jlowder

That explains it. I thought you were talking about "different answers to the question of what natural property (if any) was being referred to," and not the supernatural vs. nonnatural vs. natural property debate. In THAT sense, you are absolutely right that there is significant disagreement among objectivists.

Yes, he is.

Jeffery Jay Lowder

jlowder

Noncognitivists maintain that moral propositions do not have truth values, in contrast to moral cognitivists who believe that moral propositions have truth values. Therefore, moral noncognitivists believe that moral cognitivism (in all it forms, including objectivism) is meaningful but FALSE. In other words, moral noncognitivists are noncognitivists with respect to first-order moral beliefs (e.g., "The proposition, 'Torturing newborn babies for fun is morally wrong,' is neither true nor false") but cognitivists with respect to *second-order* moral beliefs (e.g., "Moral objectivism has a truth value and that value is false").

Jeffery Jay Lowder

Clutch

I thought that was your point. Again, it does not seem to impinge on bd's point, which is charitably understood to be that mere semantic ascent does not make a moral principle itself truth-apt. The truth-claim itself is may or may not be considered truth-apt, depending upon whether "is true" is used as a sentential or a propositional operator.

So:

1) Boogala boogala! [meaningless]

2) It is objectively true that boogala boogala. [meaningless]

3) "Boogala boogala" is objectively true. [false]

When he wrote,
I took bd's point to be that (2) is no better than (1) with respect to being meaningful. And that seems correct.

jlowder

No, (2) is not correct. To use your examples--I like your examples, BTW--if (1) is meaningless, then (2) is false. Since (1) is a meaningless proposition, by definition it cannot have a truth value. In contrast propositions *about* meaningless propositions can have truth values. In particular, when a proposition (like [2]) asserts that a meaningless proposition (like [1]) has a truth value, then the former (proposition [2]) has a truth value but that truth value is FALSE. It is FALSE that meaningless propositions have truth values. Indeed, it is necessarily false, since that is a matter of definition.

Jeffery Jay Lowder

bd-from-kg

jlowder:

This is a fuller reply to your long post of Nov. 8.

You’re talking about two different things. Positivists say that moral objectivism is meaningful but false, and that it is false because it is meaningless to say that a moral principle is objectively true.

In your latest post you say that if S does not express a proposition it is not meaningless but false to say that S is objectively true. This is just a linguistic quibble, but I don’t agree with this usage. “S”, “S is true” and “It is objectively true that S” are generally taken to be logically equivalent, so if one of them is meaningless, all of them are. Thus I would say that if “Stealing is wrong” is meaningless, “It is objectively true that stealing is wrong" is also meaningless. But if you prefer to say that if “Stealing is wrong” is meaningless, “It is objectively true that stealing is wrong” is false, so be it, as long as we understand one another.

As should be clear by now, I am not saying that moral objectivism is meaningless. Moral objectivism, as such, does not assert that any particular moral principle is objectively true, but that a moral principle is the sort of thing that is capable of being true or false – i.e., that it expresses a proposition. According to noncognitivists (according to my usage at least) a statement of the first sort is meaningless, whereas the second assertion is false.

But we aren’t talking about what it means to say that a moral principle is objective, but what it means to say that a moral principle is objectively true. It’s one thing to agree that a statement expresses some proposition, and quite another to say what proposition it expresses. To explain what it means to say that a statement is true is to explain what proposition it expresses. I suppose that this goes to one’s concept of “truth”. As far as I’m concerned, what “‘Snow is white’ is objectively true” means is not that it has some mysterious property called “truth”, but simply that snow is white. Similarly, “’Stealing is wrong’ is objectively true” does not mean that “Stealing is wrong” has the “property” of being true, but that .... well, this is where the problem lies. Moral objectivists do not agree (to put it mildly) on how to fill in the ellipses.

