THIS BLOG IS NOW IN STASIS.


PLEASE VISIT MY NEW WEBSITES:


My other projects include:


TrueFreethinker.com


My side projects are:


Worldview and Science Examiner


Fitness Trends Examiner (wherein I review individual exercises and workout routines, diet and nutrition, supplements and healthy snacks)


My YouTube channel

7/17/08

The Evolutionary Argument Against Naturalism

This will probably be an ongoing discussion here at AID as long as I'm around for the simple fact that I believe it is such a crafty, persuasive and sound argument. Once understood properly, there are significant problems that the naturalist will have to encounter. While this particular post uses Plantinga almost exclusively, there are related issues and arguments that will undoubtedly crop up in future explorations that will deviate from Plantingian waters.

With that in mind, let me offer a few words of advice for those of you that are encountering the argument for the first time. Some of these sound rather silly, but all of these have been lobbed against my presentation of the argument, and some I've seen already tossed at my mini-explanations on this blog.

+/-

1) The nature of R.

First of all, what is R? According to Plantinga, it is the proposition that "our cognitive faculties are reliable." This does not mean that we are always rational, or always produce true beliefs or any such related idea, but rather that our cognitive equipment provides enough accurate beliefs such that we are warranted, in general, to trust our cognitive equipment.

So then, R will not be attacked by anyone who wishes to debate this argument, as R is simply an assumption of every person (especially people that like to debate things like the existence of God). Without R, there is no justification for holding that ANY belief produced by our cognitive faculties (which is all our beliefs) which would include our belief that God does or does not exist, naturalism is true, evolution is true or I am reading a particularly wonderful entry on Atheism is Dead. Hopefully that is that.

2) The nature of N.

N is fairly simple to understand but possibly hard to pin down in print. N stands for "naturalism", and the hope is that N here is broad enough to cover whatever non-theistic position one might hold. Naturalism does not entail that everything that exists is physical, even though most naturalist might hold that. For example, a thinker I'd consider a naturalist but not a physicalist is W.V.O. Quine. Quine believed that numbers were abstract objects that had a kind of real, non-physical existence. However, he was most certainly would scoff at the idea of a transcendent Creator-God. Minimally, naturalism entails that theism is false. But what is it about theism that naturalists spurn? Let me offer the following definition of Naturalism (N in Plantinga's argument) tailored specifically for this argument:

N) The world, including our cognitive faculties, were designed by fundamentally non-teleological forces.

Alright, a few words of explanation here. What I am really driving at is the naturalists belief that the causes of our belief-creating-faculties were not designed with some far-reaching goal by any kind of intelligence. Don't let "design" trip you up, as I am using it in a neutral kind of way (much like Dennett does). In other words, there isn't some kind of agent that directed the universe to create and design minds that have the overriding purpose of producing true beliefs. I don't know of any self-respecting naturalist that would reject my charitable definition of N.

3) The nature of E.

I don't think that this one is quite important, though I've been complained to over the years for not taking the time to really flesh out a sophisticated account of E. What E stands for is simple- our modern account of evolution. I realize that there are various accounts of evolution, and that most of it is an in-house debate. Some groups place enormous weight on natural selection, while others lessen natural selection's explanatory power in favor of other factors (see genetic drift). To be honest, I haven't seen how the various interpretations of E have an impact on the argument, and for the most part a discussion on it serves as a distraction at best (though quite an interesting one). Besides all of that, the work of Stephen Stich (who is no friend to the argument) has preemptively answered all of the possible distinctives E could possibly harbor.

*If you are interested in this part of the debate, I kindly redirect to Stich's own work. If you have access to JSTOR, I recommend the precis of his book (otherwise, this link ought to work) If you are really brave, you can pick up the book which contains his work on the topic of evolution and rationality here. I'm quite familiar with the work itself as I presented a 30 page paper on his book less than a year ago for a local philosophy and science group. I would love to talk about this some other time, but unless the comments are tangentially related to Plantinga's argument, I'd ask that y'all hold off for now.

With that out of the way, we can proceed to the meat of the argument. Here is Plantinga's formula:

P(R/N&E)

This reads something like "The probability of our cognitive faculties is reliable given the conjunction of naturalism and evolution." Now, naturalists are going to attempt to explain using the resources available to them. Per our definition of N, we do not allow things like God into the mix. Rather, the naturalist will opt for another mechanical, non-teleological explanation for R. The following sentences are a token of such an explanation:

Naturalist: The reason our cognitive faculties are reliable is because of our evolutionary history. Creatures that had reliable cognitive faculties tended to survive better than those with unreliable cognitive faculties.

