Pages

Thursday, December 13, 2012

Why Plantinga's EAAN Fails (I think).

Last Sunday, the Seattle Seahawks beat the Arizona Cardinals 58-0.  Isn't a rout by your team a grand and marvelous thing?  I called my younger brother up, and we agreed Dad, a dedicated Seahawks fan, would have really enjoyed that game. 

Something seems to have somehow been programmed into our brains that delights in crushing, easy victories.  The New Atheists have a dozen sure-fire, slam-dunk arguments against Christianity that they regularly use, and John Loftus is inventing new ones every month or so.  When I was a boy, I thought Noah's Ark had been found on Mount Ararat.  Wouldn't that be wonderful if it had been?  See!  There's the elephant's stall!  What's an elephant doing at 15,000 feet? 

Slam dunk!  Christianity is utter foolishness.  The Bible is proven beyond all doubt.

In Annie Hall, Woody Allen is standing in the line for a movie by Marshall McCluhan, and gets into an argument with a Columbia professor who is also in line.  Just then, McCluhan himself wanders by, and jumps in on Allen's side, telling the prof, "You know nothing about my work!"  Allen turns to the camera and says, "Don't you wish life was like that?" 


Alvin Plantinga
Greater minds than mine seem to succomb to this desire.  Take, for example, a "slam dunk" argument against naturalism, put forward first by C. S. Lewis, then by Alvin Plantinga, in more detail.  I'm one of Lewis' biggest fans.  And there is no doubt that Plantinga is a great thinker, with a perspective on faith quite different from that of Lewis.  If the two agree on anything, how likely that what they agree on is wrong?

Yet I think what Plantinga calls the Evolutionary Argument Against Naturalism is, in fact, mistaken. 

I am neither an evolutionary (nor any other kind of) scientist, nor a professional philosopher.  It may be that my own arguments against the EAAN themselves fail, and fail badly.  I would be happy for Plantinga's argument to be proven successful, and welcome supporters of that argument to defeat my defeaters, and rout the naturalistic opposition, along with me.


Plantinga's argument

I am currently debating the EAAN with half a dozen or more fellow apologists elsewhere, all of whom so far appear to take the opposite viewpoint from me.  One of my reasons for posting this argument here, is so people can find my disputed arguments in a readable form easily, instead of searching for them through a long and shifting thread.  (About twenty new posts have appeared, since I started writing this post.)  So to save time, let me simply quote Wikipedia's description of the EAAN (I will not read criticisms of the EAAN on that page -- for better or for worse, what follows will be my own thoughts about the argument):

Plantinga defined:
  • N as naturalism, which he defined as "the idea that there is no such person as God or anything like God; we might think of it as high-octane atheism or perhaps atheism-plus."
  • E as the belief that human beings have evolved in conformity with current evolutionary theory
  • R as the proposition that our faculties are "reliable", where, roughly, a cognitive faculty is "reliable" if the great bulk of its deliverances are true. He specifically cited the example of a thermometer stuck at 72 °F (22 °C) degrees placed in an environment which happened to be at 72 °F as an example of something that is not "reliable" in this sense.
and suggested that the conditional probability of R given N and E, or P(R|N&E), is low or inscrutable.

Plantinga's argument began with the observation that our beliefs can only have evolutionary consequences if they affect behaviour. To put this another way, natural selection does not directly select for true beliefs, but rather for advantageous behaviours. Plantinga distinguished the various theories of mind-body interaction into four jointly exhaustive categories:
  1. epiphenomenalism, where behaviour is not caused by beliefs. "if this way of thinking is right, beliefs would be invisible to evolution" so P(R/N&E) would be low or inscrutable.
  2. Semantic epiphenomenalism, where beliefs have a causative link to behaviour but not by virtue of their semantic content. Under this theory, a belief would be some form of long-term neuronal event.  However, on this view P(R|N&E) would be low because the semantic content of beliefs would be invisible to natural selection, and it is semantic content that determines truth-value.
  3. Beliefs are causally efficacious with respect to behaviour, but maladaptive, in which case P(R|N&E) would be low, as R would be selected against.
  4. Beliefs are causally efficacious with respect to behaviour and also adaptive, but they may still be false. Since behaviour is caused by both belief and desire, and desire can lead to false belief, natural selection would have no reason for selecting true but non-adaptive beliefs over false but adaptive beliefs. Thus P(R|N&E) in this case would also be low. Plantinga pointed out that innumerable belief-desire pairs could account for a given behaviour; for example, that of a prehistoric hominid fleeing a tiger:
Perhaps Paul very much likes the idea of being eaten, but when he sees a tiger, always runs off looking for a better prospect, because he thinks it unlikely the tiger he sees will eat him. This will get his body parts in the right place so far as survival is concerned, without involving much by way of true belief. ... 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. ... Clearly there are any number of belief-cum-desire systems that equally fit a given bit of behaviour.

Thus, Plantinga argued, the probability that our minds are reliable under a conjunction of philosophical naturalism and naturalistic evolution is low or inscrutable. Therefore, to assert that naturalistic evolution is true also asserts that one has a low or unknown probability of being right. This, Plantinga argued, epistemically defeats the belief that naturalistic evolution is true and that ascribing truth to naturalism and evolution is internally dubious or inconsistent.


My objections

Editing note: The first two arguments here are to Plantinga's argument itself, the second two, more to objections that came up to them in the discussion.

1.  Argument from simplicity.  The simplest solution would be for evolution to produce a brain that perceived reality fairly accurately, as the simplest solution would be to produce eyes that see reality somewhat accurately.  These Byzantine alternatives, like wanting to pet a tiger and thinking the best way to do so is to run away from it, would never work in practice. Clever, but highly improbable, what a philosopher would think up, but I doubt an engineer.

Nice kitty!
For one thing, running and evading a tiger is a highly complex action, involving the coordination of many muscles and body parts, hiding behind branches, climbing up trees, jumping into life-rafts, feeding it fresh fish, etc. How often do you engage in the exact motions that would save you in that situation? Once in ten years? EVEN IF one had some ridiculous notion, like the way to pet a tiger is to run away from it, among the whole set of possible beliefs, how likely is it your deluded mind would light upon that particular delusion, so as to take all the specific actions that would be required to save your bacon?

Vanishingly improbable.

And suppose that weird belief about petting tigers did save you once. How if you were Pi, on a boat with a tiger for weeks? Or suppose next time it was a snake, or a cliff, or a tribe of cannibalistic pygmies, or a flood, or an oncoming semi with a drunk driver and black ice on the road? In each case you are just so lucky as to hold a belief that is totally wrong, but that inspires you to take just the right actions to rescue yourself? All your life long?

This is what I mean by the need for simplicity. A brain that tells us things that really are, is much simpler than the endless chain of fortuitious chances that would be required for false beliefs to lead to adaptive behavior. 

A brain that created beliefs that were utterly tangential to reality, all things equal would get us killed quickly. Plantinga's proffered example of the tiger is ludicrous, I think, because the brain would have to come up with false reasons for doing the right things each and every time one's surival was at stake. That might happen once, but consistently? Simplicity here is absolutely essential: the skin protects the body because it has a set of characteristics that ward off nettles, insects, radiation, bacteria, and also allow evaporation, flexibility, and other important functions. If it were a computer software that fed in random traits at every encounter with objects in its environment, we would quickly be toast. 


2.  Analogy to the eye.  The simplest way to construct an eye that helps us survive, is to make the eye see what really is. The simplest way to construct a brain that helps us survive, should by analogy also be to make it perceive what really is. Plato recognized the correlation between these two: 

It was the sun, then, that I meant when I spoke of that offspring which the Good has created in the visible world, to stand there in the same relation to vision and visible things as that which the Good itself bears in the intelligible world to intelligence and to intelligle objects . . . Apply this comparison, then, to the soul.  When its gaze is fixed upon an object irradiated by truth and reality, the soul gains understanding and knowledge . . .  (Republic, vi. 508)

 A "sight" here is analogous to an "insight:" neither is a body part, both might be thought to arise from the action of body parts.

By this analogy, why should our eyes accurately see objects that don't directly affect our survival?  It does not often aid survival to see the dots on a monarch butterfly's wings.  Nor did it much help, before the Voyages of Discovery, to count stars.  But the ability to see objects like tigers, snakes, mosquitos, and clouds, could also be applied to stars or, with mechanical aids, distant galaxies or tiny bacteria. 
 
