G. E. Moore’s “Four Forms of Scepticism” (selection)

[I got this paper from an epistemology anthology, and it appears to be a selection rather than the whole thing.]

The evil demon argument says that it is a logical possibility that a malicious demon exists and is causing us to perceive objects, and therefore we cannot trust our sense perceptions. Since we cannot be certain that such a demon doesn’t exist, we cannot know anything that we receive through our senses. Russell’s claim isn’t that the demon is a physical possibility, nor is he claiming that the demon actually exists. Instead, his claim is that the evil demon is a logical possibility. What does he mean by this?

Moore thinks that understanding this will show us the difference between Russell and him.

The three questions that have to be answered are: (1) What does it mean that this is a logical possibility? (2) Is it true that it is logically possible? (3) If it is true, then is it true that I cannot be certain that a perception isn’t caused by an evil demon?

Moore can think of three things that might be meant by saying that such a demon is a logical possibility. First, you might say that the idea doesn’t involve a logical contradiction. But this isn’t what Russell means, because even if the idea of such a demon doesn’t involve a logical contradiction, then I can still know it to be false.

Second, Russell might mean that the idea of an evil demon doesn’t contradict what I already know. But that is begging the question, because the very thing I’m claiming to know is that my perception is caused by the external world, not the malicious demon, and surely that contradicts the idea of the malicious demon.

Third, and this is most likely what Russell means, is that the idea of an evil demon doesn’t contradict anything I know *immediately* or anything that is derived from immediate knowledge. Moore admits that he doesn’t *immediately* know that, for example, his perception of a pencil isn’t caused by the evil demon, that a certain person is conscious, or anything else logically incompatible with being caused by a evil demon. (And Moore says that some philosophers will disagree with him and Russell, and assert that one can know these things immediately.) But, where he does differ from Russell is that Moore believes that even if we don’t know something immediately  we can still be certain of it. “Where is Russell’s argument against this?” Moore asks.

To the best of Moore’s knowledge, Russell bases the idea that I cannot be certain of what I don’t know immeidately or of what I derive from immediate knowledge upon two ideas: (1) that what cannot be known immediately or derived from immediate knowledge must be based upon analogical or inductive arguments; and (2) we cannot be certain of what we know through analogical or inductive arguments. Moore thinks the first is true, but he doesn’t believe the second assumption. He thinks we can be certain of things known through analogical or inductive arguments.

But Moore really wants to ask Russell the following: Russell’s idea that we cannot be certain of beliefs such as “The other person is conscious” or “I hold a pencil in my hand” is based upon four assumptions: (1) That I don’t know such things immediately; (2) that I don’t deduce knowledge of them from things I know immediately; if both of these are true, then (3) I must know these things from analogical or inductive arguments; and (4) we cannot be certain of what we know from analogical or inductive arguments. So, Moore wants to ask Russell, are you more certain of any of these assumptions than you are that you have a pencil in your hand or that a certain person is conscious?

Moore is more certain that he is holding a pencil in his hand than he is of any of these assumptions, much less all of them, even though he agrees that assumptions 1-3 are true.

Moore believes that it is not rational to be more certain of all of those assumptions than you are that you hold a pencil in your hand or that another person is conscious.

 

5 Key Points

1. Russell believes that you cannot be certain of things that are based upon analogical or inductive arguments.
2. Moore thinks you can be certain of things based upon analogical or inductive arguments.
3. Russell and Moore think that things you don’t immediately know or base upon immediate knowledge can only be known through analogical or inductive arguments.
4. For an argument to be objectively convincing, then you must be more certain of all the premises and assumptions than you are of the negation of the conclusion.
5. Moore thinks memory and much of one’s knowledge about oneself is immediate knowledge.

G. E. Moore’s “Proof of an External World”

Moore begins by saying that, even though Kant claimed that there could be only one possible proof of the external world (the one Kant gave), to Moore it seems that many perfectly rigorous arguments can be given. Suppose he holds up his right hand and says, “Here is one hand”; and then he holds up his other hand and says, “Here is another hand.” To Moore, this is a perfectly rigorous proof of the proposition “There now exists two hands.”