I assume that this would be the most widely used explanation of why we believe R is a justified belief. It is Plantinga's opinion that this explanation is entirely faulty. The naturalist must try to show how there is a natural connection between beliefs (mental content as opposed to merely cognitive structure) and action, and that the beliefs must be true in order to aid in survival. How is it that evolution would promise us true beliefs? Doesn't evolution give us reason to believe that we will be survivors of some sort, regardless of the verisimilitudinous nature of our beliefs? Consider the following concession by Patricia Churchland:

"Boiled down to the essentials, a nervous system enables the organism to succeed in the four F's: feeding, fleeing, fighting and reproducing. The principle chore of the nervous systems is to get the body parts where they should be in order that the organism may survive...Improvements in the sensorimotor control confer an evolutionary advantage: a fancier style of representing is advantageous so long as it is geared to the organism's way of life and enhances the organism's chances of survival. Truth, whatever that is, definitely takes the hindmost.

Patricia Churchland, "Epistemology in the Age of Neuroscience," Journal of Philosophy 84 (October 1987): 548.

Churchland's driving intuition is that adaptive behavior is king, and "true" beliefs are only desirable in a conventional, pragmatic sense, if they are desirable at all. Not all naturalists would just bite the bullet like this, however. Perhaps the naturalist would say that say something like this:

1) Evolution produces agents that tend to survive.
2) Agents that tend to survive tend to have true beliefs.
.: Evolution produces agents that tend to have true beliefs.

(1) is not controversial (perhaps it is merely tautologous). What reasons do have to think that (2) is true? Mightn't a creature have largely false beliefs and get along just fine? We have a few options here. Either:

A) Our behavior is determined by the structure of our beliefs (See here).
B) Our behavior is determined by the content of our beliefs.

If (A) is correct, then having beliefs is what is important rather than what those specific beliefs are. The color of a basketball does not matter as long as the basketball is present. Likewise, having a neuronal event takes care of the behavior of the creature without reference to to actual content of the belief. To explain this a bit more, let's say that neuronal event N is sufficient for adaptive behavior B. Would it then matter what attending thoughts are present with N? Not in the least. It is clear, then, that if (A) is true, premise (2) from above is false (unless we just get lucky and happen to have true beliefs correspond with each neuronal event). It seems to me that this is the most likely situation given naturalism, as it refers to physical happenings for an account of behavior rather than mental ones. (As Plantinga puts it, our beliefs are invisible to natural selection.) It is also the logical position of anyone who accepts the view that our cognitive faculties weren't "designed" to confer true beliefs; what reasons could one provide for a necessary link between beliefs and behavior? Surely it isn't logically impossible that false beliefs produce adaptive behavior. Why then posit such a strong connection between the two? It has the look of an ad hoc escape route written all over it. But even if (A) is false, (B) does not give us any more of a rationale for accepting the truth of (2).

Let's look at option (B) again:

Our behavior is determined by the content of our beliefs.

Granting this, does it follow that it is more adaptive to have true beliefs rather than false ones? Plantinga lists a number of beliefs that he believes are survival-conducive yet as false as can be:

*Everything is conscious
*God exists (Even some atheists admit that belief in God, while false, has been adaptive)
*Paul believes that petting tigers is the best way to survive, but he also believes that running away from tigers as fast as he can is the best way to pet a tiger.
*Paul enjoys the idea of being eaten by a tiger, but always flees in order to find a better prospect.

The point being, their are many appropriate behaviors that could possibly follow from a number of false beliefs. This means that (2) is false as it stands and furthermore, P(R/N&E) is low.

What follows from P(R/N&E) being low? It means that evolution does not suffice as an explanation of R, and that naturalists will have to supplement their explanation of R with something that provides a reason to think that R is true. Theism provides that supplement.

10 comments:

  1. If natural selection selects for survivability rather than truth, then how can we trust our beliefs?

    This is an important question. How can truth arise in a system without being injected with truth to begin with?
    (As a side note, you can derive just about every theistic argument by replacing the word "truth" with "life", "morality", "consciousness", "free will", etc. They all make the same mistake of failing to recognize the possibility of emergent properties.)

