Wednesday, July 08, 2009

Can false beliefs be helpful?

Can false beliefs help us sometimes? If your goal is to pick up as many girls as possible (not a goal I endorse by any means), doesn't the belief that you are irresistible make you more likely to succeed, even though it's wishful thinking?

This may have some relevance to the EAAN.

21 comments:

PersonalFailure said...

I would say that everyone holds false beliefs and sometimes they are helpful and sometimes they are hurtful.

For example: if I believe myself to be spectacularly gorgeous, and I'm not, that's helpful to me, and hurts no one.

On the other hand, if I believe the Rapture is right around the corner, so why even try to save the environment (and I happen to be a Senator from Oklahoma), that hurtful to everyone.

Doctor Logic said...

Do we have to keep reading about the EAAN? It's a dead argument.

Forgetting for the moment that theism is far more vulnerable to this sort of attack than naturalism...

In our culture, it may well be better to (slightly) overestimate one's own attractiveness if one wants to have lots of sex. However, there are many cultures and subcultures around the world, and you cannot succeed in all of them by using the same strategy.

IOW, THERE'S MORE THAN ONE ENVIRONMENT, PEOPLE!!!! How many times does this have to be brought up? If you're locked into thinking there's only one environment, you're doomed never to understand evolution.

The issue isn't whether it's possible for false beliefs to be a survival benefit in a particular environment. Of course it's possible. In fact, it's downright likely. The question is whether it's better to have accurate or inaccurate beliefs when averaged across many environments! And if it's better to have accurate beliefs in the grand scheme, then it's better to have a general intelligence that's truth-guided.

Citing cases in a single environment isn't going to advance the EAAN.

DocShaw said...

The sheer hubris of believing humanity, an extant species for what? 10,000 years maybe, can endanger a whole frackin' planet that's been here for 4.54 billion years during which there have been hothouse periods, ice ages, cometary and meteoric impacts, caldera explosions, several mass extinctions...

priceless.

Chad McIntosh said...

Doctor Logic,

Which refutations of the EAAN do you have in mind that you think are so incontestably lethal?

Steven Carr said...

Of course false beliefs can sometimes be helpful.

Anonymous said...

"Do we have to keep reading about the EAAN? It's a dead argument.

That's a curious assessment for an argument that received so much attention and has been defended against all sorts of criticism by Plantinga an others in published papers and several dissertations.

Your argument concerning different environments suggest that you're not familiar with later formulations of EAAN.

I suggest you read Plantingas latest and imho best formulation of the argument:
http://philosophy.nd.edu/people/all/profiles/plantinga-alvin/documents/CONTENTANDNATURALSELECTION.pdf

While I'm not sure irrationality is a good charge against naturalism, EAAN surely shows that naturalism is a very bad explanation.

The Uncredible Hallq said...

Vic--If you haven't read Steven Pinker's book How The Mind Works, I strongly recommend it, because it has a section on this. Basically, the way evolutionary psychologists think about this issue is that usually, you want to have accurate beliefs, but under certain special circumstances, inaccurate beliefs will be helpful. This seems to match how people actually behave: we're pretty rational about a lot of things, but when it comes to assessing our own worth or the correctness of our ideological shiboleths, we get quite irrational.

The EAAN, in contrast, assumes that evolution is indifferent to beliefs. Your example, because how specific it is, fits more with mainstream evolutionary psychology than the EAAN.

Doc Logic--The "which environment" thing doesn't work so much for this example. Human psychology is everywhere the same. The mechanism behind what Vic is talking about seems to be believe you're irresistable --> confident behavior --> people think "oh, he must have something going for him." That logic works everywhere. Of course, you also risk doing things that make you look foolish through over-confidence, and maybe the costs of occasionally looking foolish are higher in cultures without urban annonymity.

IlĂ­on said...

"Which refutations of the EAAN do you have in mind that you think are so incontestably lethal?"

Considering of whom the question was asked, the answer will be, "Nah-uh!" Or, on a good day, something like, "'Naturalism' is true, therefore the EAAN is false."

William said...

This issue has been studied in detail in cancer patients, who, "objectively," ought to be pessimistic about their future life, but in whom it has been shown that a lack of pessimism increases survival:

http://cancercontrol.cancer.gov/brp/constructs/dispositional_optimism/index.html

Doctor Logic said...

