move that this response is intended to avoid. The move here is to argue that even an externalist can't sever moral opinion from action in the way suggested here, and that this shows there isn't any special problem for internalism.

But is this in fact an implausible externalist position about the connection between moral motivation and action? Does the externalist still need to allow for a very close connection between moral opinion and motivation in real-world cases? I'm not sure. I admit that I'm not very sympathetic to motivation internalism, but it seems hard to deny that there is some close connection between moral judgment and motivation in many cases. But how close, and in how many cases? Honestly, I don't know.

*. Now, of course, I admit that the interestingness of the thesis being defended will depend on just how much the defender of motivational internalism will be distancing moral opinions from action in this response. It's going to depend on the number of cases in which the motivational internalist is going to have to appeal to the "not very much motivation" response to avoid a problem of a conflict with tolerance of moral opinions. (A similar issue may come up for the next problem for this move.) If the number of cases isn't very large, then there may not be a big problem here. If it is large, then it seems to rob the thesis of being anything of much interest.
**. So, clearly, if the relevant issue here is the strength of the ordinary motivation accompanying a moral opinion and not the mere existence of that motivation, the debate between internalists and externalists may not be relevant here.

Divine Command Theory and Alienation from Morality

2004-06-10T10:48:00.000-04:00

Remember when I promised this series about objections to the Divine Command Theory? Of course you don't, since you weren't then reading my blog. Well, I did make such a promise, and I'm now going to begin making good on it. I'm not really concerned about the objections to Divine Command Theory that everybody knows about--that it faces the Euthyphro problem, that God could command us to do horrible things, that it implies nothing is right or wrong if there is no God, etc. I may say something about them eventually, but they're not going to be the focus here. Instead, I want to focus on some, as far as I know, not-much-discussed lines of objection. (But I really don't know the literature here, so maybe everybody working in this area is already aware of these objections.) I don't think that they're all good objections, or that any of them are decisive (are there decisive objections in philosophy?). I'm really more interested in just thinking through some possible lines of objection that have occurred to me, lines of objection that I think at least raise an important issue. And I begin that task now.

I want to say something about how the Divine Command Theory (hereafter "DCT") may lead to a sort of alienation from morality. (I admit I'm not going to offer an clear definition of alienation from morality here, but I hope it becomes clear what I'm talking about as I go alon.) Before I get into the details, though, I need to sketch DCT. I think the following principles are all we really need:

An action-token a is morally obligatory for a person P if and only if God commands P to a; and it is God's commanding P to a that makes it the case that a is obligatory for P.
An action-token a is morally impermissible for a person P if and only if God forbids P to a; and it is God's forbidding P to a that makes it the case that a is morally impermissible for P.

This is pretty skeletal, but I hope it gets the point across. (From here on, I'll simply use the word "command" to cover both God's what God commands and what He forbids.) I don't think the details of how one chooses to reformulate DCT really affect this objection, and so I won't say anything more about how I'm understanding DCT.

The Problem concerning DCT and Alienation from Morality

DCT seems to encounter another problem that tends to befall theories in which moral values are rooted in a transcendent source: these theories tend to alienate us from our own morality. On such theories, moral duties look to be burdens on us that are imposed from without; these duties are not ones that we're willing to accept of our own accord since they may, in some way, be in conflict with our deepest aspirations, desires, goals, etc.

It seems that we would prefer to think that morality is, in an important sense, ours. It is not that we simply want the chance to choose what is right and wrong, nor is it that it should always be easy for us to do the right thing. But we do not want our moral duties to seem alien, inexplicable, and imposed from without. We want to think that moral action, as it is so important, is something that meets our needs, that reflects and embodies our deepest aspirations, that enables us to be the sort of people we want and/or need to be and to live in the sort of communities that we want and/or need to live in. Morality, it seems, needs some connection to human flourishing, to the good life for beings like us. In short, we want to see some connection between moral action and something that we can recognize as valuable to ourselves and others.

Quite simply, it's not clear that DCT can give us this. For, if DCT is true, it's possible that we may not know exactly what underlies our moral duties (God's ways may be inscrutable to us), and that we may not identify with what lies behind our moral duties (God's ways may not be our ways). So it may be that DCT, along with other theories which locate the source of morality beyond human affairs, leads to a conception of morality in which it is more or less imposed on us, irrespective of our aims, desires, goals, etc. And thus morality, to the extent that it influences our actions, may come to seem like a foreign influence on our own life. This allows a person to see moral as shackles, as prohibitions and demands that do more to constrain us than to enable us to flourish and to live the sorts of lives we find valuable and satisfying. Moral rules can come to seem unfair and arbitrary and stultifying, and this seems to lead to problems for morality. Not only may it come to seem manifestly irrational to be moral, but we may feel that we have to sacrifice our aspirations, our goals, and our projects in order to be moral. In short, acceptance of transcendent metaethical theories may come to alienate us from our own morality.

There are, it seems, two (not necessarily exclusive) ways to interpret just why this alienation might be problematic. First, one might think it flouts something that's clearly true about morality--namely that acting morality is intimately connected to our deepest goals, aspirations, aims, etc. If you think there is some logical or conceptual connection between being moral and human good and harm, this might seem a straightforward argument against DCT. If you accept some such connection, then DCT, like any other moral theory, must account for the connection between moral action and human flourishing. Thus any theory that leads to a deep alienation from morality isn't an unacceptable account of morality; it doesn't respect this requirement that any plausible moral theory must meet.

Second, one might think that this may be true, but that it leads to practical problems if it is true. For if it's true that moral demands needn't have anything to do with our deepest goals, aspirations, aims, etc., then we're going to have quite a bit of trouble getting people to act morally. Unless people are able to recognize that behaving morally will realize some good--and a good that they can understand and acknowledge--they are unlikely to do what is moral. So recognition of alienation from morality of this sort is likely to lead to a deterioration of motivation to act morally. Why, people will (and perhaps should) ask themselves, care about morality if it conflicts with your own desires, aspirations, goals, etc. in this way? And it's not clear we're going to be ready with a good answer to this question. Indeed, it is arguable that it's irrational for some, and maybe most, people to act morally in many situations if they are alienated from morality in this way.

And there's another way in which this may lead to practical problem, a way Hume mentions in Section XI of his The Natural History of Religion. There he argues that, if one thinks the value and purpose is removed from ordinary human concerns (as it is if we're alienated from our morality), then people are likely to see not only the source but also the value and purpose of moral action as removed from the welfare and ordinary concerns of human beings in everyday life. And he thinks this is how we should understand the origin of various ascetic moral codes. Ordinary human life, for ascetics, isn't morally valuable. And since the value achieved through moral action isn't a matter of improving their own lives or the lives of others, these people come to see helping others and themselves as not closely connected to living a moral life. Instead, they come to think that apparently worthless rites and apparently pointless practices come to be the the essence of moral behavior, and they direct their energies to engaging in these rites and practices rather than improving human life in the here and now.

Solutions

I don't think we need to defend either, or both, of the interpretations of what the problem is for it to be clear that there's something significant going on here. It is not clear that the defender of DCT is obviously saddled with this unfortunate problem. I can imagine three ways she might avoid it.

(1) She could argue that morality is needed to get into heaven on DCT, and so moral action is in fact line with our deepest aspirations and greatest needs. This, though, strikes me as rather crude, as it devalues life, including moral actions, in the here and now by making them a mere means for achieving the end of getting into heaven. If this the best response on DCT, then morality may be little more than a particularly onerous way of achieving something of value; and this seems to debase being moral to a troubling extent. Fortunately for the defender of DCT, there is a deeper way to argue that her theory does not lead to moral alienation.

(2) It may be that a religious morality can overcome the possibility of moral alienation by claiming that people have the ultimate aspiration to be God-like (although not in a blasphemous way). It may be that our ultimate goal ought to be to be like God, in that God serves as a sort of ideal for human beings. God's moral commands flow from his nature, and so a human being must be moral, as God is, in order to achieve this sort of ultimate goal. This, I think, is a more plausible response, but there are problems here. I'll note just one. Obviously, this is going to require that God is person-like in some important sense; otherwise, it is unclear how being God-like could be an ideal for beings like us. Such views also run into problems if combined with views according to which God is deeply incomprehensible in certain ways. But perhaps these problems aren't so profound that God cannot serve as an ideal for normal human persons.

(3) God has set up the moral rules, human nature, and the conditions of human life in such a way that acting morally will, at least in general, benefit people in the here and now. He knows what we what our deepest aspirations and goals are like; He knows what human life is like; and He's taken these facts into account in before making his moral commands. Gods wants us to live well, and we can be sure He wouldn't command that we do things unless they were, at least usually, for our own good. So a complete understanding of God's commands, the nature of morality, and our situation would reveal that acting morally is consistent with, and perhaps required for, achieving our goals, satisfying our deepest aspirations, etc. This, it seems to me, is the best of the three responses. But there's an obvious way to raise questions about it: Does the empirical evidence concerning what human life is like, and especially concerning what life is like for those who live morally, show that moral people end up living better lives than immoral people? Or, if we're concerned with something more general, does the relevant empirical evidence show that, in general, people who live morally are better off than those who don't? If the empirical evidence doesn't bear out that being morally does tend to lead to having a better human life--and, honestly, I don't know what to think about this--then this response doesn't really get off the ground.

I've noted a particular problem with each of these responses. Now I want to note a couple of general problems with which the defender of any of these responses needs to deal. The first such problem is the inscrutability of God and His intentions. This, I think, is a problem for all three responses, though probably a less serious problem for (1). The general problem here is that DCT seems to undermine moral action's leading to goods with which we can identity--that we can understand as closely connected to our own good, to our own flourishing, to the things we care most about. And inscrutability, if we appeal to it, will seemingly make this problem even worse. If God is inscrutable, it's not just that we can't accept this morality, but that we can't even understand where it's coming from, what it's based on, and what its point is. If God is largely inscrutable, then it's going to be difficult to understand what being God-like entails, and so it may be difficult to see what is of value in being God-like. If God's intentions are largely inscrutable, it's going to be difficult to understand the ways in which he has ordered the moral rules and the conditions of human life to ensure that moral action provides us with a way of satisfying our deepest aspirations, achieving our goals, etc.

Now, clearly, the extent to which this is a problem is going to depend on just how inscrutable God and His intentions are. The less inscrutable they are, the less problematic this worry is going to be. (I should note, though, that I think God's going to have to be pretty inscrutable to deal with the problem of evil. I won't develop any argument for that position now, however.) So there are two important questions here: How much do we need to understand about God and his intentions to understand, appreciate, and identify with the God's plan here, and do we have good reason to think we understand God this much? These are big questions that I won't even try to answer here, but I think it's important to realize that they're relevant here.

The second general problem is that it seems there are going to be people who simply don't care about the various types of goods to which these response claim moral action leads. This is especially pressing for (1) and (2) since they give us a specific account of what the good resulting from moral action is here: we know some people may not care about getting into heaven, and we know some people may not care about being God-like. (This is less problematic for (3) since it's so vague in just what the goods are.) We can imagine people who simply don't care about these things, or who don't care about them anywhere near as much as they care about other things. These people, then, will still be alienated from morality, even if we grant the defender of these responses what she argues about God, His intentions, and the aims that underlie His moral commands.

Is this a real worry? Well, you might think not because you can't expect a moral theory to work for everyone. Any moral theory, one might think, is bound to run into this problem at some point. Whatever it is about moral action that is supposed to connect it to our deepest aspirations, desires, goals, etc., it may not apply to every single case. Some people have odd aspirations, desires, goals, etc., and so we can't expect everyone to acknowledge and appreciate the good that comes from moral action. This is somewhat plausible, I think. Its plausibility will, of course, depend on the empirical facts about just how many people's deepest aspirations, desires, goals, etc. are inconsistent with the DCT theorist's account of the moral demands God has placed on us. The more people who are like this, the less plausible this response seems.

Is Alienation Unimportant?

I'll end this post by discussing one possible way of responding to this last worry may reveal that we've been on the long track all along in that alienation from morality may not be all that important. The preceding worry about DCT and alienation stems from the fact that people may simply not care about the goods to which moral action is supposed to lead. But maybe we should be more concerned with what constitutes a good life rather than what it is that people actually care for. Maybe we've been assuming a sort of subjectivism about the good life that the defender of DCT may want to reject. Suppose we accept some sort of objectivist conception of the good, a conception in which the good life doesn't depend on what people actually desire. If we do, then the defender of DCT could argue that, while acting morally may be inconsistent with a person's conception of her own good (i.e. with her deepest aspirations, desire, goals, etc.), it may not be inconsistent with the correct objective conception of the good. And perhaps this is all we need. If acting morally is in line with living a good life, why should we care about its not being in line with the deepest aspirations, desires, goals, etc. of people who don't aspire to lead good lives? The people who aspire to lead good lives won't be alienated from morality, and isn't that enough?

The basic idea here is that the alienation of some, and maybe even all, people from morality may not be so problematic if acting morally is consistent with what really is the good life. Those people for whom moral action is inconsistent with their deepest aspirations, desires, goals, etc. are people who've got the wrong conception of the good life, who are living the wrong way. This doesn't show that moral action isn't alienating, but that alienation from morality might not be all that problematic. The problem of alienation from morality is that some people's aims, desires, goals, etc. will not be satisfied by living morally. So, for these people, we couldn't say that the morality wasn't burdensome, as being moral would conflict with their satisfying their aims, desires, aspirations, etc. And if we accept a subjective conception of the good life, we can then say that being moral would have a harmful influence on their lives. (It might, in fact, be irrational for them to be moral in these circumstances.)

But, if we accept an objectivist account of the good life (i.e. a conception on which standards for a good life aren't set by a person's aims, goals, aspirations, desires, etc.), then we can't move from alienation from morality to acting morally actually harming one, to its actually leading to their living a worse life. So, provided that we accept an objectivist conception of the good life and claim that God has made moral commands that, if followed, would lead to people have good lives, we can still argue that acting morally would lead the person to having a better life. Morality, then, may be alienating, but it's not burdensome to people who are alienated from morality. And if alienation isn't burdensome in this way, then perhaps it's not a big problem.

I think this is a pretty powerful response to the worry about alienation from morality. Clearly, though, it's going to depend on the plausibility of an objective conception of the good life. This isn't something I have space to cover here. Instead, I want to briefly question whether this response really does show that alienation from morality doesn't matter. Why might things not be so simple? Because it strikes me that any plausible objectivist account of the good life is going to have to include some subjective elements. It seems that any plausible objectivist account of the good life is going to make it a necessary condition for living a good life that a person achieves many of her most important goals, satisfies many of her most important aspirations and desires, and so on. If this is the case, then alienation from morality is still a problem. So it's not all that clear that alienation from morality is going to be consistent with living a good life, even if we adopt an objectivist conception of the good life.

Brayton on Popper and Laudan on the Demarcation Problem

2004-06-09T13:15:00.000-04:00

Via Brian Leiter, I've come across this post by Ed Brayton on Popperian philosophy of science. In particular, Brayton is concerned with certain objections to Popper's falsificationist criterion as a response to the demarcation problem (i.e. the problem of explaining what distinguishes science from pseudo-science). I'm not expert on philosophy of science, but I think I know just enough to say something about the issues Brayton raises.

First, a little about Popper's views. According to Popper, a theory is scientific if and only if it is falsifiable. And, on a simple version of what it is to be falsifiable, a theory is falsifiable if and only if there is some observation that would entail the falsity of the theory. So, according to the simple version of Popper's criterion, a theory is scientific if and only if there is some observation that would entail the falsity of the theory.

Brayton quotes the following passage from Laudan's "The Demise of the Demarcation Problem" in which he criticizes this criterion:

A second familiar approach from the same period is Karl Popper's "falsificationist" criterion, which fares no better. Apart from the fact that it leaves ambiguous the scientific status of virtually every singular existential statement, however well supported (e.g., the claim that there are atoms, that there is a planet closer to the sun than the Earth, that there is a missing link), it has the untoward consequence of countenancing as "scientific" every crank claim which makes ascertainably false assertions. Thus flat Earthers, biblical creationists, proponents of laetrile or orgone boxes, Uri Geller devotees, Bermuda Triangulators, circle squarers, Lysenkoists, charioteers of the gods, perpetuum mobile builders, Big Foot searchers, Loch Nessians, faith healers, polywater dabblers, Rosicrucians, the-world-is-about-to-enders, primal screamers, water diviners, magicians, and astrologers all turn out to be scientific on Popper's criterion - just so long as they are prepared to indicate some observation, however improbable, which (if it came to pass) would cause them to change their minds.