    First of all, our beliefs are not particularly reliable. If you watch a car come to stop at a red light, does it get longer as it slows down? Actually, it does. If the space shuttle is lagging behind the ISS in the same orbit should it thrust forward or backwards to catch up to it? Actually, it should thrust backwards.

    We're wrong a lot. Need more examples? Go look at some optical illusions. If God intended for us to have reliable beliefs, he did a half-assed job.

    EAAN argues that natural selection is an insufficient explanation for the generation of true beliefs. And I absolutely agree.

    Where EAAN goes wrong is where it claims that because natural selection is an insufficient explanation, there exists no natural explanation at all. After all, natural selection is how humans developed their mental faculties, so what else is there?

    The answer is that we have discovered truth-finding mechanisms that are not contingent upon our evolutionary past. These include logic, math, the scientific method, etc. For example, even if believing that 1+1=3 would have helped our ancestors survive, you and I still would not believe it. Its truth is independent of the selective pressures on our ancestors.

    But really the only belief that is important to EAAN is the belief in naturalism. Is this belief reliable?
    Or were we tricked into believing in naturalism because it was advantageous for survival? That's the real concern here.

    Except for the last few thousand years, which are a blink on the evolutionary time scale, it is unlikely that any of our ancestors even considered the proposition of naturalism. So, what we should be worried about is that we were tricked into believing something from which naturalism is derived.

    Any suggestions? Maybe the uniformity of the universe or something? I dunno. At least I feel fairly confident that the strength of EAAN relies on finding a bias introduced by evolution that leads to a false belief in naturalism. The theoretical existence of such a bias isn't nearly as compelling as a concrete example.

    ReplyDelete
  2. So finally it is here, the fabled unbeatable EAAN. What to say about it. Naturally, I am automatically skeptic towards an argument that says "because we are smart, God exists." In a way it's true of course, in the sense that we were smart enough to invent Gods when other explanations failed us.

    I would say that, as you have delivered it, the argument is basically a multi-layered straw man, dressed up in science and philosophical trappings so as to disguise its weedy nature. After all the hype, this was a bit of a surprise.

    The underlying flaw is evident in the form that your argument takes: you assume the role of a collection of naturalists and pretend to make the (best possible?) case for E&N, and (mysteriously) failing to argue for P=1 by the virtues of E&N you ultimately conclude that N must be false. This is not necessarily a fallacious way to argue, to present the best possible case of your opponent and showing that it fails; but the credibility of such an argument hinges critically on your ability to actually present your opponent's best possible case.

    That invites a couple of questions:

    1) Did you really, honestly, make the best possible case for N and P=1 by using the full explanatory power of E?
    2) If you failed to make the best possible case for N or used a spurious E, is your argument not a straw man?

    That you're not making a fair case for N by shortchanging E is obvious on many occasions, as the sequence that is kicked off by this example shows:

    "The naturalist must try to show how there is a natural connection between beliefs (mental content as opposed to merely cognitive structure) and action, and that the beliefs must be true in order to aid in survival."

    No, no, no. First off, you don't get to tell the naturalist what he must or must not do in order to kill your argument. To refute the EAAN, the naturalist only has to show that P can be greater than zero; if he can do that from first principles without answering a single one of your leading questions, the job is done, and you are left to refute it instead.

    Secondly, the quoted statement suggests a caricature of E that most naturalists would not subscribe to. You even admit as much yourself and quote a second naturalist contradicting the first and offering a different view, which however also turns out to be flawed. So what happened here? You brought in an obvious straw man, shot it down but left an overly simplistic version of E in its place - another straw man. And then what? Well you concede that even this one might not be right, so you introduce a third straw man (your #2) - but this time you don't expose it as such, but instead you go on to adopt it as the final E that you ultimately employ for the purpose of your argument. So. After deliberately introducing and shooting down straw men, you wish to leave the reader with the impression that you have exhausted all possibilities and leave the last one standing, although it is still a straw man.

    In no way, not by any stretch of the imagination, could this procedure be regarded as making a fair case for E&N, and ultimately this means that your entire argument is buried under a pile of straw men.

