Tuesday, 16 October 2012

Are science and naturalism compatible?

This may seem like an odd question to ask, especially given that the usual argument of compatibility is between science and religion, but it has recently been posed by Christian apologist Alvin Plantinga in his book, "Where the Conflict Really Lies: Science, Religion, and Naturalism", and in more detail in this lecture:

Religious apologists are known for their ridiculous arguments, especially when venturing into discussions on science, but it is also wise for us to consider the fact that people like Plantinga and Craig are not stupid men; they are well-educated and often have impressive philosophical and logical skills. It is for this reason that when I read this quote:
“There is indeed a science/religion conflict, all right, but it is not between science and theistic religion: it is between science and naturalism. That’s where the conflict really lies”
I decided to try to deduce what possible rational line of thinking could give us such a conclusion. For those who are not sure why this would be a particularly strange claim, it might help to look at one of the main assumptions of science: methodological naturalism.


This combination of terms is sometimes rejected as mere "navel-gazing" by people who enjoy the more practical benefits of science rather than analysing the philosophical foundations of science, but when we look at what the concept actually means we find that it isn't very controversial at all. The "naturalism" part refers to the type of things we study; that is, we study things which are observable, measurable, repeatable, and so on.  The "methodological" part contrasts it with a metaphysical position, so since metaphysics is the study of what is "real" then a methodological position is one that simply assumes naturalism is true for pragmatic reasons, rather than claiming it is absolutely true. To put it most simply: methodological naturalism is the claim that no matter what is "real" or "true", science should just assume the world is observable and measurable and ignore anything else that doesn't fall within that category because that is what gives us meaningful results.

This is where my first possible explanation for Plantinga's claim came from: maybe Plantinga was conflating methodological naturalism with metaphysical naturalism. It would be a valid argument to claim that science is incompatible with metaphysical naturalism as metaphysical naturalism makes claims beyond what science can demonstrate or support. For example, there is no scientific experiment that could be devised to support the claim that the world is naturalistic rather than dualistic (the idea that reality is composed of two distinct substances; mind and matter) and instead we have to rely on logical arguments to disprove the idea that the brain is simply an antenna rather than being an organ that produces thoughts. Reading through his arguments though, this is not the argument he is making.

This led me to considering a second possible explanation for his claims: maybe Plantinga was conflating the natural/supernatural distinction that is considered in science and philosophy with the distinction that is often used in common language. Unlike the first possibility, this does not constitute a strong argument, however, it would be a reasonable mistake to make given that there is still a fair amount of debate and confusion over what the terms 'natural' and 'supernatural' refer to. As I mention above, what is 'natural' is generally agreed to be that which is observable, measurable, and repeatable, and the supernatural is thus its opposite (the unobservable, immeasurable, and unrepeatable). This is not how the terms are treated in common usage though, as 'supernatural' has come to take on the meaning of 'wacky' or 'magical'. What this means is that sometimes the judgement of what is or is not supernatural is made before considering how the concept is formulated, and instead it is often just claimed that things like ghosts, psychic abilities, gods, and so on are supernatural. This isn't necessarily the case though, as psychic phenomena like the kind that Daryl Bem searches for1 is most certainly "natural". So this would be a reasonable, yet incorrect argument, but again this is not the argument he is making.


After racking my brains further I just could not think of any other rational possibilities (none that didn't involve complex conspiracies carried out by the publishers of dictionaries to fool the world anyway), and so after enjoying the mental challenge of coming up with arguments in favour of a position I disagreed with, I researched Plantinga's claims in more detail. It turns out that the possibilities that I had imagined, including both the rational and absurd, were not enough to account for his actual claim:
First, note that naturalists are all (or nearly all) materialists about human persons.  A human person is a material object through and through with no immaterial self or soul or subject.  For present purposes therefore, I’ll assimilate materialism to naturalism.  The central premises of the argument are as follows: where ‘N’ is naturalism, ‘E’ is current evolutionary theory, and ‘R’ is the proposition that our cognitive faculties are reliable.  The argument goes like this: 

(1) The probability of R/N&E is low

(2) One who accept N&E concedes that 1 is true as a defeater for R

(3) This defeater can’t be itself defeated

(4) One who has a defeater for R, has a defeater for any belief he takes to be produced by her cognitive faculties including N&E itself

(5) Therefore, N&E is self-defeating and hence can’t be rationally accepted

His conflation of materialism (the idea that everything in the world is matter) with naturalism in the prelude does show some resemblance to my first suggestion above, in that materialism is a metaphysical position and adopting materialism is something scientists would need to support with philosophical and logical arguments rather than scientific ones. This is a false claim though, as even if we were to be generous and accept that the majority of scientists were materialists (or even physicalists), it does not follow that there is an assumption of materialism in the scientific method. Unfortunately, that demonstrably incorrect claim appears to be the strongest assertion that Plantinga makes, as the premises of his argument are just ridiculous.