“Supervene on”. That’s the problem in a nutshell. Let’s say, for example, that the moral “property” of “rightness” supervenes on “healthiness”. This seems to mean that an act is right (i.e., has the property of “rightness”) if and only if it has the property of “healthiness”. But the point of saying that “rightness” supervenes on “healthiness” is that “rightness” is somehow a different property from “healthiness”; that it is a substantive proposition (and not a mere definition) that an act has the property of “healthiness” if and only if it has the property of “rightness”. If this were a matter of definition the various moral philosophers who propose different natural properties as “criteria of “rightness” would not be disagreeing; they would merely be defining words differently. But they certainly seem to be disagreeing, and they seem to be under the impression that they are disagreeing, about something more significant than what definitions facilitate communication most effectively.

But if saying that the moral property of “rightness” supervenes on the natural property of “healthiness” is not just a fancy way of defining “rightness” to mean “healthiness”, we are left with the problem of what kind of property “rightness” is. If it were a natural property, why talk about moral properties “supervening” on natural properties at all? Why not just say what natural property “rightness” is? Why not just say, “To say that an act is ‘right’ means that it has the natural property Q”, rather than saying “The moral property of ‘rightness’ supervenes on the natural property Q”?

Thus it only makes sense to speak of moral properties “supervening” on natural properties if moral properties are not themselves natural properties. And this leaves us with the vexing question of what kind of properties moral properties are. This is the problem that G.E. Moore saw quite clearly, and why he took the position that “goodness” is not a natural property at all, but a simple, unanalyzable, non-natural property.

But a meta-ethical theory should be able to give us a reasonably clear notion of what kind of property might be a “criterion of rightness”, how to determine whether a given property is a reasonable candidate for this honor, and what tests and methods we might use to help determine which of the candidates is the true criterion of rightness. A “moral theory” that is unable to give us any guidance as to whether the “greatest happiness” principle, the “greatest misery” principle, or the “maximize the number of purple turtles” principle is closer to being the “true” criterion of rightness isn’t much of a moral theory. In the same way, a philosophy of science will not tell us whether quarks exist (or more precisely, whether the quark theory should be accepted) but it should certainly be able to give us some guidance on how to distinguish between a reasonable scientific theory and an unreasonable one (relative to a given body of evidence).

When different moral objectivists disagree about such questions; when they give completely different accounts of what kind of property “rightness” is, or what kind of property a criterion of rightness would have to be, or how to distinguish the “true” criterion of rightness from the many imposters, it is fair to say that they disagree profoundly about what it means to say that a moral principle is objectively true.

It seems odd, to put it mildly, to argue that moral statements are objectively true without being able to give even a rudimentary account of what it is that they assert. The noncognitivist argument is that no such account, or no remotely plausible one anyway, is possible. To answer this argument the objectivist needs to show that there is at least one plausible account of what such statements mean.

I’ve looked at the quotations from Post that you cited in various threads, and so far I have seen nothing like an actual argument that moral statements can be interpreted as descriptive statements about the natural world. I’ve seen allusions to such arguments, but that’s it. It appears that you’re using Post as an authority rather than as a source of arguments that you’re prepared to present, either in your own words or by paraphrasing or quoting him. It’s no good (especially in a forum like this) to say “read so-and-so’s book; it explains everything.” If you can’t give us some sense of his argument, why should we waste a lot of time finding the book and trying to understand it? Until you convince me that there might actually be a reasonable way to interpret moral statements as descriptive statements about the natural world, I’m going to assume that his books are garbage. If you understand his argument, you should be able to summarize it.

One last point. In your nest-to-last last post you said:

No one maintains that any propositions do not have truth values. The defining property of propositions is that they have truth values. What noncognitivists claim is that there are no moral propositions, or more precisely that what you would call “first-order” moral statements do not express propositions.

Clutch:

Thanks for the clarifications. Your explanations seem to be much more concise than mine. In fact, if I hadn’t had further points to make, I’d have been content to just say “what he said”.

jlowder

I'm going to respond to one portion of bd-from-kg's post now, and save the rest for later.