    I would have a lot more to say on why you do not make a representative case for N by shortchanging E. You basically just define E in the most simplistic terms, and then ignore its contents and implications for N and R completely for the rest of your argument. You can not realistically discuss R with such an oversimplified idea of E. Here's a short list of what is missing in your discussion of E and R: 1) instinct; 2) pre-adaptations; 3) evolutionarily stable strategies. Toward the end of your argument you try to link survival to beliefs, beliefs to behavior, and behavior to adaptation; there is no way your argument can be even remotely realistic without these components of E included.

    If I were to attempt any more detailed critique of your argument (such as, what exactly is wrong with your (1), (2), A and B) you would have to furnish it with a lot more specific definitions. For instance, define exactly what you mean by "cognitive faculties" and "beliefs" in your section on (and definition of) R. Specifically, you need to be clear on whether instinct is included in any part of your definitions, or if it is excluded. Since instinct is intimately tied to behavior and survival, and your argument links survival to beliefs and beliefs to behavior, you must also make clear where instincts reside in your system so that one can even begin address your argument.

    ReplyDelete
  3. Ya I never could swallow the whole idea of "people who believe false things die faster". In other words, all the people who's mind didn't create true beliefs simply die off.
    If my mind produced false data, such as "70s music is cool", this in no way means I'm going to end up dead with no children while Bob over here with true beliefs "70s music sucks" is somehow bound to have 15 kids and live to be 150.

    Every person on earth believes many many things that are false, but it seems to me none of it has anything to do with reproducing or living longer.

    So I just don't buy it that having true beliefs makes you survive longer and reproduce more. Except, of course, for extreme cases, if my family believes swimming with Jellyfish makes you live longer, then we are probably not going to survive very long. But these extreme cases don't work for mundane beliefs, of which most are.

    ReplyDelete
  4. "Crafty" though Plantinga's argument may be, it's not "persuasive" to me, because it is not "sound", for several reasons.

    Firstly- I'll accept the nature of N and E for the purposes of this discussion. R, however, is a bit sticky. "Our cognitive faculties are reliable" covers a lot of ground, and most of us will probably agree that some of our beliefs are more "reliable", or reliable in different ways, than others. For instance, I regard my belief that I can judge when to duck incoming bricks as pretty reliable, and I do believe that naturalism and evolution well account for the relative reliability of that belief. On the other hand, my belief that I could accurately predict, without looking, more or less where a brick would land, given the force with which it was lobbed, its weight, the direction, air friction, wind, parabolic trajectories, and so forth, is not so easily derived from evolution and naturalism alone: it requires science. Theism is no help here.

    You (or Plantinga) have the naturalist say:

    The reason our cognitive faculties are reliable is because of our evolutionary history. Creatures that had reliable cognitive faculties tended to survive better than those with unreliable cognitive faculties.

    I'll agree with that, with the proviso that "reliable" in this case means "it works"- in other words, being able to dodge incoming bricks, rather than being able to describe the math behind it. There is a continuum between these two kinds of "truth", to be sure, but only human beings (as far as I know) can give the formulas for parabolas. You go on to say:

    It is Plantinga's opinion that this explanation is entirely faulty. The naturalist must try to show how there is a natural connection between beliefs (mental content as opposed to merely cognitive structure) and action, and that the beliefs must be true in order to aid in survival. How is it that evolution would promise us true beliefs? Doesn't evolution give us reason to believe that we will be survivors of some sort, regardless of the verisimilitudinous nature of our beliefs?

    Evolution doesn't promise us anything, but it delivers. While many people, and most dogs, can't do the math behind it, they can successfully duck incoming bricks, which requires enough verisimilitudinosity from the cognitive faculties to distinguish between the huge numbers of harmless trajectories and the few dangerous ones.

    You then quote Patricia Churchland:

    [...] Improvements in the sensorimotor control confer an evolutionary advantage: a fancier style of representing is advantageous so long as it is geared to the organism's way of life and enhances the organism's chances of survival. Truth, whatever that is, definitely takes the hindmost.

    Here, Churchland is obviously using "truth" in the sense of "an (arbitrarily) complete representation" which does not enhance an organism's chances of survival: a dog doesn't need the math to dodge bricks, and neither do we. I agree: but we do need the math to send rockets to the Moon. You comment:

    Churchland's driving intuition is that adaptive behavior is king, and "true" beliefs are only desirable in a conventional, pragmatic sense, if they are desirable at all.

