“It’s raining, but I don’t believe it.”
That sentence – called “Moore’s problem,” after its creator G.E. Moore – is interesting because it is a case of something which can be true and yet is impossible to believe. It’s perfectly possible for it to be raining and for me not to believe that it is raining. In fact, it’s happened many times. Yet, I can never believe it as it is happening.
Moore’s problem is not a very serious one. Fully accounting for it is a tricky logical exercise, but it is obvious that there is no real problem. You can’t both believe that it’s raining and believe that you don’t believe that it’s raining. Beliefs just don’t work that way.
Moore’s problem points to a broader issue, however: philosophical positions that can be true but cannot be believed.
Consider Freudianism: the belief that it is our primal, unconscious desires that govern our rationality. If this is, in fact, true, then whatever reasoning we used in order to reach the conclusion of Freudianism must have been an expression of our primal, unconscious desires. This means we have no reason to trust that reasoning in the first place. Our belief in Freudianism undercuts itself.
Or Marxism: the belief that class superstructures govern our rationality. If this is true, then whatever reasoning we use to come to Marxism is an expression of whatever class superstructures we are governed by. We have no reason to think that reasoning reliable. As with Freudianism, belief in Marxism becomes self-defeating.
Generally: any belief that entails a skepticism strong enough to reach the mechanisms by which we formed the belief is an epistemically self-defeating belief. Jim Slagle calls any argument that uses this fact to disprove a philosophical position an “epistemological skyhook” (ES). For example, if I were to add to the above paragraph about Freudianism, “therefore Freudianism is false,” I would have used an ES against Freudianism.
What’s fascinating about the ES is that it only shows that a position is impossible to believe. But, if a position is impossible to believe, do we have to call it false? As we saw with Moore’s problem, things that are impossible to believe are sometimes true. In the case of the ES, however, accepting the position entails a skepticism so strong that “true” and “false” would be rendered meaningless. If we accept such a strong skepticism, we have no reason to think we know anything, including the fact of skepticism. Let’s consider a couple more examples.
What follows is a simplified version of Alvin Plantinga’s ES against evolutionary naturalism. Evolutionary naturalism is the view that 1) the material is all that exists (i.e., there are no gods or supernatural entities); and 2) all human faculties have developed through the evolutionary process.
Evolution selects for advantageous behavior – not for truth-finding. So, if human faculties were formed by the evolutionary process, we should expect rationality to be much better suited to self-serving behavior than to truth-finding. We ought to, therefore, doubt our rationality’s ability to discover truth, including the truth of evolution. Evolution is therefore epistemically self-defeating unless we consider the possibility that God is guiding it in a way that selects for truth-finding. So, evolutionary naturalism is false.
There’s an ES against determinism. If our mental states are entirely governed by material causes (or, in any case, causes over which we have no control), then we have no reason to think that they are governed by processes that yield reliable rationality. But then we have no reason to trust the rationality that led us to determinism in the first place.
An ES can be used against many positions. Some forms of behaviorism are vulnerable to it. Habermas uses an ES to argue against Adorno’s radical critique of rationality. Heidegger’s conception of Being and Derrida’s notion of Différance both arguably fall victim to it. It should go without saying that the same goes for any form of radical skepticism.
What do we do when our best evidence and reasoning point us to a position that is vulnerable to an ES? Let’s take evolutionary naturalism. It’s perfectly possible for evolutionary naturalism to be true. Indeed, a lot of evidence points in that direction. That would seem to mean, then, that we can’t trust our ability to discover truth, including the truth of evolutionary naturalism.
Can we bite that bullet? If we do, we have to accept that we don’t know anything at all – that we’re just making noises and scratches. If that’s the case, we don’t believe or disbelieve anything, including the fact that we don’t. Biting the ES bullet seems to entail abandoning philosophy and truth-seeking altogether. That’s a really rough bullet to bite.
So do we accept the ES-vulnerable positions as false? That’s certainly the more convenient attitude. But it is unwarranted – nothing about ES necessarily suggests that they’re false. If we take that route, we do so because it is more comforting to think that our rationality is reliable than to think that it isn’t.