Alonzo Fyfe

No. bd-from-kg is offering a 'reductio ad absurdum' by asserting that my view leads to an absurdity.

In a reductio argument, one cannot stipulate the absurdity -- that would beg the question. One can only stipulate the assumptions that one claims leads to the absurdity.

The conclusion that bd-from-kg asserts that I am forced to adopt is not, as I argue, one that I am forced to adopt. My statement refers to his inferred absurdity, not his assumptions.

A typographical error. I have made all relevant corrections (i think). The argument still stands.

bd-from-kg

Alonzo Fyfe:

It’s been very hard to find time to post anything recently because of the approach of Christmas; this will probably be the last post I’ll be able to make until some time afterwards. But here goes:

First off, as to your comments about desire-fulfillment and value, I can make absolutely no sense of them. If whatever you’re trying to say here is indeed a “key part�? of what you’re saying, then I’m afraid that I’m missing a key part of what you’re saying. Perhaps I’m just too dense to get your point.

As to the second section:

This is just a flat refusal to take noncognitivism seriously, and as such does not merit a response. If you keep saying stuff like this there’s no point in continuing the discussion.

The sentence you quoted wasn’t intended to be a “response�? as such, but was an attempt to clear up a misconception embedded in the question (“However, then ask him, ‘If you were not prepared to take this stand against X, would it still be wrong?’�? The actual response is the following paragraph, beginning, “But in any case I have never said…�? But since your question here once again was based on a refusal to take noncognitivism seriously, there doesn’t really seem to be much point in discussing the matter further. If you want to critique noncognitivism, you'll just have to talk about noncognitivism, not some other theory that you claim noncognitivism is fronting for.

As for your questions about “appropriateness�? and “universalizable�?:

I’ve explained before that the term “appropriate�? is subjective; there’s no such thing as a response being “objectively appropriate�?. So no, it is not possible for a criterion for appropriateness to be “mistaken�?, any more than it’s possible for my criteria as to which movies are preferable to which others to be “mistaken�?. (However, it is possible to be mistaken about whether one’s criterion is properly universalized, or to mistakenly believe that a criterion for moral praise or condemnation need not be universalized at all.) As to what it means in a subjective sense, this is hard to pin down, just as it’s hard to pin down what we mean when we say that something is “beautiful�? or that a certain action or sight is “disgusting�?. What we don’t mean (unless we’re confused) is that the thing in question has an intrinsic property of “beauty�? or “disgustingness�?. For present purposes, we can take “Smith considers condemnation (of a certain act) to be appropriate�? to mean simply that Smith is disposed to condemn actions that are relevantly like the one in question, and therefore will ceteris paribus condemn such actions, and moreover that he is disposed to approve of others condemning such actions and disapprove of their failing to do so.

As for “universalizable�?, as I’ve also explained before, an attitude of considering praise or condemnation of an act to be appropriate is universalizable if it applies to any action (actual or potential) that anyone might take under relevantly similar circumstances. And as for what that means, this was explained in post #99.

Now let’s get to the main event: a point which seems to be crucial to your argument – and indeed, to your defense of DU – but which, so far as I can recall, you have never made before.

This is unclear because you haven’t explained what you mean by a “reason for action�?. What does it mean to say that there is a “reason�? for Jones to do X? So far as I can see there are two ways to go here:

Possibility 1: By “reason�? you mean “motivation�?. That is, Jones has a reason to do X if (and only if) he has a motivation to do X. A motivation to do X, of course (according to BDI theory) is a desire coupled with a belief that this desire will tend to be satisfied by doing X.

That’s clear enough, but there are two problems with it:

(a) There’s no reason to suppose that it’s “nonsense�? to say that someone “ought�? (in the moral sense) to do something unless he has some motivation to do it. It might be that Jones has no desire whatsoever to do X, and has no desires that he believes would tend to be fulfilled by doing X, yet that he morally-ought to do X. This is pretty uncontroversial, and in fact is just as true in DU as it is an almost all other moral realist theories. So if it’s nonsense to say that Jones morally-ought to do X unless he has some motivation for doing X, your theory is nonsensical, since it says precisely that in a great many cases. (Of course in a noncognitive theory “Jones morally-ought to do X�? is not truth-apt, so we can’t talk about whether it’s true that Jones ought to do X, but rather about whether the speaker considers that Jones ought to do X. But in this type of theory it again makes perfectly good sense to say that one considers that Jones ought to do X even though one knows that he has no motivation for doing it.)

I note once again that the vast majority of the desires that figure into the “desire-fulfillment calculus�? envisioned by DU are not desires of the agent, but are the desires of others. And someone else’s desires are simply not a motivation to do anything. In fact, in many cases the agent doesn’t even know or believe in the existence of a great many of the desires in question, so that many of these desires can’t even figure into some other motivation through the “B�? part of the BDI paradigm. And if you’re serious about including all desires in this calculation (because all desires “exist�? after all) then many of the desires involved are desires of lower animals whose mental processes may be so alien to humans that we can’t even imagine them – that is, we can’t imagine “what it’s like to be�? a snake who desires X. And if we can’t even imagine what it’s like to have a certain desire, it’s hard to see how such a desire can motivate us in any sense, even if by some miracle we knew of its existence.

Possibility 2: By “reason�? you mean a moral reason. In that case it is of course perfectly true that it’s nonsensical to say that Jones ought to do X unless he has a “reason�? to do X. The problem in this case is that it doesn’t follow that the only sort of thing that can be a reason for doing X is a desire of some kind. This is merely an assertion which (combined with some related claims) begs the question in favor of a version of DU. (Unfortunately for your argument the version of DU that this line of reasoning leads to is not your version of DU, a point that I’ll get to a little later.) Naturally if you start with the premise that your moral theory is the correct one you can derive the conclusion that your moral theory is the correct one, but that’s not very interesting.