Sigh. I of all people am not asking anyone to accept a conclusion because some authority says the conclusion is true. On the contrary, I have quoted and summarized his argument on this very board. It's no good (especially in a forum like this) to say "I haven't seen the argument quoted on this thread in direct reply to my post; therefore you haven't produced an argument." For your convenience, I will quote the argument again later in this post.

Apparently you missed my post where I provided an argument, quoting some portions and summarizing others. I will reproduce the entire argument below.

You can believe whatever you want, but it is very rare for books by professional philosophers in general to be "garbage." I do not even consider those books in moral philosophy I strongly disagree with, including Mackie's Ethics: Inventing Right and Wrong, to be "garbage." Even as someone who rejects Mackie's error theory, I find Mackie's book to be very scholarly, full of insight, and quite valuable. (Hell, I strongly disagree with William Lane Craig about the existence of God, but I would never say that his books are "garbage.")

Returning to the subject of Post and his book, I know that Post is well-respected by other philosophers. (<a href="http://www.infidels.org/library/modern/quentin_smith/" target="_blank">Quentin Smith</a> is one of the Internet Infidels Supporters and a very highly regarded atheist philosopher who also just happens to highly endorse Post's book. I think that should give anyone pause who thinks Post's book is "garbage.") Since moral philosophy is a branch of philosophy in general, I value the opinions of other professional philosophers, especially morally philosophers, regarding the quality of philosophy books (including Post's book) much more than I value the opinions of non-philosophers. (By analogy, if a non-mathematician were to declare that Einstein's mathematical theories were "garbage," who cares what the non-mathematician thinks? Why should we even regard the non-mathematician as competent on the subject?) The fact that you would dismiss a book you have never read as "garbage" sounds like the sort of comment a non-philosopher would make, a comment I do not place much value in. (Are you a philosopher? Have you at least taken any college-level courses in moral philosophy? And just to pre-empt a possible reply: No, I am not a philosopher but I have a taken a college-level course in moral philosophy. Moreover, I am not calling books by professional philosophers "garbage.")

Moreover, although appealing to Post as an authority in an attempt to settle a debate on this board would be fallacious, Post nevertheless is an authority who defends moral realism, and critics of moral realism should at least be familiar with his arguments. Anyone who wants to reject moral realism (or, more broadly, moral cognitivism) should at least be familiar with Post's arguments, just as any creationist who wants to reject evolution should at least be familiar with authorities on evolutionary biology (e.g., Richard Dawkins), just as any moral realist who wants to reject moral subjectivism should at least be familiar with authorities on subjectivism (e.g., J.L. Mackie). Indeed, one could just as easily reply, "Until you convince me that you are actually familiar with the arguments of the leading moral realists, I’m going to assume that your arguments against moral realism are garbage."

If <strong>I</strong> "understand" his argument? Is that supposed to be some sort of veiled insult?

Again, I have already summarized his argument on this board. For convenience, I will summarize it again. John Post describes his argument as follows:

Post then describes precisely what the existence of objective values entails:

Post then explains what he means by the "determinacy of valuation":

More formally, Post formulates the determinacy of valuation as follows:

Thus, on the assumption that moral properties are determined by natural (physical) properties, moral properties supervene on natural properties in the sense that nothing can differ in its moral properties without differing also in its natural properties. That is what Post means when he says that moral properties are determined by natural fact.

So that's Post's thesis. What is Post's argument for his thesis? The first premise of the argument appeals to the "weak supervenience" of the moral on the natural:

The second premise is simply the application of the law of noncontradiction to moral judgments:

(MEA stands for meta-ethical antirelativism)

The determinacy of valuation follows from the conjunction of EP and MEA. And if Post's thesis about the determinacy of valuation is correct, then it follows that moral realism is true.

As a syllogism, Post's argument looks like this:

(1) EP
(2) MEA
(3) If EP and MEA, then the determinacy of valuation is true.
(4) Therefore, the determinacy of valuation is true. (from 1, 2, and 3)
(5) If the determinacy of valuation is true, moral realism is true.
(6) Therefore, moral realism is true. (from 4 and 5)

You may disagree with this argument. Hell, you may think it is the worst argument you have ever seen. But don't ever tell me I am hiding behind some authority, whose arguments I am unwilling to summarize, since I clearly have summarized the relevant arguments.

