Thursday, April 21, 2011

Apparently gratuitous evil

Consider this piece of evidence:

  • E: There are apparently gratuitous evils.
Does E favor naturalism (N) or theism (T)? To answer that, we need to ask which theory E is more likely to occur on.

Now, given theism, the mere existence of evil is perhaps not all that surprising—there are lots of great goods, such as heroic self-sacrifice, that cannot exist without evil, so God has good reason not to eliminate certain evils. But what about apparently gratuitous evils? An evil is gratuitous provided that God would have no moral justification for permitting it. If there are gratuitous evils, there is no God. An apparently gratuitous evil is an evil that appears to be gratuitous. That there are apparently gratuitous evils is surprising on theism, but not overwhelmingly surprising. After all, there may be goods of trust in God that are enabled by apparently gratuitous evils. And God has reason to create beings with fallible intellects, since intellectual limitations enable various important goods, and such beings are apt to make misjudgments on occasion. So it is not too surprising that some evil would look gratuitous to us despite not being so.

What about on naturalism? There is a lot packed into E, and much of what is packed in there is not friendly to naturalism. The evidence E entails such facts as:

  1. There are evils.
  2. There are beings that have a moral concept.
  3. There are beings that have a theological concept.
Now, the existence of instantiated value properties is unlikely on naturalism. It is hard to reduce such properties to natural ones. So, (1) is unlikely on naturalism. Moreover, (2) and (3) entail that there are beings that have concepts, and that is unlikely on naturalism, since (a) it is hard to reduce the property of having a concept to natural properties, and (b) the existence of beings with concepts is subject to the fine-tuning argument—it is unlikely we'd have constants in the laws of nature such as to permit there to be such beings. Furthermore, that we have genuinely moral concepts (as opposed to concepts of what promotes fitness, say) is perhaps not so likely on naturalism.

So it is plausible that once we take into account E's entailments, P(E|N) will be smaller than P(E|T), and hence on balance E supports theism over naturalism.

36 comments:

Louis said...

Isn't this something of a false dichotomy? What of atheistic supernaturalism? Just imagine the God of classical theism stripped of traits like personality and behaviors like miraculous intervention. It could serve to ground things like Platonism about numbers, moral realism, etc. And many of the features of the actual world that you reference as being less likely on naturalism than on theism could be equally likely on atheistic supernaturalism. Why posit all of the baggage of classical theism when a slimmer view could do the trick?

(Ontological arguments aside.)

awatkins909 said...

^ But his argument is an argument against naturalism, after all?

Even so, the existence of gratuitous evil doesn't make the probability of atheistic supernaturalism greater than theism, does it? Or at least, it doesn't make theism *less* plausible?

awatkins909 said...

I mean, *apparently* gratuitous evil.

Louis said...

Good point.

Alexander R Pruss said...

The question I was interested in is whether E confirms T over N.

It's an interesting question why in the West there are so few atheistic supernaturalists, by the way.

James Bejon said...

The potential problem I have with this argument is that the nature of E seems to depend on what it's being conditioned on.

To see this, remove E a step further back, making it something like

(E) We apprehend apparently gratuitous evils.

Now, why couldn't a naturalist say that, given T, E is most likely an apprehension of objective evils grounded in God, meaning the theist needs to explain their existence. But, given N, E is just our subjective interpretation of the world around us, so the naturalist needn't explain (1).

Christopher said...

I always think of Walter Kaufmann's critique of theology when I read arguments like this. Think of what you have done, Mr. Pruss:

You have begun by admitting that there are appearances such that, should those appearances map to reality, God's existence is highly unlikely or impossible. You have inferred from these appearances that God's existence is more likely than his nonexistence. Something is amiss here, and it is this:

Your critique creates a distinction between the claims of theism (evils, beings with moral concepts, etc.) and the claims of naturalism (no such things, possibly) that is easily collapsed when you realize that the argument about gratuitous evils is not a description of how the universe actually is or might be, as you appear to construe it. It is an attack on the internal coherence of theism.