To illustrate the problem here, suppose that (for whatever reason; we need not get into the details) the preponderance of desires would be fulfilled by hanging Green for murdering Hopkins, but that Green did not kill Hopkins (in fact, he committed no crime at all) and that you’re in a position to let the hanging go forward or to stop it.. Now many people would say that regardless of the balance of desire-fulfillments involved, you have a reason not to hang him – namely, that it would be wrong to hang an innocent man for a crime he did not commit. Clearly this is not a desire, but many - in fact the vast majority – of people would say that it is a reason.

Again, a classical utilitarian will say that a good reason for doing something is that it would produce more happiness (or a greater balance of happiness over suffering) than any alternative. But this again is not a desire, nor is it necessarily the fulfillment of a desire: the classical utilitarian will say that it makes no difference whether the happiness in question was desired or not, or whether it was the result of fulfilling a desire.

So in saying that the only “reasons for action�? are desires, you’re clearly not engaged in describing how moral language (including moral-ought) is actually used, regardless of the examples you cited. In reality a great many people (including non-religious ones) will offer all sorts of purportedly moral reasons other than desires for doing things.

Apparently you have some other justification for saying that only desires are reasons. But I have no idea what they are.

Again this is unclear. To illustrate, let’s look again at the case of Green. A person who believes that Green should not be hanged because doing so would violate the deontological rule against hanging an innocent man would say that there is a concrete, existing reason for not hanging him – namely the objective fact that he is innocent. Of course, what makes this a “reason�? is precisely the deontological rule. Or, the classical utilitarian will say that there is a real, concrete reason for doing the “right�? thing – namely, that it will produce a greater excess of happiness over suffering than any alternative. Of course, what makes this a “reason�? is the rule that one should always act so as to maximize net happiness. But then, what makes the existence of desires that are likely to be fulfilled by an act a “reason�? for doing it other than the rule that one should (ceteris paribus) act so as to fulfill as many desires as possible? (Actually this isn’t a rule according to your theory, but for the sake of simplicity we’ll pretend for the moment that it is.) How are these cases different? In the first case we have a real objective fact - the innocence of Green – which is claimed to be a reason for not hanging him. In the second, we have another real, objective fact – the fact that the act in question maximizes happiness –which is claimed to be a reason for doing it. In the third we have yet another (supposed) real, objective fact – a certain relationship between existing desires and the results of the act in question – which is claimed to be a reason for doing it. The role of real, objective fact is the same in both cases; the only difference is in the (purported) moral principle which is appealed to.

I’ve already covered the claim that other “reasons for action�? do not exist, but here I want to point out that some moral theories do not “invent�? reasons for action, but appeal strictly to reasons for action that do unquestionably exist even by your standards.

For example, many people say that the criterion for whether one “ought�? to do something is whether it conduces to survival (meaning presumably the survival of humans as a whole) more than any alternative. This criterion rests (at least arguably) on actual desires – namely the near-universal desire to survive. But it ignores all other desires except insofar as they promote this particular desire. This kind of theory gives no weight to the fact that an act tends to fulfill desires that are neutral with respect to survival, whereas DU will count this in its favor. Similarly, this type of theory will count it against an act that it fulfills desires that tend to be inimical to survival, even if the “anti-survival�? effect is very weak whereas the desires that are fulfilled are numerous and very strong. So this kind of theory doesn’t “invent�? reasons for actions, but it does ignore things that according to your theory are “reasons for action�?, and this (so far as I can see) is the only significant difference between the two kinds of theories. On what grounds do you claim that your theory is “correct�? in this regard whereas a survival-based theory is in error? Because your theory take into account all “reasons for action�? whereas a survival-based theory doesn’t? But on what grounds do you claim that all desires are “reasons for action�? when “moral=ought�? is being used? Again it’s clear that you’re not merely describing how moral language is actually used, because (as we just saw) even among those who count only desires as “reasons for action�? do not all count all desires as “reasons for action�?.

Besides, this answer gets you get into serious trouble, because your theory also ignores some desires – namely any desires of a kind that on the whole tend to thwart more desires than they tend to fulfill. But regardless of the overall tendency of a certain type of desire, it may be that acting on such a desire in a specific instance actually tends to fulfill more desires than it thwarts. But that won’t save it from being ignored in your version of DU: what matters is the overall tendency of that kind of desire.

Which brings us to the last major problem with this argument: it’s not an argument for your theory at all, but is an argument for a much simpler version of desire utilitarianism – one that counts an action as “morally right�? if it tends to fulfill more desires (or produces a greater preponderance of desire-fulfillment over desire-thwarting) than any alternative. This is obviously the theory that you argument supports, but it’s not your theory. Your theory counts an action as morally right is it’s “the sort of thing that a ‘good’ person would do�?, where a “good�? person is defined as one who has only the kinds of desires that tend, on the whole to fulfill more desires than they thwart. And this is clearly not the same as the theory that counts an act as morally right if it tends to fulfill more desires than any alternative. So you’re going to have to justify your theory against the very same kinds of objections that you raise here against other theories.

