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.
petrushka,
I resent lumping “theology” in with “philosophy”!
The tools of philosophy — analyzing concepts, describing experience, devising thought-experiments, and so on — are certainly no substitute for explanations of how things are. In my view, science tells us what is, and philosophy tells us (among other things) what ‘what is’ is. So the value of philosophy depends on whether that second inquiry is worthwhile. I don’t see why it wouldn’t be.
So I agree that philosophical investigations are useless for determining empirical issues — how things really stand in the world, as available to us in experience — but they can do all sorts of other things — such as propose criteria for distinguishing empirical questions from other kinds of question. But just because philosophy doesn’t deal with first-order, empirical questions, it doesn’t follow that testing for internal consistency among statements is the only other things that there is to do.
Telling us what “is” is would be useful, but as far as I can tell, allnphilosophy can do is tell us whether statements are coherent. I can’t think of a moment in my life — and I’m pushing 68 –when I did anything as a result of a philosophical thought.
Unless you count applied pragmatism and utilitarianism. But to me those are just words describing my behavioral tendencies.
While I often ask myself if what I am about to do is the most productive thing available, I never think about meeting some idealized standard of metaphysical consistency.
I’m probably missing something subtle in how you are using the word “proposition”, but why don’t recreational or therapeutic drugs count? Or how about those parasites that some scientists have affected human behavior (although I am not sure if these parasites would taste good on pizza).
A related point is the capability of brain surgeons, as first reported by Penfield, to create beliefs (about memories currently being experienced) in people by stimulating certain neurons.
Coincidently, I am reading from Dennett’s latest book and it includes an argument that it would not be possible to use some future technology to insert a single belief into a rational person’s brain if it contradicted his or her web of other beliefs. Being a single belief is the key point. Dennett’s example is trying to insert the belief “your brother’s name is John” into a person who grew up an only child. He argues such a belief would either be rejected as contradictory or accepted and lead to a web of confabulations.
Epistemic non-reductionism of biology to physics is the acceptance wisdom, as far as I know. But ontological emergentism is controversial at the least, I would think. Any references, preferably layman-friendly (on evidence it exists, not on what it is)?
We can easily block memory formation, even when the person is awake and fully conscious. Memory blocking is done all the time in surgery.
There’s also techniques to erase memories Skilled therapists can induce false memories. People fill in stuff to integrate the false memory with the rest. There have been some major scandals involving false memories of child abuse.
I spent some years investigating child abuse, and I constantly worried that children might be saying what we wanted to hear.
Thanks. I won’t vouch for the soundness of the inference from low/inscrutable P(R|N&E) to the defeat of N&E, even with the addition of Plantinga’s proper functionalist account of warrant, but from what I remember, it does play a role. And you can sort of see how it would.
IMO, philosophy is not just an addendum to “empirical” sensory input, it is a means of examining the assumptions you make about about that input, examining how you organize it, and examining the models you make from that input. The idea that philosophy is an unnecessary addition to sensory input, IMO, only held by those who don’t recognize that how they interpret, organize and model sensory experience is a belief structure – a philosophy, whether they realize it or not.
Those that dismiss philosophy, IMO, are actually simply holding their own unrecognized philosophy as true, and dismissing other philosophies simply because they differ from what they mistakenly believe is the only “real”, or the “natural”, way to interpret, organize and model sensory data.
Philosophy has been invaluable to me in that through it I realized that there were other ways to interpret, organize and model sensory input than the way I just happened to do so. Changing my philosophy changed my entire life for the better.
Here is a parallel example to illustrate why low P(R|N&E) does not, pace Plantinga, undermine our confidence in the rationality as applied to ourselves.
Take some specific feature of the natural world, for example the shape of the maple leaf – call it M. Suppose also that we don’t have a plausible account, from naturalistic evolution alone, of how it came to be. It would seem then that P(M|N&E) is low or inscrutable: as far as we can tell, based only on our belief that the leaf was formed as a result of naturalistic evolution, the leaf could take a wide variety of shapes and sizes (within reason). Our prediction is very unspecific, and therefore the probability of the actual shape, given by the theory, is low.