Suppose you want to construct a machine that represents a three-dimensional scene in two dimensions. There may be many ways of doing this, but the simplest are a mirror or a camera, that "reflect" or "fix" light as it actually comes in, rather than building up an image from scratch with some complex machine. It's the same way with "seeing" reality mentally. It would seem that the simplest way of navigating life cognitively, would be with a brain that "perceives" objects as they really are, rather than creating some complex mental machinery that happens to react to diverse situations successfully, while misperceiving actual states of affairs, as Plantinga suggests.


3. Teleology?  Some object to my argument, saying unguided evolution is not teleological.  It has no aim or purpose.  So how could it succeed in creating a brain that perceives the truth about things? 

Well, maybe it can't.  I'm not committed to the theory that unguided evolution can produce much of anything that is very complex.

But by hypothesis, evolution IS a blind engineer looking for simple solutions, in as much as it is posited to have formed all of life. If evolution created the eye, then the purpose of the eye -- to see things clearly -- is built into the function which the "blind watchmaker" is struggling to create, mechanisms that work to enhance survival and allow fecund reproduction. Of course this anthropomorphic and rather misleading language, but not on the key point: that if evolution works at all, one must be able to see apparent, if not real, planning in retrospect.  

4. Why don't our brains work any better?  Some of my critics claim that I simply failed to understand the EAAN, and therefore my arguments were hardly worth answering.  It seems to me that when someone misunderstands, a good thing to do is explain what they got wrong, and how to put it right.  But also, if the critics are right, and I am wrong, why is that so? The brains of both Marshall and his critics were, by hypothesis, divinely designed by an omniscient being to find truth.

But one side or the other, maybe both, has failed to find truth.

So perhaps this is (a) a consequence of the Fall or (b) of sin on their part or mine, (c) evidence that the human brain was not directly designed by an omnipotent God, or that (d) for some reason, a perfect God created brains prone to error.

In any case, it seems empirically obvious that our minds are far from completely reliable.  It is not self-evident that that state of affairs, or say Alzheimer's Disease, is easier to explain under theism than under naturalism.  (Though being somewhat skeptical of naturalistic arguments for evolution -- as Plantinga also is -- we for the sake of the argument ignore general objections to purely materialistic neo-Darwinian theories). 


Conclusions: I haven't really come to any definite ones, yet.  I am still willing to be corrected about the EAAN.  I certainly see some of Plantinga's points as interesting, and some of the distinctions he makes about neurology and belief useful.  But I don't think the "slam dunk" goes through.  This is not so unusual: even in the NBA, one can occasionally witness an All-Star center "slam dunk" the ball, and the ball come flying off the rim, the laws of physics doing their honest work even in the face of occasionally misfiring basketball genius. 

61 comments:

Thomas Henry Larsen said...

David, could you please clarify which (if any) of the following claims you reject?

1. On materialistic naturalism, beliefs are neural structures with (i) neurophysiological properties and (ii) propositional content.

2. On materialistic naturalism, the evolutionary process selects for adaptive neurophysiological properties and does not take into account true propositional content.

3. On materialistic naturalism, there is no good reason to suppose that beliefs with adaptive neurophysiological properties correspond to beliefs with true propositional content.

David B Marshall said...

I'm not sure it matters to my argument, nor that I would be the "go-to" guy for what materialists think. (I'll be asking a few skeptical philosophers to read this post and offer their own views.) But I would be inclined to dispute all three, if possible.

Thomas Henry Larsen said...

// But I would be inclined to dispute all three, if possible. //

Okay, you have me fascinated now. Why would you dispute all three claims?

David B Marshall said...

How can a proposition be a neural structure? 2+2=4 is a proposition, and it may be arrived at by means of neural structures, but clearly it is not itself a neural structure.

But as I said, I'm going to be asking some atheist philosophers for their opinions. I've already e-mailed two, and plan to e-mail at least a couple more, who can speak for themselves. One at least, I'm pretty sure will also dispute all three of your propositions, probably much better than I can.

Crude said...

A brain that created beliefs that were utterly tangential to reality, all things equal would get us killed quickly.

Well, not really. That's the point of Plantinga's 4 categories: it's not the beliefs that evolution selects for and against, but actions.

A helpful way to put this in perspective is in the following way: do bacteria have beliefs? I think most scientists would say no - and the same would go for a variety of other animals. Yet evolutionary processes presumably act on these things. They 'seek food', they 'avoid predators', etc. Yet there are no beliefs involved.

If you accept that, then you can accept that populations of organisms can flourish with no beliefs at all, because the beliefs don't matter - it's the actions. From there it's a small step to realize that populations with wrong beliefs can flourish, so long as their actions promote survival - and clearly, some do.

That's why your response is actually ones that moves you towards what Plantinga is saying: what would 'get you killed pretty quickly' aren't the beliefs, but the actions. Specifically, the survival-harming actions. If you had inaccurate beliefs that prompted survival-benefiting actions, well, then you wouldn't be getting killed. In fact, your genes would, all things being equal, spread.

In any case, it seems empirically obvious that our minds are far from completely reliable.

Plantinga's argument, I believe, is entirely a priori, not a posteriori. So responses like this just miss the mark.

A brain that tells us things that really are, is much simpler than the endless chain of fortuitious chances that would be required for false beliefs to lead to adaptive behavior.

I suppose simpler still would be a brain that doesn't care what it 'tells us', and just does whatever action in response to a given stimulus. And evolution would have weeded out the ones whose reactions are either poor, or poorly executed.

Crude said...

Also, Lewis and Plantinga's arguments are extremely different.

Lewis' argument didn't rely on reference to evolution. Plantinga's does. Lewis' argument has more to do with fundamentals of rationality, intentionality, etc that Plantinga is willing to by and large put aside for the purposes of his argument, and instead argue more in a probablistic manner.

David B Marshall said...

Crude: Lewis generally assumed evolution for his arguments to unbelievers, he just didnt' take much account of it. Lewis' argument was cruder, more purely philosophical, but I doubt evolutionary biologists would find Plantinga's argument very sophisticated biologically, either.

Some animals get by without eyes, too. That doesn't mean it's a matter of indifference whether eyes produced by evolution see what is. Neither would it be a matter of indifference whether minds perceived what is, as explained, above. So the analogy does not even begin to show that "it doesn't matter" what the mind thinks -- obviously, human beings have found survival strategies that bacteria have failed to "perceive."

A priori arguments can, as a matter of fact, be undermined by after the fact observations.

Crude said...

Lewis' argument was cruder, more purely philosophical, but I doubt evolutionary biologists would find Plantinga's argument very sophisticated biologically, either.

Sure, I just mean that at the end of the day, they are different arguments with different assumptions, even if they have pretty similar goals. And even there, not quite so similar (I think Reppert would say that Lewis' argument doesn't get one to theism even if it works totally, but it does rule out naturalism and raises the likelihood of theism as a result.)

So the analogy does not even begin to show that "it doesn't matter" what the mind thinks -- obviously, human beings have found survival strategies that bacteria have failed to "perceive."

All the analogy is meant to do is illustrate that you can have natural selection operating even when your organisms are lacking beliefs. It's because actions are what natural selection selects for/against.

A priori arguments can, as a matter of fact, be undermined by after the fact observations.

Some, not all.

Do you understand why Plantinga is saying that a posteriori considerations aren't going to work here? I'm curious what you've actually read of Plantinga's argument.

Also, keep in mind that the EAAN isn't supposed to establish theism, but undermine the conjunction of naturalism + evolution. It can well be the case that evolution + theism (or just theism) undermine our rationality as well.

I'm also not saying that the EAAN is a 'slam dunk'. I think Lewis' argument is better, and Ed Feser's summaries are better still. But I'm at least trying to explain where Plantinga is coming from here.

David B Marshall said...

Crude: I've read the whole argument, of course, in a couple different places, and found it unconvincing. I didn't memorize it, so if you can show where my arguments are wrong, rather than battle them with snark about my alleged lack of understanding, that might add to the conversation.

I've given four arguments against EAAN, and have yet to see a credible attempt at showing any of them wrong yet, here or in the original forum.

Sure you can have natural selection without beliefs. I never implied otherwise, and I've shown why that fact doesn't much help Plantinga's argument. (As if it needed to be shown.) Clearly, human being strategize as well as flight-response, and it is not absurd to suppose that ability to plan has something to do with our success as a species.

Crude said...

rather than battle them with snark about my alleged lack of understanding, that might add to the conversation.

I didn't battle you with snark. I asked you a sincere question out of honest curiosity, because I'm trying to understand what your exposure to this argument has been so I can fill in details if necessary. If you tell me you've read 'the whole argument', I'm left wondering if that means Plantinga's books/articles, the wikipedia entry, or something else.