Three things are necessary for a proof to be considered rigorous. (1) The premises must be known; (2) the conclusion must be different than the premise(s); and (3) if the premises are true, then the conclusion must be true. 

Moore says that the first requirement is met, because certainly he knows, when he is holding up one hand and then another, that “here is one hand” and “here is another”. He knows this as well as he knows that he is standing there. Surely there is nothing more certain than that he is holding out his hands.

The second requirement is met, according to Moore, although he recognizes that there might be some people who would think the conclusion “There exists a hand” to be no different than the premise “Here is a hand”.  Moore claims that the conclusion is different from the premise because the conclusion could be true even if the premise is false. So, the premise and the conclusion are not identical.

The third requirement is also true, because if the premise “Here is a hand” is true, then surely the conclusion “A hand exists” is true.

He points out that, if this argument is a perfectly rigorous one, as he thinks it is, then it should be obvious that many more can be given.

He says that his argument is the kind that we would accept as a perfectly good argument in everyday circumstances. If someone were to question whether there were three typos on a certain page in a book, it would be a perfectly acceptable proof to open the book and say “Here’s one typo, here’s another, and here’s the third.” That’s be acceptable proof.

Something people are interested when they want to prove the external world is to not only prove that external objects exist now, but that they existed in the past as well. But, Moore claims, many proofs can be given of things existing in the past. The proofs will resemble the proofs of things existing now, but they will also have important differences. One such proof could be: You remember that a few minutes ago I held up one hand, and then the other, therefore, two hands existed in the past.

But Moore is perfectly aware that many philosophers will not find his proof convincing. They want something more than this, but it can be difficult to know exactly what else they want. One things that they probably desire is a proof of the premises of Moore’s argument. But, Moore says, he won’t give one, nor does he think one can be given. To prove that Moore really held up his hands would require, as Descartes pointed out, that Moore prove he isn’t dreaming. Moore doesn’t think he can do that. But he can know, he can have conclusive evidence that he is holding up his hands, without being able to prove it. That’s all he needs. 

But another reason they would be dissatisfied with Moore’s proof is that, not only would they want a proof of Moore’s premises, but they think that if he can’t prove his premises his proof isn’t conclusive (so either they think that a rigorous proof requires that the premises be proven or, at least, provable; or they think that to know something means that you can prove it. –MAC). Moore, however, rejects this idea, even though it is common among philosophers. People claim that if you can’t prove something, you can only accept it on faith. But Moore says this isn’t so. You can know something you can’t prove. So, the dissatisfaction of these people is wrong.

 

KEY POINTS

1. According to Moore, to prove that he was holding up his hands would require him, as Descartes pointed out, to prove (not just know) that he isn’t dreaming, which Moore doesn’t think he can do.
2. Moore thinks you can know things you can’t prove.
3. It has been a common view among philosophers, however, that you can’t know things you can’t prove–you can only take them on faith.
4. A test to see if a conclusion is different from a premise (or, if two propositions are different) is to  see if one could be true while the other false.
5. To say that something might be proven to exist (or be true) is to say that that thing might be certain to exist or be true.

Barry Stroud’s “The Problem of the External World”

[Originally published as ch. 1 in The Significance of Philosophical Skepticism]

In this chapter, Stroud examines the Cartesian argument for skepticism about the external world. Reflecting on our knowledge is a common, everyday task, e.g. when we examine if what we know about the common cold can be true. Descartes did this common, everyday task, except he did so over all his knowledge. He did this by finding a common source to most of his beliefs: the senses. He found the best possible case that one could be in to get knowledge through the senses: sitting in front of his fire, paper in hand. If this case is truly representative of how we get knowledge (or, more accurately, how we think we get knowledge) of the external world, then Descartes’ findings about this case will be representative of sense-knowledge. He found that he couldn’t distinguish reality from dreaming, and you don’t *know* things on account of your dreams; so he couldn’t *know* that he was sitting by the fire. 