Consider your argument slightly rephrased:

1. If theism is true, then there are evils.
2. If theism is true, then there are beings that have a moral concept.
3. If theism is true, then there are beings that have a theological concept.
4. If theism is true, then there are no gratuitous evils.
5. There appear to be gratuitous evils.
Therefore, either these appearances are false, or theism is false.

This is consistent with your original argument, it just doesn't omit the critical terms of the critique of gratuitous evils like yours does.

And remember that it does not follow from "these appearances are false" that "theism is true." Both options are live, but one makes theism no more likely at all, and the other makes it false. These things will not render theism more likely. At best, the defender against gratuitous evils will be able to show that possibly theism is true.

But if one assumes your parody version of naturalism in which there are no evils, then the problem simply disappears. The problem of gratuitous evils will not render naturalism more likely true or false on your own characterization of it.

Your argument, when properly construed as a defense against a charge of internal incoherence on theism's part rather than assuming the problem of gratuitous evils to be a kind of strange natural science that describes the universe and we infer our beliefs up from them, shows the real problem of gratuitous evils for you.

Remember now, the events that might be described as gratuitous evils will occur under either theism or naturalism. The problem for theism is capturing all of the moral language of theism, especially of the deity's presumed perfect goodness, which includes terms like gratuitous evils, while still giving us good reasons to think that that term 'gratuitous' (part of the baggage of theism) has no actual instances despite and abundance of apparent instances. The naturalist can say that "if theism is true, then x appears to be a gratuitous evil" whether theism is true or false. She can further say "if theism is true, then there are no gratuitous evils" and "if theism is true, then actual fact x is an instance of gratuitous evils" perfectly sensibly- she would simply be saying that theism is internally incoherent. Which I think is probably her goal. You have done little to hinder her.

Alexander R Pruss said...

"You have begun by admitting that there are appearances such that, should those appearances map to reality, God's existence is highly unlikely or impossible. You have inferred from these appearances that God's existence is more likely than his nonexistence."

There is something paradoxical sounding here, I admit that. To relieve the paradoxicality, remember that when we talk of "apparently gratuitous evils" we are not talking of evils that ultima facie, after all evidence has been weighed, appear gratuitous. It's tricky to specify exactly the sense of "apparently" here--it's more than prima facie and less than ultima facie.

'The naturalist can say that "if theism is true, then x appears to be a gratuitous evil" whether theism is true or false.'

On a standard understanding of the compositionality of language (though I myself am not so sure of that compositionality), in saying this, the naturalist is using of the concept of a gratuitous evil. If she does not have the concepts involved in the consequent of the conditional, she does not grasp the conditional.


In any case, it is surely a really serious cost to the naturalist to have to deny that there are evils and that some of them appear to be gratuitous. If the naturalist has to deny that rape and torture are evil, then naturalism stands refuted.


Note also that the question here isn't of internal consistency strictly speaking, because I am dealing with the inductive problem of evil.

Christopher said...

Mr. Pruss, I appreciate your reply, but I am not so certain of some of the things you have said. To wit:

"If she does not have the concepts involved in the consequent of the conditional, she does not grasp the conditional."

Now this is obviously true, but I fail to see its relationship to the naturalist I posited in my comment. Surely you will concede that one can have certain concepts without assenting to instantiations of those concepts. You and I both probably have very vivid concepts of Greek gods - if this is what you mean by concepts - without consenting to instances of them in the world. If by concepts you here mean to refer to properties or their sets, the problem is a footstep away- you and I both probably have concepts of what it might mean to have Greek godliness, or perhaps to have a relational property like "Son of Zeus," and yet neither you and I will assent to instances of those things. Same here- as is at the heart of my original critique, it is not enough for the theist to say that naturalists who reject evil (I will not accept your contention that naturalism entails eliminative materialism, but that is off topic) cannot abuse theism with gratuitous evils since it is on theism that certain collections of actions and events amount to "gratuitous evils."

"In any case, it is surely a really serious cost to the naturalist to have to deny that there are evils and that some of them appear to be gratuitous. If the naturalist has to deny that rape and torture are evil, then naturalism stands refuted."