Does this finding undermine our belief M, i.e. the belief that the maple leaf has its characteristic shape? Well, if N&E were the only reason for holding that belief, then low probability would be grounds for doubt. However, not only N&E is not the only reason for believing M, it has nothing to do with our reasons for believing M. Our belief was formed quite independently from any metaphysical or scientific views that we might hold. In this case it is due to the evidence of our senses. If anything, the implication of the low/inscrutable probability might go the other way, towards N and/or E. Clearly though, the low probability doesn’t demonstrate much other than that the evolutionary account is incomplete: it doesn’t make specific predictions for every single feature that we might observe (it has made plenty of other successful predictions, of course, which is why we have confidence in it).
If you set out to find the ground of being or the necessary being and you end up with the universe, then you went astray somewhere, because the universe as the sum of all objects and things is the stock example of contingency, not of necessity. The other option is that you actually did not set out to find the ground of being, but you are up to something else instead.
Also, if you ascertained the ground of being and then you tried to deduce its characteristics, you’d find it logically impossible to prove that it’s something less or other than God. If you think it possible that the ground of being is something else, it’s doubtful you have given the matter any thought.
Did you notice how you equated classical theism (for which Plantinga famously has little respect) with theism as such? You jumped from “all of the classical theistic arguments are logically invalid” to “there are no good arguments for theism”. In truth, Plantinga thinks that his arguments for the properly-basic-belief theism are very good.
Example? Reference?
What are the entailments of properly-basic-belief theism.
What is the point?
There are several, but here are three that I’ve read and think are quite good:
Paul Churchland, Is Evolutionary Naturalism Epistemologically Self-Defeating?
Geoff Childers, What’s wrong with the evolutionary argument against naturalism?
Feng Ye, Naturalized truth and Plantinga’s evolutionary argument against naturalism
The crux of all three is that Plantinga’s argument fails to show that naturalism is self-undermining because he does not begin with a naturalistic conception of what semantic content is.
Granted, naturalizing semantic content is a huge issue, and there’s no consensus about which proposal is the best one for naturalists to adopt. At the same time, we have really excellent proposals on the table that have been developed by Ruth Millikan, Fred Dretske, Paul Churchland, Andy Clark, and Joe Rouse. It’s not as if we have no idea how content could possibly be naturalized; rather, we haven’t yet come to consensus about which exact proposal is the right one.
My own view is that a synthesis of Clark and Rouse is most likely to be correct, and that adopting that position would be decisively refute both Plantinga’s EAAN as well as Alex Rosenberg’s eliminativism about intentionality because (he claims) it cannot be naturalized.
I don’t think it’s sound. It was unsound when Anselm and Aquinas put it and remained unsound when Leibniz and Spinoza put it. It didn’t magically become sound when Plantinga made it seem more complicated.
Ferchrissake, look who’s back.
Zombie thread.
As petrushka says, What is the point?
That’s missing the point of Plantinga’s challenge. Plantinga’s challenge is to say that we cannot say that natural selection will tend to favor having true beliefs over having false beliefs.
He says this because (a) natural selection can only target behavior, and so only determined whether a trait is adaptive or maladaptive and (b) the causal relationship between the content of a belief (and hence its truth-value) and overt action is too complicated for us to ever know if an action is adaptive because it was caused by a true belief. There are other beliefs in the cognitive system, and also desires, that cause action.
What Plantinga doesn’t seem to understand — in fact, I will say quite explicitly that he does not understand — is that a naturalist can say that the content of a belief is its role in coordinating perception and action, which are targets of natural selection.
Thinking is a behavior. It has no content.
So when I’m thinking, I’m not thinking about anything (= the content of the thought)? And thoughts are neither true nor false?
hotshoe_ writes:
Indeed. Erik, there are still questions from hotshoe_ and walto in the Varieties of Religious Language thread that you have not answered.
I don’t think petrushka answered this, so I’ll give it a try.
When you are thinking, you are thinking about possible future behavior, usually in order to evaluate and select the best option. I don’t see how to apply “true” or “false” to that.
The possible future behavior might be linguistic behavior. But typically people are evaluating that linguistic behavior for well it might work in a discussion or as persuasion. Truth could be one aspect of that evaluation, but it is far from the only criterion used.
hotshoe:
Patrick:
And in fairness, I should point out that there are some unanswered questions waiting for KN, as well.
I didn’t answer because I haven’t seen the question. Things have been scrolling pretty fast. Thinking is covert behavior. It could be linguistic, but it could also involve covert actions and imagined consequences.
Truth is a possible attribute of verbal propositions and assertions, but I don’t lose much sleep over truth. Unless the assertions have important consequences, I don’t see that it matters one way or another. Truth about imaginary entities seems dubious. Truth about assertions that can’t be empirically tested seems more political than rational. Self consistency is a desirable attribute in discourse, but it doesn’t add any truthiness.