    I wouldn't call it an "intuition": it's fairly obvious that evolution favors what works. Being able to do math requires expensive brains, and if an organism can get by just being "pragmatic", then it will do so. That says nothing, however, about the value of higher truths such as math for humans in modern societies, and Churchland was not talking about that here.

    Now we're getting to the nitty-gritty. You have the naturalist say:

    1) Evolution produces agents that tend to survive.
    2) Agents that tend to survive tend to have true beliefs.
    .: Evolution produces agents that tend to have true beliefs.


    As a naturalist, I'll go along with that, as long as we keep in mind that "truth" here means "whatever works": the beliefs that keep our heads brick-free. You go on:

    (1) is not controversial (perhaps it is merely tautologous). What reasons do have to think that (2) is true? Mightn't a creature have largely false beliefs and get along just fine? We have a few options here. Either:

    A) Our behavior is determined by the structure of our beliefs (See here).
    B) Our behavior is determined by the content of our beliefs.


    Yes, 1) is uncontroversial. As to whether it's a tautology or not, that's a discussion for another Bat-time, another Bat-channel.

    But 2)- you must carefully define what you mean by "false". In the first place, do all creatures have "beliefs" that can be "true" or "false"? For instance, although bacteria can follow a nutrient gradient upstream to their death in hot water, it would be overreaching to say that they held a "false belief". Of course, there's nowhere to draw a line, and it's all just chemicals swirling around anyway, but if we're considering the evolution of beliefs, we must be clear about what a "belief" is: there is a continuum between very simple action-reaction (turning into the direction of increased sugar concentration, say), the IRM's (innate release mechanisms, such as the scent of a potential mate) and FAP's (fixed action patterns, such as the initiation of courtship behavior) of higher animals, up to the y = x squared of junior-highed humans. There is no line between "structure" and "content" here: content is fancy structure. So your dichotomy is false.

    Going on. You say:

    If (A) is correct, then having beliefs is what is important rather than what those specific beliefs are. The color of a basketball does not matter as long as the basketball is present. Likewise, having a neuronal event takes care of the behavior of the creature without reference to to actual content of the belief. To explain this a bit more, let's say that neuronal event N is sufficient for adaptive behavior B. Would it then matter what attending thoughts are present with N? Not in the least. It is clear, then, that if (A) is true, premise (2) from above is false (unless we just get lucky and happen to have true beliefs correspond with each neuronal event). It seems to me that this is the most likely situation given naturalism, as it refers to physical happenings for an account of behavior rather than mental ones. (As Plantinga puts it, our beliefs are invisible to natural selection.)

    Yes, our beliefs are invisible to natural selection, just as our genotype is invisible to natural selection. But the outworkings of our beliefs and our genotypes, that is our behavior and phenotypes, are quite visible to natural selection, and the connection between belief and behavior, like that between genotype and phenotype, is not random or "lucky". Sure, you can have a "false" belief that works- up until Galileo, people thought that thrown bricks described an arc (or some curve or other), but not a parabola, but somehow they still managed to dodge them. Their neuronal events were "false", but not too false, or they wouldn't have worked. As I said, your dichotomy between "content" and "structure" is false, and the correspondence between the "attendant thoughts" and the "adaptive behavior" is anything but mere "luck". If medieval people seeing incoming bricks thought about eating watermelons, instead of thinking about dodging, they would have been dead.

    Last but not least, Plantinga's list of "adaptive false beliefs":

    *Everything is conscious
    *God exists (Even some atheists admit that belief in God, while false, has been adaptive)
    *Paul believes that petting tigers is the best way to survive, but he also believes that running away from tigers as fast as he can is the best way to pet a tiger.
    *Paul enjoys the idea of being eaten by a tiger, but always flees in order to find a better prospect.


    Need I point out that such beliefs can only arise in a culture, which is subject to rather different selective pressures than evolution in the biosphere? Not only that, but Paul is mad and obviously needs a reality check: holding such beliefs is not likely to be adaptive, even in human culture, because it betokens a degree of confusion that would probably also entail an inability to dodge incoming bricks.