Now this is plainly false. I reiterate my objection to the notion that naturalism makes moral terms meaningless, but your own language provides the stronger objection here. You have simply missed the point- I have informed you that the problem of gratuitous evils is a problem of internal coherence for theism, so one may adopt its language and its assumptions for the purposes of criticizing it without assenting to its truth. We do this all the time whenever we entertain a hypothetical that, like theism, probably is false.

Let me see if I can simply it for you.

Consider some event, or action, or happenstance, called x.

On naturalism, x occurs. On theism, x occurs.

So far so good. Now: on theism, x appears to constitute a "gratuitous evil." On naturalism, x does not appear to constitute a gratuitous evil (playing along with the conflation between naturalism and eliminative materialism).

On theism, if x is a gratuitous evil, then theism is false. As such, an instance of x renders theism internally incoherent.

This entire argument perfectly captures the premises you ventured in your post, it more accurately characterizes the argument from gratuitous evils, and does not meet any particular difficulties even if one is an eliminative materialist, since x only constitutes a "gratuitous evil" if theism is true, whether theism is actually true or not.

If you concede that we can entertain concepts and even deploy their language meaningfully without assenting to their truth, then you might see the meat of this point. If you do not, then we literally are not having a conversation at all, since I am employing a concept (the falsehood of theism) to which you do not assent which, one in a parodying mood might say, from your argument implies that you do not have a concept of the falsehood of theism and so you do not grasp it, any conditions on which it might be true or false, or even its language; on the most overstated version of your reply, you simply do not grasp naturalism and so cannot critique it.

Consider that you frame your objection around "having" the concept of gratuitous evils. Re-run your comment with "having" replaced by "believing" and you will see something of what I mean.

Christopher said...

To further clarify your errors:

"If the naturalist has to deny that rape and torture are evil, then naturalism stands refuted."

Now this is plainly false, but also not a problem because the naturalist does not have to deny that rape and torture are evil- I digress. It is false because two things, not one thing, follow from "If the naturalist has to deny that rape and torture are evil, then naturalism stands refuted." They are:

1. Rape and torture are not evil.
2. Naturalism is false.

Since you seem to know so little about naturalism that you do not understand the rich body of philosophical work on morality beyond the Divine Command Theory model that seems to undergird your entire argument, I will have to be brief. It could simply be that rape and torture are not evil! (Again, consenting to your profound misunderstanding of naturalism.)

But you might object- rape and torture are obviously evil. Is it obvious in the sense that it is something like an analytic truth? Then your argument is false, since an analytic truth is always true whether anything else is true or false, and since naturalism is possibly true, then necessarily possibly naturalism is true and rape and torture are evil. If it is obvious because it seems obvious to you because it's just so obvious, then you have descended into a radically subjective moral system that I believe you wish to avoid- at best, your beliefs about moral truths are guided by what appears to be moral truth to you.

Alexander R Pruss said...

I am inclined to disagree with divine command theory. Nor did I say that the existence of evils is incompatible with naturalism. And (1) and (2) don't follow from the conditional that precedes them--at most their disjunction does.

I take it that I and many naturalists agree that there are evils. But my claim is that the fact that there are evils is a surprising fact given naturalism. If we had no information other than that naturalism holds, we would expect a world with no evaluative properties being instantiated, i.e., P(some evaluative property is instantiated | naturalism) is low. Low, but not zero.

Obviously, there is a difference between us having the concept of gratuitousness and there being things that fall under that concept. But I still contend that it is a surprising thing, given naturalism, that we would have the concept of gratuitousness. We only have the concept of gratuitousness because we engage in highly abstract and morally charged philosophical and theological reasoning. It is at least prima facie surprising, given naturalism, that we would engage in such reasoning--it is, after all, prima facie unexpected that beings would naturalistically evolve that would be interested in philosophy.

Nightvid said...

A. Pruss, you claim

" P(some evaluative property is instantiated | naturalism) is low. Low, but not zero."

This is absurd. If organisms did not assign value, they wouldn't be able to survive. Clearly in sentient organisms you should expect evolution to favor the ability to apprehend some sort of value.