Brayton finds two arguments here:

A. Some "singular existential statements" are not falsifiable, yet they are a part of science; and,

B. Many "crank claims" can be made falsifiable, so Popper's criteria would allow many pseudo-scientific ideas to be considered scientific ideas.

He then criticizes both of these arguments. I'm not sure what I think about these arguments, but I don't find his criticisms all that compelling. I want to explain why.

Here's his criticism of argument A:

I think he misunderstands how falsification is applied by Popper. Popper is not arguing that "existential statements" - by which I assume he means observations or potential observations - must be falsifiable. Using one of his examples, we know that there is a planet closer to the sun than the Earth because we can observe that there is one. In Popper's view it is not observations that must be falsifiable, it is explanations. Theories - explanations for observed phenomena - must be falsifiable.

First, as far as I can tell, Brayton isn't disputing Laudan's suggestion that these claims can't obviously be falsified. So I'll ignore that issue, though I think it's true that they can't be--at least not in the sense that is relevant here. Second, I'm pretty sure that what Laudan calls an "existential statement" isn't supposed to be an observation; it seems it's just supposed to be a claim that a particular thing, or type of thing, exists. Third, I don't think I buy the distinction between existential statements and theories that Brayton is drawing--though, of course, he frames it as a distinction between theoretical claims and observations since he takes an existential statement to be an observation statement. For instance, I really don't see any reason to think that the singular existential claim that Brayton discusses can't be taken to be theories (on his account of what a theory is). The claim that there is a planet closer to the sun is supposed to explain some of our observations: it explains certain things we see when we look up at the night sky, it explains certain things we see when we look through a telescope, it may explain certain things we see when we look at the orbits of other planets, etc. So it is, in some sense, an explanation of observed phenomena, and it is then, in some sense, a theoretical claim. Consequently, if we accept the falsificationist criterion and we're going to take this claim to be scientific, it had better turn out that this claim is falsifiable. I don't think it is.

Now Brayton's response to argument B:

I think that Laudan is correct that in a Popperian view, most of that laundry list of pseudo-scientific ideas could be formulated in ways that are testable and falsifiable. In fact, I'd say many of them have been formulated in those ways and have been falsified, regardless of whether their proponents will admit it or not. But this doesn't strike me as a compelling criticism of Popper. Of course it's true that the falsification criterion would allow for ideas to be admitted as science that ultimately are falsified. The demarcation question is not concerned with determining, a priori, which ideas are true and which are false, it is concerned with determining which ideas are capable of being answered using the tools of science. Falsification isn't the end of the process. If an idea is testable and falsifiable, then it still must be tested, by attempting to falsify it in most cases.

I guess the idea behind Laudan's criticism is supposed to be that Popper counts far too much as science. If we take the simple falsifiability criterion (i.e. a theory is scientific if and only if it is falsifiable), then we're committed to counting as science everything that can be falsified by observation. But you might think that this is too broad a conception of science. Suppose, for instance, that we falsify religious views about the deluge by doing scientific inquiry. Does it then turn out that those views about then deluge were a scientific theory (albeit a false one)? If we think a theory is scientific if and only if it is falsifiable, then it seems we have to say yes. But that seems odd, and so there's something wrong with the falsifiability criterion. And, in motivating this problem, we don't have to draw on any worries about claims that might be taken to be pseudo-scientific. Suppose I claim that it's raining right now in New York City. That's not a pseudo-scientific theory and it's a claim that can be falsified, but am I doing science when I say this? It seems prima facie odd to say I am.

So the worry isn't just that Popper claims that pseudo-science is science, but that he counts absolutely anything that's falsifiable as scientific. Is there something to this worry? Brayton doesn't seem to think so, for he appears to have a pretty liberal attitude about what should and shouldn't count as science.

Is there something wrong with a liberal attitude of this sort? Maybe, and maybe not. You might think that all that's really going on if we find a problem with this attitude is that we don't want to give just any old claim that can be refuted the title of a scientific theory. Perhaps we use "scientific" as a sort of honorific, and we only want to give it to theories of a certain sort. We don't want to claim that any old astrologer or creation scientist or whatever is doing science simply because they manage to say something that we can falsify. And if being falsifiable is a sufficient condition for being scientific, we'd have to say this. That, I think, isn't a very compelling reason to think there's something wrong with the falsificationist criterion.

Is there more to it than this? Well, you might think that it's problematic that Popper's account of what is and isn't science is purely logical. And you might think that there are more stringent restrictions on subject matter or methodology that determine what does and does not count as science. Perhaps scientific theories can't appeal to supernatural entities, or perhaps there is some scientific method that we have to arrive at scientific theories by employing, or something like this. If anything like this is true, then a theory's being falsifiable won't be sufficient for its being scientific. For it seems that there are going to be lots of falsifiable theories that appeal to supernatural entities and that aren't arrived at through the scientific method, whatever that is.

Honestly, I don't know what to think about this. But it does seem that there may be some reason for not wanting to call any old falsifiable theory scientific.

Blogging Issues

2004-06-09T09:58:00.000-04:00

It's probably a little premature for me to be doing some meta-blogging, but here goes. Some points about blogging, and about this blog.

1. I'm not sure that my comments are working. I suspect that they are, but no one yet has left a comment. I suppose that isn't too surprising, as I've only had about twenty visitors (roughly ten of the hits on my counter are from me, since it adds hits when I work on my blog from a different computer). So I'd be grateful to anyone who came by and left a comment of some sort. It could be anything really--I just want to be sure they're actually working for other people.

2. I'm having a formatting problem, and I'm wondering if anyone knows how to fix it. Whenever I use a blockquote tag, it messes up the rest of the text in my post. After the blockquote, the remainder of the text in my message is more bunched up than the text before the tag. (I notice the effect isn't as prominent when I view my blog with Internet Explorer instead of Mozilla.) Does anyone know what's causing this and how, if at all, I can fix it?

3. I'm aware that I'm not very good at keeping a consistent pace in my posting. The problem is that I have notes for several posts that I'm working on at any given time, and I want to publish what I write as soon as I finish it. So, on days when I spend some time here, I end up posting several different things; and then I go days with nothing. Is this a problem? I figure that by keeping up a steady pace, you keep people coming back. But I'm not sure about that, and I suppose that, with my low traffic, it's not something I should be too worried about.

4. I realize I don't yet have the ability to write pithy blog posts, and I'm working on it. (This is a general problem for me--a common complaint about the papers I write is that they're too wordy.) I doubt many people are actually going to read a 2,000-word blog post. But maybe anyone who's actually interested in the material I'm discussing would; I don't really know.

5. And when I did a Google search for "the tribunal of experience" today, I was the 21st entry. Not too bad.

6. Oh, and does anyone know how I can place only the start of my post on the main page and link to the rest of it on the permanent link for the post? I haven't been able to figure this out, and maybe somebody can tell me. Please. It can be quite a hassle to go down the length of my main page since I have a penchant for the inordinately long blog post, and I figure most people are too lazy to scroll all the way down and see everything I have to offer on the main page when they notice the length of the first few entries.

Kant's Categorical Imperative and Rules of Etiquette

2004-06-09T09:56:00.000-04:00

I want to raise a question about the way in which Kant applies the categorical imperative in the Groundwork. I think there's a problem here, though I'm not sure that it's one to which Kant isn't able to provide an answer. Indeed, it's possible that there's an obvious answer to the problem I'm going to develop here. But I don't know what it is.

The problem is that it's not clear to me that there aren't maxims that fail to meet the requirements of the categorical imperative, but that it wouldn't be morally wrong to act on. That is, the problem is that there seem to be maxims that we can't universalize in the right way, but that it wouldn't be immoral to act on.

A Test for Moral Impermissibility

I first want to talk about how I'm going to understand the general issues here. I'm going to be concerned with the following formulation of the categorical imperative:

Formula of Universal Law: always act in ways such that you could will that the maxim of your action become a universal law.

Drawing on this formulation of the categorical imperative, we can derive a sort of test for the moral status of various actions. We being by isolating the maxim or motive on which the agent acts. Let's take the maxims to have the following form:

I ought to do action x in circumstances C in order to bring about result r.

So there are three relevant parts here: the action, the circumstances, and aim. Since we're using the formula of universal law here, what we need to know about the relevant maxim is whether it's one that we can universalize. The second step in our test, then, is to universalize the maxim. Here's the form of a universalized maxim:

Everyone ought to do action x in circumstances C in order to bring about r.

Now we just need to ask ourselves whether we can consistently will this. And the particular question we'll need to ask for the cases I'm going to be concerned with is: Is it possible to conceive a world where this maxim is universalized? If it isn't, then acting on this maxim would be morally wrong.

Let's see how this works in one of the cases Kant discusses--namely the case of a person making a false promise. (I won't follow the details of Kant's example exactly, but I'll use the same general type of example.) Suppose you want to know whether it's morally permissible to make a promise while knowing that you aren't going to be able to keep your end of the bargain. Suppose you're thinking of promising to give someone $100 within two weeks if she loans you $50 now, even though you know you won't have that money soon enough to pay her back then.

Is this permissible? We need to begin by isolating the maxim on which you're planning to act. Let's take the maxim to be:

I ought to make a false promise in circumstances where I need money in order to get the money I need.

Again, since we're using the formula of universal law here, what we need to know about the relevant maxim is whether it's one that we can universalize. The second step in our test, then, is to universalize this maxim. Here's the form of a universalized maxim:

Everyone ought to make a false promise in circumstances where they need money in order to get the money they need.

Now we just need to ask ourselves whether we can consistently will this. And the particular question we'll need to ask for the cases I'm going to be concerned with is: Is it possible to conceive a world where this maxim is universalized? Is it possible to conceive a world in which everyone makes a false promise in order to get money whenever they need some?

Kant says no, and so he says that this is morally impermissible. Making a false promise is a good way to get what you want only because there are certain conventions in place, conventions that assure people that a person making a promise will do as she promises to do. So making a promise is a good way to get what you want only because there is a general institution of promising in place. But a world in which this maxim is universalized is a world in which there is no such general institution in place. If people didn't keep their promises even when they'd benefit from not doing so, there wouldn't be any general institution of promising. For the relevant conventions of people keeping promises wouldn't be in place. So you cannot imagine a world in which this maxim is universalized; such a world would be a world without promising.

There's a straightforward contradiction in what you're willing if you will that this maxim be universalized: you will that the convention of promising be in place (this is required to make a false promise), and you will that it be undermined (this is what would result from this maxim being universalized). The universalization of the maxim undermines the relevant conventions, and yet those very conventions are assumed in willing in accordance with this maxim. This is enough to show that the action is morally impermissible.

A Problem for the Test

Let's grant Kant this test. If a maxim fails to be universalizable in this way, then acting on it is morally impermissible. So it's a sufficient condition, though it need not be a necessary one.

The problem here is that it seems there are maxims that will fail to be universalizable in this way but that it seems pretty counterintuitive to think it would be morally impermissible to act on. Let me give an example.

Suppose you want to know whether it's morally permissible to make a joke by flouting some rule out etiquette. Suppose, for example, that you're considering making a joke by using the wrong fork to eat your salad in front of a rather stuffy group of people.* (This, I admit, would make a lousy joke, but bear with me.)

Is this permissible? Let's use Kant's test here. We need to begin by isolating the maxim on which you're planning to act. Let's take the maxim to be:

I ought to use the wrong fork in circumstances where I'm surrounded by stuffy people in order to make a joke.

Since we're using the formula of universal law here, what we need to know about the relevant maxim is whether it's one that we can universalize. The second step in our test, then, is to universalize this maxim. Here's the universalized maxim:

Everyone ought to use the wrong fork in circumstances where they're surrounded by stuffy people in order to make a joke.

Now we just need to ask ourselves whether we can consistently will this. Is it possible to conceive a world where this maxim is universalized? Is it possible to conceive a world in which everyone uses the wrong fork to make a joke in this way?

I think it turns out that you can't, for what are very close to the same reasons that you can't will that everyone makes a false promise when they need to get some money. Using the wrong fork allows you to make a joke only because there are certain conventions in place, conventions that assure people that a person will be using a fork of a particular sort to eat a particular course of her meal. So using the wrong fork is a good way to make a joke only because there are certain general conventions of etiquette in place. The joke, of course, comes from openly and surprisingly flouting these conventions. But a world in which this maxim is universalized is a world in which there are no such general conventions of etiquette in place. If people didn't normally use the 'right' fork in these sorts of situations, there wouldn't be any conventions of etiquette of this sort in place. So you cannot imagine a world in which this maxim is universalized; such a world would be a world without the conventions needed to make a joke in this way.

Again, if we focus on what you'd be willing in willing in accordance with this universalized maxim, we find a straightforward contradiction: you will that the convention of using the right fork be in place (this is required to make a joke by using the other fork), and you will that the convention not be in place (this is what would result from this maxim being universalized). The universalization of the maxim undermines the relevant conventions of etiquette, and yet those very conventions are assumed in willing in accordance with this maxim. This, if we're granting Kant this test, should be enough to show that the action is morally impermissible.

But, I take it, it's clear that acting in this way isn't morally wrong action--at least if we don't suppose that there's something very odd about the circumstances. It's simply not morally wrong to use the wrong fork and thereby make a joke. So there's a problem with this test: it tells us that actions are morally impermissible even though they aren't.

What's Going on Here?

I think the upshot--the point we should draw--is that this problem arises from interpreting Kant as having a purely formal requirement for moral duties. The formal requirement is that the maxim of one's action have the form of a universal law. If it lacks that form, it's an impermissible action. So acting on any maxims that one could not consistently will to be universal laws is immoral action. And there is no constraint on the content of the relevant maxims; there is no requirement that they pertain to some sort of subject-matter. Thus one needn't draw any distinction between moral and non-moral maxims before applying this test. And this, it seems, leads to the problem, as we don't have grounds to limit the relevant maxims to those whose content is of moral relevance; we don't have a way to rule out the maxims concerning, say, etiquette as opposed to morality.**

*. I'm too unrefined to know what the relevant rule about salad forks is, but I believe there is some such rule. And if there isn't, just imagine I'd been appealing to a similar sort of rule.
**. I think Hare's purely formal account of moral judgments runs into a similar problem, and Foot's arguments that he runs into such a problem have influenced my worries about Kant.

Even More Mackie Stuff

2004-06-08T14:09:00.000-04:00

One more response to this post.

More about how the epistemological argument from queerness is supposed to work.

Mackie should suspend judgment about the special epistemic capacity with which we acquaint ourselves with objective values. He denies that we have the epistemic capacity to know of objective values. In denying that we have the epistemic capacity, however, he must know what the epistemic capacity is. Knowing what the epistemic capacity is permits Mackie to deny that we have it. If he has denied the existence of the epistemic capacity without knowing what it is, then he risks making the same error. He has called on some unusual faculty of mind that allows him to intuit what epistemic capacity does and does not exist. If he is doing this, then he commits the same error as the person who accepts the existence of objective values.

This, I think, rests on a misunderstanding of the nature of Mackie's argument. I'm pretty sure Mackie doesn't think he has, or needs, any special epistemic capacity to discern which epistemic capacities do and do not exist. All he seems to think he needs to know this is the account of human psychology that ordinary empirical inquiry gives us. But doesn't he need to know something about what a faculty for intuiting objective moral facts would be like? Doesn't he need to know something about the faculty of moral intuition? Sure, but he doesn't seem to think he needs to know a great deal. He seems to think he only needs to know that it would have to provide us with some sort of access to objective moral values. He's already told us what objective values are like, and we're already supposed to know that they'd be awfully strange--so strange, in fact, that they don't seem to have a place within the natural world. Thus, given what objective moral values would have to be like and what empirical inquiry tells us about our intellectual capacities, it's not clear how beings with psychologies like ours could be in contact with such things. Where's the causal influence going to come from? How are objective moral values going to have some impact on us that allows us to detect or apprehend them? It's hard to see how they could, and so it's hard to see how we could have a faculty of moral intuition.

