I may have been too hasty in my acceptance that Plantinga’s Evolutionary Argument Against Naturalism was not what I had originally given. As commenter Nick Matzke pointed out, Plantinga seems to be doing a bit of historical revision here. In the version he gives now, this is an argument against a metaphysical claim, but the original (or at least the 1993 version in Warrant and Proper Function ) appeals to empirical claims like “there is a tiger here about to attack”. Moreover, Darwin’s doubt was never about the intentional content of such empirical claims – like Quine and Popper, who Plantinga also mentions, Darwin’s evolutionary epistemology is fine for giving reliable empirical beliefs. As Paul Griffiths and I argue in our chapters on evolutionary debunking arguments [2, 3], the problem arises when trying to support or arrive at non-empirical claims, like moral realism or the existence of God and purpose.
I concede that I misread the argument as rejecting evolutionary theory (although given some of Plantinga’s comments about God’s ability to intervene in the course of evolution I feel that he’s not entirely happy with it, on which see my paper on how theists can be good Darwinians ), when he is arguing against his definition of “Naturalism”. But I think that this is understandable given how convoluted the presentation of the argument is in the 1993 chapter. So here are some claims made in that chapter (I thank Helen De Cruz for sending me the text; all quotations lack page numbers but it’s a short enough chapter). Here is Plantinga’s own summary of the argument (or two arguments, one of which extends the first):
… the first argument is for the falsehood of naturalism, the second, and more developed, is for the conclusion that it is irrational to accept naturalism. Crucial to both arguments is the estimation of the value of a certain conditional probability, P(R/(N&E&C)), where (roughly) R is the proposition that our cognitive faculties are reliable, N is metaphysical naturalism, E is the proposition that our cognitive faculties arose by way of the mechanisms of evolution (i.e., the mechanisms to which contemporary evolutionary thought directs our attention),and C is a complex proposition stating what cognitive faculties we have and what sorts of beliefs they produce. In the second argument I contend that (1) it is quite plausible to think either that the rational attitude to take towards the conditional probability mentioned above is the judgment that it is low or that the rational attitude is agnosticism with respect to it, and, (2) in either case, the devotee of N & E has a defeater for any belief he holds, including N. Further, since this defeater is an ultimately undefeated defeater (as I argue), it is irrational to accept N, since it is irrational to accept any proposition such that one knows one has an ultimately undefeated defeater for it.
He there spends time criticising the view that our cognitive faculties are reliable from evolution, since so long as they deliver the fitness enhancing outcomes, the content doesn’t matter:
But isn’t there a problem, here, for the naturalist? At any rate for the naturalist who thinks that we and our cognitive capacities arrived upon the scene after some billions of years of evolution (by way of natural selection, genetic drift, and other blind processes working on such sources of genetic variation as random genetic mutation)? … If our cognitive faculties have originated as Dawkins thinks, then their ultimate purpose or function (if they have a purpose or function) will be something like survival (of individual, species, gene, or genotype); but then it seems initially doubtful that among their functions—ultimate, proximate, or otherwise—would be the production of true beliefs. … The principal function or purpose, then, (the ‘chore’ says Churchland) of our cognitive faculties is not that of producing true or verisimilitudinous beliefs, but instead that of contributing to survival by getting the body parts in the right place. What evolution underwrites is only (at most) that our behavior be reasonably adaptive to the circumstances in which our ancestors found themselves; hence (so far forth) it does not guarantee mostly true or verisimilitudinous beliefs. Of course our beliefs might be mostly true or verisimilitudinous (hereafter I’ll omit the ‘verisimilitudinous’); but there is no particular reason to think they would be: natural selection is interested not in truth, but in appropriate behavior.
So, then he sets up the “Darwin’s Doubt” frame:
… perhaps Darwin and Churchland mean to propose that a certain objective conditional probability is relatively low: the probability of human cognitive faculties’ being reliable (producing mostly true beliefs), given that human beings have cognitive faculties (of the sort we have) and given that these faculties have been produced by evolution (Dawkin’s blind evolution, unguided by the hand of God orany other person). If metaphysical naturalism and this evolutionary account are both true, then our cognitive faculties will have resulted from blind mechanisms like natural selection, working on such sources of genetic variation as random genetic mutation. Evolution is interested, not in true belief, but in survival or fitness. It is therefore unlikely that our cognitive faculties have the production of true belief as a proximate or any other function, and the probability of our faculties’ being reliable (given naturalistic evolution) would be fairly low.
