Plantinga’s private language

One of the great things about arguments for gods and the supernatural, is that you can always look back at them and find new problems. Alvin Plantinga’s arguments are especially lovely in this regard. Having been recently been thinking of Wittgenstein’s private language arguments, it occurred to me that somewhere in there is a rebuttal to Plantinga’s evolutionary argument against naturalism.

The evolutionary argument against naturalism argues that if both evolution and naturalism are true, then we cannot trust our own rational faculties, and therefore cannot trust our belief that evolution and naturalism are true.  The reasoning goes that naturalistic evolution does not specifically produce true beliefs, but rather produces adaptive beliefs. An adaptive belief does not need to be true, it just needs to produce adaptive behavior. For example, rather than believing that you should run away from a tiger because it will eat you, you might believe that you should run away from a tiger because that’s the best way to pet the tiger (Plantinga’s example). The number of false beliefs that produce adaptive behavior is much larger than the number of true beliefs that produce adaptive behavior. Therefore, most beliefs are probably false.

There are numerous issues with this argument, a few of which you might be shouting at the screen. From a scientist’s perspective, Plantinga appears to be ignorant of how evolution actually works. Evolution does not necessarily produce the most adaptive traits, certainly not immediately. If you have false but adaptive beliefs at one point in time, it is questionable whether those beliefs would continue to be adaptive when your descendants find themselves in slightly different environments. Also, Plantinga ignores that brain efficiency is an adaptive trait. I would imagine that a brain which produces true beliefs via reasoning is far more efficient than a brain that produces false but adaptive beliefs via some mysterious yet reliable process.

Ah, but Plantinga is a philosopher, and specifically the kind that will ignore practical details in favor of abstraction. Plantinga can just say it’s theoretically possible that we have brains that reliably and efficiently produce false and adaptive beliefs. It’s theoretically possible (and therefore all but certain) that naturalism will produce such brains. Somewhere out there in the multiverse are countless Boltzmann brains that know the best way to respond to tigers in all situations, and yet believe that the ultimate goal of all this behavior is to pet the tiger.

Here is where private language arguments come in. If a brain somehow produces correct behavior reliably and efficiently, on what basis can we say that its beliefs are false? According to the private language argument, the content of a belief or experience is defined by the behavior it produces.

For example, you might point at a strawberry and say that it’s “red”. And suppose you’re learning English for the first time, you look at the strawberry and identify “red” with the color sensation you experience when you see a strawberry. It doesn’t really matter if I instead called it “roja” or “赤”, it’s just a word we use, which is dependent on our language. It also doesn’t matter if your experience of “red” is actually different from my experience–for instance, if your red looks like my green. What matters is behavior, such as the behavior of being able to group items together by color.

So suppose that you had a brain that reliably decides how to behave around a tiger, and believes that the purpose of this behavior is to pet the tiger. And say that you communicate this belief to someone else, saying you’re “trying to pet the tiger”. In this language, “petting the tiger” clearly does not at any point involve literally touching the tiger softly with your hands. Instead it appears to involve avoiding danger from the tiger. So if that’s what “petting the tiger” means, are we so sure that this adaptive belief is false?

I suspect that in practice, if we have false but adaptive beliefs regarding danger avoidance, it would look like a phobia. Why am I afraid of spiders? That’s an experience I can’t express. Maybe the reason I can’t express it is because I am trying to pet the spider, and believe the best way to pet it is to recoil from it and smash it with a paper towel.

Of course, private language arguments are not universally accepted among philosophers, and I suggest regarding them skeptically. However, they certainly add to the problems in Plantinga’s evolutionary argument against naturalism.


This argument was inspired in part by The Barefoot Bum, who in distant memory made a similar argument.


  1. EveryZig says

    There are some types of belief that are frequently false but common, but WHAT kind of belief those tend to be is also an argument against Plantinga. That type of belief is approximations; the processes that lead to things like relative value perception tricks and pareidolia. What those things have in common is that as approximations they give a result that is quick while still being relatively close to the truth. That is because the further a belief is from the truth, the more likely it is to have disadvantageous side effects once actual reasoning gets involved rather than operating on weird rigid scripts like Plantinga seems to assume. What the above would suggest is that a great many of our sensory experiences and beliefs are technically incorrect while being very close approximations of the truth, which is pretty close to a scientific view of theories being very certain but not certain beyond all possibility of being disproved.

