Where the Conflict Really Lies: Science, Religion, and Naturalism (42 page)

Returning to the main line of our argument, we are considering nonreductive materialism and asking about P(R/N&E), given nonreductive materialism. (Another way to put this: we are considering P(R/N&E& nonreductive materialism).) In order to avoid automatically introducing into the argument our ordinary assumptions about our own mental life, suppose we conduct a thought experiment. Consider a hypothetical species that is cognitively a lot like us: members of this species hold beliefs, make inferences, change beliefs, and the like. And let us suppose naturalism holds for them; they exist in a world in which there is no such person as God or anything like God.
Our question, then, is this: what is the probability that their cognitive faculties are reliable? Consider any particular belief on the part of one of these hypothetical creatures. That belief is a neural structure of a given sort, and one sufficiently complex to generate content. We may add, if we like, that this structure occurs or takes place in response to something in the environment; perhaps it is a certain pattern of firing of neurons in the optical portion of the brain, and perhaps this pattern arises in response to the appearance of a predator. Suppose further that a certain content, a certain proposition, is determined by the NP properties of this structure. This structure, therefore, will be a belief, and will have a certain proposition p as its content.

But now for the crucial question: what is the probability (given N&E) that this proposition is true? Well, what we know about the belief in question is that it is a neurological structure that has certain NP properties, properties the possession of which is sufficient for the possession of that particular content. We are assuming also that this structure arises in response to the presence of that predator. We can assume further, if we like, that this structure is a reliable indicator of that kind of predator: it arises when and only when there is a such a predator in the middle distance. But why think it is a
true
proposition that is determined by those NP properties? These NP properties determine a proposition: but why think that proposition is true? Natural selection selects for adaptive NP properties; those NP properties determine content; but natural selection just has to take pot-luck with respect to the propositions or content determined by those adaptive NP properties. It does not get to influence or modify the function from NP properties to content properties: that’s just a matter of logic or causal law, and natural selection can’t modify either. Indeed, the content generated by the NP properties of this structure, on this occasion, need have nothing to do with that predator, or with anything else in the environment. True: the structure is correlated with the presence of a predator and indicates that presence; but
indication is not belief. Indication is one thing; belief content is something else altogether, and we know of no reason (given materialism) why the one should follow the other. We know of no reason why the content of a belief should match what that belief (together, perhaps, with other structures) indicates. Content simply arises upon the appearance of neural structures of sufficient complexity; there is no reason why that content need be related to what the structures indicates, if anything. Indeed, the proposition constituting that content need not be so much as
about
that predator; it certainly need not be true.

What, then, is the likelihood that this proposition, this content, is true? Given just this much, shouldn’t we suppose that the proposition in question is as likely to be false as true? Here’s the picture: the NP properties of a belief are adaptive in that they cause adaptive behavior. Those NP properties also determine a content property. But as long as the NP properties are adaptive, it doesn’t matter, for survival and reproduction, what content is determined by those NP properties. It could be true content; it could be false content; it doesn’t matter. The fact that these creatures have survived and evolved, that their cognitive equipment was good enough to enable their ancestors to survive and reproduce—that fact would tell us nothing at all about the
truth
of their beliefs or the reliability of their cognitive faculties. It would tell something about the
neurophysiological
properties of a given belief; it would tell us that by virtue of these properties, that belief has played a role in the production of adaptive behavior. But it would tell us nothing about the truth of the
content
of that belief: its content might be true, but might with equal probability be false. So shouldn’t we suppose that the proposition in question has a probability of roughly .5? Shouldn’t we estimate its probability, on the condition in question, as in the neighborhood of .5? That would be the sensible course. Neither seems more probable than the other; hence we should estimate the probability of its being true as .5.

The probability we are thinking of here is objective, not the personalist’s subjective probability, and also not epistemic probability.
²⁴
(Of course there will be a connection between objective and epistemic probability, perhaps a connection in the neighborhood of Miller’s principle; presumably epistemic probability will in some way follow known objective probability.) But then, in suggesting the first attitude above, am I not relying upon the notorious Principle of Indifference? And hasn’t that principle been discredited?
²⁵
Not really. The Bertrand paradoxes show that certain incautious statements of the principle of indifference come to grief—just as Goodman’s grue/ bleen paradoxes show that incautious statements of a principle governing the projection of predicates or properties comes to grief. Still, the fact is we project properties all the time, and do so perfectly sensibly. And the fact is we also regularly employ a principle of indifference in ordinary reasoning, and do so quite properly. We also use it in science—for example in statistical mechanics.
²⁶

Given that the probability, for any belief on the part of these creatures, is about.5, what is the probability that their cognitive faculties are
reliable
? Well, what proportion of my beliefs must be true, if my faculties are reliable? The answer will have to be vague; perhaps a modest requirement would be that a reliable cognitive faculty must deliver at least 3 times as many true beliefs as false: the proportion of true beliefs
in its output is at least three-quarters. If so, then the probability that their faculties produce the preponderance of true beliefs over false required by reliability is very small indeed. If I have one thousand independent beliefs, for example, the probability (under these conditions) that three quarters or more of these beliefs are true will be less than 10
^–58
.
²⁷
And even if I am running a modest epistemic establishment of only one hundred beliefs, the probability that three-quarters of them are true, given that the probability of any one’s being true is one half, is very low, something like .000001. So the chances that this creature’s true beliefs substantially outnumber its false beliefs are small. The conclusion to be drawn is that it is very unlikely that the cognitive faculties of those creatures are reliable. But of course the same will go for us: P(R/N&E) specified not to them but to us, will also be very low.