But that's fine. Consider the question withdrawn.

I have no idea why you seem uptight at my commenting here lately, but I will say this: if you'd rather I not comment on your site, just say so and I will depart without complaint or ill feeling.

I've given four arguments against EAAN, and have yet to see a credible attempt at showing any of them wrong yet, here or in the original forum.

Then let's focus on what I think is the most problematic objection: number 4.

In any case, it seems empirically obvious that our minds are far from completely reliable.

Here are the problems.

1. Plantinga's goal with the argument isn't to show that our minds are completely reliable, or even close to completely reliable.

2. Nor does he demand that E+N show our brains to be '(close to) completely reliable'. That's why he argues that our cognitive reliability on E+N is either low or inscrutable, rather than merely 'flawed'. Because 'flawed' wouldn't be a problem.

3. The argument does nothing to establish that our brains would be reliable on theism, nor does it attempt to. It's not a comparative argument such that 'our cognitive reliability is low compared to theism or supernaturalism.' You can drop all reference to theism, and the argument will perform unscathed.

So by my estimation, 4 doesn't even get off the ground as an objection to the EAAN.

David B Marshall said...

Crude: I enjoy having you comment. You're welcome to come any time. But one of your virtues (sometimes vices) is that you tend to be very forthright: I tend to respond to people as I perceive them as posting, and am therefore inclined to do you the courtesy (sometimes discourtesy) of responding with equal frankness.

I read Plantinga's recent book on science, and read an article previous to that.

I don't see the relevance of your first two points. The point of (4) is to show that the same argument can be used to argue against divine creation of the human brain. It is not my main argument, but I think is an important consideration for Christians to "keep in mind."

Your third argument is more to the point. But obviously, our brains were either designed, not designed, or partially or indirectly designed. Evidence against design is therefore evidence for undesign. (I am using more specific teleology here than in the sense in which non-directed evolution can be said to design.) Plantinga's argument cannot, I think, be divorced from considerations that undermine his own position, whether implied or stated.

Crude said...

Either way, I intended no snark. Pardon my question, but I've been wondering about it for a while.

I don't see the relevance of your first two points. The point of (4) is to show that the same argument can be used to argue against divine creation of the human brain. It is not my main argument, but I think is an important consideration for Christians to "keep in mind."

The problem here is that it's not 'the same argument'. What you give in 4 is a general a posteriori argument against divine design (okay, maybe not an argument, but the beginnings of a concern) irrespective of whether our origin was via evolution or any other method. Plantinga's argument is a wholly a priori argument that relies on the assumption of naturalism and (a particular form of) evolutionary theory.

Evidence against design is therefore evidence for undesign.

But the EAAN does not attempt to argue that our brains are either designed or undesigned. Even if the EAAN is successful, that doesn't get you to a conclusion of 'therefore, our brains are designed!' At the absolute strongest, if successful, it provides a defeater for having a strongly warranted belief in E+N. (Note: it's the conjunction. At least in principle, just holding E or just holding N wouldn't be affected by this argument, as odd at it would seem nowadays to hold N without E.) Beyond that, it leads to some interesting skeptical results. But that's about all.

I think understanding the EAAN as an argument that functions as 'evidence against undesign' doesn't work, because at no point does the argument show or attempt to show that our minds are designed or that they're not undesigned. It's showing a consequence of holding two particular beliefs at the same time, due to the way they interact with each other.

I think that's the best way to view the EAAN. It's not an argument for or against naturalism, it's an argument about what can be rationally believed.

Crude said...

I think that's the best way to view the EAAN. It's not an argument for or against naturalism, it's an argument about what can be rationally believed.

I'll add - I know that seems odd as a way to put it. 'It's not an argument against X, it's an argument that you can't rationally believe X!' Doubly odd, since it's The Evolutionary Argument Against Naturalism.

But really, at no point does the argument attempt to undermine evolution (this is one thing I see many people, not here, assume that it does), nor does it attempt to undermine naturalism on its own. It functions as a kind of skeptical argument.

Ron said...

This is an interesting discussion.

I think Crude is right about this argument. It is a defeater for N+E. It isn't about God, design or even evolution or naturalism per se but rather the conjunction of those two.

To draw an analogy from apologetics that annoys me consider the Argument from Evil which is a defeater for theism arguing that believing in 1) Evil and 2) an omnipotent and omni-benevolent God exists is logically contradictory. Some apologists try to shift the issue to the atheist saying that "if you believe that evil exists, then this undermines your argument against me because evil has no meaning on the atheist worldview." Of course, this response is besides the point. The theist has to respond to this argument by showing that these two things aren't contradictory.

The same goes for the EAAN. You can argue that evolution really doesn't undermine the confidence one can have in naturalism by saying that it provides for true beliefs more often than not, I suppose. And to be fair you try to argue on the positive side that evolution would produce true beliefs if they were more likely to help with survival. I think Plantinga's tiger analogy is unfortunate for going off on a tangent really. Even if we grant this, naturalism isn't connected to survival in a basic way as one's beliefs about tigers, cliffs or rattlesnakes for example.

Anyway, just my two cents.

Victor Reppert said...

Plantinga's argument is one of several that can be thought of as versions of the AFR. For the longest time I was inclined to be skeptical of it because I saw it as a kind of Skeptical Threat Argument (if you have CSLDI you can see my explication there between a Skeptical Threat and Best Explanation argument), but I am not entirely sure what to think now.

Crude said...

Ron,

I think the problem with describing the argument is that, well... even the title of the argument presents it as an argument against naturalism. But the actual target isn't naturalism itself, but people's reasoning capabilities.

I wonder if calling it the Evolutionary Argument against Confident Naturalism, or the Evolutionary Argument against Naturalist Beliefs, would make it clearer. Probably not!

Unknown said...

Crude: That may be, and maybe I should have described four as "an important additional consideration for apologists inclined to use EAAN to support the Christian faith in some way." But that seems a little pedantic. And it's striking to me how many dismissive "responses" to my arguments have been given now, here and on the other site, without addressing the primary ones in any meaningful way.

So I still have no reasons to credit EAAN as successful.

Unknown said...

Sorry, Victor, I'm new to this field, and not following your acronymns. Can you explain a little?

Crude said...

And it's striking to me how many dismissive "responses" to my arguments have been given now, here and on the other site, without addressing the primary ones in any meaningful way.

I'm not sure what other site you're even referring to.

As for here... really, the conversation so far is between you and myself by and large, and I chose to start with point 4 of your list. I don't think I've made a merely pedantic criticism of it - I think 4 is removed entirely as an objection to the EAAN. (You may disagree - alright, well, I think we'll just have to move on in any case.)

So I'll move to 3. The problem is, I don't see how 3 works as an objection to the EAAN anyway. Maybe it's a response to someone on whatever other site you're discussing this, and would make more sense in that context. As it stands, it doesn't seem to be an EAAN criticism, and if it's a response to a particular defense of the EAAN, I don't see that defense spelled out. (The EAAN doesn't argue that a brain that 'perceives the truth about things' is logically impossible. It's just that the likelihood of such a thing is either low or inscrutable. Rather like how it's not impossible for a broken clock to be right about the time - hey, it may be right twice a day. But most of the time, it probably won't be.)

And I think Victor's referring to CS Lewis' Dangerous Idea.

David B Marshall said...

Crude: Thanks for clarifying Victor's reference.

As I explained, this article was originally posted to bring together and clarify some of my comments on another site, debating with eight or so Christians who like the EAAN. It was a bit of a rush job, since that debate was raging at the time, and I posted it in part as a means by which to think through the EAAN, and allow input from others, both Christians and non-Christians.

I grant that point 4 may not rebut Plantinga's argument per se. But you may notice that I was also arguing against how it was being used by my partners in that dispute. So with that much clarification, I think it's still worth mentioning (4), also as a warning against hubris on the part of Christians who do quite easily read EAAN as a weapon against materialism.

I do explicitly frame (3) as a rebuttal to an objection I faced with the first two points. Granted, this may not be clear from my introduction -- but as I said, I was typing fast.

It is worth keeping in mind, though, that evolution can in some limited sense be seen as teleological. And if evolution is as good at creating organs as evolutionists suppose it to be, I think that is a blow against Plantinga's claim that the reliability of our minds need be "low or instrutable" under naturalistic evolution, especially in light of points (1) and (2).

It's funny that you begin at the end of the argument, but perhaps that's a helpful procedure.

Crude said...

I do explicitly frame (3) as a rebuttal to an objection I faced with the first two points. Granted, this may not be clear from my introduction -- but as I said, I was typing fast.

