Alvin Plantinga’s Evolutionary Argument Against Naturalism has attracted a great deal of serious critical discussion (e.g. Naturalism Defeated?) and has had a substantial impact on ‘popular’ appraisals of naturalism. (For example, William Lane Craig frequently uses it, and it also appears in the dismissal of naturalism in The Experience of God.) Many philosophers have pointed out various problems with the EAAN, and in my judgment the EAAN is not only flawed but fatally flawed. Nevertheless, it’s a really interesting argument and it could be worth exploring a bit. I’ll present the argument here and then we can get into it in comments if you’d like — though I won’t be offended if you’d rather spend your time doing other things!
The EAAN has gone through various iterations, but here’s the latest version, from Where the Conflict Really Lies: Science, Religion, and Naturalism (2011). Intuitively, we regard our cognitive capacities — sense-perception, introspection, memory, reasoning — as reliable, where “reliable” means “capable of giving us true beliefs most of the time” (subject to the usual caveats). Call this claim R (for ‘reliable’). But how probable is R?
Suppose that one accepts evolution (E) but also affirms naturalism, defined here as the belief that there is no God or anything like God (N). What is the probability of R, given N & E? One might think it’s quite high. But Plantinga thinks that, however high the probability of R, nevertheless the probability of R given N&E is low or inscrutable. Why’s that?
Now, here’s the key move (and in my estimation, the fatal flaw): beliefs are invisible to selection. Why? Because selection only works on behavior. If an unreliable cognitive capacity is causally linked to adaptive behavior, then the unreliable capacity will be selected for (i.e. not selected against). Even a radically unreliable capacity — that one never or almost never yields true beliefs — can be selected for. Selection only “cares” about adaptive behaviors, not about true beliefs. (More precisely, we have no reason to believe that the semantic content is not epiphenomenal.)
So, Plantinga thinks, given N&E, the probability of R is very low. But, if the probability of R is low, given N&E, then that should ‘infect’ the likelihood of all of the beliefs produced by those capacities — including N&E themselves. So, given N&E, we should it think it extremely unlikely that N&E is true. And so the initial assumption of N&E defeats itself. (Here I’m being much too quick with the argument, but we can get into the details in the comments if you’d like.)
Anyway, it’s a really cool little argument, and it’s not immediately clear what’s wrong with it — and I thought it might be worth discussing, given how influential it is.
I really don’t understand what sort of “truth” might be inaccessible to evolved cognitive processes. As best I can tell, the example I’ve posted elsewhere, about proto-cavemen whose cognitive processes lead them to think of different substances as food, and the brute practical consequences of such beliefs, constitutes a more-than-adequate demonstration that evolved cognitive processes bloody well can and do have access to truth… and a fellow name of Willard van Orman Quine would probably agree with me. Quoth Quine:
If there’s any sort of “truth” which that “pathetic but praiseworthy tendency” does not grant evolved intellects access to, it is a sort of “truth” which, to my eyes, appears to be nothing but a kind of reified delusion.
Perhaps it might be best to use a different term for the “truth” of which religious apologists (Erik, Plantinga, etc) are so enamored, than for the “truth” which gets you killed in accordance with that “pathetic but praiseworthy tendency”. I propose that apologist-truth be referred to by the term “TRVTH”, and the existing term “truth” be used in reference to truth-for-every-purpose-other-than-religious-apologetics.
cubist,
I think we still need an explanation of the difference between a creature that does generally respond in goal-satisfying ways to sensibly detectable and motivationally salient features of its environment and a creature that can regard itself as having correct or incorrect beliefs about that environment. The former sort of creature doesn’t really have correct (or incorrect) beliefs; it just responds as it does, and not as it should.
In other words, what the naturalist needs is a theory of normativity, or intentionality, as such.
He might think that. But he won’t say it.
The sort of truth that come via a direct pipeline from God. Many theists seem to think that’s what they are using.
Won’t say what?
If you are a direct realist, then it (i.e. your position) makes sense. Thanks for being open about it.
I have no problem whatever position people hold. I have a problem when they are inconsistent so that their position cannot be pinned down. In this case you are nicely straightforward.
In my point of view, direct realism is untenable because its epistemology is naively optimistic. For example, how does direct realism account for misperception and hallucinations? How do you detect and correct them?