That’s how things stand for nonreductive materialism: P(R/ N&E&nonreductive materialism) is low. We can deal more briefly with P(R/N&E&reductive materialism), the probability of R given naturalism and evolution and reductive materialism. On reductive materialism, mental properties are complex combinations of physical properties; more briefly, taking complex combinations of physical properties to be themselves physical properties, mental properties just are physical properties. What is the probability of R on N&E and reductive materialism?

Here we get the very same results as with nonreductive materialism. To see why, consider, again, any given belief on the part of a
member of that hypothetical group of creatures—say the belief
naturalism is vastly overrated
. That belief is a neuronal event, a congeries of neurons connected in complex ways and firing away in the fashion neurons are wont to do. This neuronal event displays a lot of NP properties. Again, we may suppose that it is adaptively useful for a creature of the kind in question to harbor neuronal structures of the sort in question in the circumstances in question. The event’s having the NP properties it does have is fitness-enhancing in that by virtue of having these properties, the organism is caused to perform adaptively useful action—fleeing, for example. Since the event is a belief, some subset of these NP properties together constitute its having the content it does in fact display. That is, there will be some proposition that is the content of the belief; the belief will therefore have the property of having that proposition as its content; and that property, the property of having such and such a proposition as its content, will be a (no doubt complex) NP property of the belief.

Now what is the probability that this content is
true
? What is the probability that this proposition, whatever it is, is true? The answer is the same as in the case we’ve already considered. The content doesn’t have to be true, of course, for the neuronal structure to cause the appropriate kind of behavior. It just happens that this particular adaptive arrangement of NP properties also constitutes having that particular content. But again: it would be a piece of serendipity if this content, this proposition, were
true
; it could just as well be false. These NP properties, including those that constitute its having that content, are adaptive just as long as they cause adaptive behavior. They also constitute the property of having that particular content; but it doesn’t matter at all, so far as adaptivity goes, whether that content is true. So take any particular belief on the part of one of those creatures. We may suppose (given that these creatures have come to be by way of evolution) that having this belief is adaptive; its NP properties cause adaptive behavior. These NP properties also constitute the property of having such and
such content; but, clearly enough, it doesn’t matter (with respect to the adaptivity of these properties) whether the content they constitute is true. It could be true: fair enough; but it could equally well be false. If these properties had constituted different content, they still would have had the same causal effect with respect to behavior. Hence the probability that the content of this belief is true would have to be rated at about one-half, just as in the case of nonreductive materialism. If this is true for each of the independent beliefs of the organism in question, however, the probability that the cognitive faculties of these creatures are reliable (on N&E& reductive materialism), would have to be rated as low. The conclusion to be drawn so far, then, is that given N&E (N including materialism), it is unlikely that these creatures have reliable cognitive faculties.

Isn’t it just obvious that true beliefs will facilitate adaptive action? A gazelle who mistakenly believes that lions are friendly, overgrown house cats won’t be long for this world. The same goes for a rock climber who believes that jumping from a two-hundred-foot cliff will result in a pleasant and leisurely trip down with a soft landing. Isn’t it obvious both that true beliefs are much more likely to be adaptive than false beliefs? Isn’t it obvious, more generally, that true beliefs are more likely to be successful than false beliefs? I want to go from New York to Boston: won’t I be more likely to get there if I believe that Boston is north of New York than if I believe it’s to the south?

Yes, certainly. This is indeed true. But it is also irrelevant. We are not asking about how things
are
, but about
what things would be like if both evolution and naturalism (construed as including materialism) were true
. We are asking about P(R/N&E), not about P(R/the way things actually are). Like everyone else, I believe that our cognitive faculties are for the most part reliable, and that true beliefs are more likely to issue in
successful action than false. But that’s not the question. The question is what things would be like if N&E were true; and in this context we can’t just assume, of course, that if N&E, N including materialism, were true, then things would still be the way they are. That is, we can’t assume that if materialism were true, it would still be the case that true beliefs are more likely to cause successful action than false beliefs. And in fact, if materialism were true, it would be unlikely that true beliefs mostly cause successful action and false belief unsuccessful action.

Here you may ask, “Why think a thing like that? What has materialism to do with this question?” Here’s what. We ordinarily think true belief leads to successful action because we also think that beliefs cause (part-cause) actions, and do so
by virtue of their content
. I want a beer; I believe there is one in the fridge, and this belief is a (part) cause of my going over to the fridge. We think it is by virtue of the
content
of that belief that it causes me to go over to the fridge; it is because this belief has as content that there is a beer in the fridge that it causes me to go to the fridge rather than, say, the washing machine. More generally, we think it is by virtue of the content of a belief B that B part-causes the behavior that it does cause.

But now suppose materialism were true: then, as we’ve seen, my belief will be a neural structure that has both NP properties and also a propositional content. It is by virtue of the NP properties, however, not the content, that the belief causes what it does cause. It is by virtue of
those
properties that the belief causes neural impulses to travel down the relevant efferent nerves to the relevant muscles, causing them to contract, and thus causing behavior. It isn’t by virtue of the content of this belief; the content of the belief is irrelevant to the causal power of the belief with respect to behavior.