    In any case, we do require a supplement to naturalism and evolution, in order to get as close to the truth as we can: science. Theism, far from explaining the relative reliability of our cognition, is itself the source of many (not all, of course) false beliefs, such as the second one in Plantinga's list.

    cheers from cool Vienna, zilch

    ReplyDelete
  5. This seems like a fairly solid argument to me also, but I have yet to read any attempted refutations of the argument. I'm thinking especially of Richard Carrier, I'm sure he would have something to say about this.

    ReplyDelete
  6. I think this is a very strong argument, personally, and I find the atheist responses here interesting. It seems to me to be a whole lot of sidestepping of the issue. The response is either nitpicking at technicalities that don't change the argument so as to ignore the argument, or else an insertion of an atheist faith mechanism. The fact remains that naturalism fails to provide any basis upon which to argue anything.

    ReplyDelete
  7. Kuhlmann:
    Where EAAN goes wrong is where it claims that because natural selection is an insufficient explanation, there exists no natural explanation at all. After all, natural selection is how humans developed their mental faculties, so what else is there?


    I'm leaving y'all up to think of another explanation. Until a successor to naturalistic evolution comes along, the EAAN is successful insofar as that goes.

    Adonais:
    The underlying flaw is evident in the form that your argument takes: you assume the role of a collection of naturalists and pretend to make the (best possible?) case for E&N, and (mysteriously) failing to argue for P=1 by the virtues of E&N you ultimately conclude that N must be false. This is not necessarily a fallacious way to argue, to present the best possible case of your opponent and showing that it fails; but the credibility of such an argument hinges critically on your ability to actually present your opponent's best possible case.


    You are more than welcome to lay out specific instances of my straw-men. The reason I don't assign specific values to this Bayesian formula is because any designation would be wholly arbitrary. Certainly P(R)=1 would be good for the naturalists, but I don't see it producing a good reason for setting the bar that high (can any have a probability that high- 1 is equal to certainty).

    To your questions...

    (1) I believe so. My definition and usage of (E) was gleaned directly from naturalists; ones I respectfully disagree with. If you want specifics, you can look at the works of Lawrence Shapiro and Elliot Sober. Both have been ardent critics of the argument I've presented here.

    (2) Probably. By the very nature of it, (N) is notoriously difficult to pin down. If I have erred in my presentation of it, please offer your thoughts on what I did wrong.

    No, no, no. First off, you don't get to tell the naturalist what he must or must not do in order to kill your argument. To refute the EAAN, the naturalist only has to show that P can be greater than zero; if he can do that from first principles without answering a single one of your leading questions, the job is done, and you are left to refute it instead.

    I disagree. I DO have the right and obligation to inform my opponent (the naturalist) of what we are going to be arguing about and what I believe the goals of the debate are. Curiously enough, you seem to believe that the probability of (R) needs only be greater than 0 in order for the argument to be defeated. This is not the case. The probability of (R) given (N&E) needs to be REASONABLY HIGH. If it is low, we have a defeater for naturalism, just like one ought to believe that you can walk outside on a sunny day without being struck by a meteor. The premises in this argument are probabilistic, not all-or-nothing. Greater than zero probability only entails that something is possible, not reasonable.

    As for your second objection, the only thing I can think of is that you are confusing Plantinga with a naturalist. I do not have two direct quotations by naturalists in this article.

    Zilch:
    I'll agree with that, with the proviso that "reliable" in this case means "it works"- in other words, being able to dodge incoming bricks, rather than being able to describe the math behind it. There is a continuum between these two kinds of "truth", to be sure, but only human beings (as far as I know) can give the formulas for parabolas.


    You are confusing complexity with reliability. Reliable cognitive faculties entails that it generally delivers true beliefs/knowledge, which may include mathematics and physics, but the latter certainly aren't a requirement for reliability.

    On the other hand, my belief that I could accurately predict, without looking, more or less where a brick would land, given the force with which it was lobbed, its weight, the direction, air friction, wind, parabolic trajectories, and so forth, is not so easily derived from evolution and naturalism alone: it requires science. Theism is no help here.

    This comment was hopelessly off course. Science is a methodology. It is not a competitor to naturalism or theism. The point is, would one need to have TRUE beliefs about where a brick would land in order to dodge it. It seems perfectly possible.

    Evolution doesn't promise us anything, but it delivers. While many people, and most dogs, can't do the math behind it, they can successfully duck incoming bricks, which requires enough verisimilitudinosity from the cognitive faculties to distinguish between the huge numbers of harmless trajectories and the few dangerous ones.