The first premise is really where the argument collapses in on itself but we'll look at the whole argument. He asserts that the probability of our cognitive abilities being reliable (R), given the truth of naturalism and evolution (N&E) is low. That is, since our mental faculties are a product of evolution, and evolution does not necessarily favour "true" beliefs or accurate cognitive processes, it is possible that our beliefs could be wrong. Using some fancy mathematics, where Plantinga assumes that there is a 50/50 chance of a belief being true, he demonstrates that if we held a hundred beliefs, then the probability of three-quarters or more of them being true would be less than one in a million. So what he concludes is that if we accept that naturalism and evolution are true, and naturalism and evolution produce inaccurate beliefs, then the belief in naturalism and evolution could be flawed therefore naturalism and evolution cannot be rationally accepted given the high probability of them being false.

I bring up the mathematical example because it's hilarious and not because it needs debunking, given the nonsensical attribution of a .5 probability to a belief being true. The rest of his argument is not much better though, as we know that the probability of evolution being true is approaching 1 (i.e. practically true). To avoid the issues with naturalism, we adopt the position of methodological naturalism rather than metaphysical naturalism, so we don't assume it's true at all meaning that it can have no probability value of being true. And whilst it's true that our beliefs about the world can be false (in part due to the processes of evolution), this fallibility is not 50/50. The probability that nearly every biologist in the world are all misreading the data, misrecalling facts, fabricating data, and generally acting under a delusion, is near-zero (i.e. practically false), therefore Plantinga's first premise of P(R/N&E) being low becomes false - the probability instead approaches 1; practically true.


For shits and giggles, I'd like to point out that I wasn't too far off with my second suggestion either, in that Plantinga defines naturalism as: "...for any study of the world to qualify as "scientific," it cannot refer to God's creative activity (or any sort of divine activity)"2. This is an incredible definition of naturalism but of course it is not one that is adopted by science. A phenomenon of the world could be studied by science even if it was created by a god, it would simply require that the god was a natural god using naturalistic processes - like the Greek and Roman gods. What Plantinga means to say is that since the Christian god is not natural, and since science only studies the natural, then it must ignore the possibility of a Christian god. This is similar to the Intelligent Design arguments that try to appeal to some vague 'designer' as an explanation for the development of organisms, but rather than accepting the more realistic possibility of aliens creating life on Earth, they prefer to fall back on to a very specifically defined supernatural creator as defined by the holy texts favoured by their parents.

So is the main assumption of science incompatible with science? As you might have guessed, it's not. Any argument that requires us to assume something as clearly true as evolution to be false must be treated with extreme skepticism. The question of whether science and religion are compatible is an immensely interesting one, and I could not do the topic justice by briefly discussing it in my conclusion here so I'll leave the details to another post, but it is one that can't be handwaved away by trying to distract us with nonsensical claims like the ones Plantinga presents. His argument can be summarised as a childish tu quoque, and is really not much better than an educated philosopher's reiteration of "I know you are, you said you are, but what am I?".


1. Bem, D. J., (2011). "Feeling the future: experimental evidence for anomalous retroactive influences on cognition and affect.". Journal of personality and social psychology, 100 (3): 407–25.

2. Plantinga, A. (1997). "Methodological Naturalism". Philosophical Analysis: Origins and Design, 18:1.


  1. Replies
    1. Hey Larry - thanks for those links. I should have made it clear that my article wasn't intended to be an absolute refutation of all the ways in which Plantinga is wrong as, of course, that would be a lengthy diatribe. Your articles do a great job of comprehensively dissecting the flaws in his position though and they were very well done.