In fact, as I’ve pointed out before, your theory counts some actions as “morally right�? that not only fulfill fewer desires than some alternative, but in extreme cases counts actions as “morally right�? that fulfill no desires. The fact that a desire is of a type that on the whole, in general tends to fulfill more desires than it thwarts doesn’t mean that it always does so; occasionally acting on such a desire won’t fulfill any desires (other than the desire in question, of course). In such a case it might well be “morally right�? because it’s the “sort of thing that a person with only ‘good’ desires would do�? (since such a person will have this desire and might therefore act on it). But now suppose that the person actually faced with this choice is not a “good�? person, so that in fact he has no desire to do the thing in question. Then according to your theory this action is “morally right�?, yet doing it would fulfill no desires whatsoever; it might even thwart a good many desires. Of course such an action would never actually be done (since the agent would have no motivation for doing it), but that’s not the point; the point is that your theory identifies this act as the “right thing to do�?.

So what you need to do is to show how someone who adopts the “simple�? version of DU is “mistaken�? while someone who adopts your version is “correct�?. To make this point as sharply as possible, let’s imagine that in my scenario Smith is an advocate of this simpler version of DU, and that the act in question is one that this simpler version says is morally wrong while your more convoluted version says that it is morally right. Once again Smith (using your terminology) says ““This action would be morally right; Jones should do it; he ought to do it. But if he does do it, I’m going to condemn him for doing it, because I consider condemnation to be the appropriate response to an act of that sort. And this will be a moral condemnation; it will be based on all-things-considered, universalizable grounds.�? What reasons for condemning this action is he “inventing�? that don’t really exist? How is his criterion not properly universalized? In short, how is he making a “mistake�?? And if he isn’t, doesn’t it seem strange to you that he can be using moral terminology correctly and say such a thing? Doesn’t this seem just a little paradoxical to you?

Bomb#20

Definitely true. We've had it since our first exchange in 2004. It's why I switched from attempting to engage you in discussion to throwing peanuts at you from your peanut gallery.

When you wrote

"It's part and parcel of moral and intellectual responsibility to try and get at the underlying methodological problems as a part of pointing out the 1st-order theoretical problems when you're disagreeing with someone. Again, following the golden rule, it's something I would hope everyone else would do in the reverse position.",

you were explicitly asking for a conversation about the conversation. That's why I gave you one -- if you hadn't said that, I would have simply pointed out that what you said about Alonzo wasn't correct and left it at that. Perhaps I was wrong to take your statement seriously.

Yes, you repeatedly mischaracterized me and my arguments.

Which is nuts, because I never challenged you on parsimony. I challenged you on otioseness, and you diverted it into a discussion of whether parsimony was a good thing, at the end of which you declared us to be in agreement about parsimony. This failed to address my challenge.

That pattern is a figment of your imagination. You might want to consider not telling your opponents what they realize and what they're trying to do -- you aren't any better at it than you are at telling when your opponents think they're Newton.

I was not "trying to find something to disagree with". Disagreeing with you doesn't take trying. There was no such time as "when I realized that we had been violently agreeing all along". That is an entity you believe in the existence of without evidence. We have not been violently agreeing. I have been strenuously disagreeing with you all along, and you systematically fail to notice this.

The pattern I've noticed is that you say something I think is unreasonable in aspect A, I challenge you on A, you seize on some uncontroversial aspect B of what you said, and respond as though I'd challenged you on B. Then we're forced to talk about B for a while in order to correct your misunderstanding of what I said; at the end of which, you pronounce us in agreement on B -- that's the part of our discussion that's the wasted effort. Then you get annoyed at me for wanting to keep the subject alive. But your original claim still has aspect A to it, and we never had a chance to reach agreement on that aspect because you wanted to talk about something else, and at this point in the discussion you've become doubly resistant to talking about A, both because you think we should move on and because you'll doggedly interpret everything I say about A as though it were about B. So we never actually come to grips with each other over the original bone of contention. It's a curious and regrettable debate dynamic, and it's one I don't know how to fix; but what it isn't is evidence that you got aspect A of your claim right.

That's not me switching gears. That's me trying to get the gears back to where they were supposed to be, after you switched them. You argued we should dispense with moral facts in our explanations because they were otiose. The point of the consciousness example is that consciousness is otiose AND parsimonious. It's proof positive that otioseness isn't the correct criterion for parsimony. It's therefore proof that moral facts being otiose doesn't show they're unparsimonious. Show moral facts are unparsimonious and you've got a problem for realism. Show they're otiose and you've got nothing.

Nonsense. You list two of the four possible combinations, find we agree about those two, and declare on that basis that we were always in agreement?!? What about the other two combinations: the unparsimonious nonspooky and the parsimonious spooky? Going by your stated criteria for spookiness, there is a parsimonious spooky way to conceptualize consciousness. And that way is the best way, as far as I can see. And this shows that either your criteria for spookiness are wrong, or else spookiness is not undesirable. Either way, finding moral facts "spooky" by your criteria is not a legitimate criticism against them. You think it is. That is what I disagree with you about. That is what I always disagreed with you about. But you wouldn't talk about that.

No, it hasn't been. You've left out most of the things it was the intent of my question to ask you. Do you think this property is otiose? Do you think human behavior cannot be explained without it? Do you think you are applying the same standard for inclusion in your world view to consciousness that you are applying to moral facts? The evidence so far produced in this thread strongly suggests that you are not.

And it happens yet again -- I challenge you on point A, and your reply is "I never said B". I criticize you for falsely accusing me of falsely accusing you of claiming things can't exist, and your response is to simply do it again. If you put any critical thought into checking whether I'd actually accused you of denying anything's existence, you show no sign of it. I'm about ready to scream.

Is it possible you interpreted my charge

"You have a standard for parsimony that automatically excludes high-level entities."

as meaning

"You have a standard for parsimony that automatically excludes high-level entities from existing."?