That's fine. I'm just stating them as I comprehend them.

It is worth keeping in mind, though, that evolution can in some limited sense be seen as teleological. And if evolution is as good at creating organs as evolutionists suppose it to be, I think that is a blow against Plantinga's claim that the reliability of our minds need be "low or instrutable" under naturalistic evolution, especially in light of points (1) and (2).

I agree, evolution can be seen as teleological. (Strongly so, in my opinion. Stronger than it should make naturalists comfortable, really - but that's a whole other argument, or it will be until a certain point.)

But 'as good as evolutionists suppose it to be'? Here's where things get tricky. For one thing, just how good evolution is supposed to be at this kind of thing is up in the air. Can you run an 'argument from poor design' against evolution? I don't think so, because no matter how rotten the design is, so long as it's either functional in the most basic way, or it's 'neutral' enough of a feature, or... etc, you'll hear that this fits perfectly with evolution.

So I'm not sure how good evolution is supposed to be. In fact, when I hear it talked about, it's put as 'good enough for the population to survive'. If it's better than that, that's not evolution working - that's a happy accident. Just like, if a given mutation happens to become beneficial in the face of a new environmental threat/opportunity, suggesting that the mutation was provided in advance of the need is utterly verboten. Evolution does not look forward.

That's more a response to something you're saying now than addressing your points 1 and 2 (I think 3 and 4 can be put aside right now, I've said what I wanted to about them.) I'll get to that later.

James M. Jensen II said...

I can't accept the EAAN because it's incompatible with Wittgenstein's meaning-is-use theory, which I accept. Donald Davidson and Richard Rorty pointed out that when you accept meaning-is-use, one of the implications is that most beliefs have to be true.

Truth can't be determined apart from reference, and reference can't be determined apart from truth -- if I really thought hairdressers are old men who meet in windowless lodges every Monday night to practice esoteric rituals while wearing funny aprons, I'm confused more than I am wrong.

And if we have a basis for saying most beliefs are true, we have a basis for trusting our cognitive faculties, at least most of the time, no matter how those faculties came about.

(This is also why I don't feel even neoclassical foundationalism is very interesting. Any actual belief will necessarily have a web of other beliefs surrounding it that justify it, so taking it to also be properly basic seems superfluous.)

ozero91 said...

"But as I said, I'm going to be asking some atheist philosophers for their opinions. I've already e-mailed two, and plan to e-mail at least a couple more, who can speak for themselves. One at least, I'm pretty sure will also dispute all three of your propositions, probably much better than I can."

Will you be posting their responses?

Crude said...

if I really thought hairdressers are old men who meet in windowless lodges every Monday night to practice esoteric rituals while wearing funny aprons, I'm confused more than I am wrong.

The EAAN seems entirely compatible with people's thoughts being confused nonsense rather than their making coherent statements which are, in fact, wrong.

Cornell Anthony said...


"Plantinga defined:

N as naturalism, which he defined as "the idea that there is no such person as God or anything like God; we might think of it as high-octane atheism or perhaps atheism-plus."

E as the belief that human beings have evolved in conformity with current evolutionary theory

R as the proposition that our faculties are "reliable", where, roughly, a cognitive faculty is "reliable" if the great bulk of its deliverances are true. He specifically cited the example of a thermometer stuck at 72 °F (22 °C) degrees placed in an environment which happened to be at 72 °F as an example of something that is not "reliable" in this sense"

What book or article are you quoting from and what exactly is the date?

I would think this is the best layout

P = probability

N = Naturalism (not necessarily atheism, as an atheist can commit to supernaturalism)

R = proposition that are cognitive faculties are reliable

E = proposition that we and our cognitive faculties have come to be in the way proposed by the contemporary scientific theory of evolution.

From there I will proceed, but I want to hear your thoughts on this layout

Cornell Anthony said...

I don't think any of these objections work at all, though I'll start with number 1

"Argument from simplicity. The simplest solution would be for evolution to produce a brain that perceived reality fairly accurately, as the simplest solution would be to produce eyes that see reality somewhat accurately."

It was my friend Wade who pointed out to me the fact that even if it is true that the simplest causally-relevant set of beliefs is truth-conducive, that factor will not matter at all if beliefs are not causally relevant to behavior.

On SE, thoughts are like smoke to the fire. The fire does the real work, what color the smoke is is irrelevant. On SE, the semantic content of a belief (e.g. "snow is white") is a useless byproduct that doesn't cause anything.

Whether it's simple or complex doesn't matter if it's causally irrelevant, and so it's unlikely that the belief will be true. Because it could really be anything at all and it wouldn't matter.

Just like what beliefs I have in my dreams don't matter because they won't affect my behavior.

This is just conceding to the reliability of the senses anyways,

"For one thing, running and evading a tiger is a highly complex action, involving the coordination of many muscles and body parts, hiding behind branches, climbing up trees, jumping into life-rafts, feeding it fresh fish, etc. How often do you engage in the exact motions that would save you in that situation? Once in ten years?"

How often do I do this, you say? I don't think making an example which just considers the daily life of OUR time period will do your objection much justice either, as we'd have to make examples from earlier times.

David B Marshall said...

0991: Yes, I haven't had time to ask everyone I had in mind yet, and one whose thoughts I hoped for said he's pretty busy right now, but will see what he can do . . .

David B Marshall said...

Cornell: You say you don't think my arguments work "at all," but your rebuttal implies otherwise. Obviously, a big part of Plantinga's argument is that beliefs on naturalism (fine, fine) are not likely to be true. Conceding that there is reason to believe they would be (even tentatively) is to concede that my first argument does work, which I agree it does.

I don't know what SE is -- I don't see any "S"s in your glossary -- but I see no reason at all to think "thoughts are like smoke to fire." Aside from which, smoke DOES have an effect (ever smoke a salmon?), and dreams do too (read Brother Yun's Heavenly Man, or the Book of Genesis, or maybe Freud).

Aside from which, I see no reason whatsoever to grant your major premise, that human beings are totally unlike what we know from experience every single waking day of our lives they really are -- homo sapiens, thinking man, whose thoughts (for better or for worse) certainly does influence our actions. If we got that wrong, we all live in the Matrix.

I don't know what your final comment is responding to; obviously my examples mostly are relevant to the life of early man. I, personally, have never had to run from a tiger.

RD Rauser said...

David, I too find myself joining the ranks of the unpersuaded. (This is perhaps not surprising since I have my own form of the argument from reason in "God or Godless.")

First, a tangential point: the argument from reason (or, rather, a family of arguments from reason) is much older than either Plantinga or Lewis. Descartes presents his own argument from reason.

Second, as for your rebuttal I don't think that you've demonstrated that cognitive faculties which produce largely true beliefs are more likely to be selected for through undirected processes than are cognitive faculties which produce largely adaptive but not true beliefs. And that's the core of Plantinga's argument.

David B Marshall said...

Well, my arguments may or may not work, and I appreciate all the answers, but I'm a little nonplussed as to how to respond to so many autobiographical reports. I've given two substantive reasons why evolution (if it works creatively at all, as supposed) ought indeed to select for cognitive faculties that produce true beliefs. No one yet has shown why those arguments fail. I'd be happy to change my mind, but mere assertion should not overthrow what appear to me to be strong, and certainly remain unrebutted, considerations.

ozero91 said...

Marshall, if you are looking for some responses to the EAAN in literature, here's one.

This is Plantinga's recent paper on the issue:

http://www.andrewmbailey.com/ap/Content_Natural_Selection.pdf

And here is a reply that says, if the naturalist adopts "teleosemantics" they have a way to deflect the EAAN:

http://ling.uni-konstanz.de/pages/home/leahy/documents/TeleosemanticsEAAN.pdf

David B Marshall said...

0091: Thanks, but I've avoided academic rebuttals of Plantinga so far, and would like to see how my intuitive criticisms fair, first. No reason has been given, here or in the other thread, to reject those arguments, yet. And I'm a little leery of jargon like "teleosemantics," which would seem to refer to something like "meaningful language," which is a clear enough concept.

James M. Jensen II said...

Crude: The EAAN seems entirely compatible with people's thoughts being confused nonsense rather than their making coherent statements which are, in fact, wrong.

But it's not confused nonsense. In the case I used, I'm confused about what I'm referring to: what I call "hairdressers" is what other people call Freemasons.

The point is you can only be wrong about what you're mostly right about. Otherwise you're not even wrong.

The EAAN argues that evolution could result in us having generally-unreliable beliefs. Meaning-is-use implies that can't happen, because on any topic you like, at least as many beliefs about it will be true as are needed to determine what I'm referring to. And as my example shows, even a variance of a small number of beliefs can be enough to demonstrate that, rather than being wrong, I'm mostly-right about something else.

ozero91 said...