Have you been overly influenced by evolution denialists like William Lane Craig, or what?

Alexander R Pruss said...

I am not interested in this argument in the probability that we attribute value, but in the probability that there really is value there.

Nightvid said...

A. Pruss,

If value is attributed, that implies that it exists. Therefore, if the probability of attributing value on naturalism is not low, neither is the probability of value existing on naturalism low. So my argument still stands.

Alexander R Pruss said...

Would you accept the parallel: if divinity is attributed, then divinity exists?

Nightvid said...

A. Pruss,

I meant to distinguish between attribution and imagination, because I thought that is how you meant it (although for all I know, I could have been mistaken). Using this definition, if something is attributed, it exists (but not if it is merely imagined). I would accept your parallel if you had evidence, but in order to show that something is attributed by this definition rather than simply imagined or (falsely) claimed, you have to first show that it exists.

For the purpose of this discussion, if one can accept that behavior beneficial to the group can have an observable effect on the real world via its evolutionary effects, then the property or properties of that behavior beneficial to the group in virtue of which it is selected for in evolution can be identified with morality, and whatever these properties are is independent of what anyone thinks they are, and thus objective. Thus by construction, there must be some such factor(s) that exist(s) for some such factor to operate which exists within the naturalistic framework, it follows that objective morality exists in the naturalistic framework and thus cannot be used as evidence or premise for the existence of a god.

Alexander R Pruss said...

I don't think you can do that identification. What is morally obligatory is directly relevant (understatement) to what I should do. What behaviors helped (i.e., contributed to survival and reproduction) the human community tens of thousands or even hundreds of thousands of years ago is not directly relevant to what I should do. It's just pre-history.

Suppose it turns out, as it may very well, that treating members of other tribes as not fully human helped communities maintain cohesion and survive and reproduce. It does not follow that it is morally right to treat members of other tribes as not fully human.

some kant said...

How is that distinction between attribution and "imagination" useful or valid?

I just had some ice cream soda with my friends. It was, according to all of us, delicious. There was attribution, so accordingly deliciousness exists. Fine. I have never eaten mammoth meat. But it ought to be delicious (or maybe not). What did I just do? Did I attribute it, did I "imagine" it?

If I "imagined it," then what is exactly being "imagined" here? Deliciousness itself, or the deliciousness of mammoth meat? If it is the former, then deliciousness doesn't exist, which is absurd. If it is the latter, then the issue is not about the existence of an evaluative property; it is about the possibility of attributing that property to something, which renders the distinction meaningless.

Nightvid said...
This comment has been removed by the author.
Nightvid said...
This comment has been removed by the author.
Nightvid said...
This comment has been removed by the author.
Nightvid said...

A. Pruss,

Your first argument invokes false or misleading implicit pretenses. The fact that individual actions we do today have an effect on a larger group of others is as true today as back then, and as central and relevant to morality today as in the distant past. Stealing today harms the owner of a thing today for the same reason as always. Torture is harmful to the victim today for the same reason as always. Taking care of children is necessary for them for the same reason as always. Maybe the minute details of the behaviors involved are not the same, but so what? The principles are still applicable.

Even putting that issue aside, I am not talking about pre-history, I am talking about now (Notice everything I said in that part is in the present tense.) People tend to think of evolution as a past event, but this is mistaken. Evolution is ongoing and continuous, and exists today just as much as "back then". So even if my first rebuttal failed, you appear to have a gross lack of understanding concerning biological evolution and your mis-representation of it makes your argument fail.




Your thought experiment with the tribes is a complete misrepresentation of social benefit theory. While some may benefit from others being treated as sub-human, the ones treated as sub-human would themselves not benefit, but would "disbenefit". Thus my definition of morality does not, in your hypothetical case, mean that what you describe would be moral. Put another way, you are straw-manning my argument by assuming the other tribes don't count in my analysis. By "the group", I meant this in the most inclusive way possible.

You have yet again failed to make a convincing argument, just like every other theist I have come across the arguments of.

Alexander R Pruss said...