Is this a/the naturalistic conception of semantic content that you referred to above? This makes naturalism look worse, not better. And unrealistic too – people often enough hold beliefs that they keep hiding and they reveal them at the worst possible moment. Questions like “Are you lying?” should be impossible in the real world, because it should always be possible to read beliefs from behaviour.
Further muddling the concepts of truth and belief does not help naturalism against EAAN. It only makes it worse. If intellectual honesty matters, that is, but there’s no inherent reason why it should matter for naturalists.
Thanks for the links, KN. I will at least click them.
(The abstract of) Feng Ye’s criticism of EAAN is more interesting than the other two. The other two are easily dismissible.
Feng Ye says that, among other things, EAAN states that “(3) the most popular naturalistic theories of content and truth are not admissible for naturalism.” I was not aware of this aspect of EAAN, but if this thesis is included, then it is false to claim that EAAN fails to consider naturalistic theories of truth.
Feng Ye’s response, “I further argue that Plantinga’s argument for (3) is not successful either, because an improved version of teleosemantics can meet his criticisms. Moreover, this version of teleosemantics implies that the truth of a belief is (probabilistically) positively related to its adaptiveness, at least for simple beliefs about physical objects in human environments. This directly challenges Plantinga’s claim that adaptiveness is indifferent to truth.”
Last thing first, EAAN does not claim that adaptiveness is indifferent to truth. Instead, the claim is that insofar as evolution (natural selection, adaptive and reproductive success of life forms) matters, then truth doesn’t (i.e. more like an inverse relationship instead of a relationship of indifference). It’s correct that this claim depends on a particular theory of truth, but this is why EAAN includes the thesis that the most popular naturalistic theories of content and truth are not admissible to naturalism. So, the job for the critic of EAAN is not to devise yet another particular theory of truth, but to demonstrate that the theory of truth, devised whichever way, is compatible with naturalism.
In my opinion, consistent naturalists can only say “there’s no truth”. This is the only theory theory of truth that is compatible with naturalism. In any other case, we are dealing with inconsistent naturalists.
Why? Let me guess… because without god there’s no truth maybe?
Assuming truth is something along the lines of that will make you sick, so don’t eat it, or Thag makes things up, so don’t trust him, how is the ability to distinguish truth at odds with survival?
That’s a simplistic misrepresentation of what “truth” would be like for evolving creatures. Truth, understood as the ability to understand the environment, and other members of your society is definitely an advantageous feature to have. There’s nothing contradictory or self defeating in evolving that ability to figure out how things work in the natural world, and morality as means to thrive in society, just like many social animals do to some extent. Linking truth to morality is your arbitrary definition, not necessarily the kind of definition that naturalism needs to deal with. And even by your own definition, your theistic position can’t help you know any of that kind of moral “truths” you talk about. As a matter of fact, in a world where people are fallen, imperfect and born sinful, you can’t claim that any of the truths coming from that theistic view can be reliable, because you believe truth can only come from god. So it’s actually your definition of truth in your worldview / framework of knowledge that is self defeating.
This sort of pathetic argumentation may seem new-ish, but it’s the same old question begging crap “you can’t account for truth/morals without god”
Can you name a single truth that emanates from your theistic worldview and is incompatible with a naturalistic view? You can only have:
1. “Truths” about the natural world and it’s features, necessarily tied to naturalism
2. “Truths” about the supernatural world that you can only know through your imperfect material nature, hence can’t be considered truths in the absolute, infallible way you attribute to that supernatural stuff
So besides pretending, you know no truths that are not material, naturalistic and your truths are most definitely not absolute
Any god? You mean I can’t make up my own? bummer
The woefully unfit remark is a giveaway that theism is ultimately about power, about forcing other people to do what you want.
Eventually the guns come out.
Very good reasoning, but irrelevant to EAAN. EAAN supposes arguendo that naturalism is true. You suppose that whatever general worldview is true, specific beliefs have nothing to do with it. Hence your reasoning is not providing a parallel.
By this definition, truth value is inapplicable to naturalism and to any other general worldview, insofar as they don’t motivate or impact survival. And they don’t – you may believe God keeps you alive or you may believe natural selection does it, either way you don’t believe it’s yourself keeping yourself alive, therefore ultimately your own actions (and, on naturalism, beliefs are a sort of action, so I keep hearing here) don’t matter.