Griffiths and I argue against this view. We can distinguish between two (or three) kinds of beliefs: environmental beliefs, social beliefs and possibly, if they are different in kind from social beliefs, metaphysical beliefs. There is every reason to think that beliefs about the environment are reliable and true. While evolution also provides behaviours not underwritten by beliefs (flatworms and bacteria probably have no beliefs of any kind and yet they evolve by behavioural fitness), if you do have cognitive faculties, and thus can reason from them, the prior probability is high that those cognitive faculties that deliver beliefs that are false will end up causing unfit behaviours (if you can’t tell that it’s a lion and that it wants to eat you, you will, as Quine put it, demonstrate a pathetic tendency not to reproduce your epistemic preferences).
As to social beliefs, here the content of the belief is largely decoupled from the fitness effects of the belief. If you fail to believe in the God of the surrounding society, you could be excluded from socially remunerative activities, from business deals and employment up to and including breathing. So argumentum ex consensu gentium, or appeal to the consensus of the people, fails to give us any confidence in the truth of the beliefs, although it clearly gives the conditions under which such beliefs will be fit or unfit.
Metaphysical beliefs, however, lack any kind of fitness value, except as far as they are socially acceptable. While some people have been killed for failing to accept the metaphysics of, say, the Host (read up on trans- and con-substantiation sometime), usually what causes the lack of fitness is deviation from the consensus, not a deviation from the true metaphysics.
So Plantinga’s attack on “naturalism” as he defines it, or positive atheism as I define it, rests on a conflation of ecological fitness of beliefs and metaphysical fitness, which so far as anyone can tell is entirely lacking. His two-pronged attack – on evolution producing true beliefs; and on evolution warranting metaphysical beliefs – fails to connect. Evolution does produce true beliefs of many kinds, and he accepts this:
The issue, then, is the value of a certain conditional probability: P(R/ (N&E&C)). Here N is metaphysical naturalism. It isn’t easy to say precisely what naturalism is, but perhaps that isn’t necessary in this context; prominent examples would be the views of (say) David Armstrong, the later Darwin, Quine, and Bertrand Russell. (Crucial to metaphysical naturalism, of course, is the view that there is no such person as the God of traditional theism.) E is the proposition that human cognitive faculties arose by way of the mechanisms to which contemporary evolutionary thought directs our attention; and C is a complex proposition whose precise formulation is both difficult and unnecessary, but which states what cognitive faculties we have—memory, perception, reason, Reid’s sympathy—and what sorts of beliefs they produce. R, on the other hand, is the claim that our cognitive faculties are reliable (on the whole, and with the qualifications mentioned), in the sense that they produce mostly true beliefs in the sorts of environments that are normal for them. And the question is: what is the probability of R on N&E&C? (Alternatively, perhaps the interest of that question lies in its bearing on this question: what is the probability that a belief produced by human cognitive faculties is true, given N&E&C?) And if we construe the dispute in this way, then what Darwin and Churchland propose is that this probability is relatively low, whereas Quine and Popper think it fairly high.
The bolded passages highlight some problems I have with this argument. First, Darwin never rejected the possibility of a God of traditional theism, and I challenge Plantinga to show otherwise. He certainly didn’t believe in such a God, but he takes great care to his dying breath not to reject it as a possibility. Armstrong, Quine and possibly (but I think actually not) Russell are positive atheists, but not Darwin (and not Russell). This is just poisoning the well. I have already attacked the view that naturalism requires rejecting the possibility of that deity (Plantinga dismisses agnosticism as leading to the same conclusion, but I won’t address that here). Let us focus on Plantinga’s R.
Remember that the version he gave to my account was that the beliefs in question were metaphysical beliefs. So the reliability (R) of beliefs that are not metaphysical is, basically, a side issue. If we can show that empirical beliefs are reliable, and I think we can, this says nothing about our ability to produce metaphysical beliefs. And if N is a metaphysical belief, then the probability of N&E&C given R is simply undetermined. But if N is not a metaphysical belief (as I argued for last post), then it is probably reasonably high, since our knowledge claims regarding E and C are also empirical beliefs.