Ah, I thought your view was that the morally right is defined by the attitudes and behaviors that have had evolutionary benefits.

But I guess I misunderstood. So, it seems that view is that morality is defined by the attitudes and behaviors that presently have evolutionary benefits, right?

But no present behavior has present evolutionary benefits in the literal sense. Evolution is always diachronic--it is about passing genetic information to the next generation. So, the view would have to be that those behaviors that will yield evolutionary benefits are the right ones.

Alexander R Pruss said...

I just removed a comment some of the content of which was not conducive to discussion.

The basic point made by the comment was that if a highly intelligent and healthy man impregnates 10,000 women, that is for the evolutionary good of the human race, and that the Ministry of Truth might decide that opposition to evolution should be shut down for the evolutionary good of the human race.

But along with the basic point, there was a mildly offensive remark perpetuating a stereotype about internet atheists, and hence I removed the comment.

I would be happy to see the comment re-phrased in a more academic fashion.

some kant said...

Sorry about that one, Prof. Pruss.

You summarized my point. My problem with those evolutionary accounts of ethics (whether based on previous or future events) is that they can be used to justify some particularly nasty things, eugenics being the example that always comes to my mind.

And I don't think redressing them with more and more utilitarian objectives solves the problem. It just keeps on moving it to another level, as you showed with your example of the tribes, which can be treated as another instance of the utility monster, "benefits and disbenefits" aside.

Again, I apologize for the off-handed remark. I guess the wine from last night is still running around, but then again that's no excuse. Cheers.

Nightvid said...

A. Pruss,

I find it a bit hypocritical to demand that a naturalistic account of morality must have everything occur in exactly the present moment. It's your worldview, not mine, that tries to entice people by claiming benefits after death which for all we can tell don't even exist. That said, the evolutionary effects, however miniscule, DO occur within a single generation, which IS "now" in the sense that was relevent to your previous comment.

You also have attempted to sneak in a moving of the goalposts, hoping I wouldn't notice. You first claimed that any form of objectively describable value whatsoever is surprising on a naturalistic worldview, regardless of time or temporal localizability, and now you're complaining about details concerning its application, which do nothing to show what you had at first claimed. I'm not fallin' for that, sorry. Nice try, though!

Alexander R Pruss said...

I wasn't conceding anything. :-) I was just getting a clarification of the target view you're presenting.

Here are some options. Which of these, or maybe none of these, are you advocating:
1. x's particular action of A'ing is wrong provided that it in fact makes those genes that are responsible for A'ing less likely to be passed on.
2. x's particular action of A'ing is wrong provided that actions of that sort in x's species tend to make the genes responsible for these actions less likely to be passed on.
3. x's particular action of A'ing is wrong provided that most actions of that sort in x's species will in fact harm the passing on of the genes responsible for these actions.

Nightvid said...

A. Pruss,

None of the above. Rather,

4. x's action of A'ing is morally wrong if, when one includes both direct and indirect effects, it does something to one or more other members of the species, which, if they were done to x and his or her kin, would both (1) tend to decrease the genetic fitness of x, assuming ignorance for which member of the species "X" is, and (2) result in a state of affairs of the sort that "X" would, by its nature, seek to avoid, in a way that can be explained on the basis of evolutionary advantages and/or pressures (which may be past, prsent, or both).

Alexander R Pruss said...

Bracketing the merits and demerits of the particular account, I just don't see much reason to think that, if naturalism were true, one would have a moral obligation to avoid doing "something to one or more other members of the species, which, if they were done to x and his or her kin, would both (1) tend to decrease the genetic fitness of x, assuming ignorance for which member of the species 'X' is, and (2) result in a state of affairs of the sort that 'X' would, by its nature, seek to avoid, in a way that can be explained on the basis of evolutionary advantages and/or pressures (which may be past, prsent, or both)."

Granted, one might well have a natural inclination to avoid such actions, and society might have a natural inclination to punish such actions, but why think that, over and beyond these facts, the action would be wrong?