Can we expand our picture of how we arrive at knowledge of objective moral facts?

Second, the real question for Mackie is how we arrive at our moral beliefs. There may be alternative ways of arriving at moral beliefs than through our intuition. For instance, I am reminded of the way Aristotle proposes one becomes virtuous. According to Aristotle, acting virtuously is a matter of habit. We learn how to behave appropriately and we habituate this sort of activity. We can learn what the moral principles, and we have the capacity for moral sentiments. We have to dedicate ourselves to constantly revising the way we act and how to act so that it is in accord with the moral principles.

But I wonder if this really helps. For it seems to me that, if you interpret Aristotle in this way, you're really denying that Aristotle thinks we can arrive at moral knowledge of the sort that Mackie thinks we ordinarily believe we have (or could have). On this interpretation of Aristotle, it looks like he thinks moral knowledge is a matter of knowing how to act, and not of knowing some special objective moral facts. Moral knowledge, then, isn't knowledge of objective moral facts, as it isn't knowledge of facts at all. So knowing the moral thing to do in some situation is a matter of knowing-how rather than know-that. There aren't, if this view is correct, moral states of affairs 'out there' in the world to be discovered by us.

However, Mackie clearly doesn't think that this is what we ordinarily think of morality. He thinks it does involve knowing objective moral facts, and that these are special facts that are practical (i.e. that have a necessary connection to action). He thinks that we ordinarily think that there are moral states of affairs 'out there' to be discovered by us. Indeed, it is the fact that these putative objective moral facts essentially practical that makes them moral, and that makes them queer.

So it doesn't seem that Aristotle's views gives us a picture of a non-intuitive way of arriving at the sort of moral knowledge that Mackie denies we have. Of course, one might think that Aristotle's picture of how we arrive at moral knowledge is very plausible, and this, one might think, shows that Mackie fails to understand what moral knowledge is like. Moral knowledge isn't a matter of knowing some peculiar moral facts; it is, rather, a matter of knowing how one ought to act in various types of situations. So Mackie is wrong about what objective moral knowledge would consist in. It wouldn't consist in knowing that some objective moral states of affairs obtain; it would consist in there being objective moral reasons to act in certain ways in certain situations. And some people (e.g. Thomas Nagel) think this is the best way to respond to Mackie.*

I think that the author is getting at something similar when he says

The underlying problem here is how we arrive at our moral beliefs and that they are justified moral beliefs. The way that I want to explain away the underlying problem is that moral beliefs do not guide our action, but it is our judgment that guides our action. So, it doesn't really matter whether we have moral beliefs that correspond to something but it does matter whether we judge that there are moral beliefs.

I guess it's not clear to me what is meant by 'judgment' here, but he seems to be making a similar point. Having the correct moral beliefs isn't a matter of one's beliefs corresponding to reality, but a matter of having good moral judgment, where having good moral judgment isn't a matter of having beliefs that correspond to moral reality. I don't see what role moral beliefs then need to play, though, and it seems that we should just give them up. There may be beliefs that are relevant to how one makes moral judgments, but they aren't moral beliefs.

So the real dispute here is with Mackie's conceptual analysis of ordinary moral language and thought. Mackie believes that the best conceptual analysis reveals that moral language purports to describe objective moral facts, and that moral thought involves having beliefs that attempt to mirror these moral facts. This conceptual analysis underlies his conception of what moral objectivity would consist in--it would consist in there being objective moral facts to accurately describe in our moral language and to accurately represent in our moral thought. If moral language is a discussion of how to behave rather than a discussion of what is the case, and if moral thought is thinking through how act rather than thinking through what's the case, then Mackie's conceptual analysis is confused and he's got a faulty conception of what moral objectivity would consist in.

But it's not clear that this is of any help to defenders of moral intuition, unless they have a conception of intuition on which we can intuit something other than facts. They'd have to present a conception of intuition as a faculty revealing how we ought to act, and not as revealing certain moral facts. I don't think I've seen a conception of intuition of this sort. The intuitionists of whom I'm aware always treat intuition as revealing certain moral states of affairs.

Finally, the last point:

Third, the emotivist or prescriptivist may come to the rescue of the objectivist. Mackie's argument depends upon the truth of the premises guaranteeing the truth of the conclusion. He writes that we cannot be aware of the truth of these ethical premises without some special sort of intuition. The objective emotivist may claim that Mackie has not considered that statements like, "lying is wrong" fails to be a proposition with a truth-condition. This sentence may only express the sentiments of the speaker. If the statement is merely an expression of the speaker's commitments, then it fails to concern the truth or falsity of the premises. The clear-headed objectivist then need not resort to a special sort of intuition. Objectivism turns out true.

This looks similar to what I was suggesting above. Noncognitivists can argue that Mackie's arguments rely on an unacceptable account of what moral objectivity would amount to, since Mackie is assuming that some form of cognitivism is true. Cognitivism is false, and so Mackie has an unacceptable account of moral objectivity.

But how, exactly, is this supposed to help an objectivist about morality? If you're going to concede this much to the noncognitvist, how are you going to end up with an objective account of morality? Sure, you may no longer need to appeal to some faculty of intuition that allows us to have access to the moral facts, but it's not clear that you can still offer an objective account of morality. What is the objectivity supposed to consist in? Perhaps you can give an account of objectively good or bad, right or wrong, correct or incorrect commitments for the speaker to have. But how is that supposed to work?

*. I believe Hare also thought something like this. He thought that Mackie's argument really revealed that there was a sort of conceptual confusion in thinking that moral objectivity would consist in there being objective moral facts we could know, as he thought it was clear that moral language didn't purport to describe facts at all. Mackie thought he was making an empirical argument that morality wasn't objective, whereas he was simply showing that a particular conception of moral objectivity was incoherent. Moral language doesn't purport to describe moral facts at all, and so it couldn't describe objective moral facts.

More on Mackie

2004-06-08T11:20:00.000-04:00

Here's some more stuff about Mackie and Mackie-related topics in response to this post.

First question: Can we just hold fast to the assertion that we have moral intuitions?

Perhaps the easiest objection to Mackie's argument is that he begs the question. For instance, we can deny that we do not have a special sort of intuition which enables us to discern different objective values. We have this intuition, and nothing that Mackie or any of his cohorts says will convince us otherwise.

I guess I don't see this as very plausible. I can imagine two things that might be going on here, and neither of them seems to work against Mackie.

First, one might have in mind some conception of intuition according to which there is a phenomenological difference between intuiting something and coming to know it in some other way. Then you can assert that, through careful introspection, one can figure out that moral thought has the phenomenological qualities that are distinctive of intuitive thought. That is, one could argue that there is a special phenomenology to intuiting some moral fact, and that we recognize in our own experience that we sometimes arrive at moral beliefs through such thought.* The problem with this response, though, is that it just doesn't seem that there is anything distinctive of moral thinking in this way. It's not clear to me that there is any special phenomenology that is involved in intuiting that something is the case as opposed to simply believing it to be the case.

So let's not suppose there is any distinctive phenomenology to moral intuitions. They're just intuitions like we have in other areas of philosophy; they're immediate reactions to a case that is presented to us. And if this is all an intuition is, then I don't see that Mackie is really denying that we have them. He can agree that we have intuitions of this sort, and yet he can deny that they provide us with access to anything like objective moral values. So the issue isn't about whether we have intuitions in this sense, about whether we arrive at some moral beliefs in an immediate way. It's instead about the following thesis: that at least some of our moral intuitions are arrived at in a way that gives us immediate access to the objective moral values.

It seems, then, that the debate here is going to be about the origin of these intuitions. And we know Mackie has a theory about their origin--we have the moral intuitions we have because we have been socialized into various groups. He thinks, moreover, that this isn't an explanation of our having the moral intuitions we have that provides us with good reason to think that those intuitions provide us with access to objective moral values.

It should now be clear that simply asserting that one has intuitions isn't going to be enough to deal with Mackie's argument. The defender of the view that intuition gives us access to objective moral values is either going to have to directly rebut Mackie's suggestion about the origin of our intuitions or provide reasons for thinking that our intuitions best explained as involving access to objective moral values. Either of these options, it seems, is going to require telling us something about just how we get these intuitions. Mackie assumes this is going to require attributing some special mental faculty to us; others, like Price and Ross, think moral facts are intuited reason in the same way that other truths (e.g. mathematical facts) are arrived at through reason. Regardless of which tactic is chosen, though, we're going to need a good deal more than the flat assertion that we have moral intuitions.

*. Of course, there's going to be the additional question of why we should think that thought with these phenomenological characteristics does in fact sometimes reveal the facts to us. It would seem possible for intuition to be misleading in all cases. But we needn't worry about this here.

Mackie Stuff

2004-06-07T23:12:00.000-04:00

I'd like to say some things about the argument of the first chapter of Mackie's Ethics: Inventing Right and Wrong in response to the posts here and here at Oohlah's Blog-space. I'll talk about one of these posts now, and I'll get to the other in the near future.

First, I want to say something about Mackie's epistemological argument from queerness. I should note first that it seems to me that there are actually two distinct, albeit interconnected, issues in the epistemological version of argument from queerness. For it seems to me that Mackie thinks two things about knowledge of objective moral values would be queer: the faculty that would allow us to detect metaphysically queer objective values, and the sort of knowledge (viz. intrinsically action-guiding knowledge) that knowledge of objective moral values would be. Here I'll only be talking about the former sort of epistemological queerness.

Here's the worry about the epistemological argument from queerness:

No matter what I do or what I say cannot convince the anti-realist of the existence of numbers or the moral anti-realist of the existence of moral properties. What I can do is provide a negative argument against the anti-realist position. Even if we had some sense of moral intuition, we would still wonder whether that allowed us access to moral properties. Our sensory apparatus allows us to perceive the world. Sometimes this apparatus fails us. For instance, we think that we see a round tower in the distance when it is actually a square tower. When we submerge a straight edge into the water, it appears to bend. Despite the clarity with which our senses present the material world, we have no reason to believe that it presents us with what is actually there before us. Thus, no faculty is able to give us access to the material world, and no explanation of a faculty (regardless of whether it is a natural explanation) permits us to say that x is a moral property.

Admittedly, I'm not sure that I fully understand the argument of this passage. The argument, if I understand it, turns on the claim that Mackie's argument rests on the assumption that if we had some faculty of moral intuition, we wouldn't have any doubt about whether it provided us with access to moral facts. Mackie then argues that we do have such doubt--we do have good reason to doubt that what we ordinarily think of as moral intuitions provide us with access to moral facts--and so we don't have a sense of moral intuition.

The problem with this argument, it seems, is that we can provide an analogous argument that to undermine our usual thinking about sensory perception. If we're going to accept Mackie's premises in his argument against our having a faculty of moral intuition, then we ought to accept a premise that if we had some faculty that provided us with access to natural facts, we wouldn't have any doubt about whether it provided us with access to natural facts. The faculties that are supposed to provide us with access to natural facts in this way are our sensory faculties. So, if we have reliable sensory faculties, we shouldn't have any doubt that they provide us with access to natural facts. But skeptical scenarios and ordinary sensory illusions show us that we can doubt that our senses provide us with access to natural facts. So if we accept a premise of the sort on which Mackie is relying, we have to give up the idea that our sensory faculties provide us with access to the natural facts. We don't want to say that, and Mackie certainly wouldn't.

Moreover, any faculty we possess is likely to run into errors of this sort. There's always room to doubt whether a given faculty provides us with accurate information about the world, and so no faculty could possibly pass this test. Thus, if cogent, Mackie's style of argument can be used to the thesis that we have any faculties providing us with access to facts.

Now, this argument looks pretty good to me--if we agree that Mackie's epistemological argument from queerness rests on an assumption that we wouldn't have any doubt about our access to moral facts if we had a faculty of moral intuition. But I don't think it does rest on any such assumption. The problem with knowledge of objective moral values isn't that we can doubt that we have access to a realm of objective moral facts; rather, it's that we can't really explain how beings like us could possibly have access to such facts. For Mackie, the argument has to begin with a commitment to some sort of philosophical naturalism. And, in particular, it has to understand with a sort of naturalistic conception of human intellectual faculties. The best account of human psychology we have is one that we acquire through the natural sciences, and this account seems inconsistent with our having a faculty that provides us with knowledge of objective moral facts?

Why is this a problem? Well, because of the metaphysically queer sorts of things that objective moral values would be. Mackie seems to think that their combining objectivity and their intrinsic and categorical action-guidingness shows that they would have to be non-natural facts of some sort. And the epistemological problem is that we have no account of how natural beings like we are, beings whose psychologies are best described by the natural sciences, can have access to such facts. We can't understand how these objective moral facts could influence us in a way that allowed us to detect or apprehend them.

So Mackie's argument does rest on some general assumptions about what our faculties have to be like if they're going to provide us with access to facts. But one of those assumptions isn't that we have to be sure that they provide us with access to such facts. It's fine to doubt whether the faculty actually provides us with access to facts, but we need to be able to understand how it could provide us with access to facts; we need to be able to understand how it could have some influence on natural beings like us. The problem with a putative faculty of moral intuition is that we simply don't understand how we could have access to objective moral facts. So the challenge for the objectivist, Mackie thinks, is one of explaining how the faculty of moral intuition works.

Hume's The Natural History of Religion

2004-06-04T16:25:00.000-04:00

Before you read the remainder of this post, go out and read Hume's The Natural History of Religion if you haven't already read it. Here is the best copy I can find online

All right, I'm sure no one went and read it. But you should. Until last night, I hadn't read it since I was in high school and I'd forgotten just how great it is. Here are three reasons you should read it. First, it's hilarious. It's just about the only work of philosophy I've ever read that is laugh-out-loud funny in places. I'll provide some evidence of this below. Second, it's an interesting application of Hume's general methodology of searching for a wholly naturalistic account of the operations of our minds. Rather than apply this theory to general metaphysical and epistemological issues as he does in the Treatise and the Enquiry concerning Human Understanding, he here applies it to ordinary human thought about religious issues. So it provides some considerable evidence for the view that interpreting Hume as some sort of naturalist is the proper way to understand his general philosophical project. Third, it's Hume, what more do you need to know to be convinced that reading it is worth your time?

If you're still not convinced, maybe I can provide some textual evidence that this is worth you're time. I'm just going to quote some of the funnier passages that I found here. We begin with Hume, man of the people:

The doctrine of one supreme deity, the author of nature, is very ancient, has spread itself over great and populous nations, and among them has been embraced by all ranks and conditions of men: But whoever thinks that it has owed its success to the prevalent force of those invincible reasons, on which it is undoubtedly founded, would show himself little acquainted with the ignorance and stupidity of the people, and their incurable prejudice in favour of their particular superstitions.

Hume on communion:

A famous general ... having come to Paris for the recovery of his wounds, brought along with him a young Turk, whom he had taken prisoner. Some of the doctors of the Sorbonne ... thinking it a pity, that the poor Turk should be damned for want of instruction, solicited Mustapha very hard to turn Christian, and promised him, for his encouragement, plenty of good wine in this world, and paradise in the next. These allurements were too powerful to be resisted; and therefore, having been well instructed and catechized, he at last agreed to receive the sacraments of baptism and the Lord's supper. The priest, however, to make every thing very sure and solid, still continued his instructions, and began the next day with the usual question, How many Gods are there? None at all, replies Benedict; for that was his new name. How! None at all! cries the priest. To be sure, said the honest proselyte. You have told me all along that there is but one God: And yesterday I eat him.

Some more of Hume's distinctive brand of religious pluralism:

How can you worship leeks and onions? we shall imagine a Sorbonnist to say to a priest of Sais. If we worship them, replies the latter; at least, we do not, at the same time, eat them. But what strange objects of adoration are cats and monkeys? says the learned doctor. They are at least as good as the relics or rotten bones of martyrs, answers his no less learned antagonist. Are you not mad, insists the Catholic, to cut another's throat about the preference of a cabbage or a cucumber? Yes, says the pagan; I allow it, if you will confess, that those are still madder, who fight about the preference among volumes of sophistry, ten thousand of which are not equal in value to one cabbage or cucumber.