So I think that Plantinga plays a little card trick here. He shifts from talking about ecological beliefs to metaphysical beliefs after gerrymandering the territory so positive atheism is the core content of the metaphysical beliefs and cannot be anything else (like an empirical estimate of the likelihood gods etc exist).
Can I revise my original argument form then? Here’s what I put up:
P1. If evolution is true, then we have modified monkey brains.
P2. Modified monkey brains are not evolved to find out the truth
P3. Evolutionary naturalism (the view that everything about humans, including their cognitive capacities, evolved) is the output of modified monkey brains.
C1. Therefore evolutionary theory is unreliable and should be rejected
C2. Therefore evolutionary naturalism should be rejected
Here’s how I would now frame it:
P1. If evolution is true our cognitive faculties evolved (we have modified monkey brains)
P2. Our cognitive faculties (modified monkey brains) do not give a high probability for our beliefs
P3. Our positive atheist (“naturalist”) beliefs that gods do not exist is the output of modified monkey brains
C1. Therefore our beliefs that gods do not exist are unreliable and should be rejected
C2. Therefore we should reject positive atheism (“naturalism”)
And I challenge P2, and argue there is an equivocation on the step from P3 (only some of our beliefs are unreliable) to C1 (even if our beliefs there are no gods are metaphysical, the argument doesn’t establish that empirical beliefs are unreliable), and reject the identification of positive atheism with naturalism. In Plantinga’s terms, we do not have a defeater for all evolved beliefs, nor do we have a defeater for beliefs about the nonexistence of gods where that is an empirical, not a metaphysical, conclusion.
In fact, as I argued in the first post on the EAAN, evolution provides a good reason to dismiss all metaphysical beliefs as not truth tracking, including (and especially) about the existence of gods and the supernatural. There is no fitness value to tracking such arcane entities and propositions as “there are universals” or “supernatural beings exist” apart from their social value. This means that, in the absence of any other source of reliable information, belief in gods is contrary to evolved cognition and so there is a defeater, in Plantinga’s terminology, for theism, not atheism. Plantinga writes:
The traditional theist, on the other hand, isn’t forced into that appalling loop. On this point his set of beliefs is stable. He has no corresponding reason for doubting that it is a purpose of our cognitive systems to produce true beliefs, nor any reason for thinking that P(R/(N&E&C)) is low, nor any reason for thinking the probability of a belief’sbeing true, given that it is a product of his cognitive faculties, is no better than in the neighborhood of 1?2. He may indeed endorse some form of evolution; but if he does, it will be a form of evolution guided and orchestrated by God. And qua traditional theist —qua Jewish, Moslem, or Christian theist—he believes that God is the premier knower and has created us human beings in his image, an important part of which involves his endowing them with a reflection of his powers as a knower.
This is simple assertion, a form of special pleading. Why isn’t the theist who accepts evolution forced into the same views as the atheist? Evolution doesn’t happen differently if God exists than if he doesn’t, from the evolutionary science point of view. If cognitive functions evolved by natural selection and other unguided processes when God exists, then our cognitive faculties are just as unreliable as when he does exist. The final sentence indicates that Plantinga thinks one cannot believe we are the product of evolution by unguided processes at all. So he is, in the end, rejecting evolutionary biology. I may have misread the structure of his argument, but I don’t believe I misread the intent.
1. Plantinga, Alvin. 1993. Is Naturalism Irrational? in Warrant and Proper Function. New York: Oxford University Press.
2. Griffiths, Paul E, and John S. Wilkins. In Press. When do evolutionary explanations of belief debunk belief? In Darwin in the 21st Century: Nature, Humanity, and God, edited by P. R. Sloan. Notre Dame, IN: Notre Dame University Press.
3. Wilkins, John S., and Paul E. Griffiths. In Press. Evolutionary debunking arguments in three domains: Fact, value, and religion. In A New Science of Religion, edited by J. Maclaurin and G. Dawes. Chicago: University of Chicago Press.
4.Wilkins, John S. 2012. Could God create Darwinian accidents? Zygon 47 (1):30-42.