If you read it that way, I guess that would account for your bizarre response. It's a far-fetched reading -- if that were what I meant, I'd have to have been referring to your standard for existence, not to your standard for parsimony. But let that pass -- maybe what I wrote was ambiguous, the miscommunication is all my fault, fine, whatever. What I meant to assert was

"You have a standard for parsimony that automatically excludes high-level entities from being parsimonious."

It's because I meant this that I called it a "standard for parsimony" -- I'm funny that way. Since this isn't the first time we've gone through this process, we have all long since clarified that it really is possible for unparsimonious entities to exist. It follows that what I'm accusing you of clearly isn't a denial of anything's existence. So, even if you're done with talking to me, can you please finally understand that it's unresponsive when you interminably reply to my questions about the desiderata of explanatory theories with assertions about what you are or aren't "advocating denying the existence of"? :banghead:

Why on earth do you think you're entitled to a methodology that doesn't produce disagreement? We disagree. And that's obvious, since you're a noncognitivist and I'm a cognitivist. Why do you think people argue with one another? I argue with you because I disagree with you -- you keep saying things I think are wrong. A methodology that leads to the conclusion that we agree is patently erroneous.

Why do you react to my challenges by finding some strained way to conclude that I agree with you? Do you argue this way because you evolved your argument style, and you selected for techniques that gave you the feeling you'd won? What you're doing could achieve that, by making your positions unfalsifiable -- if you don't admit to yourself that an objection to your position exists, then you won't have to refute it or know you failed to. It's like arguing with a Christian who refuses to admit that I'm an atheist. Try accepting my apparent disagreement at face value, and thinking about which implication of your words my disagreeing with best accounts for my words.

Okay, fine, we have bad conversational chemistry. You probably think if you got my meaning wrong it's because I stink at explaining myself. Maybe so. But every time I challenged you, it's because you said something I disagreed with, whether you're willing to believe me when I tell you that or not. Any time we argue for 30 pages and you conclude I agree with you, all it shows is that you didn't understand which implication of your words I had a problem with. We can break our impasses any time you become willing to abandon the idees fixes you form about which bone is the bone of contention.

I had to look up "ostensive". If you intended your quotation marks to indicate that you were quoting me, you weren't. Have you been talking to mirage?

I don't recall you ever comparing either moral properties or gremlins/angels to entropy and wave functions, and I can't find anything in my posts alleging you did, so I have no idea what I said that you think that's an accurate rendition of. I can only guess that I must have torn into you for some A, and this is the B that you've somehow convinced yourself I was tearing into you for. Whatever.

If you find it difficult to credit, why are you charging me with it? "This thing is not like this other thing" isn't even jargon; and when I complained about your use of actual jargon it was because I spend too much time trying to figure out what it means and half the time I can't. You don't appear to be trying to baffle the audience; you appear to have a problem writing arguments that make sense, and you appear to incorrectly believe the jargon helps.

But you didn't. You tried to steer it towards the discussion of specific moral facts, not moral claims. If you'd tried explaining how a moral claim like "I think I've earned Y" can be noncognitive, that would have been pertinent to our debate. But until we've established what if anything is being claimed, whether such a claim can be a fact isn't even a well-posed question.

So it seems. My bad, I guess. If I may be forgiven for borrowing a debating tactic you used on Alonzo, it's about as unproductive a strategy as responding to

"Evolution can't explain where the first cell came from, so evolution is a failed theory, so God must have specially created each species."

with

"Evolution is a theory of how one species comes from another; it isn't its job to say how the process got going; that's a job for a separate theory of abiogenesis."

I.e., it's a perfectly reasonable response; it instantly refutes the creationist argument; and it's plowing the sea. A creationist is apt to be utterly unmoved by it, because he sees that first-cell problem as settling the matter. Many people are intuitively convinced their own world view must be right as long as there are any unanswered questions in world views they see their own as being in competition with -- and whether there are errors in their own reasoning is an insignificant matter to them as long as their conclusions are right.

Noncognitivism stands or falls on its own merits as a linguistic theory, regardless of what ethical theory anyone has up his sleeve. If you have a problem with that concept then you are thinking like a creationist.

Enough of the conversation about the conversation. To summarize the metaethical points on which we disagree...

I disagree. MN seems to me to be an unparsimonious theory. Worse, it seems to me to be a poor predictor/retrodictor of moral speech. Postulating that "It's bad." means "THHBBBTHTTT!!!" doesn't appear to lead to the conclusion that people would want to express "I think my having done X caused THHBBBTHTTT!!! on you not doing Y."; yet MN appears to be committed to translating moral claims into that sort of construction. I think it is clearly inferior in both parsimony and predictive accuracy to another theory that makes essentially the same philosophical claims: error theory. It is therefore dominated. So it should be discarded, either in favor of error theory, or else in favor of some third theory that's better than error theory in either parsimony or explanatory capability.

Obviously I think realism is parsimonious. To say an action is bad is to say it's something a person would only do if the part of the mind for making people be good isn't performing its function in him. For moral realism to be right, there has to be such a part. Based on observation of brain-damage patients it appears there is one, in the ventro-medial prefrontal cortex. And postulating that natural selection evolved a complicated anatomical feature with no function seems less parsimonious than postulating a function for it, since organs with no function are the exception, not the rule.

But the bigger problem with your argument is that it's simply not a good reason to subscribe to noncognitivism. To make the leap you're making is a sense/reference fallacy. If moral realism indeed captures the essential conceptual content of what people mean when they predicate goodness of something, that circumstance instantly rules out noncognitivism, no matter how unparsimonious moral realism is. If you want to support noncognitivism, you have to support it on the semantics front, not the existence-of-moral-facts front. Any attack on realism's parsimony is only an argument for error theory, not an argument for noncognitivism.