A theory of truth is a crucial point in criticism of EAAN. Have no theory of truth (preferably flowing from naturalism itself) and you have nothing to stand on.
So it’s not about whether what is perceived and interpreted is true or false, but how can it be considered true or false if one doesn’t think absolute truths exist?
Firstly, I’m a disjunctivist — I think that either one is actually, in fact, perceiving, in which case one is perceiving a physical object (and not sense-data) or one is hallucinating or dreaming, in which case one is not actually perceiving. In other words, hallucinating and perceiving are not species within a genus, but different kinds of cognitive/perceptual activity in which similar capacities are actualized in quite different (and incompatible) ways.
Secondly, I take the key distinction to be that in perceiving, perceiving is necessarily linked to moving. In perceiving the coffee-mug, I perceive it as something that has a bottom I do not actually observe, a back to it that I do not actually observe, and so on — but my awareness of these non-facing sides is nevertheless present to me in my perceptual encounter of the object because I am aware of what I would see if I were to move the mug or my body were in a different spatial location relative to the mug. By contrast, there is neither spatiality nor movement in hallucinating, which is precisely why to hallucinate is not to perceive.
If one thinks that we are directly aware of “sense-data,” then it falls to the intellect to do the hard work of sifting through the sense-data to determine which are likely to be veridical and which are not.
On the direct realist view that I favor, this misconstrues the role of the intellect in human life. Bodily movement carries out the perceptual task that the intellectualist assigns to the intellect. I’ll happily insist that discursive practices play an indispensable role in resolving perceptual discrepancies between perceivers, and more fundamentally, that discursive practices allow us to organize different perceptual encounters from differently embodied and situated perceivers into a more coherent, organized model of perceiver-transcendent reality. But the intellect can carry out these distinctive epistemic and semantic tasks only because the work of embodied perception is already being carried out by the sensibly aware, constantly moving lived bodies that are embodied cognitive/conative agents.
You will notice that it is crucial to my version of pragmatic naturalism that the human epistemic and semantic condition be described in a way that forecloses the very possibility of Cartesian skepticism as an intelligible option, while still preserving the necessary conditions for the fallibility and corrigibility characteristic of all empirical inquiry.
I think so, though I’ll confess I don’t entirely understand the question.
I’m interested in how a naturalist can explain the evolution of creatures that have a concept of truth, and can ask themselves and each other whether what they believe is true or false, correct or incorrect. Even our most intelligent and sophisticated primate siblings and cetacean cousins cannot do that, even though they are extremely clever at problem-solving, adept at tool-use, and so forth. Rationality, in the sense that I’m interested in, and awareness of objective reality, are very distinct cognitive capacities that pose a real challenge to evolutionary theory and hence to philosophical naturalism. Plantinga isn’t wrong to see that rationality is a challenge to naturalism; he’s only wrong to suggest that the challenge can’t be met.
If nothing is either true or false then it can’t be selected for based on whether it is true or false, and it can’t be stumbled upon by accidental mutations or random sampling or via any other physical mechanism.
That’s true.
I get the distinct impression that people making Plantingoid arguments think that our cognitive processes are fallible but theirs aren’t. I am not sure that they would admit that this is what they are thinking.
My cognition is reliable; you have lapses; he is delusional.
al la Sidney J Harris.
I think you may have misunderstood the question. It wasn’t whether something IS true or false, but whether one doesn’t think absolute truth exists. Clearly, natural selection (descent with modification) occurs even among species unable to consider any questions of truth.
I have no difficulty with the idea that one might accept relative truth, contingent on knowledge or perceptual limitations. It’s possible to accomplish a great deal using logic, observation, and intersubjective verification without ever wrestling with the notion of Absolute Truth, if it exists.
Nobody thinks “nothing is true or false”. The question is if absolute, objective truth can exist or can be known. I don’t think one can have a framework of knowledge without assuming a minimum number of postulates, so nothing can be known to be absolutely true. I honestly don’t see the problem with that kind of tentative truth science deals with.
Absolute truth may or may not exist but I think we can all agree it can’t be known, even if one believes it exists, it’s just a belief, a philosophical position.
Is the fact that we can conceive those “absolute” ideas at odds with naturalism? I don’t know why would that be or why any other position can account for absolutes any better than say, ethical naturalism. At the end of the day it seems to me all it takes to account for objective morals or absolute truth is to call dibs and roll with it.