That's fine, the papers are kind of lengthy anyway. I have been looking into teleosemantics, and it seems to be a fairly under-represented idea. But from what I have read, it seems broadly Aristotelian, which flies in the face of typical naturalist "mind as a computer program" theories.

Can you direct me to the "other thread?" Is it referring to the comment you posted on Reppert's blog?

Crude said...

The point is you can only be wrong about what you're mostly right about. Otherwise you're not even wrong.

Sure, but being 'not even wrong' seems entirely compatible with the EAAN. Plantinga uses examples of actions being correlated to false or erroneous beliefs, sure, but they could also be correlated to what is actually gibberish.

Now, you can always assign meaning to gibberish to make it meaningful - but so what?

I'm mostly-right about something else.

Even if that's granted, it doesn't seem to help much. Though that's a funny reply to imagine. 'You were right all along, X! It's just that you gave answers to questions no one, including yourself, ever asked you.'

Mike D said...

Can this be? A post where David Marshall and I... agree? Why yes!

I think this is a solid argument. I also think that one of the failings of Plantinga's argument is that he overestimates the reliability of our cognitive faculties. While our sensory information is generally reliable in a crude sort of way, we're also subject to a litany of errors. We see or hear things that aren't there, construct false memories, experience sensations simply because we expect to do so, etc. etc.

That's why we need science. Science essentially starts with an assumption, then tests that assumption to see if it is reliable and valid. Does it allow us to make predictions about what we will observe? Can the results be independently reproduced using the same parameters? That methodology provides us with a way to gradually cut through our many cognitive errors and biases.

You may also be interested in Stephen Law's take on it here:

http://analysis.oxfordjournals.org/content/early/2011/12/08/analys.anr130.full

Crude said...

I also think that one of the failings of Plantinga's argument is that he overestimates the reliability of our cognitive faculties.

Plantinga's argument at no place relies on our cognitive faculties being reliable.

That's why we need science.

If Plantinga's argument is successful, then for the E+N believer, trust in science is undermined.

Again, Plantinga's argument is a priori, not a posteriori.

Crude said...

David,

I think we can put 3 and 4 aside as objections to Plantinga's EAAN. Let's move on to 2.

The simplest way to construct an eye that helps us survive, is to make the eye see what really is. The simplest way to construct a brain that helps us survive, should by analogy also be to make it perceive what really is.

But this is not obvious. If anything, the opposite seems to be more defensible.

I brought up the example of creatures that thrive despite having no beliefs whatsoever (I think it's a helpful example, because we tend to think about natural selection in terms of humans and larger animals, where questions of belief may come up.)

For your brain example, I'll lodge a counter-claim: the simplest way to construct a brain that helps us survive is to construct a brain that reacts appropriately in situations of danger.

If you think I just repeated what you said, I didn't. Here's the key difference: your example relies on perception. Mine makes no reference to it, and it doesn't obviously factor into things whatsoever. A brain that is wired to avoid dangerous places, *even if there is no subjective perception going on*, benefits survival. Likewise, a brain that is wired to avoid dangerous places, *regardless of what subjective perception is going on*, benefits survival.

You can leave beliefs entirely out of the loop.

You also give this example in 2:

But the ability to see objects like tigers, snakes, mosquitos, and clouds, could also be applied to stars or, with mechanical aids, distant galaxies or tiny bacteria.

But the problem is that these abilities actually don't directly aid survival. What's going unsaid is the following: 'perceiving these phenomena and reacting appropriately aids survival'.

What's the survival advantage of seeing tigers, snakes and mosquitoes, detached from any action? The answer, apparently, is nothing. But if it's the action that's valuable rather than the belief, then the usefulness of the belief is obscured.

You can even see this in part by your camera example. What's simpler: a computer that can spit out '2 + 2 = 4' for you mechanically? Or a computer that knows 2 + 2 = 4?

David B Marshall said...

Crude: I've revised the OP to explain the functions (3) and (4) play in my argument more precisely.

I don't think your critique of (2) works, though.

"I brought up the example of creatures that thrive despite having no beliefs whatsoever (I think it's a helpful example, because we tend to think about natural selection in terms of humans and larger animals, where questions of belief may come up.)"

I don't. By hypothesis, the brain is just one of many biological structures that has been formed by naturalistic evolution. It happens to be the most complex, also the most useful, if you ignore Loftus' last thread.

"If you think I just repeated what you said, I didn't. Here's the key difference: your example relies on perception. Mine makes no reference to it, and it doesn't obviously factor into things whatsoever. A brain that is wired to avoid dangerous places, *even if there is no subjective perception going on*, benefits survival. Likewise, a brain that is wired to avoid dangerous places, *regardless of what subjective perception is going on*, benefits survival."

This doesn't help. The obvious fact is, true beliefs about dangers and opportunities do help us maximize our survival and reproductive opportunities, if we act on them. And your counter is too vague to rebut the specific examples I gave earlier.

We can be wired to fly at "scary sounds." But now Aragorn and his crew are contemplating a Balrog. Is their best strategy to fight it at the bridge? They feel an impulse to try to save Gandalf, and feel another to hide under a rock at the approach of scary danger. But Gandalf rationally assesses the situation -- the Balrog outmatches any of them except him, the orcs and trolls vastly outnumber them, he has bought them time, the gates out of Moria are close by. "Fly, you fools!" They recognize that he has encapsulated the best survival strategy given the objective realities of their situation -- all the variables flash through their minds, along with the reflection that it would do no honor to Gandalf to waste his sacrifice -- and they take off.

This is way too complicated for mere instinct. And if it doesn't involve a true belief about the outer gate, the company is doomed.

"You can leave beliefs entirely out of the loop."

Obviously not.

"What's the survival advantage of seeing tigers, snakes and mosquitoes, detached from any action? The answer, apparently, is nothing. But if it's the action that's valuable rather than the belief, then the usefulness of the belief is obscured."

You're missing the point. Seeing is NOT detached from action, it is a highly useful, if not absolutely necessary, prerequisite too it. The action will not occur, or not occur as usefully, if it is not preceeded by true sight, or insight.

Crude said...

I don't. By hypothesis, the brain is just one of many biological structures that has been formed by naturalistic evolution. It happens to be the most complex, also the most useful, if you ignore Loftus' last thread.

Sure. But what is it about complexity that requires beliefs to enter the chain? Your computer is quite complex. So is the software. Does your computer have beliefs, much less true ones?

I know you're aware that the brain does many, many things with no beliefs apparently in play.

The obvious fact is, true beliefs about dangers and opportunities do help us maximize our survival and reproductive opportunities, if we act on them.

Look what you just said. 'True beliefs help us, if we act.' But if we perform those acts with zero beliefs or false beliefs... it still helps our survival and reproductive opportunities. And if we have true beliefs and perform the wrong acts, we're harmed anyway.

As I've said: it's the actions that matter.

You're missing the point. Seeing is NOT detached from action, it is a highly useful, if not absolutely necessary, prerequisite too it. The action will not occur, or not occur as usefully, if it is not preceeded by true sight, or insight.

This statement and your Gandalf example seems to me to make a mistake about how evolution proceeds.

First, even if Gandalf and company's actions are the best ones they can reasonably take, they may well be doomed. In fact, they may for all purposes be certainly doomed. What matters in evolution isn't just the particular acts of one individual, but the outcomes of population events over time. A good way to think about it would be, what if Gandalf and company had the opportunity to blindly try a wide variety of strategies again and again in evolutionary history. Many, even most, may fail. But some may well work - even if they were blindly tried. And if they work, well, there's the more likely source of your next generation.

Seeing is NOT detached from action, it is a highly useful, if not absolutely necessary, prerequisite too it. The action will not occur, or not occur as usefully, if it is not preceeded by true sight, or insight.

And again, what matters is the action triggered by the sight. If I repeatedly avoid likely predators, it doesn't matter to naturalistic evolution if I'm doing so because I'm afraid they'll eat me, or I think they're ghosts, or I find them annoying, or any number of wrong/superfluous beliefs. So long as I get out of there.

Crude said...

Just to try and make things clearer on this point.

Pointing out situations where an organism needs to act in a very specific way or else they'll be doomed aren't, in my view, going to help out your argument very much here, precisely because deaths of organisms are no problem for evolutionary theory - in fact, it thrives on that. That's how 'problems' get solved: random mutations and selection operating over time. 'The Blind Gandalf' instead of 'The Blind Watchmaker', in this situation.

