Plantinga's Evolutionary Argument Against Naturalism, revisited

I thought of another objection to Plantinga's "Evolutionary Argument Against Naturalism."  If you'll recall, I blogged on this before.

Plantinga's argument hingest on the fact that for any kind of adaptive behavior you can think of, you can think of a combination of a false belief and a desire that will result in that behavior.  Since false beliefs can result in adaptive behavior just as well as true beliefs, and since natural selection is determined by beliefs only insofar as those beliefs affect behavior, then the probability that evolution would result in reliable belief-producing cognitive faculties is low or inscrutible, provided that naturalism is true (in which case there are no gods who preordained or directed with the evolutionary process).

One of the primary objections people raise against Plantinga's argument is that while false beliefs can result in adaptive behavior just as well as true beliefs, you cannot have systematically false beliefs that will result in adaptive behavior.  My objection is similar but with more detail.

All of the beliefs and desires we might imagine that could produce adaptive behavior are derived beliefs.  That is, they are inferred from observation, experience, laws of inference, and foundational a priori beliefs.  It does not seem possible that all of our behavior-inducing beliefs could've been built into our brains from the get go and still be adaptive since the environment is in a constant state of change.  So the only way beliefs and desires could result in adaptive behavior is if they are derived (at least in part) from our interaction with our environment through some process of reasoning.  The only kind of knowledge that could be built into a brain to enable it to form beliefs are a priori knowledge, including some kind of laws of inference.

Right now, it seems like we are able to learn true information about the world because our built in beliefs about the laws of inference are true.  The laws of inductive and deductive reasoning are reliable guides to truth when combined with true premises and reliable observations.

But there are only a couple of ways things could have been different.  One way is that we could've been born without having any built in knowledge of laws of inference.  Another way is that we could've been born with completely different laws of inference.

If we were built without any laws of inference, then there would be no way for the brain to produce consistent beliefs.  One day, I may see a tiger and think tigers are dangerous.  Another day, I may see a tiger and think tigers are safe.  All of my beliefs would be random, arbitrary, and in a constant state of change.  So there is no way that I could survive.  I'd be weeded out of the gene pool.  I couldn't have consistently adaptive beliefs because I wouldn't have consistent beliefs at all.

If we were built with different laws of inference, then the types of beliefs we came to might be consistent, but they couldn't be consistently adaptive.  The reason is because the only way to come up with combinations of desires and beliefs that are adaptive is to be random.  It's easy to think of random combinations of beliefs and desires that are adaptive for any situation. But it would be a great coincidence if, given some fixed laws of fallacious inference, the beliefs we came to usually turned out to work in our favor even though they were false

It is understandable why sound laws of inference would tend to produce beliefs that are adaptive.  It's because, in general, true beliefs are adaptive.  But false beliefs are not adaptive in general.  So if we were built with fallacious laws of inferences, they could not consistently produce adaptive beliefs, even if some of the wrong beliefs were adaptive.

Since almost all true beliefs are adaptive, and since most false beliefs are not adaptive, it stands to reason that natural selection would be more likely to produce minds with built in valid laws of inference rather than invalid laws of inference.

Plantinga's witch illustration doesn't address this objection.  The witch illustration is meant to be an example of a systematically false beliefs that are nevertheless adaptive.

**Edit: Jan 13, 2013**

I came up with this today, posted it on facebook, then decided to add it here:

I was just thinking about how all animals seem to have a survival instinct.  But what if, instead, we all had a suicidal instinct?  You'd think our species would become extinct.  We could survive if there were a variation in our species in which a few people here and there had a survival instinct.  Over several generations, that instinct would become dominant, and very few people would want to commit suicide.

But I was just thinking that's not necessarily so.  What if, in addition to having a suicidal instict, we also had an innate belief that (1) the best way to commit suicide is to eat poisonous food, and (2) all the food that we would ordinarily think is good for us is actually poisonous.  In that case, we would instinctively be drawn to food that is good for us, thinking it was poisonous and hoping to die by eating it.  But instead of dying, we'd survive and flourish.  So it seems possible that we could be born with unreliable belief-producing falculties, and still survive.  Natural selection wouldn't necessarily select for true beliefs or reliable brains.

Of course, you might think if that were the case that people would be able to learn from experience that their innate beliefs were false.  But we could only learn that if we were also built with the innate belief that past experience can tell us what to expect in the future.  As David Hume showed in An Inquiry Concerning Human Understanding, it's not possible to prove that supposition without engaging in circular reasoning.  So maybe if we were born without that belief, then our past experience of always remaining healthy after eating "poisonous" food would not be able to tell us that the food wasn't poisonous after all.   So it still seems possible that natural selection might result in us having unreliable belief-producing cognitive faculties.

But I question whether it would be possible to survive without the belief that the future will resemble the past.  I'm having a hard time coming up with a hypothetical scenario in which people are unable to learn from past experience, yet still survive and thrive.  Any ideas?