That’s not at all what they are trying to say. They are saying that if naturalism is right, then the probability that our cognitive processes are reliable at all is either low, or inscrutable (no probability can be assigned to it). Conversely, we are rationally justified in considering our cognitive processes to be reliable at all only if the evolutionary process was guided by God or something like God.
That’s the claim.
Obviously I think that the claim is deeply mistaken but it’s important to understand correctly the views that one is arguing against.
Yes, I know that that is the Plantinga argument. I’m just talking about what they think on dark and stormy nights when they get all upset at others, and start taking comfort in their own sensus divinitas.
So they are seriously arguing that the reason I am able to accurately put my food into my mouth so often, rather than missing altogether most of the time, is because their god is guiding my hand? Really? Or are they regarding “cognition” as something very unlike my idea of what it might mean?
I don’t know about Alvin Plantinga’s senses, but my senses are definitely not absolutely reliable. But they are reliable in some statistical sense, and my perceptions can be improved if I argue with people who themselves have not-absolutely-reliable senses, but whose senses err in ways somewhat different from mine.
And evolution can improve the reliability of senses, and that affects fitness. I don’t know why Plantinga requires us to have probably-absolutely-reliable senses for us to pay attention to them at all.
Hazard a guess. Hint: Plantinga’s foregone religious convictions have certain philosophical entailments…
Fortunately you’re the only unbiased person on the planet and you just happen to be posting here at TSZ! I call that miraculous myself.
Fixed that for you. And you’re welcome.
This is, I think, quite close to the heart of the issue, and one that is generally neglected in the controversy around the EAAN: what does Plantinga mean by “reliable”, and — more importantly — is this a conception of “reliable” than naturalists ought to endorse? (Bear in mind, the EAAN only successfully shows that naturalism is self-undermining if it begins with premises that naturalists endorse!)
Firstly, Plantinga uses a version of “faculty psychology” that no naturalist would accept. He think that we have several faculties: perception, memory, a priori intuition, sympathy, introspection, testimony, moral sense, and sensus divinatis. These faculties can function independently or together.
(Plantinga basically writes as if philosophy of mind has not advanced since Thomas Reid!)
Secondly, a faculty is reliable only if it produces mostly true beliefs. By “mostly true” he means that more than 2/3 of the beliefs generated by that faculty are true. So my memory is reliable only if more than 2/3 of my memorial beliefs are true, my perception is reliable only if more than 2/3 of my perceptual beliefs are true, etc.
Thus he defines “mostly reliable cognitive capacities” in terms of “generating true beliefs about 2/3 of the time, more or less”. (In one version of the EAAN he specifies that a capacity must produce true beliefs more than half the time in order to be a reliable capacity.)
So here’s the first move: he doesn’t think about cognitive reliability in terms of guiding successful action in the first place! That’s the first big anti-naturalistic assumption that he smuggles into what is supposed to be a demonstration of how naturalism is self-refuting.
And the rest unfolds from there.
By contrast, if we start off from thinking about cognitive reliability in terms of guiding successful action, especially if we take care to note that action need not be optimal in order to be successful, then we can very easily see how natural selection will tend to promote reliable cognitive capacities!
This comment of Erik’s was classified into Guano. Here’s the non-Guano bit:
There’s a very serious misunderstanding of evolutionary theory here, and since it is a widespread misunderstanding — especially among creationists — it is worth a moment to examine it.
The first misunderstanding is between what an organism values and the explanation of that valuation. Different organisms value different things — what a gazelle values is not what a lioness values, after all! A tick’s values are not those of a deer, and so forth. The explanation of how those different organisms have come to have the characteristically different kinds of values that they do is an explanation in terms of evolutionary theory.
The second misunderstanding about the reduction of evolutionary success to “survival”. Correctly put (I think, but Alan or Joe will correct me here), the trajectory of phylogenetic change is driven by differential reproductive success, which in turn is explained in terms of how selection and drift act on morphological diversity, and where genetic changes are one of the causes of morphological diversity.
(How much of a cause is the main point of contention between, say, Dawkins at one extreme and Brian Goodwin and Susan Oyama at the other. Just in case I have not made the point sufficiently clear, I’m on the Goodwin/Oyama side of that debate.)