Here Hume demonstrates grudging respect for other cultures:

There occurs, I own, a difficulty in the Egyptian system of theology. ... It is evident, from their method of propagation, that a couple of cats, in fifty years, would stock a whole kingdom; and if that religious veneration were still paid them, it would, in twenty more, not only be easier in Egypt to find a god than a man, which Petronius says was the case in some parts of Italy; but the gods must at last entirely starve the men, and leave themselves neither priests nor votaries remaining. It is probable, therefore, that this wise nation, the most celebrated in antiquity for prudence and sound policy, foreseeing such dangerous consequences, reserved all their worship for the full-grown divinities, and use the freedom to drown the holy spawn or little sucking gods, without any scruple or remorse.

And one more quick one:

Is it strange, when mistakes are so common, to find every one positive and dogmatical? And that the zeal often rises in proportion to the error? Moverunt, says Spartian, & ca tempestate, Judæi bellum quod vetabantur mutilare genitalia. [Trans.: The Jews at this time initiated war because they were forbidden to mutilate their genitals.]

Yglesias on Kant

2004-06-03T21:09:00.000-04:00

[I hereby warn the reader that I'm not very satisfied with this post. I doubt it lives up to the (pretty low) standards I've set with my previous posts. Still, I'm going to post this, because (i) I spent some time working on it and (ii) I think I may be on to something here and (iii) this way I can say that I did in fact post something today. Perhaps what I'm really trying to get at will come to me over the next few days, and I'll be able to write another post that more clearly expresses what I'm on about here. Or perhaps someone else will stop by and show me what I'm actually getting at--and why it's way off the mark. Who knows. Also, I wonder if all I end up saying here is that I don't much care for Kant's absolutism. I hope it's something more substantial than that, but I have my doubts.]

Quite a while ago, Matthew Yglesias wrote the following post on Kant. What he said there has stuck in my mind, and I think I've finally figured out why. Here's a problem he thinks Kantian ethical theory encounters:

One problem with the Kantian approach, I think, is that it becomes hard to explain what's bad about certain events that don't have agents behind them. If a giant comet were to slam into Tokyo one day, I don't think any of us would say that the comet acted wrongly -- it's a comet, no agency, no culpability, no blame. And yet, we want to be able to say that it's a bad thing that all of those people died. In a consequentialist scheme it's easy to say why that would be bad even if no one did anything wrong -- it's bad because the world is now worse. On a Kantian scheme, though, I'm not sure what there is to say. The comet hasn't "failed to respect the people of Tokyo as ends in themselves" because it could hardly have respected them any differently by cruising past without harming anyone.

As several people pointed out in the original comments on Yglesias's post, Kant was only trying to give an account of morally bad actions, and so it's not a problem for his view if we can't use it to explain the non-moral badness like the badness of a comet's killing a large number of people. Moral badness seems to be limited to badness that is somehow dependent on the actions of rational beings; the badness of a comet's killing many people isn't badness resulting from the actions of rational beings; thus it's not clear that Kant is intending to provide an account of the nature of the badness of a natural event like a comet's killing people.

The Problem

This response seems right to me. Nevertheless, I think there is something that Ygelsias is onto here. (I should note that I'm no Kant scholar, and so you should take all this with a considerable grain of salt.) It's not that Kant can't explain why a comet's killing people is bad; he can provide some account of non-moral badness that explains why this is bad. It is, rather, that it's not at all clear that he can allow for the common-sense notion that considerations concerning the non-moral value (i.e. non-moral goodness and badness) that will result from certain courses of action is relevant in deciding what we ought to do in cases where moral value (i.e. moral goodness and badness) are also in play.* For Kant, it seems that moral value always trumps non-moral value, however great the non-moral considerations might be and however miniscule the moral ones might be. In particular, he seems to think that you may never create moral badness in the interest of avoiding non-moral badness or securing non-moral goodness. And, in your deliberations, you needn't even consider how much non-moral badness can be avoided or how much moral goodness can be secured by doing something immoral. It doesn't even matter, as non-moral value is always trumped by moral value.

So Kant can allow for a notion of non-moral value, and so he can provide some account of how a comet's killing a large number of people is a (non-morally) bad thing. But this notion of non-moral value doesn't really have any teeth when we're also concerned with moral value. In an important sense, non-moral value doesn't really matter when we're also concerned with moral. Non-moral value just doesn't have all that much weight. The problem for Kant, then, is not that he can't explain why the comet's killing people is bad, but that he can't explain why the nature and extent of the non-moral badness that would result from this happening is something that should concern people in making their moral decisions.*

And this might seem rather counterintuitive, as it does seem part of common-sense morality that you need to weigh moral and non-moral value against one another. Indeed, it seems to be part of common-sense morality that sometimes it may be permissible (and maybe even obligatory) to create some moral badness in the interest of avoiding non-moral badness or securing non-moral goodness.**

Kant and Moral vs. Non-Moral Value

Let's return to the comet example to see what I'm talking about. Suppose you become aware that the comet is about to smash into your town, but that no one else knows about this. You have time to warn some people, and for them to escape with you. But suppose that the only way you can convince them quickly enough for all of you to escape safely is by making a false promise. If you tell them the truth about the comet, they simply won't believe you about the comet and it will take too long to convince them that you know what you're talking about. But you suspect that they'll be happy to leave the city immediately if you promise to give them $1,000 apiece to leave with you. And suppose you can't think of any other way to insure they get out of the city in time. Do you make this false promise in order to avoid the considerable non-moral badness of these people dying?

It seems Kant has to say no, as there's an absolute prohibition on making false promises. Making the false promise would entail bringing into existence a sort of moral badness that cannot be outweighed by the non-moral badness you avoid by getting the people out of town safely before the comet strikes. According to Kant, you create moral goodness or badness by adopting certain maxims as the maxims of your action: that is, by acting on certain principles. Your action creates moral badness if it is action on a maxim that you cannot universalize, and he thinks you can't universalize this maxim in the appropriate way. (For the details, see Kant's second example in section II of the Groundwork. There he's talking about a case of false promising in the interest of securing some non-moral goodness for oneself as opposed to others, but I think the same argument he uses there will work perfectly well in this case.)

But, one might wonder, isn't there some point at which creating some moral badness will help to avoid enough non-moral badness or create enough non-moral goodness that it overrides the moral badness? Suppose it was a matter of your false promise saving the entire town of people? What about an entire city? Isn't there some point at which the balance is tipped in a way that makes the false promise at least permissible? So far as I can tell, Kant's answer to all of these questions is going to be no. Why? Because he can't allow for any notion of non-moral value tipping the balance here. This, of course, underlies his absolutism, his belief that moral prohibitions are absolute prohibitions (i.e. prohibitions that a person may not flout in any circumstances). If being able to secure enough non-moral goodness or avoid enough non-moral badness made it permissible to ignore these moral prohibitions, then they wouldn't be absolute in the way Kant thinks they are.

The Relation between Moral and Non-Moral Value

What, exactly, is the Kant's view about the relation between moral and non-moral value that underlies this view? What justifies these absolute prohibitions on sacrificing moral value for non-moral value? I can think of two possible views that Kant might have in mind here, and I'll explain them by discussing what they tell us about the comet example.

(1) The first view is the one I was assuming in my discussion above. This is that view that Kant thinks that there is both moral and non-moral value, and that moral value always trumps non-moral value. When considerations of both moral and non-moral are relevant, non-moral value always losses out to moral value. It is never permissible to sacrifice some moral value in order to secure any amount of non-moral goodness or avoid any amount of non-moral badness.

What does this view tell us about the comet case? It tells us that making a false promise and getting people out of town to avoid the comet would be a good thing (i.e. it would lead to some non-moral goodness), even though it is done through making a false promise. We can still say that this leads to some non-moral goodness. Nevertheless, this is not permissible. For making the false promise would also bring into existence some moral badness, and that outweighs any possible non-moral goodness it might bring into existence. The moral badness created by making a false promise trumps any non-moral value you can secure by saving people from the comet, and so it's not permissible to make a false promise in this case.

(2) The second view is one that denies that there is the sort of non-moral value that I've been assuming in my descriptions of the case above. According to this view, there is no coherent notion of non-moral value that is independent of the notion of moral value.*** If this view is correct, immoral action can never lead to any sort of goodness, nor can it allow us to avoid badness. The only things that have value are those things that result from what is morally permissible.

What does this lead to in the comet case? Allowing people to avoid dying from the comet would not be a good thing if it were done through making a false promise, since the moral badness of the making a false promise would rule out the possibility of any value coming from helping people to avoid the comet. So we cannot say, as we could on the previous interpretation, that making the false promise would lead to non-moral good. If we take this interpretation, then the problem with making the false promise is not that the moral badness trumps whatever non-moral goodness we secure. The problem is instead that making the false promise leads only to the moral badness that false promising results in; it leads to no goodness of any sort.

I certainly don't know enough about Kant to determine whether the first or second view is his actual view. But I certainly think he should argue for the first, as, for reasons I'll now discuss, I find the second implausible. For Kant, moral goodness and badness come through acting on maxims of rational beings. If the second view is true and no other sort of value can be made sense of independently of moral value, then we can't makes sense of the notion of value independent of moral value. So, if there are no rational beings, there will be no value.

But this just doesn't seem plausible to me. First, it seems to imply that any two worlds where there are no rational beings are of equal value--nothing in those worlds has any value. But imagine two worlds--one where non-rational sentient beings suffer terrible pain frequently, and one where they don't. Doesn't it seem that the latter world, at least insofar as it is a world with less pain, is of higher value than the latter? It does to me. Second, the very same problem arises within worlds where there are rational beings in cases where rational beings play no role in the situation. Imagine two situations not involving rational agents in any way--one in which non-rational sentient beings suffer terrible pain frequently, and one where they don't. Again, it seems the latter situation is better than the former. It seems to me, therefore, that we do have a notion of non-moral value that can be understood independently of moral value, and that any theory denying this is an implausible one.^

So let's assume Kant is going for the first view.

Conclusion: The Problem Again

I've now described what I take Kant's view to be, and I've described why his absolutism might seem to lead to a problem when we compare what it tells us with our everyday moral thought. I hope his position that moral value always trumps non-moral value at least seems somewhat strange. While there need be no obvious method for weighing moral value against non-moral value, it does seem that it's part of moral common sense that it is sometimes acceptable to trade off moral value for non-moral value. It seems we think it is sometimes acceptable to do something wrong in the interest of securing a good deal of non-moral goodness or in the interest of avoiding a good deal of non-moral badness.

Moreover, this problem seems especially pressing for Kant. According to him, moral value comes into play whenever a person is acting in accordance with maxims; and that seems to be pretty much whenever a person is acting. So, given the extent to which Kant's conception of moral value will play a role in the ordinary circumstances of human life, I'm tempted to say that Kant can't really account for non-moral value and its genuine importance. Non-moral value, we think, really does matter when we're deciding what we should and should not do. That is, things that don't result from human action in accord with maxims really do matter when we try to determine how we should act in some situation. But it doesn't seem that Kant can really allow for this. Since moral value comes into play whenever people act on maxims and acting on maxims seems to go on pretty much all the time, moral value is going to be in play pretty much all the time. And since moral value always trumps non-moral value, non-moral value is going to be left with no role to play pretty much all the time. In most real world cases, it might as well be the case that Kant has no notion of non-moral value at all, for, if he is right, the non-moral value in which our actions result will rarely have any role in determining how we ought to act. The problem, then, is that the accords non-moral value a secondary status that makes it almost negligible in most cases, and, at least to me, that seems pretty counterintuitive.

*. I admit that this is a little too crude. It certainly seems this non-moral badness is going to matter when you consider the duties a person might have to help others avoid being killed this way by warning them of the impending disaster or providing them with transportation away from the area will the comet will strike and so on. I hope what I say below will make clearer just what I'm talking about.
**. Now, according to some forms of consequentialism, there is going to be nothing wrong, there is no moral badness created, in any action that maximizes non-moral value. There won't be any moral badness created by acting in this way, provided that it maximizes non-moral value. The reason is that there is only non-moral value; they deny the distinction here. It's all a matter of maximizing non-moral value, and so there is no special moral badness created by doing something in order to maximize non-moral value.
***. For a paper arguing for a position remotely like this, see Philippa Foot's "Utilitarianism and the Virtues." Reading her paper suggested to me that this is a possible position.
^. I suppose it's possible to argue that we're really appealing to some notion of moral value in the cases I'm considering. Here's how one might argue this. What's really going on here, they could claim, is that we're thinking in the following way: if there were a rational being (a sort of ideal observer) who had to create one world (or one situation not involving rational beings) rather than the other, which should he choose? The issue, then, is one of what maxim(s) of the action of a being choosing between the worlds (or situations) would be universalizable. So this accounts for what we think of as a notion of value that is independent of moral value. It is a notion of value, and it is a notion we can apply in all situations in our world and in worlds where there are in fact no rational beings. But it is not a notion where we can understand this type of value independently of the notion of moral value. I really don't know what to say about this response, but it doesn't strike me as prima facie plausible. The problem, of course, is that it turns all value into moral value, which just strikes me as being wrong. (I admit that's not really an argument against the view--but, hey, I can't do everything here, now can I?.)

Hume's Dialogues concerning Natural Religion

2004-06-02T17:16:00.000-04:00

I finished reading the Dialogues here recently, and they were as satisfying as I remembered them being. I'm largely in agreement with pretty much everything Hume has to say, and I think his case against the argument from design is about the most convincing negative case against some argument that I've seen. It seems clear to me that Hume has more or less shown that the argument is worthless as support for anything resembling traditional theism. So, needless to say, you should read the book if you haven't, and especially if you find the argument from design to have some sort of prima facie plausibility.

I'll say that I'm also in agreement with the section of the book that discusses something like the problem of evil. When we're in the context of the argument from design, it's probably better to talk about the 'problem of imperfect'; but the main idea is the same. If God's so great and He's the creator of everything, why does the universe seem so awful in so many ways? Why are people so rotten to one another, why is there the amount of natural evil that there is, why is life so difficult in so many ways, etc.? It just doesn't look to be the sort of universe that a God-like being would create, and so maybe the argument from design ought to lead us to conclude that the universe as a whole was created by some being who isn't quite God-like.

I think the response to this worry to which Hume seems sympathetic is the only one that works: God's intentions and plans are largely inscrutable to us. We can't fathom what God is up to, and, even if we could, we don't know enough about the universe to see how his plan is realized in it. I admit that this response isn't terribly satisfying. One would certainly like to be able to say something more substantial about this problem, but it's not clear what it would be. Nevertheless, I don't think it's totally fair to claim that this response is ad hoc, or that it's a cheat. There are very good reasons to think that the nature and ways of a being like God would be largely incomprehensible to beings like we are. And, our ever-increasing scientific knowledge notwithstanding, there's no doubt that we're profoundly ignorant about a lot of what goes on in the universe. So, if you start out a theist, the fact that you'd accept some sort of inscrutability doesn't strike me as being all that surprising.*

One more issue about the Dialogues seems worth discussing here. The edition that I've been reading is the one edited by Richard Popkin and published by Hackett. And, in his short introduction, Popkin claims that, from the evidence we have to go on, Hume accepted some very minimal belief in a deity of some sort. Having just read the text again, it seems to me that there's some plausibility to this thesis. Here's why. None of the characters ever seriously entertains a form of atheism or agnosticism, and Philo, who's given most of the longest passages and best arguments in the Dialogues and whose general philosophical views seem closest to Hume's, seems to accept something of this sort. Philo never denies the existence of a deity of some sort; he simply denies the ability of human reason to discover anything substantial about what such a being is like. That Hume agrees with this is, I think, the most we can glean from this text about Hume's own religious views. It seems clear that he has no sympathy for organized religion, or for any religious views that purport to describe the nature of God, His intentions, or how and why He created the universe as He did. Nevertheless, it may seem that he accepts that we have reason to think there is a God, a creator of the universe who is, in some ways, similar to a human intelligence.