Descartes assumes: (1) you can’t know something on account of your dreams; (2) to know X through your senses, you’d have to know you weren’t dreaming when you thought you were learning X through your senses; and (3) you can’t distinguish dreaming from being awake.

The first assumption is widely regarded as correct; the way we use “know” in everyday life excludes us from knowing something on account of our dreams. The second assumption is more difficult to discuss. To know X, we don’t have to know the truth of every proposition that must be true if X is true; if this was true, we’d have to know an indeterminate number of propositions. But we do have to know the falsity of some things that contradict what we know, and we have to know the falsity of things that undermine our evidence for believing what we believe. But it’s not clear how much we have to know in order to know X. Descartes idea that we have to know we aren’t dreaming to know X has been intuitively forceful throughout history, however, so it’s at least plausible. Thirdly, if you accept that you must know that you aren’t dreaming to know anything, then you can never have knowledge of anything. Because, how can you know the evidence/information you use as the basis of your belief that you aren’t dreaming is reliable, if you can’t *know* anything unless you know you aren’t dreaming? This quickly turns into an infinite regress, because to use evidence to prove the reliability of the evidence you are using to prove you aren’t dreaming would mean that you’d have to know that this secondary information is reliable. And if you aren’t certain about the evidence/information you use to prove you aren’t dreaming, how can you be certain you aren’t dreaming?

So, prima facie, Descartes’ arguments are compelling. We are left behind a veil that separates us from reality. We have lost knowledge of chairs, tables, rocks, etc. We’ve even lost knowledge that other people exist. We can’t dismiss Descartes’ conclusion on the basis that it is some kind of philosophical knowledge, not the normal knowledge people care about, because Descartes reflected upon knowledge in the manner that people reflect upon knowledge everyday.

5 KEY POINTS

1. Dreaming and knowing aren’t incompatible; you can know X to be true, and dream X, but by dreaming you don’t *thereby* know X.
2. It’s not the fact that you are dreaming that keeps you from having knowledge, it’s merely the fact that you *might* be dreaming and you don’t *know* you aren’t.
3. Descartes assumes that anything that can occur in real life can occur in your dreams.
4. To establish that he’s not dreaming he would need something more than just experience or information alone that shows he’s not dreaming, he would also need to know that these experiences or that that information is reliable.
5. “A possible deficiency in the basis of my belief can interfere with my knowledge without itself rendering false the very thing I believe.” 

Is Merlyn the Magician a Libertarian?

I came across this passage from T.H. White’s The Once and Future King and I thought you might find it interesting.

Merlyn was still explaining. 

“When I was a young man,” he said, “there was a general idea that it was wrong to fight in wars of any sort. Quite a lot of people in those days declared they would never fight for anything whatever.”

“Perhaps they were right,” said the King.

“No. There is one fairly good reason for fighting–and that is, if the other man starts it. You see wars are a wickedness, perhaps the greatest wickedness of a wicked species. They are so wicked that they must not be allowed. When you can be perfectly certain that the other man started them, then is the time when you might have a sort of duty to stop them.”

“But both sides always say that the other side started them.”

“Of course they do, and it is a good thing that it should be so. At least, it shows that both sides are conscious, inside themselves, that the wicked thing about a war is its beginning.”

“But the reasons,” protested Arthur. “If one side was starving the other by some means or other–some peaceful, economic means which were not actually warlike–then the starving side might have to fight its way out–if you see what I mean?”

“I see what you mean,” said the magician, “but you are wrong. There is no excuse for war, none whatever, and whatever the wrong which your nation might be doing to mine–short of war–my nation would be in the wrong if it started a war so as to redress it. A murderer, for instance, is not allowed to plead that his victim was rich and oppressing him–so why should a nation be allowed to? Wrongs have to be redressed by reason, not by force.”

(part 2, chapter 4)

The bolded line is interesting to me because Merlyn seems to hold governments to the same standards of justice and morality as individuals. That is a central idea of libertarianism.

And if Merlyn is libertarian, is there any doubt that we’re right?