You are correct about this; I was mistaken when I earlier denied this and I retract my objections. By definition, a proposition has a truth value. Since moral noncognitivists deny that there is any such thing as moral truth or falsity, moral noncognitivists also deny that there any such things as moral propositions.

Jeffery Jay Lowder

[ November 09, 2002: Message edited by: jlowder ]

[ November 09, 2002: Message edited by: jlowder ]

(edited a third time to discuss the relationship between propositions and truth values)

[ November 10, 2002: Message edited by: jlowder ]</p>

jlowder

Here is my reply to another portion of bd-from-kg's post, a portion which dealt with the concepts of supervenience and meta-ethics in general:

Your point is well-taken. I agree that "'Snow is white' is objectively true" means that snow is white, not that snow has some mysterious property of "truth." I also agree that "moral objectivists do not agree (to put it mildly) on how to "fill in the elipses."

Bloomfield discusses the very issues you are raising, when he mentions the different "strengths" of supervenience. (As an aside, I would recommend Bloomfield's book to anyone looking for a good, easy-to-understand introduction to the concept of supervenience.) Here is part of Bloomfield explanation:

So if one accepts Bloomfield's account of the different strengths of supervenience (as I am inclined to do), then the supervenience relationship is not necessarily reductive. Only the strongest form of supervenience is reductive.

Quite honestly, I had wondered about the exact same questions. Again, I found Bloomfield's answers to be helpful. In a nutshell, Bloomfield's answer appears to be that moral properties are not reducible to natural properties. (Bloomfield makes an analogy with the physical property of entropy, and his book contains an entire appendix arguing that entropy is a nonreducible but fully physical property. Since I'm still reading the book, I haven't gotten to the appendix yet so I can't summarize that yet.) Bloomfield writes:

Bloomfield's reply would be that it makes sense to speak of moral properties supervening on natural properties if moral properties are nonreducible but fully natural properties.

By "kind of property," do you mean natural vs. nonnatural vs. supernatural? Or do you mean the exact identity of moral properties (e.g, "moral properties are just the natural property of healthiness," "moral properties are supernatural properties expressed through God's commands, etc.)? And whatever you have in mind by "kind of property," on the assumption that moral realism is true, why should we expect moral realism to "give us a reasonably clear notion of what kind of property might be a “criterion of rightness”?

Again, why should we expect this?

Why?

A normative ethical theory that can't give us that guidance isn't much of a normative ethical theory. However, I have no idea why you think a meta-ethical theory should provide such guidance.

Agreed.

I don't see the problem. In the first place, it is false that moral objectivists are unable to provide a "rudimentary account of what it is that they assert." On the contrary, as you yourself admit, many moral objectivists have provided detailed accounts on what you call the identity of the criterion of "rightness." It's just that they don't agree on the identity of that criterion. But that doesn't mean they can't provide a "rudimentary account."

Moreover, notice that when they do provide such an account, they have stepped out of the domain of meta-ethics and into the domain of normative ethics. From the perspective of meta-ethics, why can't the moral objectivists simply say that moral propositions are true just in case they correspond to the facts. If one then asks, "Yes, but which facts?", the moral objectivist could then respond, "That depends upon which normative ethical theory is correct; normative ethical theories specify the exact identity of moral properties. According to my normative ethical theory, the criterion of rightness is X," where X is determined by the normative ethical theory of the speaker.

That's the noncognitivist position. What is the argument for that position?

What do you mean by "plausible"? I think there are several credible normative theories that are objectivist in nature, including those advocated by Brink, Smith, Martin, and Bloomfield.

Jeffery Jay Lowder

(edited by adding the discussion of what a meta-ethical theory should be able to provide)

[ November 09, 2002: Message edited by: jlowder ]</p>

Clutch

JJL,

I think you are making a use-mention error.