What, then, is it that Hume thinks? The only positive religious claim that is given respectful treatment here is the bare claim that we have reason to think that the cause of the universe as a whole is somewhat similar to a human intelligence.

But does acceptance of this minimal thesis mean he ends up being a theist? It's very hard to tell. First, of course, one might wonder whether this fairly vague positive view is really enough to amount to some form of theism. But let's put that issue to one side. Even if it is enough to support some form of theism, it's often difficult to tell whether Hume means to be advocating such a position here--or, if he is advocating it, just what he means to be advocating. The problem is that it often seems Hume's advocating this position amounts to little more than a description of what he thinks is an inevitable human tendency to think this way. What he often seems to claim is simply that, given how our minds actually work, we're bound to think something like this about the origin of the universe; we're bound to think the universe was created by something like a human intelligence.

Yet it's at best somewhat unclear that he thinks forming beliefs in this way is reliable. It may simply be that he thinks we have a brute instinct to think in a way that insures we'll see the world as resulting from some human-like intelligence, and it's not clear that that isn't a debunking account of the plausibility of theism.** So his position might be something like: Sure, we bound to think this way about the creation of the universe as a whole, but that doesn't mean that we have good reason to do so. For more support that this is a debunking explanation, see his The Natural History of Religion, where the explanations of various religious beliefs certainly seem to be ones that suggest those beliefs simply aren't plausible. There Hume's explanation of religious beliefs is that they ultimately result from certain human frailties, from our fear and ignorance of the larger world. That sort of explanation certainly looks debunking to me.

So I'm not sure I buy that we have good reason to think that Hume believed there is good reason to think the universe as a whole was created by something resembling a human intelligence. I don't think there's convincing evidence that he thought this way false, but I don't think there's convincing evidence that he thought it was true, either.

*. Importantly, though, appealing to inscrutability doesn't get the theist off scot-free. I think there are problems with inscrutability, and these problems are especially pressing for someone who also wants to argue that morality is somehow based on the will or activity of God. I'll get around to explaining why I think this when I start my little series on the Divine Command Theory. But there's something you can look for.
**. Of course, if we think this is a debunking explanation, it's going to have consequences for what we think about the rest of his views in metaphysics and epistemology. For Hume often presents himself as providing a wholly naturalistic account of the mind. That is, he claims to be more interested in describing the psychological principles that lead us to think as we do than he is discussing whether we're justified in thinking as we do. And what I take him to be claiming here is that there are certain psychological principles that lead people to think in such a way that they're pretty much bound to find theism (or something similar) plausible, in much the same way that there are certain psychological principles that lead people to think things stand in causal relations, to think there is an external world, to reason inductively, and so on. If we take the naturalistic explanation of our religious thinking to be debunking, then it seems we're going to have good reasons to take his naturalistic explanations of much of our thinking about the world to be debunking.

Tendencies and Dispositions

2004-06-02T16:43:00.000-04:00

Is having a tendency to do something the same as having a disposition to do it? I'm genuinely curious about what people think.

Here's how this came up. I was having one of these fruitless internet debates with someone about the permissibility of homosexuality. Eventually, the problem of just what constitutes homosexuality came up, and this person went to a dictionary. Here's the main entry he come up with for the adjective 'homosexual':

having a tendency to direct sexual desire toward people of the same sex.

I doubt that's verbatim, but it's close. Appealing to this definition, I claimed, and thought true at the time, that being homosexual is perfectly consistent with never having had homosexual intercourse, and indeed with never having had the desire to have sexual intercourse with someone of the same sex. I took this to be true since I was understanding a tendency to be a disposition, and it's perfectly possible for something to have a disposition to do something and yet never do it. A fragile glass (i.e. a glass with a disposition to break if dropped, etc.) can be fragile even if it never in fact breaks. Similar, I thought, a person can be homosexual even if they never manifest a desire to have sex with someone of the same sex--perhaps they just never get into the right situation. (Admittedly, sexual desire doesn't seem to work this way. But I don't see why it couldn't in some situations.)

Now I'm not sure that this is right if we're talking about tendencies as opposed to dispositions.* If we're going to say that someone has a tendency to do something, we may need to claim that it's something they in fact do, at least from time to time. Why? Well, it seems to me that you have a tendency to do X just in case you tend to do X. And it seems to me that saying that you tend to do X implies that you do in fact do X, at least from time to time.

Suppose I say that Bob tends to wear a red shirt on Mondays. Does this imply that Bob does sometimes actually wear a red shirt? It seems to me that it does. Even if Bob is disposed to wear a red shirt on Mondays but never does so (perhaps something's always coming up and interfering with his manifesting that disposition), I don't think I'd say he has a tendency to wear a red shirt on Mondays. If I say that he has a tendency to wear the red shirt, then it seems I'm saying he does wear a red shirt sometimes (and maybe even most of the time). And if a person has a tendency to do X just in case she tends to do X, then Bob has a tendency to wear a red shirt on Mondays only if Bob tends to wear a red shirt on Mondays. Furthermore, since Bob's tending to wear a red shirt on Mondays implies that he does in fact sometimes wear a red shirt on Mondays, Bob's having a tendency to wear a red shirt on Mondays implies that he does in fact sometimes wear a red shirt on Mondays. But Bob's having a disposition to wear a red shirt on Mondays doesn't imply that he does in fact sometimes wear a red shirt on Mondays, and so Bob's tendency to do this is distinct from his disposition to do it.

So here's our short argument that tendencies aren't dispositions.
(1) If you have a tendency to do X, then you tend to do X.
(2) If you tend to do X, then you do in fact sometimes do X.
(3) If you have a tendency to do X, then you do in fact sometimes do X. (from 1 and 2)
(4) In fact sometimes doing X is not a necessary condition for having a disposition to do X.
(5) Having a disposition to do X is not the same thing as having a tendency to do X. (from 3 and 4)

What do people think? Is there something wrong with this?

*. If it turns out that there is this difference between tendencies and dispositions, then I think the definition of 'homosexual' will need to appeal to a disposition rather than a tendency.

Getting Clear about Moral Obligations

2004-06-02T12:38:00.000-04:00

Moral obligations can be puzzling at times, as this post reveals. The problem isn't so much that we aren't sure where we have obligations and where we don't, thought that can be a vexing enough problem at times. I think the real problem is that there's some deeply confused folk thinking about obligation that just doesn't hold water, and a good deal of it is manifest in the post linked to above. (I'd have restricted this to a post in comments on that post, but there's this annoying 1000-character limit on comments posts; and I crank out well over 1000 characters just getting warmed up.)

First up, the connection between responsibility and choice.

But how is it that a person can simply be "born into" a responsibility without their having made a choice to freely accept that responsibility or actually do something to come into it? What if you were born into citizenship of a country that demanded one or two years of military service of you as your 'responsibility', and that military service happened to include your participating in genocide? If you can have some 'responsibility' automatically, by birth, just because the people in control say so, wouldn't that mean that you would have a responsibiltiy to do whatever they say?

Now, of course, this is a straw man. No one is going to be arguing that your responsibilities are determined in all details by the fiat of the person or group to whom the duty is owed. But this isn't to say that the person or group to whom you have a duty doesn't have some say about how that duty should be discharged, but simply that they don't have complete control here.* So, even if you think that people can have responsibilities that they haven't explicitly chosen to accept, you needn't think that the nature and extent of those responsibilities are somehow set by the person or group to whom you owe the relevant duty.

But is there really anything so mysterious about your having a duty that you didn't explicit choose to place yourself under? I certainly don't have any general theory of how one comes to have such duties, but it doesn't seem to me that there's anything odd about the idea that we have them. That we have such duties seems to be part of moral common sense. The usual examples are examples of duties that we owe to our parents. Surely, you never choose your parents; nor, at least before you hit adolescence, do you have much of any control over the various things they do for you that are for your own benefit. Nevertheless, if your parents are halfway decent human beings who are minimally competent in performing their parental duties, it seems you have certain obligations to your parents that are based on things they've done for you before you were able to do anything that could be construed as consenting to be placed under an obligation. Good parents provide you with benefits that you need, and for that you owe them something. Again, though, there are limits. You certainly don't have to do absolutely anything they tell you, and you don't owe them whatever it is that they may want.

And there are more prosaic examples of the same sort of thing. Suppose your friend gives you a gift that you didn't ask for and weren't expecting to receive. It seems to me that you're then under some sort of moral obligation to return the favor--and, at the very least, you're obligated to thank her for the gift. However, it's not clear that, in a case of this sort, you have done something that places you under either of these obligations. What places you under the obligation is the fact that your friend did something, and not that you did anything.**

Can we take any lessons away from these cases and apply them to the case of duties you owe as a citizen? It seems that maybe we can. What these examples have in common is your being in a special relationship with someone. The person to whom you owe this duty is your parent or your friend, and so you owe them something even if you didn't act in some way that placed you under the obligation. It's not clear that you'd be under a similar obligation if the people didn't stand in this relationship to you. (If, say, some total stranger comes up to you and hands you a gift, do you own them anything in return? Maybe you do, maybe you don't.)

Now all we need to argue is that being a being a fellow citizen with other people is a special relationship of this sort. That is, we need to argue that citizenship is a special relationship to your fellow citizens. Does it look that way? It's arguable that it does. For being a citizen provides you with certain special rights and abilities that you wouldn't have otherwise. Thus it may also place you under an obligation to your fellow citizens, without whom you wouldn't possess these rights or abilities. You receive certain benefits from being a citizen, and you're going to need to repay them--irrespective of whether you've done anything that we can construe as an explicit act of accepting citizenship and its attendant responsibilities. Still, as in the cases above, this doesn't mean that other citizens may demand whatever they want from you, anymore than the duties you owe to parents and friends allow them to demand whatever they want from you. There are limits to the extent to which the people to whom you owe the duty are able to control what you need to do to discharge that duty, but nevertheless the duties are genuine ones.

Now, something about the worry that unchosen demands are going to be unacceptably demanding or arbitrary in their extent.

And what if the obligations placed on you did not force you to engage in immoral action[?] Perhaps the State decides that every citizen has the responsibility of working in government factories until they are 65--would that mean that you would actually be obligated to do so because you were born in an area controlled by that State? I would think that most people wouldn't agree that you can be obligated to be a slave to the government just because it is demanded of you. But is it possible to draw a line between instances that seem unacceptable, and this grand old American way you're promoting in which I am supposedly responsible to influence the government if I don't approve of what it's doing?

I guess I really don't see the force of this worry. It's clear enough that the proposed demand is far too much to ask of someone for the rights and abilities they receive by being made a citizen. Then the only real worry is that we can't draw a line between this and other cases. Can we draw a clear line between one sort of case and another? I doubt it. Is this a problem for the very idea of unchosen obligations? I doubt it.

For it certainly doesn't seem to be such a problem in the cases of the examples I gave above. For instance, you seem to owe your parents something for what they've done for you before you were able to choose to remain their charge, and it's clear certain things are owed and certain things aren't. It's clear that a not negligible amount of respect and gratitude is owed to them for what they've done for you. And it's clear that giving them all the money you make for the rest of your life isn't something that you owe to them. So there are clear cases, and thus we know that there's a line and that some things fall one side of it and other things on the other side. But there will, of course, be borderline cases. If your mother becomes seriously ill and can no longer take care of herself, do you have a duty to provide her with a place to live in your home and with around-the-clock care? There may be no clear answer. It's going to depend on how good a mother she was, on how much of a burden your doing so would be for you, on what the other options are, and so on. The difficulty of these cases suggests that we're not going to arrive at any simple formula for drawing a line here. But I don't see how the difficulty of cases of this sort shows that there's something illegitimate about unchosen obligations to your parents. It seems to show, rather, that there is something illegitimate about the demand for a clear line.

Now for some stuff on responsibility, action, and causation.

I'm always wary of claims that a person can be responsible for something by not acting--after all, isn't responsibility generally understood as something that comes about when you are the cause of something? You are certainly not the cause of something if you refuse to even become involved in the situation (assuming you had no part in setting up that situation in a way such that your earlier actions were the cause of later occurences). Causing something require[s] action, not inaction.

This, I take it, is the basic argument of the passage:

(1) A person is responsible for a state of affairs only if that person causes that state of affairs.
(2) A person does not cause a state of affairs simply by refusing to intervene and stop that state of affairs from coming to pass.
(3) A person is not morally responsible for a state of affairs simply because she refuses to intervene and stop that state of affairs from occurring. (from 1 and 2)

The basic idea here is that causing a situation is a necessary condition for being held responsible for that situation, and that a person who doesn't act to stop a situation doesn't thereby cause that situation. So a person isn't responsible for a situation simply in virtue of the fact that she failed to step in and keep it from occurring.

I've got objections to both premise (1) and premise (2). Let's look at premise (2) first. The philosophers out there will probably notice that there are some dubious assumptions about causation being made in this premise, and the problem is that there are accounts of causation according to which premise (2) is false. Take a crude counterfactual analysis of causation, according to which the correct analysis of "event c caused event e" is the following:

(C) If event c had not occurred, then event e would not have occurred.

It seems clear that this account of causation conflicts with the claim that inaction never amounts to causation. Consider a case in which a person didn't act and in which we stipulate that her action would have prevented the relevant event from occurring. Let's call the event of your not acting A, and let B be the event we might hold you responsible for. We're supposing it's the case that if you had done not-A (i.e. if you'd acted), then B would not have occurred. According to a counterfactual analysis of causation, that's all you need to truly say that A caused B. If event A had not occurred, then event B would not have occurred. So A caused B. So there you have the causal causation, and there you your inaction meeting the necessary condition for resulting in your moral responsibility for its effects. You can, if this analysis is correct, cause a state of affairs simply by failing to intervene and keep it from coming about.

Of course, we don't want our rejection of the argument above to hinge on a contentious account of causation. So let's just suppose that we have some special notion of causation in mind that rules out inaction of this sort counting as causation. Let's suppose, that is, that we accept some theory of causation according to which causation by a person requires that person to act. For short, let's call this person-causation. And this brings us back to premise (1). Naturally, we'll need to reformulate the argument slightly because of our focus on person-causation. In particular, premise (1) should now read:

(1) A person is responsible for a state of affairs only if that person person-causes that state of affairs.

And similar alterations will need to be made throughout the argument. I leave making those alterations as an exercise for the reader.

Admittedly, I'm not sure what person-causation would be--but, fortunately, it doesn't really matter. For it seems clear to me that you can be responsible for all sorts of things that you don't person-cause, and so it seems clear to me that our reformulated premise (1) is false.

Here's an example to show this. You're taken to the emergency room with a bullet wound, you die as a result of that wound. The wound wouldn't have killed you if the doctor on duty had done something--and suppose it's something pretty minor--to stop the bleeding. But he didn't do anything but stand by and watch you bleed to death. This process took a few hours, and he wasn't working with another patient or unaware that he was supposed to be treating you or anything of that sort. He knew how to fix your wound, and he was simply more interested in watching someone bled to death than he was in doing what he could to keep you from dying of your would. Now, the doctor didn't person-cause you to have this wound; he wasn't the shooter, after all. But is he responsible for your death? It certainly seems so, and yet he didn't act in a way that person-caused your death.*** So it seems that your person-causing an event isn't a necessary condition for your being responsible for it. It seems, therefore, that premise (1) is false.

Unsurprisingly, there's a special relationship involved in the doctor case as well. But, importantly, it's not a personal relationship; it's a relationship that arises from the particular role that the doctor is supposed to play. Because of the role the doctor, qua doctor, is supposed to play in helping out his patients, it's possible for him to be responsible for their deaths even if he doesn't cause whatever ends up resulting in their deaths. If gross negligence on the part of the doctor ends up preventing him from keeping his patients alive, he's responsible for their deaths. (I'll just point out that similar obligations are going to be present in cases where people stand in personal relationships to one another. Take a case where a parent stands by and watches her child drown, even though she could have stepped in and kept this from happening with little or no risk of being injured herself. Is she responsible for the child's death? It seems so. More responsible than a stranger would have been in the same situation? Again, it seems so.)