Anonymous,

I looked at the Plantinga paper you cite. In it, Plantinga says:

Natural selection, in modifying the content properties of beliefs in the direction of greater adaptiveness, will probably not be
modifying belief-producing processes in the direction of greater reliability.


Really?!!! Why is that? Because Plantinga assumes that natural selection has been operating in a single environment with fixed rules. That way, the genes can arrange themselves over many generations into a mechanism suited to that one fixed environment. And, in that case, mental faculties don't have to be able to find the truth in any general case, but only what is useful in a particular case.

More of the same:

P, the property of having this particular content in the relevant kind of situation, was selected for—not because that content was true, but because the behavior P causes (caused) in that situation was adaptive. P was selected for because in that sort of situation it caused adaptive behavior; but the adaptivity of that behavior doesn’t depend on the truth of the content P comprises.

"because the behavior P causes (caused) in that situation was adaptive"? Adaptive to what? Clearly, Plantinga thinks that it is adaptive to a fixed environment/situation. And therein lies his error - an error that is pretty obvious to anyone familiar with evolutionary biology.

I've pretty much lost hope that theists on this blog are going to address this issue at all. If they did, they'd see that the EAAN is simply wrong, and Plantinga's remaining claims degenerate into the plain old AfR, i.e., the claim that physical systems cannot possess intentionality.

Anonymous said...

Doctor Logic, given that in any possible enviornment, and even across possible enviornments, there are more (indeed, far more) false belief/adaptive behavior combinations than there are true belief/adaptive behavior combinations, I fail to see how your point in any sense refutes the EAAN.

Take Plantinga's 'everything is a witch' example. This sort of false belief, attached to adaptive behavior, will work across enviornments. And, since there are countless possible false beliefs like this (i.e. that operate without a problem across possible environments), it seems to me that your objection fails.

Doctor Logic said...

Eric,

First, I want to set the record straight on the burden of proof.

In every rational argument, there's an implicit assumption that one's rational faculties are pretty reliable (R), e.g., that after several attempts, you'll converge on a correct conclusion. If, at a later stage in the argument, I again assume R, it costs me nothing because the assumption was already implicitly made in step zero. There will only be a problem if there is a necessary logical contradiction between the implicit assumption and the conclusion (or one of the other premises).

Plantinga tries to argue against this deflection by suggesting that a similar deflection for the problem of evil would be illicit. He says that the theist cannot simply claim that P(G/G&E) as a basic assumption. However, Plantinga's deflection deflection doesn't work. P(G/G&E) isn't a necessary or implicit assumption for every rational argument. In contrast, assuming R *is* necessary and implicit to every argument before the first premise is written down.

The implicit assumption of R helps the theist in the same fashion. Why think God made you rational? You can't devise a rational (non-circular) argument for this if your rationality has been compromised by God. Fortunately, re-assuming R costs the theist nothing either. So you don't need to argue for your own rationality if God exists because it was assumed at step zero. You would only be in trouble if we could show that God always led to a contradiction of R. Even I don't think that's true.

In other words, if R is a problem for naturalism, it's a problem for theism too. And to show it's a problem for naturalism, you have to show that R is *impossible* under N.

Doctor Logic said...

Eric,

I googled "everything is a witch" and found this exchange at Debunking Christianity in which you (or another Eric) say(s):

[T]he notion that evolution acts directly on our behavior and not on our beliefs is pretty obvious.

I don't agree. Behaviors close the feedback loop for evolution, so they're important. However, in the case of genetic evolution, one can say that the genes and their regulation is what evolution acts upon, and these in turn cause the behaviors.

Under naturalism, thinking mechanisms play the role of fast-evolving genes. They adapt to new environments and conditions much faster than genes. This makes thinking species better adapted to changing environments, e.g., migration from one environment to another, surviving catastrophe, etc. Moreover, our thinking processes create complex predictive models of our environment, those predictive models guide behavior, and the results of the behavior close the loop.

Given that the two mechanisms (thinking and genetics) are almost the same, why doesn't Plantinga go after genetics? Why doesn't he say that genetic information has nothing to do with survival, and only behavior counts? The answer is that he sees that genes aren't epiphenomenal, and that genetics determines behavior. On the other hand, he believes that conscious beliefs are epiphenomenal and don't cause anything. He thinks they're out of the feedback loop.