(2) is indeed meaningless. It uses the string 'boogala boogala'. Since that phrase is meaningless, there is no compositional interpretation of the sentence that returns a complete proposition, hence no truth-apt interpretation of it.

(3), by contrast, places that string in quotation marks, creating a name for it. So (3) mentions the string 'boogala boogala'. It says of a named item that it is true. Since that item is not meaningful, and thus not true, the sentence is false.

bd-from-kg

jlowder:

Well, I have to say that I have still not seen any quotations from Post that you have cited that I would consider an argument that moral statements can be reasonably interpreted as descriptive statements about the natural world. More on this later; in the meantime a number of my comments are based on my opinion that the quotations you’ve cited do not present an argument for moral objectivism.

As I indicated, I was not relying only on your posts in this thread. In fact, I saw all of the stuff you quoted here elsewhere, and in addition found some long quotations on the “C. S. Lewis and the Moral Standard” thread.

Note: Once again I was being sloppy in describing the issue as being whether value judgments are determined by physical facts. In the first place, it may well be that value judgments are physical facts. The real issue is whether the correctness of value judgments is determined by natural facts. But even this could be misinterpreted. Actually almost everyone agrees that if there are moral facts they do supervene on natural facts. As someone put it, once we have the natural facts, the moral facts come for free. The difference between naturalistic and non-naturalistic moral objectivists is not about whether moral facts are determined by physical facts, but about the nature of moral facts. For example, some philosophers have claimed that moral properties such as “goodness” and “rightness” are non-natural properties which have the peculiar feature that whether an act or state of affairs has such a property follows from its natural properties, but cannot be identified with any combination of or relationship between them. Although such theories entail that moral facts are determined by physical facts, they are not naturalistic theories. On the other hand, the position that “goodness” and “rightness” not only supervene on natural properties, but are themselves natural properties, defines moral naturalism.

Now as to my “garbage” comment, it should have been obvious that I am not familiar with Post’s work since the whole point was why I did not intend to become familiar with his work unless you convinced me that it would be worth my time. So this could not have been meant as an evaluation of his work. As to whether you’d call the work of this or that philosopher “garbage”, this is beside the point; we don’t all use language the same way. What I mean by “garbage” in this context is that it’s not worth my time, and therefore is worthless to me. All that I meant was that I would assume that it would not be worth my time to read his book (because as far as I know it has nothing interesting and original to say) unless you could show me something that would convince me otherwise.

Really? No one is allowed to have an opinion on the subject unless he has read the works of every philosopher of any significance who has written anything about it? Isn’t that a rather high standard, especially for a board like this one? I suspect that you’ve just ruled out at least 99.9% of the people who might be interested in participating in a discussion here as being unqualified to join in. If this is really what you think, why don’t we just close this forum? There obviously aren’t enough qualified participants to make it worth the trouble and expense.

Anyway, from what I’ve seen so far it appears that I am familiar with his arguments; I’m just not familiar with his particular exposition of them.

That’s because creation is rejected by all but a tiny minority (if that) of qualified scientists. Anyone who chooses to reject such an overwhelming consensus does indeed have a responsibility to give a good account of why he does so. Unless one has a great deal of expertise in the subject it is irrational to reject such a consensus. But this is not the case with regard to moral realism. There is nothing like an overwhelming consensus among qualified philosophers that moral realism is correct. In fact, it may well be a minority position among professional philosophers today, especially among nontheists.

It might or might not be reasonable for you to assume this, but if you do there is little point in continuing with this thread. My point was that you haven’t presented Post’s arguments for moral objectivism, so finding out what they are would involve a good deal of time and trouble for me. By contrast, I have presented some arguments (even though the original point was merely to give a very rough sketch of what the key issues are between competing moral theories). to find out what these arguments are, you need merely read this thread (as you’ve already done). At that point you needn’t assume anything about them; you can evaluate them for yourself.