So when children starve in third-world countries, unless a person has somehow caused the food shortage, it is ridiculous to say that he or she is responsible for their deaths. It might be praiseworthy, even morally good, to help people in such situations. But that does not mean that you have an obligation to help, or that inaction is immoral and responsible. Why should it be impossible for a person to simply live their lives without interfering with the freedoms of others, desiring that their own freedoms not be infringed upon as well, without their being responsible for every act of evil that they didn't try to prevent?

Again, there's simply not enough appreciation of the details of human interconnections here. There are lots of people whom almost anyone would hold responsible for inaction in the deaths of these children even if those people didn't do anything that caused the food shortage in the first place. Suppose Mary works for Oxfam, and she's been embezzling money that would have gone to help children in some far away country. With the money she steals, dozens (or hundreds) of children could have been saved. Is Mary responsible for their deaths? It sounds harsh, but I don't have much trouble saying that she is. Did she cause the food shortage? Nope. And suppose Bob flies the plane that is supposed to deliver the food to the children. Bob gets the cargo, and, even though he's supposed to make his run immediately, he just doesn't feel like delivering it for a few weeks. In the time between when Bob could have gotten the cargo there and when he did, dozens (or hundreds) of children who could have been saved pass away. Is Bob responsible? Yep. Did he cause the shortage? Nope.

So the moral here is that you can be held responsible for all sorts of things you don't obviously cause as long as you play some role or stand in some relationship that provides you with an obligation to step in and do something that is necessary to keep those things from happening.

It seems we've now made some progress towards providing an answer to the following question:

So unless you are prepared to argue that people are, through their inaction, responsible for a great amount of evil (far outweighing the good they could possibly do, I'm sure), how can you claim that responsibility can come about through inaction?

Cases where people are responsible for inaction often seem to be cases in which people stand in special relations to one another or cases in which people play a role that provides them with special burdens to help others. So it seems we can allow for some responsibility through inaction without accepting that people are responsible for absolutely everything they could have prevented. At least in general, they're going to be responsible only when one of these special roles or relationships comes into play. That doesn't happen all the time, and so we really don't have to worry about constantly being under the obligation to do things.^

The remaining questions are: What sorts of roles and relationships matter here, and what, exactly, does one owe to others in virtue of playing particular roles and standing in particular relationships? I hope it's not surprising to find out that I won't be answering these questions here.

*. If, for instance, I borrow my friend's car for the day and he says that he'll ask me for a favor in the near future as repayment, then it seems that I have a duty to do what he asks me, provided it fails within certain, not always clear-cut, limits. So it does seems he's going to have some control over how I ought to discharge my duty to compensate him, as it's probably going to be the case that I should do what he asks me to do. If he asks me to loan him ten bucks or to let him use something of mine for a few days or something of this sort, then it seems this is what I should do. But, of course, my borrowing his car doesn't obligate me to hand over my first-born child or to murder someone who's been annoying him if either of those is what he asks for in return. There are limits here, obviously.
**. I recognize that there is a possible objection to this example. Someone might argue that you have in fact done something that places you under the obligation--namely you've accepted the gift. You could have rejected it, and so maybe there is something like a choice here. So the idea here would be that you're under the relevant obligation only if you choose to accept the gift. I'm not convinced by this, for a couple of reasons. First, the obligation you have to thank the person for the gift doesn't go away if you don't accept it. (OK, maybe you want to say it's then actually an obligation to thank them for the offer, and not for the gift. Fine, it's still an obligation that you're under even though you've done nothing to bring about your having that obligation.) Second, it's not even clear to me that not accepting the gift exempts you from being obligated to give them a gift. It may be that the mere fact that they sincerely offered you a gift places you under an obligation to make a similar offer. I'm not sure what I think about this, but there are cases where it seems plausible to me. Suppose you turn down an invitation to a party from Bob. Even if you turn down the obligation for good reason (e.g. you had to be out of town that day), it seems that, if you're going to be having a similar party in the future, you ought to at least return the favor and invite Bob. (This, of course, will partly depend on why you've turned down the invitation. If Bob is irredeemably boorish and rude, then maybe you haven't any such obligation. But that doesn't show anything about an ordinary case where Bob's a stand-up dude, and you simply couldn't make it to his party. All it shows is that these situations have lots of nuances and can be extremely complex.)
***. This, of course, doesn't imply that the person who actually shot you isn't also responsible for your death. There's more than enough responsibility to go around here.
^. There may also be cases in which one is still responsible for inaction even if one doesn't stand in some special relationship to the relevant people or play a speical role in the relevant situation. I'm not denying that this is the case; I think it probably is. But I don't doubt that these cases are fairly rare, and that they often involve a lesser responsiblity than cases in which there is a special role or relationship involved.

Abduction and Harman's Argument against Moral Realism

2004-05-31T17:00:00.000-04:00

It occurs to me that my misgivings about the response to Hume's arguments in the Dialogues have a connection to my misgivings about Harman's prima facie case against moral realism in the first chapter of his The Nature of Morality. What frustrates me about that chapter is similar to what frustrated me about the suggested response to Hume's criticisms of the argument from design: that it fails to properly respect the ways in which abductive arguments need to rely on empirical considerations. Now, this isn't something you'd expect a hardline empiricist and naturalist like Harman to fail to property appreciate, but it's hard for me to understand the argument of that chapter in a way that doesn't suggest he's making such an error.

The conclusion for which Harman argues in that chapter is that putative moral facts appear to be isolated from empirical testing, and that this raises problems for understanding them as objective facts (or, perhaps, even as real facts at all). And, like many people, he means to be motivating the worry here by drawing a contrast with science. Science is our paradigm of an objective area of inquiry; there's this important dissimilarity between science and morality; so, it seems we have good reason to suspect that morality isn't objective. That, I take, is the basic argument. (It's basically just a variation on the standard reason for suspecting that there's something suspicious about ethics--namely that ethics doesn't seem enough like science.)

Now, for Harman, this argument turns into an argument about the best explanation of certain facts because of his account of how empirical testing of theories, both scientific and otherwise, works. Empirical testing, he argues, works by seeing how well the facts posited within the theory explain our experience of the world. We test a theory by seeing how well its truth, along with the rest of our picture of the world, would explain our sensory experiences. (Naturally, for the explanation we get to offer any confirmation of the relevant theory, that theory itself is going to have to do some of the explanatory work.) Harman isn't very explicit about the details here, but I suppose they're something like this. If the theory can play some role in explaining a certain group of our experiences, then that group of experiences provides some, perhaps not very strong, confirmation of the theory. If the theory plays some (irreducible) role in the best explanation of a certain group of experiences, then those experiences provide very good evidence that the theory is true.

And, according to Harman, the problem with theories positing moral facts (i.e. the state of affairs of something's instantiating a moral property) is that these facts don't seem to have a role to play in our best explanation of any of our experiences. This suggests that we don't have very good evidence for any theory positing the existence of moral facts. But Harman goes even further, and indeed he claims that moral facts appear to be totally irrelevant to explanations of our observations. Then problem for moral realism, then, really turns out to be one arising from the apparent explanatory impotence of moral facts.

At some points, this seems to be the whole of Harman's argument. But his contrast of explanations appealing to putative moral facts with explanations appealing to putative physical facts makes more clear just what the argument is supposed to be. It seems that Harman is especially concerned with causal explanations of the mental states involved in sensory perceptions. This allows us to see what Harman's real point is supposed to be: that, given a naturalistic account of the world and of human beings, it's not clear how moral facts can have any effect on our sensory apparatus.*

The problem I have with the argument of Harman's chapter is that there's really no good argument for this irrelevance. It's just asserted, as though it were something we could determine just by thinking about it. But we certainly can't do this; we can't figure out what might explain our appearances without doing the empirical work of figuring out what the best theories of the world are. Whether Harman is right here is going to depend on just how well theories appealing to moral facts do in helping us to explain our experiences, and so far as I can tell, that's an empirical question, and not one to be settled from the armchair.

He could, of course, argue from some sort of meta-inductive claim about what our best scientific theories currently look like. These theories, he could argue, don't appeal to moral facts, and so we have good reason to think that the best explanation of our experiences won't turn out to involve appeals to moral facts. But this isn't what it seems Harman is up to. He has us look at a particular case where a moral fact is posited, and he claims it doesn't look to do any explaining of the relevant sort there. And he doesn't appeal to some general claim about what scientific theories look like in making this argument. Given that, it's hard to see why Harman thinks moral facts are unlikely to play some role in the best theory explaining our experiences of the world. We'd need to know what a moral fact would look like in order to know whether it can have the relevant sort of causal influence. And, even if Harman thinks we have some intuitive grasp of what a moral fact would be like, he can't think that we have a similar grasp of whether or not such facts have any role to play in the best causal explanation of our experiences. For the proper naturalistic theory of the world is one we arrive at by empirical inquiry, and so we can arrive at a picture of what sorts of facts are and are not part of the fabric of the world only by doing science. Furthermore, it's not clear to me that Harman can argue that we possess any a priori metaphysical understanding of what's going to count as natural and what isn't. If we had something of this sort, then maybe we could determine whether moral facts are going to fit into a properly naturalistic conception of the world. But, given Harman's conception of our thought about the world, what is natural and what isn't will depend on what our best scientific theories end up appealing to. That, and nothing else, is what determines what is natural and what isn't. So, again, he needs some sort of meta-inductive argument for this to go through.

In short, Harman hasn't brought forth the empirical evidence that would be required to draw the contrast he needs to draw, and, since he hasn't done so, he hasn't provided us with much of an argument against realism.

*. It seems Harman needs to go further than this, though. What he really needs to argue is that it seems putative moral facts would have no influence on anything that would eventually result in influence on our sensory apparatus; it needn't be a direct causal influence on our sensory apparatus. As long as we can get a causal chain from moral fact to our sensory apparatus, even if we need some mediating facts, that would seem to be enough to allow for empirical testing of the sort Harman wants.

Kinsley on the (Embryonic) Stem-Cell Debate

2004-05-31T10:46:00.001-04:00

Atrios links to a interesting editorial by Michael Kinsley on the issue of the permissibility of destroying embryos in stem-cell research. (I think I've seen Kinsley make much the same argument before. And, looking back, I see that he made some similar points in the following piece from Slate.) Now, Kinsley's focus in the former piece, from Time, isn't on the moral argument; it is instead on the fact that, for whatever reason, stem-cell research is considered controversial, whereas in vitro fertilization isn't. But there's a particular sort of moral argument underlying this, and that argument comes out in the following passage from the original Slate article. I want to talk about that argument, which I think is a pretty good one.

Here's the relevant passage:

If stem-cell research is morally questionable, the procedures used in fertility clinics are worse. You cannot logically outlaw the one and praise the other. And surely logical coherence is a measure of moral sincerity.

This seems to be the argument that Kinsley is making:

(1) You can consistently make different moral judgments about two types of action only if there is some morally relevant distinction between them.
(2) There is no morally relevant distinction between the action of destroying an embryo for stem-cell research and the action of destroying an embryo as a result of the process or providing someone with in vitro fertilization.
(3) Therefore, you cannot consistently make different moral judgments about the action of destroying an embryo as a result of the process or providing someone with in vitro fertilization.

The basic argument here is a simple one, and it's an instance of a powerful and very important sort of moral argument. (Indeed, according to someone like Hare, this sort of argument is all there is to rational moral argument.) The basic idea expressed in premise (1) is that you can't reach different moral conclusions about two acts if those acts don't display any morally relevant differences. The underlying idea is that moral discriminations have to be based on reasons, and these reasons have to involve pointing to something about the particular acts being judged and using it as a basis for the judgment. So, if you want to justify treating two types of action differently, you need to provide a reason; you need, that is, to point to something that's different between them to justify the divergent moral judgments. And, importantly,

The heart of Kinsley's argument is the claim in premise (2) that there isn't anything to which you can point to justify divergent moral judgments about destroying embryos in stem-cell research and about destroying embryos in the process of providing in vitro fertilization to couples who need it. Given this, it seems he's got you: you can't consistently make different moral judgments about the permissibility of the two types of embryo destruction. If you think the destruction involved in in vitro fertilization is permissible, then you've got to think that the destruction involved in stem-cell research is permissible; and vice versa. Similarly, if you think the destruction involved in stem-cell research is impermissible, then you've got to think that the destruction involved in in vitro fertilization is impermissible; and vice versa. To fail to reach identical moral conclusions about the two types of cases, provided that you know the relevant facts, is to flout a requirement of consistency on the formation of your moral views*; and perhaps it's even to opt out of making a moral judgment altogether.

Now, it seems clear to me that the important debate here is going to center around the plausibility of premise (2). In both articles Kinsley goes after those who think both that in vitro fertilization is permissible and that embryonic stem-cell research isn't because it involves the destruction of "human being[s] with the same human rights as you and me." I'll simplify this expression by using a term of art. Let's call any being with the moral status of an ordinary adult human being a "person." (I hope this gets at what Kinsley meant by talking about human beings with rights like ours.) His argument, then, is that these people have the very same reason to think the process of in vitro fertilization, which normally involves the destruction of human embryos, is morally impermissible. If they think that the destruction an embryo in stem-cell research is the destruction of a person, then they should think that destruction of embryos in the process of in vitro fertilization is the destruction of a person. And since they, and probably almost everyone else, think the deliberate destruction of a person is impermissible in any process**, they should think that the process of in vitro fertilization is impermissible--at least if they're going to make consistent moral judgments.

There might seem to be two easy ways out of this argument. One I won't say anything about here, though I may return to it in a later post. That response would be to claim that this argument convinces you that you ought to oppose the destruction of embryos involved in in vitro fertilization. This, of course, would be one way to restore consistency to one's moral judgments.

The other easy way out would seem to be to point out that there are differences between the processes of in vitro fertilization and the process of conducting embryonic stem-cell research. All one has to do then is claim that one or more of these difference is morally relevant, and that this morally relevant differences justifies one's divergent judgments about the two cases. That, in general, is certainly a fine way to respond to arguments of this sort.

However, I'm not sure how helpful it is going to be to a person who has the views we're considering. The problem is that their reason for objecting to embryonic stem-cell research is an exceptionally strong one: that it involves the deliberate destruction of human persons. This is a very weighty reason, and no minor difference between the two actions being judged is going to outweigh its strength in suggesting that those actions is straightforwardly impermissible. While you could certainly latch on to some minor difference between the two actions and claim it has moral relevant to retain mere consistency in judgment, it's far from clear that this would result in a tenable moral position. I think Kinsley makes the clear in the Time article when he quickly dismisses the justification that the embryos that would be used in stem-cell research were never going to fully develop anyway.*** He quickly points out that this leads to a sort of ghoulish moral view that would allow the extermination of people who are bound to die anyway. And I think the point here is that a minor distinction like that can't outweigh the fact that, according to the person we're arguing with, both of these processes involve the destruction of persons.

At the moment, I can think of only one possible distinction between the two actions that a person might argue is of sufficient moral significance to justify divergent moral opinions about the two actions. Kinsley suggests this distinction, and he may be aware of it and trying to head it off, in the following passage from the Time article in which he describes what goes on in the process of in vitro fertilization:

Successful or not, the process creates many more embryos than babies. There is a built-in presumption--really, an intention--that even most of the transferred embryos will die.

So we can imagine someone appealing to the following putative distinction: the process of embryonic stem-cell research involves intentionally destroying an embryo, whereas the process of in vitro fertilization does not involve intentionally destroying an embryo.

[I'll finish this up later. I'm feeling ill.]

*. In the Slate article, Kinsley suggests that this shows that someone (in this case, Bush) who knows the relevant facts and yet draws distinct conclusions can't have genuine moral anguish about reaching the one conclusion and not the other. If you have anguish in one case, then you'd have to have anguish in the other. I guess I don't think this is fair, as it's not clear to me that a person's emotions have to be as responsive to these demands of consistency as their thinking should be. Sure, we'd hope your emotions about these cases would display the sort of consistency that your moral judgments ought to, but it's an unfortunate fact, it seems to me, that they often fail to do so. And it's really not clear what a person could do to insure that their emotions don't diverge from acceptable judgments in this way. That, of course, isn't to defend having conflicting judgments in cases like these. I'm pretty dubious about letting one's anguish play too much of a role in determining which moral judgments one reaches.
**. Of course, this is a bit crude. For we do think it is permissible to deliberately kill persons in some cases. But I think it's clear that we're not here talking about cases of self-defense or fighting in a just war or any other case where people might normally think there is justification for deliberately killing another person. Also, I don't want to rely on something like the idea that we're here talking about the killing of innocent persons. I'm dubious about applying the concept of innocence to an embryo; it's not clear to me that that wouldn't involve a category mistake of some sort.
***. I don't know that this is really a difference between the two actions, but, for the sake of argument, suppose it is.