At this point, let's distinguish between thinking processes and conscious beliefs. The epiphenomenalist naturalist (I'm not one, btw) might claim that our unconscious thinking processes are adequately rational in the sense that they map the world onto a mostly-consistent, predictive model, and this model is wired into behavior. In this sense, zombies are rational. Presumably, we all agree that there's an evolutionary advantage to zombie rationality because of its powerful adaptive capabilities.

The relevant question is then, What connects conscious beliefs to the corresponding zombie constructs about the world?

Plantinga denies there's any way to connect them. He regards conscious beliefs as epiphenomenal, and, therefore, arbitrary. But if Plantinga really thinks there's no intentionality for conscious beliefs, he's done. He doesn't need to bring evolution into the argument at all. He just needs to tell the naturalist that her conscious beliefs don't reference anything (or that they reference lots of random things in succession), and so her conscious belief that naturalism is true is meaningless.

In other words, the evolution aspect is irrelevant, and Plantinga is just regurgitating the AfR.

So the critical question is whether naturalism REQUIRES conscious belief to be epiphenomenal. It doesn't. The AfR is just an argument from failure of imagination. It's not impossible for R&N, so the argument fails to defeat the assumption of R in step zero.

Lewis Moore said...

Doctor Logic,

Need we keep hearing you say that the EAAN is dead? It really is not.

Forgetting the repititions for a moment...

I only want to respond to your last post to Eric. Not that you don't raise some interesting objections, but it seems apparent to me from reading through these posts how such a conversation would go. So, Let's look at some of what you say.

I'm quite curios how you suppose Plantinga would go after genetics. Is there a genetic parallel to truth-reliability? What precisely is the feature of genes which he would argue that naturalism could not account for?

You are right to see that the reason he goes after "thinking" has something to do with its being mental, but you are wrong to assert that it must be epiphenomenal. His argument against naturalism concerns naturalism's ability to produce reliable cognitive faculties. Thus, it is concerned not merely with behavior but also with the relation of truth and falsity in the produces of our cognitive faculties TO behavior. What is the genetic corollary? He says that evolution does not select for truth, but for behavior. What, pray tell, would be the corresponding feature relating to genetics as truth relates to beliefs?

I’m also curios about your statement that Plantinga believes mental states to be epiphenomenal. He does not. But perhaps you meant that he assumes they are for the sake of the argument. This is not true either. In earlier formulations of the argument (such as the paper “Naturalism Defeated”) he mentions four possibilities. 1. Epiphenomenalism. 2. Semantic epiphenomenalism. 3. Beliefs are causal and maladaptive. 4. Beliefs are causal and adaptive. He then argues against naturalism’s ability to provide reliable faculties in each case.

Your responses to Plantinga display a misunderstanding which is common to most responses one finds on the internet. It is astonishing how many people confidently rebut the argument who never understood it in the first place. Plantinga is not a skeptic. He does not question logic, nor does he question as a matter of fact the reliability of our faculties. He is not arguing that beliefs are epiphenomenal in any way, nor does he deny intentionality. What he does argue is that epiphenomenalists who accept evolution should refrain from being naturalists as well, as combined, these positions would undermine the reliability of one’s faculties and is self-defeating. It would perhaps be more accurate to say that he is saying ANYONE who believes in evolution, no matter their philosophy of mind, ought not to be a naturalist, as to do so undermines the reliability of one’s cognitive faculties and is self-defeating.

Side Note: epiphenomenalism as such, does not entail unreliable faculties or false beliefs, their reliability would just have to be established by some other means.

You demonstrate the misunderstanding explicitly when you say, “Plantinga denies there's any way to connect them. He regards conscious beliefs as epiphenomenal, and, therefore, arbitrary. But if Plantinga really thinks there's no intentionality for conscious beliefs, he's done. He doesn't need to bring evolution into the argument at all. He just needs to tell the naturalist that her conscious beliefs don't reference anything (or that they reference lots of random things in succession), and so her conscious belief that naturalism is true is meaningless.”
All such rebuttals seem to arise from thinking that Plantinga holds to the very position he is arguing should not be held. How this occurs is beyond me.