      As for the modal argument, you're absolutely right in your article and the word 'equivocation' rebuts most attempts to invoke it as part of an argument.

    2. I should have made it clear that my article wasn't intended to be an absolute refutation of all the ways in which Plantinga is wrong as, of course, that would be a lengthy diatribe.

      You definitely made that clear. I was just trying to be helpful. :-)

  2. I think you've missed Plantinga's main point. R is not the reliability of evolution being true (he has no qualms with evolution), it is the reliability of our cognitive faculties. It is not at all obvious that evolution would produce beings with true beliefs, let alone accurate philosophical intuitions.

    1. Hi Anon, thanks for the comment.

      Firstly, I don't believe I've missed Plantinga's point at all. I understand that R is the reliability of our cognitive faculties and that's how I've discussed it in my post. Once we account for his conflation of methodological naturalism with metaphysical naturalism (which, as methodological naturalism isn't assumed to be true, cannot be assigned a probability of truth value), and once we reject his simplistic mathematics of assuming that our cognitive faculties have an accuracy of 50%, the entire argument falls apart - compounded by the fact that the truth of evolution necessarily approaches 1.

      I agree that it's not obvious that evolution would produce beings with perfect cognitive faculties and, to some degree, they would certainly be unreliable. However, it is wrong to suggest that they are categorically 'unreliable' (to the level where the rate of our true beliefs reaches a depth of only 50%). More importantly, evolution is of course a scientific fact so Plantinga's argument is self-defeating - if our cognitive faculties are unreliable due to evolutionary processes, then his argument against MN and evolution (derived from his 'unreliable' cognitive faculties) are not guaranteed to be true beliefs. That is, whether god exists or not, we know that evolution occurred so any argument based on the effect of evolution must apply to all arguments; Plantinga does not become immune to the supposed unreliability of his cognitive faculties by claiming he was made by god.

      Finally, I disagree that he has no qualms about evolution. In the article I linked to he dedicates a section to "The Grand Evolutionary Myth" where he brings up typical creationist misunderstandings of science (e.g. criticising science writers for referring to evolution as a fact when he thinks it is 'mere theory', when of course it is both a scientific fact and a theory - there is the observation of evolution occurring, and there is the explanation of how it occurs). In his analysis, his strongest support for evolution is in granting the possibility that god may have used some evolutionary processes as part of his grand scheme but he subsequently discounts this possibility as unlikely.

      In that section he also explicitly states that he 'does not believe in evolution'.

  3. You should really pick up Plantinga's latest book, I think it would clear a lot of this up for you. Plus I think he would drop your jaw with some of his other arguments! His problem is with unguided evolution- that is the critical distinction. Unguided evolution won't produce reliability, but a Perfect Designer intent on relational flourishing would orchestrate the process such that human beings would be equipped with reliable cognitive capacities. He is deeply concerned with his image bearers ability to probe the depths of his creation and uncover its gems of truth. The argument doesn't cut both ways as you seem to want it to. The naturalist needs some impersonal mechanism to explain how evolution would be concerned with truth. That's proven to be tougher to come up with. Do you have any ideas?

    By the way, I really appreciate you distinguishing MN and N, I feel like that is so often overlooked in scientific discussion.

    Unfortunately I think this might be another case of blogger unable to defeat the arguments of a preeminent analytic philosopher. But I appreciate your dialogue on these important questions!

    1. I think he would drop your jaw with some of his other arguments!

      Can you cogently summarize his argument? Everything else I've read of his is complete bollocks, so I'm going to need more than an exclamation point from an anonymous commenter to induce me to find the book at the library.

    2. Hi Anon,

      Thanks for the recommendation but I've read a fair amount of Plantinga and honestly I'm not that interested in going through that again. I just find his argument style to be dishonest and more akin to a used car salesman trying to sell me a dodgy car rather than someone interested in stimulating intelligent discussion.

      I understand that the 'get out of jail free' clause for him is his focus on the idea that a "designer" could magically allow for reliable cognition but it still doesn't help his point. He states that given naturalism and evolution, the probability of our cognitive faculties being reliable is low. We know that MN is a reasonable assumption, and we know that evolution is (for all intents and purposes) true, which means that if his argument is valid, it applies to everyone - including the possibility that we were created by a "designer".