It simply is not the case that, according to evolutionary theory, organisms only care about their own survival. Rather, according to evolutionary theory, differential reproductive success explains how it came to be the case that different organisms value what they do, in fact, characteristically tend to value.
That said, it is an interesting question how it came about that a specific kind of organism — that is, advanced hominids — came to value truth, to the extent that we do. And the way to do that, I think, is to both (a) examine carefully the pragmatic function of truth-discourse in everyday life and (b) embed (a) within a naturalistic theory of what makes human beings different from other kinds of apes, with particular focus on the very interesting fact that great apes are very bad at cooperation, and human beings are very good at it, even from very young ages.
The preliminary result of (a) and (b) is that we care about truth because caring about truth is necessary for successful cooperation, and successful cooperation is central to the distinctively hominid ecological niche of cooperative foraging, which in turn has been just-good-enough for one particular twig of the hominid bush to (thus far) avoid extinction.
Dazz said:
I have absolute knowledge that the statement “error exists” is absolutely true. Anyone care to mount a challenge to that?
Of course certain statements are absolutely true because they make explicit constitutive rules of a language-game. “Error exists” is absolutely true, because it makes explicit the fact of our finitude and fallibility. That’s different from saying that there are claims about how the world really is which are immune to revision even in principle.
Can you defend that without resorting to logical postulates? If you’re going to invoke something like the law of non-contradiction, then that statement is true relative to the law of non contradiction. So no, that is not an absolutely true statement on itself
If the predestination, god-pre-decreed everything, crowd is correct, then error CANNOT exist. Even our “errors” were predestined to serve some purpose, meaning they aren’t errors at all. Just limitations in our perceptions.
And, it should be emphasized, the law of non-contradiction is not absolutely indispensable for logic. There are alternative logics that do not assume it.
True, the law of noncontradiction belongs specifically to classical logic.
On another note, I’ve been thinking about this and I’m not sure if this makes any sense from an evolutionary or philosophical perspective, but I think maybe truth perception has evolved the other way around: the more primitive the cognition, the more absolute the perception of reality is, even if not consciously aware of it.
Maybe what makes our cognitive abilities especial, if anything, is the fact that we’re able to question our own perception and not simply assume whatever goes on in our minds must be true. Does it make any sense?
If you mean that we do not NEED to be religious, I’d agree.
I have a conflict of interest. I was once nicknamed Old More Or Less.
I’m not one to be asking about the desirability of absolute truth.
I think there’s something right about this intuition.
We can put it a little bit differently: in non-enlanguaged cognitive systems, cognitive activity is so closely coupled to affordances that there isn’t sufficient “slippage” for the system to construct meta-representations about the discrepancy between its own representations and the representations of other cognitive systems. Non-enlanguaged cognitive systems are inveterate “dogmatists,” if you like.
That’s awesome, LOL. English is my 3rd language but I couldn’t express it half as eloquently in my own mother tongue in a million years (or 1000 YEC days)
I think a clear implication of this is that truth poses no challenge to naturalism if it’s the “default”, if absolute truth is the null hypothesis from an evolutionary perspective so to speak, and self awareness plus abstraction capabilities allows us to question truth itself and whether there are degrees of truth.
So, it’s not that we evolve in a way that makes our beliefs necessarily unreliable, we evolved to question their reliability and that’s a key part of what has made us thrive as species. Since it’s an advantageous trait, there’s no reason to think that’s a problem for naturalism/evolution
I there some subtle semantic difference between “finitude” and “finiteness”? */quibble*
I query the claim that error exists. I agree that errors exist. We, and other sentient beings presented with choices, make wrong ones – errors. But “error exists” seems like reification to me.
dazz,
Thanks! Though it’s not just that English is my first language, but that thinking about this stuff is what I do for a living. Presently I’m reading Joe Rouse’s Articulating the World, which focuses specifically on the question as to how language could have evolved, using both philosophy of language and cutting-edge evolutionary theory.
When Josiah Royce says “error exists,” all he means is “there are errors”, or “we sometimes make mistakes”, both in perceiving and in reasoning. There’s no reification in the original version of the argument, just a 19th-century rhetorical flourish.
OK, no problem. I was reminded of an exchange with StephenB at UD (can’t locate the thread now).