Regarding the sight example, you also have this problem: you mention an eye then compare it to a mind. But if your eye involves a mind anyway, what's the point of the comparison to begin with? It's redundant. But an eye's operation detached from a mind just IS stimulus and response. The eye doesn't need to, and in fact can't (without a mind) 'see what really is'. All it needs to do is trigger a good response to the right stimulus. In fact, that's all it really can do in that context.

David B Marshall said...

Mike: I invited Dr. Law by to offer his opinion. He said he might drop by, but sounded pretty busy. Maybe its the end of the term.

Dr H said...

Hi David, I'm not sure how much I can contribute to this discussion; you seem to have some intelligent commentators who have already raised the salient points. But what the hey... here's my $0.02.

Not to oversimplify an argument that has kept a particular school of philosophers busy for nearly 20 years, but I've always seen the EAAN argument as something of a strawman. It rests, as many arguments do, on certain axiomatic assumptions, but it has far from clear that some of these assumptions are supported.

First off, I don't see that naturalism is synonymous with atheism, but for me that is a minor point.

The larger problem is his proposition that "our faculties are "'reliable'". It is empirically demonstrable that our faculties are not reliable. Furthermore, his proposition rests on an assumption that evolution would necessarily produce 'reliable faculties'; evolution theory makes no such claim.

I think your"argument from simplicity" points in sort of the right direction, but ultimately misses the mark. Plantinga's "tiger" scenario is highly contrived, but to me the reason it's a bad argument has to do with a more general circumstance.

Plantinga appears to be making a couple of unstated assumptions, similar to those made by a lot of lay persons who look at evolution: that evolution has a perfered direction; and that human beings represent the pinnacle (or at least the current pinnacle) of this prefered evolutionary direction. The first of these is, I think, contraindicated by the evidence, and the second is not supported by sufficient evidence.

Human beings, with what we think we recognize as something similar to the modern human mind, have been around for maybe 50,000-100,000 years. Life has existed on Earth for around 4 billion years. All of humanity occupies less than 1/4000th of that time span; less than an eyeblink -- the eye hasn't even yet realized that it needs to blink.

In evolutionary terms, we don't really know yet that human beings are a viable species. There are other species far older than ours, and many whose members are far more numerous. Because we seem to have mastery over other species we tend to assume that we are the way we are because it confers upon us a significant survival advantage. In some ways our intellectual evolution has enabled us to enhance the chances for continued survival of our species -- we've developed agriculture; sanitation; medicine; etc. But in other ways it has enabled us to work towards our own extinction: war; environmental damage; development of nuclear weapons; deliberate development of pathogens; etc.

Whether our "faculties" actually confer an evolutionary advantage will only be seen from a vantage of tens of millions of years in the future; right now, we just can't tell.

EAAN ignores all of this, and focuses Plantinga's version of "evolution theory" on naturalism. The problem is, his version of evolution theory is not that held by most evolutionary biologists.
Hence my view of this as a strawman argument.

David B Marshall said...

Dr. Richard Field commented by e-mail:

I recall reading a similar argument some years ago. I hazard a guess that it was offered by Richard Taylor. Interestingly enough his reasons for posing the argument had nothing to do with defeating naturalism, but simply to suggest that evolutionary theory is consistent with skepticism, It is clear that Plantinga doesn’t want us to follow that path.

The gist of my comments are touched on by some earlier comments. Suffice to say that I too think Plantinga’s argument fails. First, it is based on a rather outmoded dichotomy between belief and behavior. Here my view runs along the Wittgensteinian lines already offered, although I tend to think of such issues more in terms of pragmatism. A belief is intimately tied to behavior, and by and large successful behavior. In evolutionary terms, we would say adaptive behavior. A child learns to identify colors in conjunction with behaviors that are responsive to a child’s interests and needs. If a child’s color identifications are wrong, those beliefs will quickly be discarded because they do not serve. What Plantinga would ask us to imagine, using this example, is a child whose identifications are all off the mark but whose behaviors based on those beliefs are successful, which makes little sense at all. That doesn’t mean that a true belief might fail on occasion, or that a false belief might serve, but on the whole true beliefs and successful behaviors are indissolubly linked. Second, Plantinga works on the basis of a rather atomistic conception of belief. Beliefs form a structure of semantic entanglements and implicatory relationships. It is helpful to think of Quine’s metaphor of a “web of belief.” Plantiinga can easily raise a case where a belief in isolation is false by adaptive, but the very meaning of the false belief depends on its connection with other beliefs, many of which we must presume are true. I scarcely know what it would mean to speak of a totally errant web of beliefs.

Jason Pratt said...

Crude: {{Plantinga's argument at no place relies on our cognitive faculties being reliable.}}

True, unless he has radically changed his procedure since the late 90s. At that time (I haven't kept up with his polishing since) Plantinga built his argument from an objective probability that our cognitive faculties are low given N&E, following Patricia Churchland's infamous dictum about the Four F's: the principle chores of the nervous system from the perspective of evolutionary development are Feeding, Fleeing, Fighting and Reproducing. {g} "Truth, whatever that is, definitely takes the hindmost."

This is, or at least was, the primary cognitive doubt Plantinga built his argument upon. It is not, however, the kind of cognitive doubt Lewis arrives at. For Lewis explicitly agrees that such a process of conditioning "might serve us as well as reason or in some circumstances better"! (The fact that Plantinga builds his argument from a cognitive doubt while Lewis arrives at a cognitive doubt, also illustrates large differences in their various approaches.)

The sort of conditioning Lewis is discussing may not be as sophisticated in its description as the sort being considered by Churchland, but the principle similarity should be obvious. True, Lewis is very careful to insist that this process would not result in actually rational behavior--a failure he deems important, both before and after this point--but insofar as the reliability of our cognitive processes are concerned, he must be on the same page as Churchland: as well as reason or perhaps even better in some circumstances. Plantinga, however, clearly disagreed about the likelihood of (R)eliable cognitive faculties developing in such a situation. (I expect Patricia Churchland would have agreed with Lewis that reliable functionality doesn't require rational action per se, though doubtless she would disagree on where he takes his argument in many other regards.)

Btw, Victor may recall that I regard Lewis' argument (and even moreso in MaPS 2nd edition) as a Skeptical Threat rather than Best Explanation attempt. On that I do agree Plantinga is similar to Lewis: he's also mounting a ST argument. It's just almost completely different from Lewis in most other regards. {g}

JRP

Jason Pratt said...

David,

I'm going to slightly update and send along a paper I wrote back before 2005 comparing P's argument (circa 1994) with Lewis' argument (from 2nd edition MaPS). Depending on how much Plantinga has updated his argument, you may not find it very useful, but for whatever it's worth I didn't agree with the EAAN back then either, and still don't now (at least in the form being proposed back then. As I said, I haven't kept up with developments, and have read this thread with some interest.)

JRP

Unknown said...

Jason: Good, as always, to hear from you. Everyone seems to be more than an expert than I am, which can be a good thing. Any thoughts on my reasons to question EAAN?

Jason Pratt said...

Not yet. You're probably more expert than I am on that at the moment (since I haven't really been catching up on the EAAN in several years). {g}

JRP

Crude said...

A belief is intimately tied to behavior, and by and large successful behavior. In evolutionary terms, we would say adaptive behavior. A child learns to identify colors in conjunction with behaviors that are responsive to a child’s interests and needs. If a child’s color identifications are wrong, those beliefs will quickly be discarded because they do not serve. What Plantinga would ask us to imagine, using this example, is a child whose identifications are all off the mark but whose behaviors based on those beliefs are successful, which makes little sense at all.

Plantinga doesn't ask this.

First, you don't need "all" of the identifications off the mark - the accuracy has to be low or inscrutable. Nor does Plantinga's argument require the beliefs to even really be connected with observation - seeing blue doesn't require you to belief that you see 'some other color'. You could believe you see an elephant, or complete gibberish, or... etc.

Further, I don't think it 'makes little sense at all'. It's a pretty straightforward general scenario to make sense of: it's easy to imagine a scenario where a survival-enhancing action is tied to no belief, or a false belief, or gibberish. Once you imagine one, 'low or inscrutable' is pretty simple to picture next. I think even 'all' is easy to picture, even if it's not required.

but the very meaning of the false belief depends on its connection with other beliefs, many of which we must presume are true.

I see no reason to require us to presume the beliefs true - this may be making the move of 'they have to be at least in some sense true, or else the result is gibberish'. But I don't think the gibberish claim does much to Plantinga's argument, and again, Plantinga's standard here isn't one of 'our beliefs are never true' but 'low or inscrutable'.

Crude said...

The larger problem is his proposition that "our faculties are "'reliable'". It is empirically demonstrable that our faculties are not reliable. Furthermore, his proposition rests on an assumption that evolution would necessarily produce 'reliable faculties'; evolution theory makes no such claim.

Plantinga's argument does not require that our faculties are 'reliable'. Nor is his proposition that 'evolution would necessarily produce 'reliable faculties''. It can't be, because if he thought that, there would be no point to the EAAN.

Plantinga appears to be making a couple of unstated assumptions, similar to those made by a lot of lay persons who look at evolution: that evolution has a perfered direction; and that human beings represent the pinnacle (or at least the current pinnacle) of this prefered evolutionary direction.

Plantinga makes neither of these assumptions in his argument, and neither plays any role in the EAAN. The entire question of theism or 'pinnacles' are walled off from consideration by his argument, nor does he suggest that evolution has a preferred direction.

In evolutionary terms, we don't really know yet that human beings are a viable species.

Not relevant to the argument.

I'd try to reconstruct the argument I think you're perceiving with Plantinga's EAAN, but honest to God, I can't really piece it together. Whatever it is, it's not the EAAN.

Cornell Anthony said...

"I don't know what SE is -- I don't see any "S"s in your glossary -- but I see no reason at all to think "thoughts are like smoke to fire."

SE = Semantic Epiphenomenalism


"Aside from which, smoke DOES have an effect (ever smoke a salmon?), and dreams do too (read Brother Yun's Heavenly Man, or the Book of Genesis, or maybe Freud)."

You are equivocating the word 'smoke'. Anyways I'm using Smoke as the 'cause'

I didn't understand the rest of your response.

It just seems that an evolutionist + naturalist needs to take the reliability of their cognitive faculties on FAITH in which they can only posit an ad-hoc response in the process.

This argument from Plantinga is similar to the Cartesian Evil Demon hypothesis, and I'm afraid empirical justification becomes just as circular here.

David B Marshall said...

Cornell: Everyone takes the reliability of their minds "on faith" -- rational faith, because that does make the best sense of the situation. I agree that belief in God strengthens warrant for that faith, but see no reason why warrant should be absent, even on naturalism.

The point of the tiger is that the most economical way for evolution to create cognition, given that it can do such a thing to begin with, is to connect "insight" or "perception" to the state that things take in the real world, to ontology. It seems to me Plantinga is just wrong in denying this. But of course I could be wrong, and will be happy to change my mind if anyone can just show why I'm wrong.

John Fraser said...

Hi David,

I'm going to join the fray here. I haven't read every comment in detail, but I do see you complaining that people aren't addressing your points. I'm just going to address your first objection for starters.

You say, “The simplest solution would be for evolution to produce a brain that perceived reality fairly accurately, as the simplest solution would be to produce eyes that see reality somewhat accurately.”

Plantinga’s argument is about beliefs, not perception. One can perceive things accurately but still have false beliefs. And the point is that evolution selects for survival, not simplicity. Otherwise we would all be amoebas.

You say, “EVEN IF one had some ridiculous notion, like the way to pet a tiger is to run away from it, among the whole set of possible beliefs, how likely is it your deluded mind would light upon that particular delusion, so as to take all the specific actions that would be required to save your bacon?”

Plantinga is simply illustrating the point that false beliefs can nevertheless produce adaptively advantageous behavior. His argument is not, “evolution would necessarily produce beings who think the best way to befriend a tiger would be to run away from it.” The main point is that false beliefs can nevertheless be adaptively advantageous. I think he’s clearly right about that. It matters not a bit how probable the occurrence of any particular one would be.

You say, “In each case you are just so lucky as to hold a belief that is totally wrong, but that inspires you to take just the right actions to rescue yourself? All your life long?”

Two problems: first, Plantinga’s argument does not require that every single belief that is produced by evolution would be false (“in each case”). He suggests we assume that evolution would produce faculties which deliver about the same number of false beliefs as true beliefs. Second, there would be no “luck” involved because the beliefs would be the result of faculties which have developed through natural selection. Creatures whose faculties produced beliefs which led to non-adaptive behavior would be removed from the gene pool. At best your objection only shows that creatures with non-adaptive false beliefs would not survive. But of course they wouldn’t! If you want to show that all creatures with 50/50 true/false beliefs wouldn’t survive, you need an argument for that.

You say, “A brain that tells us things that really are, is much simpler than the endless chain of fortuitious chances that would be required for false beliefs to lead to adaptive behavior.”

Again, there is no chance involved here. But I think you are again confusing perception with belief. You also have not shown that simplicity rather than adaptability is going to be preferred by natural selection. In fact I don’t think it is (or at least I don’t see why). Natural selection should be indifferent to how simple or complex something is, as long as it produces more surviving offspring than the competition.

You say, “A brain that created beliefs that were utterly tangential to reality, all things equal would get us killed quickly.”

Not if those beliefs led to adaptive behavior, and that’s the point. You are assuming without argument that beliefs “tangential to reality” are of necessity not adaptive. I see no basis for the claim and I think Plantinga shows that this is not the case.

You say, “Plantinga's proffered example of the tiger is ludicrous, I think, because the brain would have to come up with false reasons for doing the right things each and every time one's surival was at stake.”

Again, no. Plantinga does not assume that evolution would produce ONLY false beliefs, and it’s not necessary for his argument. I’m afraid you’re attacking a straw man here.

Incidentally, I was a bit surprised just yesterday to see that Thomas Nagel actually says Plantinga’s argument is a good one in his book, Mind and Cosmos (p. 27). Nagel is the first atheist philosopher I've seen to admit that it's a good argument.

David B Marshall said...

John: As pointed out, beliefs are analogous to perception: such a close analogy, that we use the one for the other, and not just in modern English. It is reasonable a priori to expect that IF evolution can produce sight, it can also produce insight. (I am not talking about physical mechanisms, nor claiming that it can, in fact, do either.)

By simplicity, I mean finding a simple solution to an engineering problem, not that no new mechanisms are required that makes the organism absolutely more complex. Obviously, if evolution can produce machinery, it would be easier to produce working machines with few working parts than machines with many working parts working together. And one mechanism that works for a variety of situations, is simpler and therefore better than a variety of mechanisms with ad hoc functions that prove fortuitous for a variety of reasons. That is the point: Occam's Razor for engineers.

From an engineering standpoint, an eye that sees truly, or a brain that perceives truly, are far simpler and more elegant solutions to the evolutionary challenge, than creating thousands of false beliefs that somehow result in lots of correct actions. This is obvious, and Plantinga's own ludicrous example of the tiger demonstrates that fact.

Of course I don't have to prove that. Plantinga is making an argument against naturalism. It's up to him to show why evolution wouldn't do such an obvious thing.

And I think Plantinga is clearly wrong, not to say false beliefs CAN produce useful action, but to imply that they are AS LIKELY to do so, or anything within many orders of magnitude as likely. Probability is what counts, since evolution is a probabilistic process.

But this is an old argument, now, and I may need to read it again.

John Fraser said...

David,

You say, “It is reasonable a priori to expect that IF evolution can produce sight, it can also produce insight.”

Even if one acknowledges that evolution COULD produce “insight,” that doesn’t defeat Plantinga’s argument. As he points out, it’s also reasonable to think that it might not. In that case one should adopt a position of agnosticism towards R. But this would also require agnosticism towards N&E. Plantinga has more bases covered that I think you realize.

You said, “From an engineering standpoint, an eye that sees truly, or a brain that perceives truly, are far simpler and more elegant solutions to the evolutionary challenge, than creating thousands of false beliefs that somehow result in lots of correct actions. This is obvious, and Plantinga's own ludicrous example of the tiger demonstrates that fact.”

Evolution doesn’t work like an engineer solving a problem. It is a “blind” watchmaker. Evolution (supposedly) works by false starts and dead ends, and the accumulation of “junk DNA,” that sort of thing. So I think this analogy is not that helpful. And again, natural selection doesn’t select for “simplicity” or “elegance,” it only selects for adaptability. If a mechanism that produces a false belief leads to adaptive behavior, natural selection does not have the ability to say, “yes, but I’m going to scrap that one in favor of this other one over here which is simpler and more elegant.” You are attributing powers to natural selection which it just doesn’t have and can’t have (and no, I don’t think your point #3 actually answers that objection).

You say, “Of course I don't have to prove that. Plantinga is making an argument against naturalism. It's up to him to show why evolution wouldn't do such an obvious thing.”

