The reliability of perception

by Neil Rickert

In a recent post about Plantinga’s argument against naturalism, John Wilkins quotes Plantinga as arguing:

If our cognitive faculties have originated as Dawkins thinks, then their ultimate purpose or function (if they have a purpose or function) will be something like survival (of individual, species, gene, or genotype); but then it seems initially doubtful that among their functions—ultimate, proximate, or otherwise—would be the production of true beliefs.

John Wilkins seems to think that this is an objection that deserves a response, arguing that selection for fitness will provide a perception that generates true beliefs.

I disagree.  Plantinga is quite right.  There is no basis for expecting that perception of an evolved organism will produce true beliefs.  However, that’s a rather  hollow “victory” for Plantinga.  For there is also no basis for expecting that perception will produce false beliefs.  Quite simply, truth or falsity is not a criterion for perception.  As Al Gore might have put it, there is no controlling authority.

This is one of the points I was trying to make in my earlier posts on truth – see, for example here and here.

There is, however, a different issue.  The evolved agent may use what he/she perceives to generate a description, and that description could be expressed as a verbal or written belief of the agent.  There is a controlling agent for descriptions, namely the language community to which the agent belongs.  And learning how to form such descriptions in a way that meets the truth requirements of that language community is part of the language acquisition that a child is expected to accomplish.

3 Comments to “The reliability of perception”

  1. Neil, I have made a slightly more subtle point than that: I think some observations are going to produce true beliefs because that is the best explanation for why they raise fitness. Others aren’t. Those that are I call environmental or ecological beliefs, which must be true if they are to contribute to fitness. Those that aren’t I call social beliefs, or side effects of psychology, depending on the etiology of that belief. There are many beliefs we acquire that do not track truth, but those that do do so because they are explicable only if truth is being tracked.


    • I gave a simplified version for brevity. If you think your “more subtle point” invalidates my post, then you failed to understand my post. I am saying that the whole idea that perception is answerable to a truth requirement is a misguided one.

      If your argument about fitness were correct, then wouldn’t it imply that there should be no near-sightedness, no far-sightedness, no astigmatism, no color blindness, no amblyopia? The fitness requirements should remove these problems from the population if it were as effective as you suggest. Yet these problems are quite common, and not only in people past their reproductive years.

      If truth is correspondence to reality, then the perceptual system sets up its own correspondence, and is therefore its own standard of truth. If it is imperfect, that does not impact the general truth of observations. The color blind person will still make mostly true observations, but with poorer ability at discriminating details. There is no standard of truth that a perceptual system must or could follow, other than its own.


  2. While I agree that beliefs can be true or false, I do not see how that impinges on naturalism, the axion to explain things without reference to supernatural causes. How, then, does Platinga jump to the conclusion that naturalism is at issue? (I know, I should have carefully read all the posts on the issue first).


%d bloggers like this: