G.E. Moore famously laid out a straightforward proof for the skeptic, namely, that he has hands. Moore’s response to skepticism maintains that we can know of everyday propositions and hence know the denials of skeptical hypotheses. It appears that Moore looks beyond Descartes’ skepticism and does not explain how he knows he has hands in the first place. In this paper, I will reconstruct Moore’s proof using Nozick’s Tracking Theory in a way that will address Descartes.
I shall first begin by discussing what Descartes wants to be proven when it comes to external things in the world. In First Meditations, Descartes gave us reason to doubt there are external things in the world: our senses sometimes deceive us and we often dream of things that happen when we are awake. Because of this, we need to first be able to distinguish which mental state we are in before we can say there are external things in the world, such as hands.
With this in mind, we can see that Moore’s proof does not address Descartes’ skepticism and even seems dismissive towards Descartes’ points. There are two main weaknesses in Moore’s argument. The first weakness is that Moore does not explain how he knows he has hands. For all he knows, there may be a hologram-projecting machine that follows him everywhere so that every time Moore looks down on his arms, he sees two hands that, unbeknownst to him, are projected hologram hands. Even if we doubt this odd occurrence and given that Moore does have hands, the premises being true do not imply that Moore knows them. The second weakness is that Moore’s argument is circular by nature. In order to know that he has hands, Moore must first know there are external things. However, in his proof, Moore concludes that he can know external things exist as a consequent of his having hands. The skeptic’s objection is that he cannot know he has hands unless he first proves that there is an external world because the premise depends on this. These two weaknesses in Moore’s argument show that he does not adequately address Descartes’ skeptical points; Moore does not know he has hands nor has he proven he does.
Yet, Moore claims that his proof of an external world is a rigorous one. After all, Moore writes, “we all of us do constantly take proofs of this sort as absolutely conclusive proofs of certain conditions” (Moore 27). Proofs of “this sort” are frequently made for everyday propositions. To illustrate further with an example, Moore pointed out that when we prove there are misprints on a page, we simply take our fingers and point at them. If we take this kind of demonstration as sufficient proof, then surely, Moore’s proof that he has hands should be sufficient. Given a hypothetical situation where we were to agree with what Moore proposes to be a rigorous proof, we can infer that Moore is not dreaming when he is making his hand gestures nor is he dreaming when he points out the misprints. Given that we know many everyday propositions, we can deny skeptical hypotheses, such as us not knowing whether we are dreaming right now. Moore’s proof takes the following form, where ‘P’ is any everyday proposition that we say we know, and ‘¬SP’ stands for the denial of a skeptical hypothesis that is inconsistent with our everyday propositions:
b) K(p à ¬SP)
Moore’s proof tells us that if we know the truth of everyday propositions, then we should also know that skeptical possibilities are false.
I will now turn to Nozick’s Tracking Theory of Knowledge and consider Moore’s argument from the perspective of this theory. Nozick proposed that in order for an agent to know that P, in addition to P being true and the agent believing that P, the agent’s belief in proposition P must be sensitive to the truth of P. That is, in the nearest possible world where P is false, the agent will not believe that P. Likewise, in the nearest possible world where P is true, the agent will believe that P.
Using Nozick’s Tracking Theory, we can see that being sensitive to the truth helps us keep tabs on our everyday propositions. Our everyday propositions are based on possible occurrences close to our actual world. We know skeptical hypotheses to go against our intuitions and are highly improbable. Going along this point, we can say that a skeptical world is probably unlikely and very far out in our universe of possible worlds. The beliefs we currently hold about everyday propositions are true in close possible worlds and safe from the world of the skeptic. If we are able to know everyday propositions, then we are able to rule out radical skeptical hypotheses as being true. As Nozick wrote, “Whether or not Q is true in P worlds that are still farther away from the actual world [e.g., skeptical worlds] is irrelevant to the truth of the subjunctive” (Nozick 256). As long as we are able to track our everyday propositions in our actual world and other close possible words, we can say that we know them.
Let’s now consider Moore’s proof of external things by reformatting it to one that would fit Nozick’s Tracking Theory. Our proposition P is “I have two hands.” We have an agent named Bob. P is certainly true. Bob believes that P. In the nearest possible world, if P were false (Bob lost his hands in a firework accident), would our agent believe that ¬P (Bob does not have hands)? The answer is yes, Bob would believe that he does not have hands because he knows this to be a fact. He experienced the accident and he knows that he did not undergo surgery for prosthetic hands. Now let’s consider, in the nearest possible world, if P were true, would our agent believe that P is true? The answer is yes, he would take this statement to be true. He has not experienced anything traumatic that cost him his hands. As such, we can say that our agent, Bob, is sensitive to the truth and has tracked his knowledge in close possible worlds.
When Moore raised his two hands up, he could be certain that he has hands because he has no reason not to think he has hands; he is sensitive to the truth in his proposition. Not only does Moore hold this belief to be true, he holds this true across a range of close-by possible worlds. Even when confronted with a skeptical hypothesis, e.g., “We do not know whether we are dreaming or not,” Moore was able to hold onto this proposition; he still knows that he has hands despite others telling him the contrary. The skeptic’s objection seems so far-fetched that this objection does not influence his knowledge of everyday propositions, such as one as obvious as having hands.
All in all, Moore’s proof of an external world may be better received by skeptics if it was to follow the format of Nozick’s Tracking Theory. With his theory, Moore would be able to test the falsity and truthfulness of a proposition and determine whether he was tracking this proposition as knowledge. This theory also helps explain how Moore is able to bypass Descartes’ skepticism – the skeptical world does not influence the belief one takes in a proposition because it is too far out in from actuality in the realm of possible worlds.