By the way, all along, there’s the unanswered question why survival itself should matter. Why not truth (as an ultimate value, i.e. truth for truth’s sake)? Why not world peace and happiness? Why not universal harmony?
Do you get bonuses for posting stupid stuff like this? One would think at this point something as obvious as the fact that NS can act upon actions of organisms would be understood. Apparently not for Erik, who seems to picture natural selection as some sort of deity that dictates one’s fate regardless of actions or behavior. Heh
I am not in possession of truth. Nor are you.
That was simple.
The primary difference between us seem to be that I am not troubled by this.
What you really mean, is that the theistic conception of truth value is inapplicable to naturalism. And that’s no surprise.
You follow Plantinga by smuggling in your theistic presupposition via the way that you conceive of truth.
Yep. It’s no surprise when theists of any stripe dehumanize anyone else who doesn’t happen to bow to the same god they do.
But I suspect Erik has some personal issues which make him worse, more hostile, more smarmy, more falsely-superior, than just theists in general.
Agreed. I wonder why they have no answer to my previous comment on this, and why they think their conception of truth is attainable under their own prerogatives, that humanity is fallen, sinful, and there can’t be truth if it doesn’t emanate from god. It must mean that anything coming from their sinful minds is highly improbable to be true, including their own belief in god. It seems to me it’s the christian dogma / conception of truth that is self defeating
I’m still waiting for the explanation for how God provides truth to us.
Even if they came up with something there (other than the “God can do anything” rot), I’d then like some evidence for it happening.
Just two crucial problems with their position that have no prospects for being answered, ever.
Glen Davidson
It is true that a fully adequate naturalism will need to explain, among many other things, why we take an interest in objective reality. “Why is truth important to us?” is a question that naturalism does need to explain.
Consider a certain truth-skeptic who says, “oh, we’re not really interested in truth — we’re really interested in Something Else, and ‘truth’ is a residue of some theistic or mystical metaphysics.” A truth-skeptic might argue, for example, that truth-as-correspondence requires a God’s-eye view, but there is no God’s-eye view for evolved finite creatures like us, so there’s no epistemic position we could ever occupy from which we could guarantee to ourselves that correspondence had been achieved. Putnam argues this way, and so does Davidson and Rorty.
The problem with truth-skepticism like this is that it doesn’t consider (a) the essential and ineliminable role that talking about truth plays in our ordinary discursive practices — a point raised again and again by Huw Price and John McDowell in their criticisms of Rorty and (b) the fundamentally groundbreaking work of earlier generations of pragmatic naturalists, like Dewey and Sellars, to show that correspondence is itself a natural relation between an organism’s cognitive maps of its environment and the salient features of that environment. The capacities of an organism to map some of the features of its environment are evolved capacities, which means that they are sub-optimal, and that in many cases accuracy of representation is sacrificed for the sake of other goals, such as not wasting energy, not exposing oneself to predation, and so on. Better to run away, even if it’s a false positive, then stick around to be sure and get eaten!
The hard question for naturalists is how to connect this fairly basic story with the more demanding story.
The fairly basic story — one that we understand pretty well in general, though many of the details still need to be worked out — is about the ecological function of generally reliable, multiple modules or maps of different salient features of the physical and social environment in which organisms of this kind tend to live.
The more demanding story about how the lives of sapient animals is structured by the space of reasons, where we are capable of taking an interest in what one is committed and entitled to assert what is the case, and capable of recognizing the similarities and differences between what I actually think and what I should think, and where what I should think is what anyone (suitably informed and suitably rational) would think, as well as recognizing the similarities and differences between what I think and what you think.
It seems to me that the EAAN assumes that there is no naturalistic explanation that can connect the easy story (about the cognitive lives of non-rational animals, which can be grounded in evolutionary theory) with the more demanding story (about the epistemic and semantic lives of rational animals).
I’ll happily admit that the non-naturalist is entitled to a certain degree of skepticism about whether or not there is a good naturalistic explanation that connects these two stories.
But that’s not what Plantinga is doing. Rather, Plantinga is assuming that there’s no way of connecting these two stories. And he’s not entitled to make that assumption, because he simply has not done the hard work of familiarizing himself with contemporary naturalism. He hasn’t. He knows nothing of contemporary naturalism. He doesn’t know what has and hasn’t been accomplished, or what the big questions are that remain. He hasn’t done his homework. He has a lot of assumptions about what naturalism is and isn’t, and what naturalists are and aren’t committed to, but he’s basically just making it up.