Incidentally, you said "what people mean when they predicate goodness of something". The whole point of noncognitivism is to insist that people don't predicate goodness of things. Freudian slip?

bd-from-kg

Bomb#20:

I wasn’t especially interested in most of your last post to Hiero5ant (except to note once again that you seem to have “chronic Bad Conversation Chemistry�? with a lot of other people besides me), but the last part deserves a response, if only because you finally dropped a hint or two about your own metaethical theory. I gather that Hiero5ant has bowed out of this particular dialogue, so here goes:

How so? Parsimony, as I understand it, refers to the simplicity and number of premises needed to explain a phenomenon. For example, the Copernican model of the solar system is more parsimonious than the Ptolemaic because the planetary orbits can be modeled (to a close approximation) as ellipses with the sun at one focus in each case (simple and elegant), whereas the Ptolemaic system requires epicycles within epicycles to get the same accuracy. (Of course after Newton the contrast was even greater since the planetary motions in the Copernican model could be predicted to very high accuracy in terms of a simple force law whereas there was no such law that could predict the planetary motions in the Ptolemaic system.) But how does this apply to realism vs. noncognitivism? So far as I can see any plausible application would favor noncognitivism since it does not postulate any new entities at all, whereas any realist theory that attempts to “capture the essential conceptual content of what people mean when they predicate goodness of something�? has to postulate the existence of something more than what’s known to exist.

In fact, I’d say that if it really wants to capture this “essential conceptual content�? a realist theory has to postulate something with properties radically unlike anything that is known to exist and which are difficult to reconcile with the conceptual framework of what exists and how it works that has been produced by modern science. To be sure, the requisite entities or properties could exist nevertheless, but the premise that they do seems to be quite unparsimonious. Of course, in your theory the postulated new entity is pretty “ordinary�?, but I don’t think it succeeds in capturing the “essential conceptual content of what people mean when they predicate goodness of something�?. Perhaps this will become clearer when I discuss your theory a little later.

Now you’re sounding like Alonzo. The kind of theory you’re talking about is the rather primitive kind of noncognitivism propounded by Ayer and others a long time ago. Hardly any moral philosophers take this kind of theory seriously nowadays. If you want to offer a serious criticism of noncognitivism, talk about more modern versions such as Hare’s or Gibbard’s or Blackburn’s. Or for starters you can critique the theory that I laid out earlier.

Well, error theory is certainly parsimonious, but “explanatory capability�?? Surely you jest. There are all sorts of things that error theory fails miserably to explain.

First off, it fails to explain why moral language and discourse is so very useful. Every human society has found it absolutely indispensable. If moral language were inherently, intrinsically nonsensical this would be completely inexplicable. (As usual I’m using “nonsensical�? to include cases where the statement(s) in question assume the existence of something that doesn’t exist, not necessarily that they have no truth value at all, or are (singly or taken together) logically self-contradictory. Thus, for example, a discussion about which virgin should be sacrificed to the volcano god this year is “nonsensical�? in my sense.)

Second, it leaves it as a complete mystery why and how these completely erroneous concepts having no connection to reality came into existence and eventually became ubiquitous.

Third, it fails to explain why even people who have concluded that these concepts are false (e.g., that there’s no such thing as “intrinsic ought-to-be-doneness�?) still find it extremely useful, and indeed unavoidable, to make use of moral language and concepts both in thought and speech. This just isn’t the way this sort of thing ordinarily works. For example, if a society that has been sacrificing virgins to appease the volcano god finally decides that there’s no volcano god to appease, they don’t keep talking and thinking in sacrificing-virgins-to-the-volcano-god terms; this language, and the concepts underlying it, soon disappear.

More seriously still, this whole line of argument involves an erroneous notion of the status of error theories. It’s only appropriate to adopt an error theory in cases (like the sacrifice-virgins-to-the-volcano-god case) in which the language and concepts in question have no useful purpose and function. If we were to adopt an “error theory�? interpretation whenever the “conceptual content�? of a certain kind of language is false, we would have to adopt an “error-theoretic�? interpretation of classical mechanics and Newtonian gravitational theory – certainly as these were understood in the 19th century – because the underlying concepts are just plain false. But this would be absurd. These theories even today have a useful purpose and function: they predict many aspects of reality with remarkable accuracy. In order to justify adopting an error-theoretic interpretation of moral language you would have to show that it does not similarly have a useful purpose and function. But it quite obviously does have a useful purpose and function, and the job of the moral philosopher is to help elucidate just what they are and how moral language helps us to accomplish it.

Basically any language that has some intelligible connection to (external) reality has some potential use. So an error theory is really only the “best�? theory if the language that it purports to offer a “correct�? analysis of does not have any intelligible connection to reality (other than the reality of what’s in the minds of the people who use it). But to say that it has no such connection to reality is to say that there’s nothing about it that’s in need of an explanation (beyond the fact that it has no such connection to reality). So error theory by its very nature has little or no “explanatory capability�?. It’s as though someone were asked to interpret some “signals�? from outer space and concluded that they were not produced by sentient beings of any kind, not intentionally created for the purpose of communication, contained no message, and so were not “signals�? at all properly speaking. This would be a perfectly good “error theory�? for the “signals�? in question, but it wouldn’t be an “explanation�? of their meaning; rather, it would be a claim that there is no meaning; that there’s nothing to explain. Any theory that finds any intelligible connection to reality (i.e., any actual content) would have superior explanatory power. In fact, demonstrating that there’s any intelligible content – i.e. that the supposed “signal�? really is a signal - refutes the error theory.