The Argument from Design and the A Posterioricity of Abductive Arguments

2004-05-29T17:07:00.000-04:00

I never really finished up my post on the argument from design. Here I'll explain a little more thoroughly what I was trying to get at in the conclusion of that post.

The point I was making in that post was that it seems to me that inferences to the best explanation have to draw on empirical considerations, since we need empirical evidence to determine what's really the best of the available possible explanations. This seems to me to be true in the case of our trying to determine what best explains the existence of the universal as a whole, and so it's not clear to me that interpreting the argument from design as an abductive argument rather than an analogical one is going to help its defenders avoid Hume's criticisms.

I think I need to say a bit more about why it seems to me that abductive arguments like these are going to need to draw on empirical considerations. Take a standard example of an abductive argument: a murder with characteristics a, b, and c is committed, and we're trying to figure out the best theory of the crime. We can think of our procedure in trying to solve the crime as one of trying to figure out which theory of the crime best explains the relevant data (viz. why the murder has characteristics a, b, and c). And it seems to me that, as long as there isn't an eyewitness to the crime or a confession by a key suspect or something similar, we've got to rely on empirical data about similar murders in order to determine which theory best explains this data. We'll need to draw on what we know from previous experience about which sort of criminals are likely to commit crimes with those characteristics, why they're likely to commit such crimes, how they're likely to commit them, etc. So we need to draw on empirical evidence about similar past crimes and their causes. And this, we hope, will allows us to infer a good explanation of the commission of a crime with characteristics a, b, and c in these circumstances.

What seems clear to me is that we'll need something similar in the universe case: that is, that we'll need to draw on empirical evidence about how things similar to the universe can into existence. And we can now see that Hume is arguing that the relevant empirical evidence just doesn't support the thesis that the universe was created by a God-like being.

The first important point is that we haven't seen universes being created in the past, and so we aren't arguing that we've observed a certain sort of causation in previous events just like this one we're trying to explain. The abductive inference, then, isn't going to be all that straightforward. It's not as though it's going to be something like the following: I know the best explanations of previous phenomena just like this one (i.e. the existence of the universe) was creation by God; if creation by God was the best explanation of previous phenomena just like this, it's gonna be the correct explanation of this one; so creation by God is the best explanation of this one.

If this isn't how the abductive inference works here, how can we go about figuring out which possible explanation is best? It seems to me that we're going to have to rely on some sort of similarity or analogy. The central point of the argument, then, would be that creation by God is the best explanation because it's an explanation analogous to the best explanation of similar phenomena.

But once we accept this interpretation of the argument, Hume's original arguments return to finish the job. All he needs to show are that there are the sorts of disanalogies he used to undermine the analogical argument, and I can't see why the original arguments he used to undermine the analogical argument won't work perfectly well here. First, he can argue that the relevant explananda aren't all that similar. The universe isn't all that much like the sorts of things human intelligences create, and so the analogy here fails. This would be analogous to a strategy we might use to argue against a particular theory in the case of the murder with characteristics a, b, and c.

Suppose that a person puts forward a theory that this is the work of a serial killer, but one with certain distinctive characteristics needed to account for characteristic a. We can think of this as arguing from analogy of a sort. The argument is that the best explanation of murders with characteristics b and c is usually that they were committed by serial killers, and so, by analogy, the best explanation of this crime would be an analogous one--namely that the murder was committed by a serial killer with the distinctive characteristics. Now, it seems that a good way to argue that this isn't the best explanation of the relevant phenomenon would be to point out that crimes committed by serial killers don't usually have characteristics b and c. This would show that the analogy isn't such a good one, and that we really don't have good reason to think the crime was committed by a serial killer at all. That alone doesn't establish this theory is false, of course; but it certainly seems to provide pretty good reason to think that the serial-killer theory doesn't provide the best explanation of the murder. And something like this seems to be a part of what Hume is up to in the Dialogues. He often points to ways in which the universe as a whole doesn't seem much like the sorts of things created by human intelligences.

And even if Hume grants that the universe is quite a bit like what human intelligences create, he has arguments that this wouldn't lead him to the conclusion that the best explanation of the existence of the universe is creation by God. For this putative explanation--namely creation by God--is quite a bit different a human intelligence, and so it's not clear this is a good explanation of what is quite a bit like what human intelligences create. Even if we grant we have similar effects, it may be that the empirical evidence suggests that inferring a similar cause isn't going to amount to inferring that the cause was God; rather, it would amount to inferring that the cause was something closer to a human being. The putative explanation offered isn't all that much like the explanation of things that are best explained by human intelligences, and so it's not clear it's the best explanation. Again, this would be analogous to another way of arguing against a theory of the crime in the murder example.

Go back to the putative explanation of the crime that we were considering above: that this murder, which has characteristics b and c, is the work of a serial killer, but one with certain distinctive characteristics needed to account for characteristic a of the crime. Now, another way to object to this theory would be to argue that we don't really need to attribute distinctive characteristics to the killer in order to account for characteristic a of the crime. So the best explanation isn't that the crime was committed by a serial killer with characteristic a, but that the crime was committed by a garden-variety serial killer. This would be to show that the analogy between this murder and ordinary murders by serial killers is close enough to suggest that the best explanation of this crime is that it was committed by an ordinary serial killer, and not that it was committed by a serial killer with the relevant distinctive characteristics. And, again, something like this seems to be a part of what Hume is up to in the Dialogues. He often points to ways in which the universe is, or might be seen as, similar enough to human creations to suggest that it was brought into existence by something much more similar to a human intelligence than a God would be.

So I think if Hume's arguments go through against the analogical argument from design, then they also go through against the abductive argument from design. I haven't directly argued that they do in fact go through in either case, however.

What lessons can we draw from this? First, it just doesn't look like the nature of the best explanation of the relevant phenomena is something you can determine from the armchair. I don't seem to have any best explanation-detecting sense; nor can I think of any a priori considerations that suggest God has to be the cause in cases of this sort. Thus I need to have some empirical data to draw on in trying to determine how we can best explain the existence of the universe. Second, I think this helps us to see something about the problems facing the argument from design. In particular, it helps us to see that, if this argument is going to be a cogent one, its proponents need to avoid two problems: (a) making the universe seem too much like a human product, as this suggests a cause more like a human intelligence than a God-like intelligence; and (b) allowing the universe to seem far too different from a product of a human intelligence, as this suggests a cause that is even less human-like than God is supposed to be. So the proponent of this argument needs to provide us with empirical evidence that allows us to understand the universe as similar to the creations of human intelligences--but not too similar to them.

Ethics Group Blog?

2004-05-29T14:49:00.000-04:00

I realize no one is reading this--and so it's probably the case that no one will ever read this idea--but I think it would be interesting to have a graduate student group blog on topics in ethics and cognate areas of philosophy. What areas am I thinking of? I guess normative ethics, applied ethics, meta-ethics, and political philosophy. (I realize meta-ethics may not fit in there all that well.) And if I didn't mention something you think belongs, go ahead and let me know.

I've seen several group blogs that are jointly authored by students at a single school (see here, here, and here), but I haven't seen one authored by students studying a particular subject or a related group of subjects. And I think it would be interesting to see how such a group blog would go. (Yes, I realize that Brian Weatherson or somebody else with a blog noted that there weren't any group blogs of this type a while ago. Again, though, I really can't remember when and where I heard this.)

So if you're a philosophy graduate student and you'd like to contribute to a group blog on ethics and related areas, contact me and we'll see if we can get to work on something. (You can find an email address in my user profile.) I'm willing to listen to pretty much any suggestion about how the blog should be organized, where it should be hosted, what program we should use to write it, what it should be named, and so on.

Coming Attractions

2004-05-28T16:55:00.000-04:00

I suspect that no one is reading this--and, having added the counter to the bottom of the page, I know no one is--but I want to say a bit about what should be appearing here the next couple of weeks.

There's going to be an extended series of (probably not very good) objections to a version of the Divine Command Theory. A couple years ago I wrote a very long (around 30,000 words, I think) essay in response to a short essay arguing that the objectivity of morality hinges on the existence of God. The paper is by William Lane Craig, and it is entitled "The Indispensability of Theological Meta-Ethical Foundations for Morality." (Yeah, I don't know why he needs that "for morality" at the end, either.) Anyway, this essay seems to be all over the web, and you should be able to get ahold of it if you want. But here's a copy of it for those too lazy to go out and find it for themselves. (If, for some reason, that link doesn't work, just google the title. You're bound to find it.) I don't think what I'm going to post here will be very closely connected to the essay, but some of it might be. Still, it's short and worth a look.

Anyway, I originally posted those objections--along with some positive stuff that was really warmed-over Foot and Cornell Realism--on a message board that doesn't seem to exist anymore. Since I think some of the negative stuff is at least worth looking at, I'm going to post what seems worth posting here.

That's going to be the most substantial thing coming up here. But there will be more. Here's a little list for those dying to know. First, I've got a post coming up about the ontological argument and the sort of perfection that is relevant to our understanding of God as the most perfect of all beings. (I don't know what the deal is with all the philosophy of religion stuff. Maybe it's that I've been reading quite a bit of modern philosophy lately, and the concerns of the moderns are starting to rub off on me.) Second, I may post something about the work I'm actually doing right now. That work is on the multiple realizability of moral properties and on moral disagreement and what consequences, if any, it has for moral realism. So I may be posting something on those topics here. Third, and finally, I may want to say something about moral laws and their relation to laws of other sorts.

So that's what you can look forward to in the near future here. If those topics don't sound appealing, I suppose you needn't check back here anytime soon. If you're reading this and you'd especially like to hear about one of those topics (or something connected to them), tell me about it in the comments. I may just take your advice and get to work on your suggested topic(s) first. Maybe.

The Argument from Design: Analogical or Abductive?

2004-05-28T15:55:00.000-04:00

I've been reading Hume's Dialogues concerning Natural Religion, and I'd like to say a bit about the argument from design.

I'll note in passing that Hume could tell Berkeley a thing or two about how to write a philosophical dialogue. His prose is, of course, lovely, and he doesn't ever turn his interlocutors into cyphers spouting the details of their respective positions. Each figure has a forceful personality, and they're all exceptionally quick and clever in argument. Moreover, his dialogues never degenerate into the pattern I noted in my discussion of Berkeley. As always with Hume, it's an exceptional performance in all respects.

But enough about that. I may have more to say about issues brought up in the Dialogues at some later point, but all I want to discuss now is the structure of the argument from design. I recall reading the following objection to Hume's argument(s) against the argument from design in the Dialogues. Hume's objections aren't very forceful since he doesn't adequately understand the logic of the argument. The argument, claims the proponent of this response, needs to be understood as an inference to the best explanation, and yet Hume understands it as an analogical argument. Hence few of Hume's criticisms are forceful. (Sadly, I'm unable to recall where I saw this argument against Hume. I believe I came across it when reading something about the current debate concerning evolutionary theory vs. intelligent design. It may come from Behe's Darwin's Black Box, which I flipped through once in a bookstore. Or, actually, it may have been something that Dawkins says about Hume when he discusses why he thinks the argument from design wasn't truly refuted until Darwin gave us a better explanation of design in The Blind Watchmaker. I really can't remember. Sorry.)

An Analogical Argument

What is it to understand the argument from design as an analogical argument? I think it's to understand it as something like the following:

(1) The universe as a whole is similar to a complicated artifact.
(2) Complicated artifacts are creations of intelligences like human minds.
(3) Similar effects have similar causes.
(4) Therefore, the universe as a whole is a creation of God. (from 1-3)
(5) God exists. (from 4)

Now, I don't deny that something like this is what Hume usually seems to have in mind. I won't, however, say that he never appeals to something that looks more like an inference to the best explanation; I haven't read the text closely enough to say anything of that sort.

What does Hume say about this argument? His major criticism is that the empirical evidence doesn't really gibe with this argument. First, the empirical evidence suggests that (1) simply isn't true or, at least, that the relevant similarity is a very limited one. Our empirical evidence suggests that the universe as a whole is so different from any artifact created by a human intelligence that it's far from clear that (1) is true. Second, he seems to suggest that (1)-(3) should really lead to a conclusion like the following:

(4') Therefore, the universe as a whole is the creation of an intelligence like a human mind.

And given the empirical facts about human minds and the ways in which they create things, it's a big and unacceptable leap from (4'), which is the conclusion we can draw from (1)-(3), to (4). So, even if this analogical argument shows something, it doesn't show the universe as a whole was created by a God-like intelligence. (I think we could say more about how we might add more to the argument above in order to provide a way from (4') to (4). I don't want to go into this here, however, as I don't think it's relevant to what I want to say.)

So Hume's central contention here is that the empirical data suggest that there are problems with the analogies that are relevant to this argument--namely, the analogy between human artifacts and the universe as a whole, and the analogy between our minds and a God-like mind. The analogies aren't very close, and so the argument isn't very strong.

Response

Now let's look at the response I mentioned above. The argument from design, claims the proponent of this response, needs to be understood as an inference to the best explanation, and yet Hume understands it as an analogical argument. Hence few of Hume's criticisms are forceful. Why wouldn't they be very forceful? Well, they're based on a misunderstanding of the argument. Hume's basic argument is that the empirical evidence suggests that the relevant analogies aren't terribly good ones. But if we go for an argument based on inference to the best explanation, then we don't have to worry about these disanalogies that Hume relies upon. For we don't have to think that the argument appeals to any putative analogies between artifacts and the universe or between our minds and a God-like mind.

Here, I guess, is what the argument from design is going to look like as an inference to the best explanation.

(i) The universe as a whole has the appearance of an extremely complicated and finely tuned artifact. (That is, the universe as a whole looks like it was designed by an intelligent being.)
(ii) The best explanation of (i) is that the universe as a whole was created by God.
(iii) The universe as a whole was created by God. (from i and ii)
(iv) God exists. (from iii)

Now, the response. Look, it goes, there's nothing about analogies in these premises, nor do we need to assume anything about analogies to make the relevant inferences here. Hence the various putative disanalogies to which Hume points aren't really relevant to the argument from design, provided that it is formulated correctly.

Does the Response Work?

So far as I can tell, this response to Hume's arguments simply won't work. For, it seems to me, Hume could use the very same empirical evidence to object to the argument in this formulation. First, he could use it to dispute the truth of premise (i). Given the empirical evidence we have about it, the universe as a whole certainly doesn't look like the sort of artifact that human beings tend to create. So (i) isn't all the plausible. Now, I don't think this is an air-tight objection, since it may look like my way of formulating (i) simply begs for this; and you might think there is some way to avoid this problem. For instance, you might give some more detailed and specific account of what suggests design instead of simply appealing to the appearance of design. Would this work? Maybe it does, and maybe it doesn't. I'm not sure that any other way of formulating (i) is going to avoid the problem, but I'm not sure it matters since I think the major objection to this response concerns premise (ii).

The major objection is that determining which is the best explanation of the phenomena mentioned in premise (i), however it is formulated, is something that is going to have to proceed a posteriori, and that the relevant evidence, it seems, is precisely the evidence that Hume marshals against the analogical argument from design. Before we can defend premise (ii), we need to figure out what is a good explanation a phenomenon like the universe as a whole. And it seems that the only way we can figure this out is to bring forward some a posteriori evidence about what constitutes a good explanation of phenomena like the existence of the universe as a whole. I don't see any way to figure this out a priori; nor do I see how it could be figured out a posteriori without appealing to the sort of empirical evidence that Hume thinks undermines the argument from design when it is formulated as an analogical argument. Furthermore, I don't see how the verdict to which the evidence Hume cites leads us will change if we understand the argument as an abductive one rather than an analogical one.

I may come back and say more about this later. But this'll be all for now.