    But why should one think that we need to have TRUE beliefs in order to duck a brick? Can't one have radically false beliefs about the brick but still but the parts of his body in the right place?

    Here, Churchland is obviously using "truth" in the sense of "an (arbitrarily) complete representation" which does not enhance an organism's chances of survival: a dog doesn't need the math to dodge bricks, and neither do we.

    You misread Mrs. Churchland. She does not believe that we need complete or true or ANY beliefs in order to survive. Neither does Dan Dennett. May I suggest a perusal of the following: http://plato.stanford.edu/entries/materialism-eliminative/

    As a naturalist, I'll go along with that, as long as we keep in mind that "truth" here means "whatever works": the beliefs that keep our heads brick-free.

    I heartily disagree. Truth is what corresponds to reality. On your definition of truth, naturalism "works" but does not correspond to reality. That is not a happy situation.

    There is no line between "structure" and "content" here: content is fancy structure. So your dichotomy is false.

    According to this position, then, brain/CNS structure would necessarily determine the mental states of human being. In other words, it would be logically impossible for mental states to be multiply realized. This is old-fashioned Identity Materialism which has long gone the way of the dodo, and for good reason. This would force someone to say that a pain is NOTHING BUT a firing of the "C-fibers" in the brain, which is experientially false.

    Not only that, but Paul is mad and obviously needs a reality check: holding such beliefs is not likely to be adaptive, even in human culture, because it betokens a degree of confusion that would probably also entail an inability to dodge incoming bricks.

    Why would it entail all that? I can only come to that conclusion by assuming causal connections between the minds (whatever that is) content and the brains structure.

    Logan:
    This seems like a fairly solid argument to me also, but I have yet to read any attempted refutations of the argument. I'm thinking especially of Richard Carrier, I'm sure he would have something to say about this.


    Gah, Carrier has mentioned something about it, but it was quite awful:

    http://www.infidels.org/library/modern/richard_carrier/reppert.html

    I'd rather you read a scholarly treatment of the work. It is well worth the money:

    http://www.amazon.com/Naturalism-Defeated-Plantingas-Evolutionary-Argument/dp/0801487633/ref=pd_bbs_sr_1?ie=UTF8&s=books&qid=1216757289&sr=8-1

    ReplyDelete
  8. "The reason I don't assign specific values to this Bayesian formula is because any designation would be wholly arbitrary."

    And therein lies part of the problem: the EAAN is a fundamentally qualitative argument, but it still makes a quantitative claim in its conclusion by saying that P(R|E&N) is "low" or "sufficiently low" - but it offers no mechanism for establishing exactly what is "low" as opposed to high in the context (your "pretty low" might be my "high enough"). To be meaningful in practice, we must know what the "low" in the conclusion really means, and that requires a quantitative investigation which we don't know how to conduct.

    It is important to keep in mind that in reality P(R|E&N) is either 1 or 0 since the event R has already happened and we are observing it after the fact, i.e., we know that P(R)=1. When we talk about P(R|E&N) having a value between 0 and 1, this is just a fuzzy way of expressing our degree of confidence in the statement given the evidence - we're not talking about probability density functions of relative frequencies in any mathematical sense.

    This is why I dislike the pseudo-scientific use of probabilities in the EAAN: it gives an impression of mathematical rigor that just isn't there. The argument could be made entirely without invoking probability jargon - what's the point of defining mathematical variables when there is no algebra or calculations?

    "By the very nature of it, (N) is notoriously difficult to pin down. If I have erred in my presentation of it, please offer your thoughts on what I did wrong."

    Well I think the opposite - there's not much to do about N, but your E needs a lot of work.

    "I disagree. I DO have the right and obligation to inform my opponent (the naturalist) of what we are going to be arguing about and what I believe the goals of the debate are."

    Then let me rephrase the objection: certainly you may tell me what you think I should be discussing (I have no way of stopping you anyway), but you must also be open to the possibility that your opponent can see ways of undermining your conclusions that you haven't thought of. Remember Orgel's Second Rule.

    "Curiously enough, you seem to believe that the probability of (R) needs only be greater than 0 in order for the argument to be defeated. This is not the case. The probability of (R) given (N&E) needs to be REASONABLY HIGH."