Your response (and especially the quote above) is a horrendous account of Plantinga’s philosophy of mind, his argument, and the point of his argument. At the very least, abstain from such mischaracterizations. If you do not know Plantinga’s philosophy, then do not make comments about it. If you do not understand the argument, then please refrain from repeating frequently and confidently that it is dead and you have killed it.

Doctor Logic said...

Lewis,

You misunderstood my comment. It's mostly my fault because when I said "[Plantinga] regards conscious beliefs as epiphenomenal, and, therefore, arbitrary," I meant that (in the context of that particular claim) he regards naturalistic accounts of belief to be epiphenomenal. He obviously doesn't believe that actual beliefs are epiphenomenal.

I know Plantinga is not an epiphenomenalist, and I know he believes our faculties are reliable. I also know he doesn't attack genetics. I thought that was totally obvious to everyone (even on our side), so I thought I had more leeway to economize in my writing. Apparently not.

You say:

Side Note: epiphenomenalism as such, does not entail unreliable faculties or false beliefs, their reliability would just have to be established by some other means.

I know that there's no logical contradiction inherent in having reliable epiphenomenal beliefs. However, I wonder how you think the reliability of epiphenomenal beliefs could be "established". Couldn't the anti-epiphenomenalist argue that any such belief in an establishment of reliability requires that beliefs be reliable in the first place?

If you think that is the case, then epiphenomenalism is just as vulnerable to a self-defeat style argument without any mention of evolution.


cont...

Doctor Logic said...

Lewis,

So, let's drop epiphenomenalism because, in addition to its being a straw man, evolution is irrelevant to the attack upon it.

Semantic epiphenomenalism is basically a lack of intentionality. No evolution needed there either, if the AfR is to be believed.

That leaves us with causal beliefs. Obviously, the maladaptive ones we can skip. Now we're down to something with minimal straw in it.

In the adaptive, causal belief picture, our belief causing mechanism is genetic, but it has the advantage that it allows us to adapt to a situation rapidly (much less than a single lifetime, perhaps minutes or seconds).

Against this scenario, Plantinga considers ways in which the correct behavior of the caveman in response to seeing a tiger (i.e., running away) might be caused by faulty beliefs. He says:

Or perhaps he thinks the tiger is a large, friendly, cuddly pussycat and wants to pet it; but he also believes that the best way to pet it is to run away from it.

How did our caveman get these beliefs?

Suppose our cavemen have never seen a tiger before. These cavemen came from a valley in which all animals like to be petted, and humans like to pet them. So caveman #1 walks up to the tiger and tries to pet it. He's cut down by the tiger, and dies a painful death. How do the cavemen witnessing this horror come to believe that the best way to pet the tiger is to run away from it? Why run? That didn't work in the valley where they grew up. In their last environment, running away was an extremely poor petting strategy. Why not do a dance for the tiger, or sing a song, or have sex, or eat a sandwich? Why would Plantinga think the cavemen would think to run away in order to pet the tiger?

Because Plantinga thinks that those who believe otherwise will, over many generations, die out. But this totally defeats the advantage of rationality as an adaptive force. If our response to tigers depends upon the trial and error of generations of cavemen, we might as well respond to tigers with genetics instead of causal beliefs.

Try this same approach with any of Plantinga's supposed faulty beliefs, and consider the mechanism by which the belief was caused. Plantinga doesn't seem to care how beliefs get changed. He treats beliefs as if they are evolved over generations rather than inferred in seconds.

Plantinga can't see the adaptive advantage of true beliefs.

Plantinga has another variety of faulty belief...

Perhaps he is an animist and thinks everything is alive. Perhaps he thinks all the plants and animals in his vicinity are witches, and his ways of referring to them all involve definite descriptions entailing witchhood.

These beliefs are unverifiable. There is, in principle, no way to detect alive things or witches in this context. If there were, it would be adaptive for us to be able to know the difference if our survival depended upon it. In effect, Plantinga is arguing that unverifiable claims are more likely to be faulty. Amen to that.

Lewis Moore said...

Doc,

I’m not going to respond to everything you said. I have no desire to write and defend Plantinga’s argument in whole, as the project is horribly time-consuming, and you seem quite intelligent enough to delve into the argument on your own if only a few obstacles are removed.

In my last post I made several points, which I would like to briefly mention. First, that Plantinga’s argument is primarily concerned with naturalism’s ability to provide not for “thinking” in general, but reliable cognitive faculties, and by this he means faculties producing mostly true beliefs. If the argument were made about genetics, I fail to see what the corollary would be to cognitive reliability.