This is not, I should add, unique to Plantinga or to theists. Naturalists routinely fabricate what they imagine theists to believe.
In terms of how I think the easy story and the demanding story are connected, it seems pretty clear to me that the big difference is language. So what the naturalist needs is a theory of the origins of language, if language is what makes the difference between the relatively simple minds of sentient animals and the more complex minds of sapient animals.
What you ignore is that every critic of EAAN acknowledges, tacitly or overtly, that naturalism has no sensible theory of truth, but needs it urgently in order to counter EAAN. No surprise that you ignore it.
As for me, I reject naturalism for many other reasons. It has nothing right in it. Even the name is wrong.
If you are right about this, then you are in possession of truth.
If you are wrong, you have no clue whether I am in possession of truth or not.
That was simple.
There is a perfectly good naturalistic theory of truth that takes into account two different features.
Firstly, “correspondence” is itself a natural relation between an organism’s cognitive maps of its environment and the salient features of that environment. Basic cognitive systems are comprised of generally reliable, multiple modules or maps of different salient features of the physical and social environment in which organisms of this kind tend to live.
Secondly, language adds to this a second layer of semantic information whereby conceptually structured representations can be shared across embodied cognitive systems. Language makes possible a “pooling” of semantic information (both referential and inferential) that would otherwise be limited to individual cognitive agents, each of which is an embodied cognitive/conative system.
Any natural language that has devices for marking or indicating first-person, second-person, and third-person voices will thereby have the pragmatic structures necessary for coordinating information across embodied cognitive systems and hence have the resources for allowing those systems to break free of mere suboptimal representation of affordances and attain some degree of awareness of objectively real causal structures.
In other words, naturalism has a perfectly sensible theory of truth just in case it has a theory of animal cognition and a theory of natural languages. And it does.
Dorm room stoner reply from Erik.
Supposedly a deep thinker.
His actual experience of deep thought seems to have stalled about two thousand years ago.
And he still thinks fossils on Everest are evidence of some kind of “global” flood.
Deep thinker, indeed!
Erik:
Last thing first, having familiarized myself with the criticism of EAAN that you pointed out, I can say that the work of familiarization was rather easy. Naturalists self-admittedly have no settled theory of truth. Once they settle on one, they will still need to ensure that it’s properly naturalistic. This work lies still ahead for naturalists.
As far as I am of aware of EAAN (which is admittedly less than you), Plantinga does not assume that naturalists have no way to connect the easy story with the hard one. Plantinga’s argument is modal, probabilistic. He does not say that on naturalism we cannot know the truth. Instead, he says that, given naturalism, the probability that we know the truth is low. Thus the argument permits much more flexibility – for naturalists – than you lead us to believe.
Naturalists are free to keep devising their theories of truth until they hit upon one that is consistent with naturalism and does not smuggle in transcendental notions. But until then, let’s acknowledge their situation for what it is.
Isn’t it about what truth is and what kind of truth (what operational definition) we can deal with? It seems to me that this debate equivocates the claim that there’s no absolute truth, or that it can’t be known, with some sort of hard solipsism. That’s not the case IMO, but then again I know next to nothing about philosophy
I would just say that low is higher than zero.
Correspondence theory of truth? I can see how you can call it naturalistic, but perfectly good, no. The theory looks good if you are a direct realist. But direct realism itself is false…
Why should it be “perfectly good”? I really wouldn’t want to smuggle “transcendental notions” here
How dare they!
Then let me recommend Richard Rorty’s Philosophy and the Mirror of Nature. It’s available free as a PDF here.
Thanks, much appreciated, will certainly read it
Erik,
That’s actually an interesting point: a naturalistic rejection of the EAAN can look good if direct realism is true. Certainly I myself think direct realism is true, so I use direct realism in my own criticism of the EAAN. I don’t know if any naturalistic criticism of the EAAN must rely on direct realism, though it’s worth thinking about.
Personally, I don’t subscribe to naturalism, if only because I have never been able to work out what it is.
As for theories of truth — as best I can tell, no sensible theory of truth has ever been advanced.
But Erik’s got one, no doubt, that he won’t advance here because you’re subhuman – errm, because he knows you don’t agree with his particular view of god and therefore you’re woefully unfit to hear it.
Pearls before swine, right Erik?