Of course not. I assume that the reason that you haven’t cited Post’s argument for moral objectivism is that it’s too lengthy. I’m asking you to summarize it for us.

Now perhaps we are ready to actually talk about moral philosophy.

As to the part of Post’s argument that you do quote, it shows only that if there are objective moral facts, they supervene on natural facts, which as I pointed out earlier is hardly controversial. For example, here’s G.E. Moore:

This of course applies only to consequentialist theories, but is otherwise essentially the same as EP. But Moore makes only the modest claim that any reasonable objective consequentialist moral theory must have this property. He certainly doesn’t imagine that anything of this sort could be used as a premise of an argument that a non-objectivist would find at all convincing. At any rate, the very statement of EP plainly shows that it assumes that morality is objective. For example: “... an act A has a certain moral status for person P in W1 iff ...”. This doesn’t make sense unless one assumes that there is such a thing as a “moral status” of an act. And it further says “... a substantive moral rule, principle or first principle R is true ... iff...” But to give conditions for a moral rule to be true is to assume that it makes sense to talk about a moral rule being true – i.e., that morality is objective.

Note: Post actually says “... for a moral rule to be true (or correct or to be followed”. I’m not sure what he means by saying that a moral rule is “to be followed” as opposed to being “true”, but this sounds strikingly like the sort of thing that I might say about a moral principle. To say that the rule “Do not break promises” is a rule to be followed is not to say that it is “true”, or even that it expresses a proposition. In fact, it’s not at all clear that “Moral rule R is to be followed” expresses a proposition; it seems to me to be a prescription, closer in meaning to “Follow rule R” than to “R has such-and-such a property”. At any rate, if Post intends “to be followed” to have a meaning such that “Moral rule R is to be followed” expresses a proposition, I’d like to know what it is, and what proposition he intends this statement to express. It certainly is not intended to express the proposition that moral rule R is true, or he wouldn’t have bothered to make this distinction. Also, what does he mean by distinguishing between a moral rule’s being “true” and “correct”?

As for MEA, I have no idea why Post says “If we pretend our moral judgments are true or false...” as opposed to “if our moral judgments are true or false...”. Why should anything of interest follow from our pretending that moral statements are true or false? But in any case, it is transparent that MEA also assumes moral objectivity; after all, it says that of various differing moral judgments, one is “correct”. But again, this is gibberish unless it makes sense to speak of a moral judgment as being “correct”.

Now as for the final argument, it seems to be valid and may be sound, but it is utterly unpersuasive because the conclusion is contained so transparently in the premises. To illustrate the problem, consider these two arguments:

P1: “God exists” is true, “God is omnipotent” is true.
P2: “God exists” is true.
C: “God is omnipotent” is true.

P1: If “God exists” is not true, “God is omnipotent” is not true.
P2: “God exists” is not true.
C: “God is omnipotent” is not true.

Now both of these arguments are perfectly valid, and one of them is sound. But neither of them is at all persuasive, because anyone who does not already accept the conclusion of one of them is certainly not going to accept the premises either. So it is with Post’s argument (or rather the argument that you offer on his behalf). Why would anyone who is not already a moral realist accept EP or MEA?

It’s hardly surprising that one can “prove” that moral realism is true from premises that presume the truth of moral realism. The real question is what arguments Post has for EP and MEA. I imagine that he must have some, but so far you have failed to present them.

bd-from-kg

jlowder:

1. Supervenience and its relationship to naturalistic ethics

You seem to be arguing here against someone else; I never said anything about whether this relationship is reductive, and I don’t see why it would matter. The point is that as I understand you, your position is that “rightness” is a natural property, but also that it supervenes on a natural property or properties. Now the fact that one natural property supervenes on other natural properties is often of some scientific interest, but I fail to see how it can possibly be of any interest to moral philosophy. If “rightness” is a natural property, no matter what kind of complicated relationship it might have to other natural properties, my point stands: it is meaningful to ask whether an act which has this property is more desirable than one which does not. And this in itself is sufficient to show that the property in question is not identical to the property of “rightness”, even if it is true, as a matter of fact, that all right acts have this property.