Second, I mentioned that Plantinga is not an epiphenomenalist, and neither did he make the argument only against epiphenomenalism. Third, I mentioned that epiphenomenalism does not in itself entail false beliefs or unreliable faculties. And finally, I said you misunderstood Plantinga’s argument and misconstrued his position.

All but the first received comment, so let’s look at your responses to them.

I apologize if I misunderstood your comments about Plantinga and epiphenomenalism. But, I still have to disagree with what you do say. You say that (“in the context of that particular claim”) Plantinga regards naturalistic accounts of the mind to be epiphenomenal. Reading from your last post to Eric, the context is how to connect conscious beliefs to their corresponding zombie constructs.

In some sense you are right here, but in what sense? Plantinga’s actual position is that naturalists ought to be eliminativists, not epiphenomenalists. So it can’t be that he thinks that naturalism in anyway entails epiphenomenalism, or that naturalists are all epiphenomenalist. However, you could be taken to mean that in the argument he allows for conscious beliefs in naturalism, but then only allows for an epiphenomenalist explanation of how to connect them to their corresponding zombie constructs. This doesn’t seem right either, for he also makes his argument against naturalism with non-epiphenomenalist positions. If he does focus on some form of epiphenomenalism, that is because the position is so widely held. It is not a straw man, and he has never pretended (so far as I know) that naturalists must have an epiphenomenalist position regarding the connection.

Cont…

Lewis Moore said...

Doc,

On to the third point. I would not argue for epiphenomenalism, and so I would not be interested in demonstrating how epiphenomenalist beliefs could be reliable, but I do not think an attempt to establish an epiphenomenalist account of reliable cognitive faculties necessarily suffers the same kind of defeater. Plantinga’s defeater takes the form that: If one accepts this philosophy, then one has a defeater for their beliefs, including the belief in this philosophy. Thus, if one accepts it, one has a defeater for it.

However, if one were providing an epiphenomenalist account of reliable faculties, then they wouldn’t have such a defeater, since the premise would not hold. If one accepted the philosophy, they would be accepting a philosophy which provides for reliable faculties (assuming, of course, that they successfully showed that this could happen). Accepting the philosophy would not provide a defeater for it. It is true that such an argument presupposes the reliability of one’s faculties, but this is not itself problematic. Here again, I think you have misunderstood Plantinga’s argument. His defeater is a consequence of accepting ideas. Thus, there is no parallel to the anti-epiphenomenalist argument you provide.

You may be correct to say that epiphenomenalism would be every bit as subject to A self-defeat style argument without evolution, but you are wrong to say that it would be every bit as subject to THIS self-defeat style argument without evolution. The problem lies in the conjunction of evolution and naturalism. You agreed in the last post that it is at least possible for there to be an epiphenomenal account of reliable faculties (that is, of faculties producing mostly true beliefs). Well, Plantinga shows that evolution combined with naturalism is not one such system. Theistic epiphenomenalism could at least escape these problems (though you may object that it suffers from others), as it could provide an explanation for how beliefs could be mostly true: namely, that God makes them that way.

You may be correct that epiphenomenalism is self-defeating, even without evolution. But that is not to say that it is self-defeating in the same way.

cont...

Lewis Moore said...

Doc,

Finally, the last point. (And the last post. I apologize for the length). You make a lot of comments regarding Plantinga’s argument, but I’m going to ignore them for the reasons stated above. Nonetheless, there are a few I would like to comment on.

You say at one point that “Plantinga can’t see the adaptive advantage of true beliefs”. I think he can see it quite well. What he can’t see is the adaptive advantage of true beliefs over false ones, and neither can I. You object that Plantinga treats beliefs as if they form over generations, but I don’t think this is true at all. What he does seem to believe is that the cognitive faculties which produce our beliefs would have to evolve over generation. I suspect you would agree with this. Beliefs may form in seconds, but the faculties which produce them, supposedly, do not. Plantinga’s argument, I repeat is against naturalism’s ability to provide reliable Cognitive faculties. That is, cognitive faculties producing mostly true beliefs. The problem here is that you focus on individual beliefs, whereas Plantinga’s concern is with the reliability of the cognitive faculties that produce them. The beliefs are illustrations that faculties which produce false beliefs can be evolutionarily advantageous, as false beliefs can be evolutionarily advantageous, but the argument is still about naturalism's ability when combined with evolution to provide reliable cognitive faculties.