Motivational Internalism and Tolerance, Part II: Some Possible Responses

2004-05-27T23:26:00.000-04:00

In the first post on this topic, I suggested that there might be some tension between accepting a form of motivational internalism and tolerating people's moral opinions. I now want to suggest some possible responses. I'm not going to argue that all of them fail; I'm not sure that all of them do.

There seem to be two general strategies for arguing that there isn't any special problem here. First, one could simply deny that we, or reasonable people in general, should tolerate moral judgments in this way; and second, one could argue allow that we should, or at least some reasonable people could, think that we should tolerate people's moral opinions, and that this is perfectly consistent with accepting motivational internalism. Let's look at these strategies in turn. (After doing some work on this post, I've come to realize that it's going to be pretty long. So I'll look at only the first strategy in this post. I'll take up the second strategy in a third post.)

Strategy I

Argument A. Is there some good reason to think that reasonable people wouldn't think tolerating the moral opinions of others is a good idea? I can think of two reasons to think this. The first would involve arguing that, whether motivational internalism is true or not, there is a close connection between moral judgment and action, and that this is enough to show that tolerance of people's moral judgments isn't a good idea. Even if the connection isn't as close as motivational internalism tells us it is, there's no denying the empirical facts about the connection between moral judgment and action. After all, externalists are going to need to argue that there is a close connection here as well. For the practical dimension of moral judgments, and in particular their connection to action, is an empirical phenomenon that every meta-ethicist has to deal with.

Given the connection between moral judgment and action, no one should be tolerant in this way. It's simply too dangerous. Go back to our racist from Part I. Everybody realizes that his having these opinions is likely to lead him to act in certain unacceptable ways, and so it's patently unwise to tolerate his thinking these things. This is true even if we don't think there's the sort of necessary connection between moral judgment and motivation that the internalist tells us there is.

The most immediate worry about this response would be that it leads to a sort of blanket intolerance of other people's moral opinions. The obvious response to this worry would be to argue that this sort of reasoning needn't justify refusal to tolerate any and all moral opinions with which we disagree, since not all moral opinions are equally likely to lead to action. We can make some distinctions by drawing on the relevant empirical evidence, and we can tolerate some moral opinions and not others. And, one might think, the ability to draw such distinctions is independent of the truth or falsity of motivational internalism. The internalist can make the same distinctions by drawing on facts about the strength of motivation that is a necessary concomitant of particular types of moral opinions, and this will allow her to determine which moral opinions we should and shouldn't tolerate.

What should we think of this response? I won't say anything about this now, but I'll return to it and raise a possible objection when I consider one of the responses that employs the second strategy.

Argument B. Available to noncognitivists, there may be other reasons to think we shouldn't tolerate moral opinions in the way this tension assumes we should. Most noncognitivists are going to tell us that moral judgments primary express attitudes or emotions rather than beliefs. And, importantly, most of them think that these attitudes or emotions have a very close connection to moral motivation. This, of course, is supposed to be one big reason for accepting noncognitivism: that it alone can account for the essential practicality of morality. And the noncognitivist can argue that the intimacy of the connection between motivation and the relevant attitudes or emotions is a good reason to reject a broad tolerance of other people's moral judgments. (I should admit that the general point of this response is pretty much the same as the main point of argument A: that there is a close enough connection between moral judgment and action that no reasonable person should accept a broad sort of tolerance of the moral opinions of others.)

Return to our racist example. As I presented the case in the prior post, I suggested that the racist's objectionable moral judgment expressed a belief. For instance, in the original case for tolerating moral opinions as we tolerate other opinions, I assimilated tolerance of the racist's moral opinions to tolerance of beliefs. And the noncognitivist can admit that maybe we would find this plausible if we could understand moral opinions as ordinary beliefs. But, of course, she's going to tell us that this simply isn't how we should understand them, that we shouldn't assimilate the racist's moral opinions to ordinary beliefs.

Now we need to ask ourselves how we'd react to the racist example if we assimilated the racist's moral opinion to an attitude or emotion. Do we find it plausible to suggest that we ought to be tolerant of attitudes and emotions in the same way many of us think we ought to be tolerant of ordinary beliefs? A noncognitivist could argue that we shouldn't.

Suppose our racist had an unacceptable attitude towards people of another race. Suppose, for instance, that he's literally disgusted by them or that the very sight of them makes him angry. Do we think we should be tolerant of attitudes or emotions of this sort? If not, then the noncognitivist could tell us that this is no real problem for her view. And perhaps they could argue that we shouldn't be as tolerant here, precisely because of the intimate connection between motivation and a person's attitudes and emotions. It's not just that the racist has some belief that people of other races are troubling in some way; instead, he has a feeling or attitude about such people. And because an emotion or attitude of this sort is what we're talking about here, it's really not plausible to claim that moral opinions of this sort ought to be tolerated. Why? Emotions and attitudes of this sort have a close connection to action that ordinary beliefs simply don't have.

The most immediate worry about this response would be the same as the most immediate worry about the response considered above: that it leads to a sort of blanket intolerance of other people's moral opinions. But, again, the proponent of this argument can appeal to empirical facts about the relevant attitudes or emotions to avoid the possible worry that this will lead to blanket intolerance of moral opinions. This sort of reasoning needn't justify refusal to tolerate any and all moral opinions with which we disagree, since not all the attitudes or emotions expressed in particular moral judgments are equally likely to lead to action. Some of these attitudes and emotions may be only tenuously related to action, and it might be acceptable to tolerate the moral judgments expressing these attitudes and emotions.

Is this a good response? Clearly, its plausibility depends on the plausibility of noncognitivism; and the plausibility of noncognitivism is an issue that's too broad to discuss here. But it's important to note that any old version of noncognitivism may not do the need work here. It seems the plausibility of the noncognitivist's response is going to depend on just what she tells us about the attitudes that are expressed in moral judgments. So what, exactly, are the relevant moral attitudes or emotions, and how are they connected to motivation? We need to know this before we can appraise this response.

Berkeley's Dialogues

2004-05-27T16:35:00.000-04:00

Last night I went and read some of Berkeley's Three Dialogues between Hylas and Philonous for the first time in five years or so. I've only made it through the first dialogue, but I'm having trouble seeing why Berkeley decided to write this in dialogue format. For, like bad dialogues of all sorts, it takes the following form:

A (i.e. Philonous, i.e. Berkeley): Do you think such-and-such?
B (i.e., Hylas, i.e. a sensible person): Yes, I do.
A: But doesn't it then follow that so-and-so?
B: I believe it does.
A: But then you can't deny blah, blah, blah, can you?
B: It doesn't seem that I can.
A: But if blah, blah, blah, then p. Thus you have to conclude that p. (p, of course, being something B denies.)
B: It seems hard to deny that p at this point. But maybe ... (and so on.)

I guess I don't see any benefit to presenting philosophical discussion in this particular format. Jonathan Dancy, in his introduction to the Oxford Philosophical Texts edition of Three Dialogues, suggests that this allows Berkeley, in the person of Philonous, to respond to the sorts of objections his view is likely to elicit and to do so in a way that seems pretty natural. I guess that idea is that we're supposed to see Hylas struggling with the same objections to, and worries about, Berkeley's position that we're likely to have, and so we're less likely to think that Berkeley's simply trying to pull a fast one on us. The suggestion, then, is that Berkeley has anticipated all our objections, and that he can tell us why they shouldn't trouble us all that much.

Now, I see no reason to think that you couldn't provide the same content in a different manner. Couldn't Berkeley provide a list of common objections to his view along with his responses? That would be more straightforward, and I don't see why it would lead to some significant loss in content.

I'm tempted to suggest that there may be a more propagandistic motive for using the dialogue format: it makes it seem to the reader that any reasonable person is going to be persuaded by Berkeley's arguments. Reading through the dialogue seems to suggest the following question to the reader: If someone as clever as Hylas is persuaded, why shouldn't I go along as well?

So this suggests a couple of strategic points for all you dialogue writers out there. If you want to make your dialogue a convincing argument for your position, you need to do at least these two things. First, make the person(s) arguing against your position, and who will eventually be persuaded of your position, appear to be smart and powerful debaters. For you don't want the reader to think that the arguments your mouthpiece is using to persuade the person(s) are the sorts of arguments that would only convince someone who isn't all that clever. I'm not sure Berkeley succeeds in doing this; Hylas is pretty slow at times, and he has a tendency to make the same mistakes time and again. Second, make it seem that these people are also extremely unlikely to accept your position at the start of the debate. For you don't want the reader to think that the person(s) being convinced are the type who could be convinced of almost anything. Here Berkeley succeeds; Hylas at least seems to detect the preposterousness of Philonous's position, even if he isn't clever enough to figure out why we ought to think it preposterous.

Hume Quotes

2004-05-26T19:41:00.000-04:00

Here are a couple great quotes from Hume's Enquiry concerning the Principles of Morals, which I happened to be reading today. I figure that whoever has shown up here deserves some better prose after reading the preceding post. Unfortunately, writing good prose isn't one of my talents, and so I had to pilfer from another source. I hope I still get credit for my generosity to my (probably nonexistent) readers. (And wouldn't generosity to nonexistent people be especially praise-worthy? It's not like I'm doing this with the expectation that I'll end up receiving some future reciprocation from these nonexistent people. My motives are as pure as any Kantian could want, baby.)

In any event, here you go. First up, a broadside against ascetics:

A gloomy, hair-brained enthusiast, after his death, may have a place in the calendar, but will scarcely ever be admitted when alive into intimacy and society, except by those who are as delirious and dismal as himself.

Next, a pretty damn inspiring passage about benevolence:

We enter, I shall suppose, into a convenient, warm, well-contrived apartment: we necessarily receive a pleasure from its very survey because it presents us with the pleasing ideas of ease, satisfaction, and enjoyment. The hospitable, good-humored, human landlord appears. This circumstance surely must embellish the whole, nor can we easily forbear reflecting, with pleasure, on the satisfaction which results to everyone from his intercourse and good offices.

His whole family, by the freedom, ease, confidence, and calm enjoyment diffused over their countenances, sufficiently express their happiness. I have a pleasing sympathy in the prospect of so much joy, and can never consider the source of it without the most agreeable emotions.

He tells me that an oppressive and powerful neighbor had attempted to dispossess him of his inheritance and had long disturbed all his innocent and social pleasures. I feel an immediate indignation arise in me against such violence and injury.

But it is no wonder, he adds, that a private wrong should proceed form a man who had enslaved provinces, depopulated cities, and made the field and scaffold stream with human blood. I am struck with horror at the prospect of so much misery and am actuated by the strongest antipathy against its author.

I'm not sure that this is as inspiring as Kant's "starry heavens" passage, but it's close.

Motivational Internalism and Tolerance, Part I: A Possible Tension?

2004-05-26T18:16:00.000-04:00

Some meta-ethicists, and especially those tempted to some form of irrealism, have argued that their meta-ethical theories make no real difference with respect to our understanding of first-order moral issues. We can, they tell us, go on thinking whatever we think about normative ethical issues, regardless of what meta-ethical views we adopt. (For a classical expression of this viewpoint, see the opening sections of the first chapter of Mackie's Ethics: Inventing Right and Wrong. And I suppose that showing this to be true is the fundamental goal of Blackburn's quasi-realist project.)

I want to raise a question about whether this is true about motivational internalism. And, in particular, I want to discuss whether there might be is some tension between accepting motivational internalism and holding a first-order normative view about tolerance. (I should note that I don't know whether this issue is discussed anywhere in the literature, nor do I know whether it even makes sense. I've never encountered any discussion of this issue, though I wouldn't be surprised to discover someone had beaten me to it. But it crossed my mind recently, and I'd like to say a word or two about it.)

I'll understand motivational internalism to be the following thesis:

(I) Necessarily, if a person p judges that she ought to do action A, then p is motivated to A.

I'll ignore all the messy details about how exactly we should formulate this thesis here. It doesn't seem to matter what sort of motivation we're talking about here or whether we think we need to add some further qualifications to (I) in order to make it plausible. All that strikes me as important is that we're sure we're talking about what David Brink calls "appraiser-internalism": that is, about a form of internalism that ties motivation to an individual's moral judgments as opposed to her moral obligations. Provided that we think some appraiser form of motivational internalism, I think some version of this tension will arise. (I should note, however, that this account of motivation internalism is similar to the simplest form of internalism Michael Smith mentions in Chapter 3 of his The Moral Problem.)

Now I need to say what sort of tolerance I'll be concerned with here. You may have already guessed it, but I'll be concerned with tolerance of another person's moral judgments. Consider the normative thesis that we ought, in general, to be tolerant of other people's moral opinions. What does this entail? Well, it seems that it would at least entail not interfering with them by trying to alter their moral opinions, provided that they don't want you to do so, that your doing so would be a significant inconvenience for them, and that they don't plan to act on those opinions in a way that is likely to harm others. So, basically, the idea is that you shouldn't hassle people about their moral opinions as long as their having those opinions isn't likely to lead them to act in objectionable ways.

I don't need to argue that the principle requiring this sort of tolerance is a plausible normative principle, but it doesn't seem prima facie absurd to me. As long as it's a principle we can imagine some reasonable people defending, a tension between this principle and (I) would suggest that it isn't true that our meta-ethical views have no bearing on what we ought to think about first-order issues in normative ethics.

Still, is there anything to be said for this sort of tolerance? It might seem there is, as we can understand it to be something that follows from a more general moral principle. The general principle would be that we ought to tolerate the opinions of others, even if those opinions are false and perhaps even objectionable, as long as we have reason to think these opinions are unlikely to lead to the person acting in objectionable ways. One justification for this would be that their opinions are simply beliefs, and that there needn't be any necessary connection between having beliefs, even objectionable ones, and acting on them. Consider a person who has racist beliefs. He may hold these beliefs and not treat members of any race as second-class citizens. And provided that he doesn't discriminate against whatever race he judges unfavorably, it's arguable that we should tolerate his having the racist opinions that he has. It's certainly unfortunate that he has them, and I'd like to see him change his opinions. But if he seems unlikely to discriminate in his actions and if we'd have to impose on him in order to change those opinions, it may seem that we should tolerate those opinions.

Can the motivational internalist allow for this sort of tolerance with respect to a person's moral judgments? As I sketched things in the case of the racist, part of the justification for this sort of tolerance is just that there doesn't seem to be any very close connection between one's judgments and one's actions. This is why we're willing to put up with the racist's opinions: we have what we take to be good reason to think that he's unlikely to act on them. If we thought he were likely to act on them, however, it's not so clear that we ought to be tolerant. So it's a sort of disconnect between opinion and action that supports tolerance in this case.

But that there is this sort of disconnect between opinion and action seems to be precisely what the motivational internalist tells us is not the case when it comes to our moral judgments. For her thesis connects moral judgments to actions in a very close way: it is, according to her, a conceptual truth that a person who judges that something is what she ought to do is motivated to do that thing. Let's return to the case of the racist. Suppose our racist judges that he ought to forbid his children to marry people of another race, that he is morally obligated to forbid his children to do this. The motivational internalist will tell us that, necessarily, our racist is thereby motivated to forbid his children to marry people of another race.

And, given this, it's not altogether clear to me that we ought to be tolerant of the racist's moral judgment. It's no longer clear that we can drive a wedge between his making a moral judgment and his acting in objectionable ways, and so it's no longer clear that we ought to be tolerant of his moral judgment. Underlying the suggested first-order principle that we ought, in general, to be tolerant of the opinions of others was the idea that we could separate judging and acting. As I said above, the basic idea is that you shouldn't hassle people about their moral opinions as long as their having those opinions isn't likely to lead them to act in objectionable ways.

It looks like the motivational internalist is going to have trouble with the qualification "as long as having those opinions ..." So it looks like she's going to have trouble justifying the moral principle suggesting tolerance of others' moral opinions--or, at least, that she's going to have more trouble than an externalist would have. For the externalist doesn't seem to encounter any obvious problems with driving a wedge between judgment and action.

Maybe, then, some meta-ethical theses do have consequences for the plausibility of certain first-order normative views.

First Post

2004-05-24T21:16:00.000-04:00

I now have a blog, and all I need is something to write about.