Talking about moral properties “supervening” on natural properties appears to me to be an attempt to get around the problem by saying that “rightness” is not identical to the natural property in question, but only “supervenes” on it. But this only helps if “rightness” is not itself a natural property; otherwise the same argument applies to the property of “rightness” itself. And if it is not a natural property, we’re back to the problem of what kind of property it is.

Again, an illustration might be helpful here. According to classical utilitarianism, an act is “right” is and only if it produces more happiness than any of the available alternatives. (This is an oversimplification, but it will do for the purpose.) Now as G.E. Moore pointed out, this can be interpreted in two different ways: (i) saying that an act is “right” means that it conduces to the greatest happiness, or (ii) acts that are “right” always in fact have the property that they conduce to the greatest happiness, and no other acts have this property. In the second case we could say that “rightness” supervenes on the property of “conducing to the greatest happiness”, while in the first we would say that “rightness” is this property.

But the first interpretation is open to the objection above: it seems clear that it makes sense to agree that an act conduces to the greatest happiness yet question whether it is right. (In fact a great many moral philosophers have questioned this, to put it mildly.) Contrast this with agreeing that a polygon has three sides, yet questioning whether it is really a triangle. This strikes us as ludicrous, precisely because saying that something is a triangle means that it is a polygon with three sides.

The second interpretation is open to the very same objection if “rightness” is just another natural property that supervenes on “conducing to the greatest happiness” in the same way that entropy supervenes on other natural properties. If it is a natural property, it can be defined in purely descriptive, naturalistic terms that do not involve any moral or evaluative terminology; call this description D. But no matter what this description looks like, it is clearly sensible to agree that an act satisfies this description, yet question whether it is right. It might be true as a matter of fact that all right actions (and no wrong ones) satisfy this description – i.e., satisfying D may be an infallible indicator that an act is right - but “X satisfies D” cannot be a correct definition of what it means for an act to be morally right.

But if “rightness” is not a natural property, but merely supervenes on a natural property, then we are not talking about a naturalistic moral theory, and we have to give some plausible, logically coherent account of what kind of property it is.

2. The role of metaethics

Since you questioned all of my statements about what a metaethical theory should be able to do, it will probably be more helpful to explain my conception of the role of metaethics.

Metaethical questions are hardly the first kinds of questions that occur to us when we start to think seriously about ethics. The kinds of questions that we start with are typically ones of “practical ethics”, such as whether capital punishment is right or wrong, or whether abortion should be legal. But in discussing such questions we find that not only do people disagree about such questions, but they offer entirely different kinds of justifications for their positions. Some might argue, for example, that capital punishment is right because it has desirable consequences; others that it is right because justice demands it regardless of consequences; others that it is wrong because it violates the sanctity of life or because it is forbidden by God’s commandments. This leads one naturally to consider the questions of normative ethics: What moral principles are valid? Or ultimately, what is the criterion of “rightness”; what quality or property of an act determines whether it is right or wrong? But here once again we encounter profound disagreements; some say that some moral principles are valid while others deny that they are and offer entirely different moral principles in their stead. Worse yet, they disagree about what makes a moral principle valid. This leads us to the fundamental question of metaethics: what does it mean to say that a moral principle is valid?

When we reflect on how we got to this point, we recall that the original purpose was to work out answers to questions such as whether capital punishment is wrong, and that this led us to ask how we can distinguish valid moral principles from invalid ones, and that this led us in turn to ask what it means to say that a moral principle is valid. So in answering the latter question we must keep in mind that the point of asking is was to help us to distinguish between valid moral principles and invalid ones, and ultimately to distinguish between right acts and wrong ones. Thus we have not done what we set out to do until we have found an answer to the question of what it means to say that a moral principle is valid that is clear and concrete enough to at least offer some meaningful guidance in answering the question of what moral principles are valid. An answer to the former question that fails to provide any such guidance is at best radically incomplete.