So again, we have evidence that you do not understand Plantinga’s argument though you claim to refute it.

One last comment and then I will close. You say that Plantinga is in effect, “arguing that unverifiable claims are more likely to be faulty”. Not at all. The fact that the examples are unverifiable is no more an effectual argument against unverifiable claims in general than his examples of verifiable claims (such as that tigers like to be pet) are effectual arguments against verifiable claims in general.

I have enjoyed our short exchange and you are a very intelligent person, but I shall not be pursuing thus farther. Be careful that you understand what people are saying before you try to refute them, Doc. Pursue the truth, not a certain kind of answer.

You may have the last word.

Doctor Logic said...

Lewis,

Sorry, but you still haven't answered my complaint.

You say:

The beliefs are illustrations that faculties which produce false beliefs can be evolutionarily advantageous, as false beliefs can be evolutionarily advantageous, but the argument is still about naturalism's ability when combined with evolution to provide reliable cognitive faculties.

Which is just restating what Plantinga thinks without facing the difficulty.

You criticize me for focusing on beliefs instead of belief-generating mechanism, but you're describing Plantinga's mistake, not mine.

If we assume that beliefs are causal and that there is intentionality, then a set of beliefs is a theory about the world. Obviously, there are more false theories about the world than true ones. In fact, there are also more false theories that produce useful behavior than true ones that produce useful behavior. Plantinga is saying that a false theory about an environment can be adaptive in that environment if it produces the correct behaviors in that environment. Agreed. However, if you move from one environment to another, producing false theories (and overly simplistic ones) is less adaptive than producing true theories.

As you move from one environment to another, and gather more data, how will you form beliefs? Plantinga implicitly suggests that you incorporate a belief at random such that the belief leads to a theory with adaptive behavior.

Continued...

Doctor Logic said...

Lewis,

Continued...

Plantinga's idea is faulty for several reasons.

First of all, no one believes that neural inference mechanisms are random. When my tried and true petting method fails, I don't simply guess at a new petting method at random. There are thousands of alternate ways I might conceive of petting the tiger, but very few of them are adaptive. If I want my learning to be adaptive, I have to do better than random guessing.

Second, there's no causal connection I can imagine that connects the invention of a new, false theory with adaptive behavior within a lifetime. You've missed this in my previous explanations so I'll try to be more explicit.

It might be tempting for you to say that the causal connection between the caveman's (otherwise random) selection of running away from the tiger on the one hand, and the adaptiveness of the behavior on the other, is natural selection pressure by tigers. That is, cavemen who don't adopt this random belief that running away is a good way to pet tigers (as opposed to singing to them), will fail to reproduce because they've been killed by tigers, leaving the run-away-from-tiger clan to survive. However, this is effectively a genetic mechanism for forming beliefs. Plantinga's cavemen survive because they have a gene, albeit a neural one, for forming the belief that running from tigers is a good way to pet them. Needless to say, Plantinga's scenario betrays a total misunderstanding of the adaptive advantage of cognitive faculties. Rational thinking mechanisms are advantageous because they don't require people to die to prove them wrong. Death is not required for learning.

Third, Plantinga thinks nothing of abstraction and generalization. One of his other examples was that the caveman likes the idea of being eaten, but when he sees a tiger he always runs off looking for a better prospect because he thinks it unlikely that the tiger he sees will eat him. Implicitly then, Plantinga suggests that the caveman's general theory is that, when you approach X, and X is unlikely to procure desire Y, you ought to run at maximum speed to evade X. Thus, when you pass a sick cow, and the cow is unlikely to provide you with good milk, you ought to sprint at maximum possible speed in a random direction in the hope of encountering a better cow?

Again, Plantinga is clearly off the rails when you consider belief formation. Plantinga can't see it because he doesn't think about belief formation. Plantinga imagines that, under naturalism, we have (through some utterly random mechanism) some peculiar set of adaptive beliefs. He then claims that the beliefs are consistent with evolution by saying the behaviors in this static environment (where learning is unimportant) are adaptive. But in doing so, he misses the point and makes beliefs genetic.