And as to the merits of the particular account, it seems easy to find cases of actions that are morally wrong if there is such a thing as moral wrongness, but that do not impair the genetic fitness of the person to whom they are done:
- causing modest gratuitous pains to people that do not impair their and their kin's fitness
- voyerism and similar violations of privacy that are done in such a way that the victim never finds out and is not harmed in any other way
- non-reproductive rape of unconscious patients
- a trusted spouse's lying to an alcoholic about having just seen a miraculous divine manifestation saying that the alcoholic should reform his or her life

Nightvid said...
This comment has been removed by the author.
Nightvid said...

Saying "I just don't see much reason to think that, if naturalism were true, one would have a moral obligation to avoid..." is both an argumentum ad ignorantiam fallacy, and an appeal to an epistemic problem to defend an ontological claim. Additionally, it is special pleading, for the atheist can equally well say "I just don't see any reason to think that, even if a Creator exists, what is in this Creator's nature is good", and IF you are allowed to take it as a brute fact, the atheist has no more difficulty taking as a brute fact the natural description of morality that I gave.

Also, you seem to be implicitly rejecting the notion of a posteriori identity. Suppose I gave a chemistry lecture in which I explained how the behavior of large numbers of H2O molecules gives rise to all of the properties we speak of when using the notion of "wetness" . If a student then exclaims "But I still see no reason to think that water really is H2O!" with the implication that "Science cannot explain water", we would rightfully think he or she is just talking nonsense. And yet your rejection of the moral account I gave is based on the exact same thing: A rejection of a posteriori metaphysical identity (in that case, between what is moral and what meets the description I gave.)

As to your last paragraph, I take it you believe that "victimless immoral acts" exist. (I do not believe that a knowably victimless act can be immoral). I think here our moral intuitions are shaped in large part by our reactions to the fact that (1) Such behaviors (or societal acceptance/toleration/permission thereof) tend to lead to corruption of the perpetrator or to act as "gateways" to victimFUL acts, (2) For all we know, they might be prone to desensitising the members of a society to its standards (akin to the "moral hazard" effect) (3) There is (in some cases) a risk of harming the victim that, from the epistemic position of the perpetrator, is real, and thus we react to it because it indicates a willingness to harm on the perpetrator's part, making it morally good to react insofar as such a reaction might discourage or prevent the perpetrator's behavior in a case for which it would have a victim,
(4) The act is not knowably victimless (for instance, in your invasion of privacy example, I would react to such a voyeur only because, for all I know, they could be plotting something more serious), or (5) Ease of enforcement within a society of "blanket" rules simply because we are somewhat afraid of arguments or conundrums arising when people disagree on the fulfillment of the criteria invoked in a more complex ruleset admitting of exceptions.

This is why I specifically said "direct or indirect effects". I know that morality has many subtle areas in it, and many things will seem counterintuitive at first about any actual reasonably complete set of meta-ethical principles, whether divine or otherwise. And yes, if we react to something as though it had a direct victim, even if there is rational justification only in terms of indirect victims or if we worry that there might be, it will cause us to feel morally queasy about it and when we forget where our intuitions come from it is easy to mistakenly believe that the existence of a victim is not needed for something to be immoral.

Alexander R Pruss said...

1. The epistemic stuff is relevant because you're responding to an epistemic point, namely my claim that P(there are evils | N) is low. That's an epistemic probability.

2. I take it you're saying that if you tweak the acts in such a way that they lack a tendency to directly or indirectly adversely affect fitness, then they are not wrong. It is easy to come up with situations where they do not. For instance, fitness has to do with reproduction. Thus, if you send two people utterly incapable of reproduction deep into space, far beyond the ability to communicate with anybody on earth, nothing that they will do to each other will tend to affect their fitness in the actual circumstances in which they find themselves. Yet they can plainly do many things that are wrong. I take your point to be just a biting of the bullet together with a bit of an explanation of why our intuitions go wrong.

I suppose you could define fitness dispositionally relative to normal circumstances, but I think that gives the wrong answers as to what we should do in abnormal circumstances. Thus, in normal circumstances, it sure harms fitness to amputate a leg. But in various abnormal circumstances, it is exactly the right thing to do.