Thus the various “levels” of moral philosophy: practical ethics, normative ethics, and metaethics, are not independent enterprises, but are closely interconnected. The purpose of each of the “higher” levels is to help us to answer questions at the “lower” levels. A normative theory should help us answer the questions of practical ethics; a theory that offers little or no such guidance is at best radically incomplete. In the same way, a metaethical theory that offers little or no guidance in helping to distinguish between true and false (or as I would say, valid and invalid) moral principles is unsatisfactory because it fails to fulfill the function or purpose of a metaethical theory.

The problem is that you seem to be determined to draw a bright line between metaethics and normative ethics, but no such line can be drawn: metaethics unavoidably overlaps with normative ethics. That is, a sufficiently clear explanation of what it means to say that a moral principle is valid will often entail that certain moral principles are valid so directly and immediately that it may be hard to say exactly where the “metaethical” part of the theory ends and the “normative” part begins. The reason for this, of course, is that both metaethics and normative ethics deal with the question of what moral principles are valid. It is very hard to argue that a moral principle is valid without getting into the question of how to distinguish valid principles from invalid ones, and it is very hard to discuss this question intelligently without getting into the question of what it means to say that a moral principle is valid. Conversely, it is very hard to give a sufficiently clear and complete account of what statements of moral principles mean to help us distinguish between valid and invalid ones without immediately implying that certain ones are valid and certain others invalid. The important thing is not to wall off metaethics and normative ethics in such a way that some important questions “fall between the cracks”. The question of how to distinguish between valid and invalid moral principles (which might also be considered the problem of distinguishing between true and false normative theories) cannot be ruled “out of bounds” for both normative ethics and metaethics. One of the two has to deal with it.

3. Objectivism and noncognitivism

Well, I thought that I’d been giving just such an argument. If one says that “moral principle M is objectively true” means that M has some natural property, one runs into the “open question” argument that I’ve discussed at some length. (Another way of saying this is that naturalistic theories ignore the “is/ought” dichotomy, or that saying that some state of affairs obtains cannot mean the same thing as saying that some state of affairs is desirable.) And if one says that it means that M has some non-natural property, one has to give some plausible account of what kind of non-natural property this could be, and how one might go about determining whether a given principle has it.

To say that such an account is plausible means that it is plausible that it is a correct interpretation of what most people mean when they use moral language, or at least what they would understand themselves to mean if they had enough knowledge and understanding. (For example, a theist might say that when he calls an act “right” he means that it is in accordance with God’s commands, but if he were to come to understand that God does not exist, or if he were to understand the significance or the Euthyphro dilemma, he would probably give a different account of what he “really” meant rather than saying that he had been talking pure nonsense. I believe that people who say such things are usually speaking truly - that they are not inventing a new meaning for moral terms, but have reached a deeper understanding of what they really meant all along.)

Now certainly there are many aspects of the logic of moral discourse that seem to imply that moral statements are intended to express propositions – i.e., to say that certain things are “objectively true”. But there are other aspects of moral discourse that seem to imply otherwise. It seems clear to me that there is no logically coherent interpretation of ordinary moral discourse that fully captures or corresponds to what most people who use it seem to intend, or in other words that there is no plausible account of what such statements mean to those who use them. (This is, after all, the real question. Anyone can invent a new meaning for moral statements. I can say that “Doing X is right” means that X is the first thing that the agent does after the sun rises, or that it involves turning to the right. But a personal, private interpretation that ignores actual usage is of no interest to anyone but its inventor.)

Well, at this point I was saying only that an objectivist has to choose such a theory and defend it in order to make moral objectivism plausible.

As for the various theories you mention, I certainly don’t intend to try here and now to refute every such attempt. Suffice it to say that I don’t think that any of these attempts are successful. Moreover, I don’t think that any such attempt can be successful, for the reasons already given and for others that I haven’t mentioned yet.

[ November 11, 2002: Message edited by: bd-from-kg ]</p>