Still, I think it could be argued from some religious perspectives that our “errors” are the Will of God, who makes no errors. So errors of logic, for example, are Divinely mandated for purposes beyond our ken. The whole idea of errors becomes a frame of reference issue, along with free will.
To be clear: Royce does intend “error exists” to be (i) immune to all skeptical doubt (much as Descartes’ cogito ergo sum was) and (ii) the first premise in an argument for the existence of God. I simply don’t think that a reification of errors into “error” is at work in how that argument is supposed to go. But I’ll look at it more carefully.
Actually, my position is that human cognition is fallible, but corrective. We can perceive or misperceive, but we can also tell, after due deliberation, which is which. Or we can be corrected by reality bigger than ourselves, which amounts to the same thing as self-correction, insofar as we are an organic part of the self-same reality.
EAAN demonstrates that on naturalism there’s not much reason to believe our cognition to be true. By this, Plantinga is making a rather modest, even trivial point. A more serious point is that on naturalism there is no (and probably can be no) concept of truth as distinguished from falsity, perception as distinguished from misperception, etc.
This more serious point is perfectly evident from the post by the poster called Kantian Naturalist. For example, he says,
On this account, on one side we have perception and on the other side we have misperception, hallucination, dream, and non-perception perfectly equated. On this account, we have dreams, which are a natural cognitive function with their own natural confines (perceived when sleeping), equated with hallucinations, which are to be kept sharply distinct from normal perception if “truth” and “reality” are to mean anything, but KN’s account does not even remotely address this task.
On this account, we also have non-perception equated with dreams (a natural function) and hallucinations (something that can be mixed up with ordinary perception, even though it must be kept apart from ordinary perception in the name of truth). Somehow non-perception, which is a non-event of no consequence, is seen as the same thing as the events of dream and hallucination which have the same consequences as ordinary perception. In reality, when you dream of a scary thing, you will be as scared as if you saw it in broad daylight, but KN thinks dream is the same thing as non-perception, even though in non-perception you don’t perceive anything and there are therefore also no such consequences as getting scared or delighted etc.
Knowing what naturalism is and what direct realism is, I can guess why KN is describing things the way he is, but the right thing for us to do right now is to recognize it for what it is – an utter convoluted falsehood and non-explanatory befuddlement. KN likes to talk as if he knew stuff even when he is totally contrary to very basic facts and logical distinctions.
KN’s view stands in contrast with mine. On my view, dreams and hallucinations are forms of (mis)perception, they have the same effects as perception and are very difficult to separate from (true) perception, but it’s possible to do it with appropriate effort. The distinction between misperception and (true) perception occurs on the level of the intellect, not of the senses. If perception were on the level of the senses (such as seeing or hearing), we would not be able to perceive e.g. a barking dog, because barking is perceived through one sense (namely, hearing) and the dog itself is perceived via another sense (namely, seeing). The proper understanding of the perception of a barking dog occurs on the level of the intellect (mind) which synthesizes hearing and seeing.
The intellect (mind) has more functions. It can deliberate a reaction. Usually people are instinctively afraid of barking dogs, particularly when the dog is loose, and particularly when the dog is attacking, but the intellect can, with some training, restrain the fear impulse and devise a different reaction, appropriate or inappropriate.
In the same manner, the intellect can deliberate on the accumulated experience of any and all perceptions and prioritize between them, assigning more relevance to some perception (such as to daily experience) and limited relevance to other perception (such as to night-dreams). This is possible because the intellect (a) in fact is a distinct function from the senses and (b) is naturally meant to organize the senses. This is known from the fact that when someone fails to exercise the intellect in the said functions, then concrete consequences follow, the individual human being becomes de-humanized and approaches animal nature.
This in contrast with KN’s view, which belittles and muddles the definition and functions of the intellect, basically equating it with the senses. On KN’s view, there’s also no basic distinction between perception and misperception (which naturally follows from his failure to properly distinguish between the senses and the intellect). He equates misperception with non-perception, which is false whichever way you look at it, logically, semantically, biologically.
To be clear, I have nothing against the fact that KN holds the view he holds. Problems creep in only when he thinks his view is somehow correct and some other view is wrong. Because, you see, on his view, taken consistently, things like “correct”, “false”, “right”, “wrong”, etc. have no meaning. When he says those words, he is uttering words without meaning.