3. I also think the account prohibits actions that are clearly permissible. Imagine that a man's sexual organs are cancerous, and unless these organs are removed, the man will die. But the man has no close relatives (so kin selection is irrelevant), though he is married and his wife is fully capable of raising a child on her own and would like to do so, all things being equal (suppose also that the wife is unwilling for moral or other reasons to go for artificial insemination). It is clearly not wrong for a surgeon to amputate the organs, even though by doing so he does an act that tends to harm the man's fitness, since if the organs were not amputated, there is a good chance that before dying the man would have a child.

Alexander R Pruss said...

Besides, it is plainly logically possible to do wrong things to members of other species. Let's say there are aliens who are person. Then it's plainly wrong to torture innocent aliens. Sure, one can talk about indirect effects like desensitization, but it is also easy to imagine cases where there is no non-negligible desensitization going on. For instance, take someone who is already as desensitized as can be. She still does wrong in torturing the aliens. So you shouldn't just restrict it to our species. I think that's a fairly minor modification.

Nightvid said...

1. Ok, maybe I misunderstood a bit, but still, how do you go from "I don't know what X could be" to "The probability of any X existing is low"? Clearly you wouldn't apply the same logic to the case of water - not having a reason that water is always H2O doesn't make the epistemic probability of water existing low!

2. You are forgetting about the " result in a state of affairs of the sort that "X" would, by its nature, seek to avoid" part. Namely, suffering. This can occur in the space example but would not occur in your amputation example if the patient is properly anaesthetized, hence the dis-analogy between those two cases.

3 and next post. First of all, in formalizing something previously described by informally defined terminology, it is common to have to split apart things that we would normally use the same word for. (A good example might be "planet vs. "dwarf planet"). Thus I don't see it as unreasonable to drop the assumption that "moral" implies "permissible". In particular, I am allowing for a situation in which, given the knowledge possessed by person "A", "A" reasonably believes it is in his or her moral duty to hold "B" responsible for doing a thing, even if such a thing was not actually immoral. So I don't really think I am biting much of a bullet, just being precise with terminology. But in addition, a definition need not be able to deal with every hypothetical case to be reasonable. For instance, many formal definitions (the context is fog and weather) of "visibility" would be inapplicable if human eyes could see X-rays. If that were the case, we plausibly would need to introduce new distinctions and/or terms, or even change a definition or two to fit ordinary usage. But that doesn't make our present definition of visibility "wrong" or "inapplicable" in the usual sense. Similarly in your contrived examples with aliens and space travel, etc. . I do admit that we would plausibly need more sophisticated meta-ethics at that point, but that is no more a problem for morality than for visibility. (I assume that latter concept doesn't seem all that weird to you...)

Alexander R Pruss said...

1. Well, the prior probability that water is H2O *is* quite low. It's only because we have good evidence for it that we think it's true. (And even so there are philosophers who question it.)

My claim was fairly modest. I wasn't claiming there are no good naturalistic theories of morality. I was only claiming that if the only thing you knew was that a world was naturalistic, you would have little reason to think that there is any such thing as good and evil in it.

2. Your definition conjoins two conditions for wrongness: (1) tends to harm fitness AND (2) tends to produce effects that people tend to avoid. So where only one condition is satisfied, you don't have wrongness on your account. Maybe you meant a disjunction?

3. I don't see how an action that isn't morally wrong could fail to be morally permissible. It's generally accepted that the two are interdefinable: an action A is wrong iff it is not permissible; it is permissible iff it is not wrong.

Nightvid said...

A, Pruss,

1. And how exactly do you calculate that probability? What is the space of possible cases you are assuming, and their corresponding probabilities, and where do they come from?

2. The criterion of tending to be avoided does go with reducing fitness under normal circumstances, otherwise the aversion would not have evolved in the first place. To separate the two in the context you are doing so in (unless I misunderstand you) is assuming an absurd hypothetical.

3. Since you seemed to completely ignore my response to that issue in the previous post, I see no need to respond at this point. Please explain what you find objectionable about the argument I gave before.