    This goes back to what I said about probabilities: you have no way of defining "reasonably high" in any meaningful way, and as long as the naturalist can show that a full account of E allows P(R|E&N)>0, the EAAN is reduced to a matter of who's personal opinion weighs heavier - i.e. it becomes an irrelevant argument.

    I should clarify something: I don't know whether the EAAN can be logically or empirically defeated, since we have no way of quantifying the probabilities it invokes; defeating it isn't my goal - it is enough to show that it is veridically worthless. And that in itself is a weakness of the argument - it invokes quantities which can not be assessed; the naturalist can argue that P(R|E&N)>0 while the theist can argue that P(R|E&N)<1, and in the end nothing of value is discovered by the argument.

    The reason for this procedure is of course that, what the theist really wants to do is to show that P(N)=0, but he does not know how to do that directly, so he tries to show that P(R|E&N)=0, which by Bayes' theorem also implies P(N)=0. But being unable to establish actual probabilities he is left with saying that P(R|E&N) is "sufficiently low," which is no longer a mathematical statement but just a statement about his own confidence in the hypothesis. Which is basically irrelevant.

    Again, I would be more interested in actually trying to make the naturalist's argument for P(R|E&N)>0 (i.e., make the argument for how evolution is eminently compatible with our having cognitive faculties that produce mostly true beliefs) - but as I said before, I can not address your shortcomings of E without the more detailed definitions that I asked for, and which you did not provide yet. So there are two options: either you improve your E to the point where we can have a more precise discussion about why you think E&N prevents R (and why I think it enables it), or, I have to do the argument from scratch with no reference to your E, in which case I would be doing all the work while you're resting on Plantinga's laurels.

    ReplyDelete
  9. Josh: if you're interested, I wrote down a few additional thoughts I had on the EAAN here, where I offer a starting point for an evolutionary counter-argument, and some (hopefully) better formulations of the ideas given in my two comments here.

    ReplyDelete
  10. Josh, you say:

    But why should one think that we need to have TRUE beliefs in order to duck a brick? Can't one have radically false beliefs about the brick but still but the parts of his body in the right place?

    Can one have false beliefs about an incoming brick but still successfully dodge it? Obviously. One might believe, for instance, that the brick was made by brickmaker Jones, whose pile of bricks is nearby, whereas actually the assailant brought his own, homemade, brick with him. I guess it depends on what you mean by "belief", and what a "radically" false belief is. I can't prove this one way or the other, but I suspect a study would show that most people having bricks chucked at them would think something along the lines of "whoa, an incoming brick on a trajectory towards my head! I'd better duck!", and not "Gee, that's a nice bluebird flying by" or "I'd better get in line for that bratwurst". Or do you have any evidence that there is no correspondence between beliefs and appropriate action, as you imply?

    According to this position, then, brain/CNS structure would necessarily determine the mental states of human being. In other words, it would be logically impossible for mental states to be multiply realized. This is old-fashioned Identity Materialism which has long gone the way of the dodo, and for good reason. This would force someone to say that a pain is NOTHING BUT a firing of the "C-fibers" in the brain, which is experientially false.

    Well, if I understand you correctly, you may color me an "Identity Materialist". Do you have neurological evidence for the multiple realization of mental states? Until you do, I will stick with the more parsimonious explanation: that our mind is what our brain does, and that our thoughts have a one-to-one correspondence with the physical state of our CNS. Naturally, the nature of that one-to-one correspondence is very complex, and might never be completely elucidated. But I'd like to see your evidence that Identity Materialism is "dead as a dodo", please.

    And what do you mean, that it is "experientially false" to say that pain is nothing but a firing of the C-fibers? Neither you nor I have direct experience of our C-fibers; our experience of pain is mediated in many complex ways, but there's no reason to believe that it is not another product of physical interactions in our CNS. Or do you have evidence for something more?

    adonais- bravo. People often make arguments based on the "probability" of one-off events which have already happened. One Christian I debated with was convinced he had an airtight proof of the existence of God. He said:

    1) We can't know that God does not exist, so there must be some positive likelihood, no matter how small, that He does exist.

    2) We know that the possibility of something coming from nothing is zero (ex nihilo nihi fit).

    3) Therefore, God the Creator must exist, or there would be nothing.

    I'll leave the fisking of this argument to brighter minds...

    cheers from overcast Vienna, zilch

    ReplyDelete