Thus an error theory isn’t really a rival to other theories in the ordinary sense, but rather a claim that there’s nothing there for any theory (on that subject) to be a theory about. And the “evidence�? for such a theory can only be the failure to find any actual content – any intelligible connection to reality – in the “signal�? being examined. The correct procedure is therefore to consider any substantive theory that anyone can think of, and only if all such theories fail, to provisionally conclude that there’s “nothing there�? – i.e., to accept the error theory as a last resort.

OK, let’s say that a “part of the mind�? – call it M - influences behavior in some way, and you want to test the hypothesis that this influence consists of “making us be good�? in some sense. How do you go about this? Don’t you need to already know what constitutes “being good�? or “being bad�?, independently of what M does, in order to make this judgment? And if so, doesn’t that mean that you already have a concept of “goodness�? – i.e., of what constitutes “moral behavior�? – that’s independent of what M does? If so, to say that an action is “bad�? cannot mean that a person would only do it if M is “performing its function�?; it must mean that it does not conform to the concept of “goodness�? that you used to determine that M (when performing its function) makes people “be good�?.

Thus for example, you say,

But what sort of observations are these, if not that brain-damaged patients (or rather, people in which M has been damaged in a certain way) act “badly�?, or noticeably “worse�? than people whose brains aren’t damaged? And doesn’t this “observation�? require, as a matter of logic, that you, the observer, have a criterion for distinguishing “bad�? behavior from “good�? before you begin your observations? And isn’t this criterion (or rather, conformity to it) what you really mean by “goodness�??

Besides, if M exists, it must have evolved. And that means that its function must have changed over time – i.e., the kind of behavior that it produces or tends to produce must have changed, and will (if we survive long enough) change further. So if what it means to say that an action is “bad�? is that it’s the sort of behavior that will only occur if M isn’t performing its function, it follows that the kinds of actions that are “bad�? must have changed over time and will change further in the future. And not as a result of changing circumstances (which might result in the same kinds of actions having different results) but merely as a result of evolving changes in M. This doesn’t sound like an objective moral theory to me. One of the criteria that a moral theory has to fulfill if it’s to qualify as “objective�? is this: Suppose that an action X1 is such that its consequences are C1 and the alternatives X2, X3, X4… have consequences C2, C3, C4… respectively. If this action is morally right, then it must also be true that another action Y1 (possible or actual, at any place or time) which has consequences D1, and whose alternatives Y2, Y3, Y4… have consequences D2, D3, D4… must also be morally right if the consequences D1 are essentially similar (i.e., the same in all morally relevant respects) to C1, and likewise the consequences D2 are essentially similar to C2, etc., and if Y1 is an action of the same nature as X1, Y2 is of the same nature as X2, etc. (so that any deontological rule satisfied by X1 is also satisfied by Y1). This is admittedly a bit complicated, but the essential point is that the criteria determining whether an action is morally right must be time and location independent in the sense that the “rightness�? of an action cannot depend on the time and location of its occurrence (as opposed to circumstances that may well differ at different times and locations). But your theory clearly violates this rule by making the criteria themselves dependent on the current state of M, which is evolving.

Another problem here is that the function of M must necessarily vary among individuals, because such variation is an essential condition of natural selection: without it there’s nothing for natural selection to “work on�?. So what if this part of Smith’s brain (call it M(Smith)) would “make�? him choose X1 over X2 in a certain set of circumstances, whereas the corresponding part of Jones’s brain (call it M(Jones)) would “make�? him choose X2? In other words, what if M(Smith) and M(Jones) differ in such a way that what they “make�? their owners do is different? Would Smith then be “doing the right thing�? by choosing X1 (in the specified circumstances) whereas Jones would also be “doing the right thing�? by choosing X2 in the exact same circumstances? If so, then again we do not have an objective theory.

Note: As you’ve probably noticed, I’ve tried to put “make�? in quotes when it appears in the phrase “make people be good�? or some variant thereof, because this phrase seems very ambiguous and I’m not at all sure what you mean by it. But none of the arguments above depends on exactly what you mean by “making people be good�?.

But how would it follow from the fact that this (supposed) feature does not have the function of “making us be good�? (where “good�? is here understood as an objective property that some behavior has) that it has no function? It seems plausible that its function (if it exists) is to influence behavior in ways that are conducive to living together in successfully functioning social groups. This would probably be quite advantageous in terms of producing descendants, and so would be selected for. Nothing in the least problematic there.

Look. No metaethical theory, whether objective, subjective, or noncognitive, seems to be entirely satisfactory. That’s why metaethics is still hotly debated. One thing that everyone does, no matter what kind of theory they’re espousing, is to attack rival theories. The idea is that even though my theory has its flaws, yours has worse. This “destructive criticism�? may not be reason enough in itself to subscribe to the theory being defended, but it can be a good reason in conjunction with other arguments to do so.

In the case of noncognitivism, much of the criticism leveled at rival theories has another function: by pointing out how other theories fail to capture essential aspects of moral discourse it calls attention to these aspects and (hopefully) prompts the listener to think about what their existence implies. Why, for example, does considering something to be “morally right�? seem to automatically act as a motive for doing it, or for praising such actions, whereas considering something to be morally wrong seems to automatically be a motive for not doing it, or for condemning people who act in that way? Ordinarily mere belief that a certain proposition about the "real world" is true does not automatically have any such motivating effect. Is there something "magical" about the property of being "morally right" or "morally wrong" that gives it this kind of power? Or again: why does it seem to always be (at best) an open question, for any natural property (such as maximizing happiness, or maximizing desire-fulfillment, or conforming to some set of deontological constraints), whether an action with that property is always the “right thing to do�?? If "X is morally right" meant that X had such a property, it would be quite surprising (to say the least) that one could be sure that an action had this property yet be in doubt as to whether it's morally right. (Not impossible, as Alonzo has pointed out endlessly, but surprising. This kind of situation is very unusual, to put it mildly.) These questions seem to be unanswerable (or at least very puzzling) in realist theories, but have obvious, straightforward explanations in noncognitive theories. So these criticisms are not just criticisms of rival theories, but are a kind of “back door�? introduction to some of the strongest positive arguments in favor of noncognitivism.