      This is further supported by the fact that we don't simply need to argue that our cognitive faculties would be unreliable given evolution, we know that they are unreliable. Any understanding of the world which assumes reliable cognitive processes (like one that assumes a "designer" could grant us such abilities) is necessarily false.

      Thanks for the kind comments though. Of course, I am just a blogger so it's good to take my word with a grain of salt - however, just have a flick through some philosophy journals and find some papers on the topic. You'll find that practically all professional philosophers reject his EAAN as being irreparably flawed; even many christian philosophers accept that it doesn't make sense.

    3. Larry, you can find out how to be a rational theist without argument! ...jaw drop
      how to solve the evidential problem of evil ...jaw drop
      how miracles and science can coexist peacefully ...jaw drop

      Honestly, I think his best work is defending Christian belief in God as properly basic. Back him up with some Paul Moser and you have yourself an intellectually robust and profoundly life-giving picture of reality!

    4. Anon, do you understand what I mean by summarizing not his conclusions but his arguments?

  4. I'm a tad confused by your last reply. I wasn't sure if I could rely on the deliverances of my visual system when I read the nice self-defeater "we know that they are unreliable". :) But that aside, perhaps we can focus our discussion to the question: Is the EAAN equally problematic for the theist?

    You say MN (Methodological Naturalism) is a reasonable assumption. Clearly this is true for the sake of science. If we are interested in understanding the physical world we need to restrict our inquiry to the observable and replicable. The theist is happy to set aside other sources of knowledge to do carry out careful science. But Plantinga's argument is not about how to do science. He targets Methaphysical Naturalism as a premise for its conclusion. Any problems a worldview founded on naturalism holds do not necessarily apply to the theist. The theist is obviously not at a naturalist, so why is it bothersome that naturalism and unguided evolution don't coexist so peacefully? He doesn't believe either of them!

    I agree that this argument is not very strong, but I think it points to some big unaddressed questions, which I would prefer not see dismissed so quickly.

    1. Hi Anon (maybe you could leave a name or a handle if you continue to comment, it feels quite rude referring to you as "Anon").

      "I wasn't sure if I could rely on the deliverances of my visual system when I read the nice self-defeater "we know that they are unreliable". :)"

      I realise this is a joke but I think it sums up a fundamental problem with Plantinga's position. That is, of course it's obviously true that our cognitive faculties are unreliable. This is why personal recollections aren't admissible in courts of law, why we're prone to a number of visual illusions and cognitive biases, why we are such incredibly poor at understanding why we perform the behaviors we do, etc.

      The problem for Plantinga isn't that this is true - the problem is that this being true does not necessitate the conclusion that anything is unknowable. The fact that I may, for example, fall victim to confirmation bias when assessing the quality of my favourite football team does not mean that I might be mistaken in thinking that my mother's name is Margaret. There are probabilities of accuracy and a relativity of wrong; so whilst it's possible that my belief that the earth is an oblate spheroid may be wrong, it is demonstrably less wrong than the belief that the earth is flat.

      "Any problems a worldview founded on naturalism holds do not necessarily apply to the theist. The theist is obviously not at a naturalist, so why is it bothersome that naturalism and unguided evolution don't coexist so peacefully? He doesn't believe either of them!"

      The problem is that his argument is clearly wrong (even ignoring for now the conflation of MN with N). There is nothing incompatible between naturalism and unguided evolution because, as mentioned above, having unreliable cognitive faculties does not mean that all beliefs are wrong.

      What it does mean is that we need to be skeptical of our beliefs, we need to eliminate confounds, biases, etc, and we need to ensure that the conclusions we reach are consistent with what everyone else agrees on so that we can at least achieve a level of inter-subjective agreement even if absolute objectivity is impossible. These methods (incidentally integral to the process of science) are necessary for counteracting our known cognitive flaws and for reaching a degree of confidence that our beliefs are well-supported enough to justify holding them.

      "I agree that this argument is not very strong, but I think it points to some big unaddressed questions, which I would prefer not see dismissed so quickly."

      Maybe you could expand on some of these problems, and preferably in your own words as, from the brief interaction I've had with you, your ability to construct a logical argument appears to be far superior to that of Plantinga so I'm interested in what you think.

    2. "your ability to construct a logical argument appears to be far superior to that of Plantinga so I'm interested in what you think."

      Way to damn him with faint praise! ;-)