He isn’t arguing against naturalism, he’s arguing against BELIEF in naturalism, which is different. There are two possible outcomes from Plantinga’s argument. One could conclude that the probability of our faculties being reliable is low given N&E. But one could also conclude that we can’t really say if it’s low or high. It MIGHT be high, but it also is reasonable to say it might be low, in which case agnosticism is the best choice. But in that case, one must also be agnostic about whether naturalism is true. To defeat the argument, it is actually incumbent on the naturalist to show that evolution would necessarily produce reliable faculties. But you also have to keep in mind that he gives 5 different possibilities for the relationship between belief and behavior (you missed one in your o.p.), and the probability of each of these also makes up part of the probability space one has to assign to R. It’s only under the fifth one that there even MIGHT be a possibility of R being high given N&E, but even there it’s only a possibility.

You said, “And I think Plantinga is clearly wrong, not to say false beliefs CAN produce useful action, but to imply that they are AS LIKELY to do so, or anything within many orders of magnitude as likely.”

Well, I sure don’t see any basis for “many orders of magnitude,” or even bare probability. The thing is, even if you think it MIGHT be the case the R would be high under N&E, Plantinga’s argument still succeeds. But to show that the probability of R is “many orders of magnitude” higher given N&E requires a lot more than what you’ve given here. Although I will say that I looked back over the last chapter of Warrant and Proper Function, and Plantinga actually says that under the scenario where beliefs have a causal effect on behavior, it would be reasonable to assume the probability of producing mainly true beliefs is “somewhat more than ½”. I was mistaken when I said 50/50, that was in reference to a different part of the argument as I went from memory (guess my faculties aren’t as reliable as I like to think!). But even given this assumption, his argument still holds.

John Fraser said...

Oh, now I see why you said it was an old argument. I didn't realize how old this thread was! Sorry!

David B Marshall said...

Don't be sorry! Thoughtful criticism is always welcome. Many of my posts are more perennial than timely in nature; this should continue to provoke debate.

You may also be interested in my last two posts, on "Fulfillment Theology" vs. its rivals (especially pluralism and inclusivism). I don't usually spend quite that much time on blog posts; there's lots of meat to them. I'm going back to China tomorrow, though, so I may not always be able to response as quickly or as thoroughly as I'd like.

Unknown said...

Interesting article and interesting discussion.

While I find EAAN compelling, I too have barriers in accepting it.

Actually, just one barrier: isn't it possible that the capacity to know truth was a random mutation which was passed on genetically?

Perhaps this mutation to know truth coincided with another trait beneficial to survival, thus it was passed on simultaneously, though not on its own merits. It was just "attached" to the other survival increasing gene(s).

Maybe I'm missing something in either evolution or the EAAN which would make this position untenable. I would love it if EAAN were valid, but as it stands I can't quite accept it.

Dave said...

I think that there are certain philosophical nuances to this argument that you may not fully be attending to.

Throughout, we assume (for reductio) that thoughts are either identical with or caused by certain neuronal firing patterns. So when you say something like "I see no reason whatsoever to grant ... that human beings are totally unlike what we know from experience every single waking day of our lives they really are -- homo sapiens, thinking man, whose thoughts (for better or for worse) certainly does influence our actions. If we got that wrong, we all live in the Matrix," I become concerned that you're missing the point. The goal of "The Argument from Reason" in all of its forms is simply to show that "being the thinking man/having efficacious thoughts/not living in the matrix" is improbable given that our thoughts are either identical to or caused by respective neuronal firing patterns. The basic instinct behind the EAAN as I understand it is that the following propositions:

1) Beliefs are wholly determined by neuronal firing patterns.

2) Natural selection only cares about inputs matching up to the right outputs - firing patterns in between don't matter at all.

...entail that evolution is entirely incapable of selecting for true beliefs. Now, it may turn out that "efficacious beliefs/not living in the matrix" considerations entail that our thoughts, beliefs, desires, and choices are not solely a product of neuronal firing patterns, and can thus have some efficacy of their own over and above the firing neurons. But that would be an even more direct argument against naturalism.

With regards to simplicity, why should it be simpler for my neuronal firing patterns to produce a conscious "mental image" (and our beliefs are ultimately founded on our mental images) of the proximate cause of the neuronal firing pattern, as opposed to a mental image of something further down the chain of causes, or even a mental image of the firing neurons themselves?

Say a tiger is coming at me from the side. This produces a projected image of of a moving orange striped thing on my retina. The projected image causes a series of firing neurons, starting with those in the retina and ending in the motor neurons in my legs. The neurons in between can have any configuration, and thus any mental image can be produced.

Why should the configuration that produces a conscious experience of a moving orange striped thing be any simpler than the configuration that produces a conscious experience of a pattern of tingling sensations with the same "shape" as the relevant neuronal firing pattern?

I think we can both agree that a person whose "world" looks like the world around them will come up with very different beliefs than the person whose "world" feels like his own brain's activity. I think we can also agree that there's no obvious reason why wiring a brain that produces the former kind of mental images (broadly construed to include odors, sounds, smells, heat, textures, pressures, etc.) would be any easier than wiring a brain that produces the latter sort. And that's just one way we can imagine conscious "sense images" being utterly divorced from their causes. We can be confident that there are many more that we can't even imagine.

Dave said...

(cont.)

We have five senses. Can we imagine having, say, eighteen? Surely it's possible, but by the very nature of the case, we cannot imagine what it would be like to have such senses, to experience aspects of reality we currently cannot. Thus, there are countless "unimaginable images" that could be produced by neuronal firing patterns, and since the patterns themselves are irrelevant, it follows that natural selection couldn't distinguish between mental images that resemble reality and mental images that don't. This is a huge problem, since our ability to form accurate beliefs is entirely dependent on the presumption that the "sense data" that comes to us resembles external reality.

In response to the example of the eye, to me, the reason why the objection misses the point seems clear: in the eye, the effect (an image projected onto the retina) is in the same order as the cause (light reflected off of objects in the environment) - both are supposed to be of the physico-chemical order. In the brain, the effect (a subjectively experienced sense image) is of the psychological order, whereas the cause (a neuronal firing pattern) is of the physico-chemical order. Why should the "laws" linking the two orders be friendly to the handful of ways for sense images to resemble the external objects that cause them, but unfriendly to the countless hordes of ways for sense images to be totally dissimilar to their causes?

Unless we want to move into a Scholastic direction - ie, allowing formal and final causes to re-enter the material world, thereby reducing distinctions between the physico-chemical aspects of things and their other aspects to mere heuristic conveniences - this is going to be an extremely problematic and important "anthropic coincidence." Naturalists ought to be wary of compromising with Scholasticism, as formal and final causes are one of the easiest ways to "let a divine foot in the door." On the other hand, having mental images that resemble their proximate causes in the external world doesn't sound like the kind of thing that could plausibly contribute to an "observer selection effect." We can be here with radically misleading ideas just fine. The multiverse only gives us more opportunities to go wrong.

Dave said...

(cont.)

So I think that the main reasons you give for rejecting the EAAN can be answered. That being said, I think that there are other versions of the Argument from Reason that are somewhat stronger.

Richard Dawkins once said, "The universe we observe has precisely the properties we should expect if there is, at bottom, no design, no purpose, no evil, no good, nothing but blind, pitiless indifference." Let us, for the moment, accept his claim. How, then, is it to be supposed that three pounds of soap bubbles and saline are capable of achieving design, purpose, good and evil? How is it that my brain is any more capable of rising above "blind, pitiless indifference" than my bowels? Why should a thought be any more "meaningful," "justified," "true," or "virtuous" than a fart or a sneeze? Posed this way, the question isn't so much how our mental states can be true as it is how they can even rise to the level of being wrong.

Imagine flatlanders with the best cranes money can buy trying their hardest to build a three dimensional structure. That's what a "naturalistic" explanation of human nature must be, by the very nature of the case. Why should we be expected to believe such nonsense?

Start with the resources of Hilbert Space, and Lineland, Flatland, Spaceland, and all the rest will be thrown in as a bonus. Start with the resources of Lineland, and you're stuck there forever. Perhaps you can encode a rather interesting message in Morse code - or even the next Great American Novel! - but you'll never make yourself a tesseract.

So it is in reality. Begin with God, and you can get a universe with as many different kinds of depth as you please. Begin with "atoms and the void" (or the modern equivalent, "quantized fields and their excitations"), and you can only ever have a flat and boring world. You might expect to find some very complex and interesting configurations of mass-energy, but you'd never in your wildest dreams expect to find anything rising above the status quo of blindness, meaninglessness, amorality, and indifference.