The problem here is that the question of what constitutes the “essential conceptual content�? of “what people mean�? when they say something can be very elusive. Often enough, when people learn more they revise their opinion of what was the “essential conceptual content�? of things that they said earlier. For example, if you were to ask someone who believes that the sun goes around the earth what was the “essential conceptual content�? of his statement that the sun would rise tomorrow at 5:46 A.M. he’d give an answer that involved as an “essential�? intrinsic component the motion of the sun around the earth (combined with the fact that this motion would bring it above the horizon, as seen from where he was standing, at 5:46 A.M. the next day). But if he were to learn that in fact this apparent motion is an illusion produced by the rotation of the earth, he would very likely give an entirely different account of the “essential conceptual content�? of this very same statement. So whose account of the “essential conceptual content�? should we take: that of the man while he was under the impression that the sun goes around the earth, or that of the very same man after he learned the true state of affairs?

Essentially the same point can be made by considering the account that would be given by a classical physicist of the “essential conceptual content�? of statements that he makes about gravity, or about electricity and magnetism, etc. We can be sure that this supposedly “essential�? content would be declared utterly false or meaningless in the light of quantum mechanics and general relativity, and that the physicist would give a very different answer if he were to be enlightened about these theories and the evidence that led to the abandonment of classical physics in their favor.

Basically the argument between realism and noncognitivism is about the “essential conceptual content�? of moral statements, but this is not to be understood (as the examples above illustrate) as a question of what account the “average person�? would give of the essential conceptual content of his moral statements, but of how these statements are best understood, based on an analysis of their internal logic and how they’re related to the “real world�?. And this analysis should not be expected to give results consistent with “theoretical�? – i.e., metaethical – statements that the person might make, any more than an analysis of how statements about gravity made by a classical physicist are “best understood�? (based on a modern understanding that includes QM, etc.) should be expected to give results consistent with his theoretical statements – i.e., statements about the scientific theory on which his statements are based. Of course a “modern�? interpretation will be inconsistent with such statements, because we now know this theory to be false, but that doesn’t mean that all of his statements are meaningless, nor that the meaningful ones are all false. If we were to impose such a strict standard of “meaningfulness�?, it’s doubtful that anyone could ever say anything “meaningful�? about the “real world�? at all.

Sure, it has to be supported on the semantics front, but in light of what we know (or what it’s reasonable to believe based on best evidence) about what “exists�?. For example, an interpretation that involves the existence of an “intrinsic ought-to-be-doneness�? (which is the kind of interpretation that the average man on the street is very likely to give if pressed) is out of court immediately (IMHO at least) on the grounds that it’s extremely unlikely that any such property exists, or that we could have knowledge of it.

By now it should be clear what I think is wrong with this statement. If (as I believe) the vast majority of people understand their moral statements in terms of some version of moral realism, but that moral realism is false, that does not imply that we should accept error theory, any more than the fact that almost all educated people in the nineteenth century understood their statements about gravitation in terms of a “gravitational force�? that acted instantaneously at a distance, and that this understanding was just flat-out, radically wrong, implies that we should adopt an “error theory�? regarding these statements.

Jinksy

Did Alonzo say anything about disappearing for Christmas? He seems unusually quiet...

Bomb#20

What's your point? I take it, since you're not especially interested in that part of my post, that you aren't prepared to argue that the bad chemistry between Hiero5ant and myself is my fault rather than his. Is there any particular person for whom you are prepared to argue that it's my fault? Such as yourself, for instance? Or are you just making an Argumentum ad Numerum?

As you note, we have bad chemistry too. Why are you responding to me? I dropped a hint or two in earlier posts and at that time you responded with contempt. Are you interested in a thoughtful debate now? If I again engage with you, but you adopt the same debating tactics that IMHO produced the bad chemistry, it will be a short and unproductive discussion. Why go there?

Sure it does -- noncognitivism postulates a new linguistic construct. Which is to say, it postulates a new subconscious mechanism for turning thought into speech, speech that sounds and operates so much like a proposition that nearly everyone is misled by appearances into thinking it's a proposition, but that is not a proposition. And this mechanism, existing side-by-side in the speech centers of human brains with the mechanism that generates genuine propositions, has the peculiar property that fluent language speakers subconsciously know to use it instead of using their proposition generators, whenever they're talking to someone about a topic that certain philosophers, without the speakers' knowledge, have decided for their own philosophical reasons not to be a fit subject matter for propositional speech. That, to my mind, is an unparsimonious hypothesis.

Somehow, ordinary language users effortlessly achieve the computationally non-trivial task of forming constructions like "Murder is wrong." and fitting them into the rest of their speech patterns in a way that appears indistinguishable from the way "Gold is yellow." fits into them. It's substantially simpler to suppose they achieve this using the same mechanism they use for "Gold is yellow.", than it is to suppose we evolved two nearly identical parallel mechanisms for doing almost the same computational task. When we consider that the evolutionary process that created the linguistic machinery in our brains cannot plausibly have known that wrongness is ill-fitted for having propositions uttered about it, the noncognitivist claim is simply extraordinary. Extraordinary claims require extraordinary evidence.