OK, let’s put that to the test: according to you, in a theistic world view you’re entitled to talk about truth and falsehood. Let’s apply that along with your paragraph above to theism itself: when two theists disagree on contradictory theistic positions (Christianity vs Islam, Theological fatalism vs compatibilism, etc…) how’s the track record for deliberation settling those issues? How do YOU determine your conclusions are correct and your opponent’s are mistaken?
Among theists, we agree on something rather crucial – there is truth to be settled. As opposed to naturalists (or whatever you call them) who keep mocking and ridiculing differing points of view, or calling them wrong, while actually having no positive account of truth themselves, so they really have no basis for ridiculing and labelling anyone wrong.
The answer to your question is: There are ways to determine who/what is right and I make use of them as appropriate. Now go ahead and prove me wrong.
What about theistic relativism? First fail on your part.
Second, you still don’t get it. It’s about defining truth, not denying it. I contend Plantiga’s argument can easily be dismissed if one considers truth tentative and provisional, which is a logical and consistent way to go about truth considering we all agree we’re fallible beings. Science has made incredible progress with that conception of truth, while you theists can only pretend that believing in absolute truth makes it real, and that positing a being to ground truth somehow grounds truth on that being when you’re actually grounding it in your premise that truth and god exist. But since those premises come from a fallible mind, you haven’t really cemented your beliefs on absolute grounds, just premises from fallen brain.
But I don’t even think absolute truth is incompatible with naturalism. One can be a natural realist and deem nature a source of objective truth. There’s absolutely no reason to believe that absolute truth must come from some supernatural being, and even if we’re not perfect and we can’t know those truths 100% if they exist, one can still believe the closest to natural truth our cognition, the more fit one organism becomes in evolutionary terms. In fact, becoming aware of our own imperfection as I contended earlier, get’s us closer to truth (if one takes the position that it exists in absolute terms) as opposed to the illusion of being in possession of it.
I disagree with almost everything in your post, but I’ll limit my response to this. Maybe, I should add that I mostly agree with KN’s stated position.
Perception is reliable, in the sense that we can rely on it. Yes, there are misperceptions, but it is generally reliable.
I’ve been studying perception (in the sense of “theorizing”). And I cannot find any way of making “perception is true” to be a meaningful statement. And, as I see it, that makes the EAAN an absurdity.
In what I quoted, your assertion was about whether “cognition is true”. I can’t make sense of that, either. If we could only think true thoughts, then the use of reductio ad absurdum would be impossible.
(Sorry to repeat – this continues to niggle)
Regarding the OP and Alvin Plantinga’s Evolutionary Argument Against Naturalism, where does this argument take us if we assume it to be correct? Indeed what if we accept Craig’s cosmological argument or the various versions of Aristotle as espoused by Aquinas and “developed” by Edward Feser?
All they seem to do is get you to a something that created the universe. That seems the easy bit. How do you get from your attribute-less creator to the tenets of any particular dogma? I don’t think the creator is wearing any clothes.
Alan Fox,
There are other arguments that work to develop the various attributes of the creator.
Would that be the creator that you don’t actually believe in? Just wondering…
If error doesn’t exist, there’s no reason for me to defend the statement. Asking me to defend it is your implicit admission that error exists.
There are certain givens that are absolutely necessary to even have anything to debate or a reason to debate in the first place. I sum those things up this way:
If you do not assume the law of non-contradiction, you have nothing to argue about. If you do not assume the principles of sound reason, you have nothing to argue with. If logic is not assumed to be a causally independent, authoritative arbiter of true statements, there’s no reason to apply it. If you do not assume libertarian free will, you have no one to argue against. If you do not assume morality to be an objective commodity, you have no reason to argue in the first place. If you do not assume mind is primary, there is no “you” to make any argument at all.
This is the great problem with naturalist arguments … they are full of sound and fury, signifying nothing. You might as well be a maple leaf arguing with an oak leaf that its shape is “wrong”.
And yet theists have been arguing in circles since the dawn of time and no agreement on anything at all has been reached.
Assume morality to be an objective commodity all you like, it seems to make no difference. Nobody agrees, nobody can agree.
Or, in other words WJM has won the argument before it started because by arguing you have already agreed with his givens. So by arguing you concede defeat.
It’s a bit sad really.
and…
Is there one that works for you? How does it proceed – this argument to get attributes for a naked creator?