  2. polishsalami says

    The obvious counter to the claim that we can’t trust our beliefs because they are the product of evolution is that the Theist’s belief in god is itself a product of evolution, in that it is a belief (obviously) that is held by a human being (a product of evolution).

    The claim by Theists that their ideas were put there by the Good Lord is nothing more than an assertion, made by human beings, who were created by evolution. There’s no escaping this.

  3. Don Cates says

    From “The Philosophical Lexicon”

    alvinize, v. To stimulate protracted discussion by making a bizarre claim. “His contention that natural evil is due to Satanic agency alvinized his listeners.”

    planting, v. To use twentieth-century fertilizer to encourage new shoots from eleventh -century ideas which everyone thought had gone to seed; hence, plantinger, n. one who plantings.

  4. says

    @Barefoot Bum,
    I definitely recall you making a similar argument one time, although I didn’t think I would be able to find it. I’m adding some credit to the post.

    The structure of Plantinga’s argument is similar to a proof by contradiction. He assumes naturalism + evolution to “prove” that our beliefs are unreliable, but Plantinga doesn’t actually believe those assumptions or the conclusion, the assumptions are only taken in order to prove a (near) contradiction. Thus, for the theist, Plantinga’s argument presents no problem–except for the problem that his argument is garbage.

    Although I agree, that evolution does produce some unreliable beliefs, usually based on shortcuts and approximations (similar to what @EveryZig said), and belief in gods is one of those.

  5. drransom says

    This version of the view that the content of a belief just is the behaviors it produces is overly simplistic. For example, it entails that the contents of a belief about the surreal numbers are just a tendency to make certain statements about the surreal numbers, which misses the point of what mathematicians are actually doing. A more appropriately nuanced view would have to do with placing a belief in a network of beliefs and actions (including involuntary responses). So we’d expect the tiger-petters and the tiger-fearers to have differing beliefs, actions, and responses related to danger, petting, other things one pets, and other things one fears, such as:

    * Making statements about the consequences of not running from a tiger that involve missed opportunities, softness, or being eaten.
    * Involuntary physiological responses associated with fear of other things (hormones, elevated heart rate, etc) or not.
    * Different descriptions of what they see if they visualize the consequences of not fleeing the tiger.
    * Production of different tiger-related artworks.

    I can keep going on like this–people who believe they’ll pet a tiger and people who don’t believe this will behave differently in lots of different ways, and that aren’t reducible to any individual statement or action.

  6. says

    I agree that beliefs with different content (tiger-petters vs tiger-fearers) will likely produce different behaviors. However, Plantinga posits that the false-but-adaptive belief produces exactly the same behavior. If you would like to reject Plantinga’s posited scenario as absurd, then I couldn’t fault you for that.

  7. drransom says

    Yes, if the would-be-tiger-petters have the exact same behaviors as tiger-fearers, including talking about their fear of being eaten, having a fight-or-flight response in response to tiger-sightings, and so forth, then I question the coherence of saying that they believe running away from the tiger is the best way to pet the tiger.

  8. Owlmirror says

    I’ve wondered if it’s worthwhile to challenge Plantinga’s scenario by flipping to the tiger’s perspective.

    1) It is reasonable to state that tigers have beliefs. That is, without suggesting that they are as intelligent or rational as humans, tigers have sensory perceptions that they generally rely on, memories of interacting with their environment which they rely on to guide their behaviors and so on.

    2) Thus, tigers will, generally speaking, ignore rocks and bushes and trees, other than as shelter or places to hide and ambush their prey. They will, generally speaking, attempt to pursue and kill prey, fight with or flee from rivals, and seek out mates in the appropriate seasons. And these behaviors arise from their beliefs about what they are interacting with.

    Does Plantinga assert that tigers have adaptive-but-false evolved beliefs, like “believing” that chasing, attacking, and killing prey is “really” just hiding behind them, and that mating is “really” what it means rest sheltered in a bush? This would presumably involve, not exactly a private language, but a private confusion in terms of memories (prior experiences of hiding and hunting and sheltering and mating, and current behaviors of hiding and hunting and sheltering and mating).

    Or would Plantinga assert that tigers do have true beliefs that, as in the case of humans, were directly granted to them supernaturally by the grace of God? Would he concur that God must then be directly responsible for the tiger’s beliefs and behaviors in hunting and killing human beings?

Leave a Reply

Your email address will not be published. Required fields are marked *