I disagree. The entity realism postulates strikes me as having properties very much like other entities whose existence is uncontroversial. If you'd like we can debate that, if you think we have any realistic prospect of improving our chemistry. But this isn't the thread for it.

Thank you!

Sorry, I shouldn't have said "MN", but rather "MN as explained by Hiero5ant". I intended my participation in this thread to be a critique only of his arguments, not of MN in general.

I didn't particularly -- for lack of anyone showing willingness to think seriously about my arguments.

That would be a reasonable way to start. We could continue this discussion in your "Modern non-cognitive moral theory" thread. Are you interested in my thoughts about it?

Did you intend that as a figure of speech meaning, roughly, "What you said is ridiculous and you're obviously a moron."? If so, we can proceed. And you won't take offense if I call you a moron in return, if I ever feel so inclined?

Because if you intend to literally call into question whether I mean my arguments seriously, well, that's the sort of thing that just ruins conversation chemistry. If you genuinely want me to take the trouble to tell you what I think and why I think it, I can do that; but if you're going to decide you know what I think better than I do, as you did in our last conversation, I'm not going to bother. Once bitten, twice shy.

So? All that means is that an error theory needs an additional psychological theory to go with it. (As does a noncognitivist theory or a realist theory or whatever.) We still get to compare how well competing models of moral speech match the phenomenon.

In this respect, it's like evolution versus creationism. Evolution needs a theory of abiogenesis to go with it, and creationism needs a theory of atheogenesis to go with it; and even though we lack those theories, we still get to compare how well evolution and creationism match observed biology.

I disagree. That policy entails that if someone can show that sacrificing virgins in the volcano serves a useful function and persists because societies that abolish virgin-sacrifice become for some reason less viable, this would imply that error theories are off-limits and we have to postulate that people who say "We do it because the volcano god demands it." don't really mean what they say. It's an intrusion of a philosophical prejudice into a scientific question, and a spurious justification for committing a sense/reference fallacy.

Good question. When and where would you like me to take it up? It doesn't really fit your non-cognitive thread, and it certainly doesn't belong in this one. Do you want to talk about my theory, or yours, or both together, or what?

Including nothing in the least problematic there for the meta-ethical theory. Those are not two competing hypotheses for the function of M. Put them together and you get the quite popular normative ethical theory that "The Good" is whatever is conducive to living together in successfully functioning social groups.

That's true, but not quite on point -- I wasn't arguing against either non-cognitivism or best-defense-is-good-offense here. I was simply pointing out a defect in the logic of a specific argument of Hiero5ant's. The reason for attacking a rival theory is that if you have a strong argument, it reduces the probability of the theory you attack -- which means the percentage points subtracted from that theory's probability are distributed among competing theories. But in what proportion? They might be distributed equally or very unequally -- the devil is in the details. Hiero5ant's argument was structured in such a way that noncognitivism didn't get any of those percentage points. Even if his argument were correct, it would have left noncognitivism no more likely than before. It is therefore not a good reason to subscribe to noncognitivism.

Quite so; which makes that a big "If". But that's a different argument from the one I was commenting on. If you want a debate with me, you're going to have to get used to seeing drawn-out arguments against specific individual steps in the reasoning I'm refuting. Don't try to interpret them as attempts to settle the whole dispute in one fell swoop -- that's a recipe for endless misunderstanding and bad chemistry. This is not a subject that can be settled in a paragraph.

Obviously. I've said that myself in earlier threads. One of the most exasperating things about attempting to reason with noncognitivists is their tendency to harp on this fact as though it were the point in dispute. But simply exhibiting some P that is not a good reason to believe Q, when the other person's argument was never "P, therefore Q", does not qualify as an argument for NOT Q. It's at best a red herring; at worst a strawman argument. Let it go.

Moreover, when we say that what matters is how the statements are related to the real world, keep in mind that this is not a philosophical debate but a debate about a science: linguistics, which is a branch of psychology. The essential conceptual content is just that: conceptual. The place in the real world where concepts reside is in human brains, not in some Platonic Concept Space. So the thing we need to analyze is how these statements are related to what is really going on in the brains that generate them. The average person's opinion about what's going on in his brain is immaterial; the average philosopher's opinion about whether there is any "Moral Reality" for moral speech to refer to is equally immaterial. The average person's account is irrelevant because he's not an expert psychologist; the average philosopher's account is irrelevant because "Moral Reality" isn't the part of real world that the argument is about.

No, it's out of court immediately on the grounds that an interpretation involving the existence of referents for words to refer to is a category error -- a theory of the meaning of "banana" that involves the existence of bananas is out of court immediately. It is not the job of a theory of semantics to make claims about what things exist outside the mind of the speaker.

The issue for noncognitivism is whether moral speech tries to refer to some property like "intrinsic ought-to-be-doneness", not whether moral speech succeeds in doing so. An interpretation that involves an attempt to refer to such a property does not as if by magic become an interpretation that involves the property's existence. It is therefore not out of court, whether the existence of the property is likely or unlikely. Arguing that the property's existence or nonexistence determines which semantic theories are viable is a sense/reference fallacy.

By now it should be clear what I think is wrong with this statement. "P does not imply Q" does not imply "NOT Q". The correct understanding of moral speech is to be discovered by observation and analysis of moral speech. It is of course not to be found by consulting the opinions of non-linguists and assuming they're right. But it is equally not to be found by consulting the opinions of non-linguists and assuming they're wrong; nor is it to be found by repeatedly reminding everyone that the non-linguists' opinions can't be relied on; nor is it to be found by examining realism's parsimony. One might as well try to settle a dispute over protein folding with a seismograph. In order to support noncognitivism by attacking realism, you have to attack realism's semantic model.