Lecture Notes, UC Davis Philosophy 102, Theory of Knowledge

Common-Sense Epistemology

G. J. Mattey, Senior Lecturer

Picture of G. E. Moore

G. E. Moore

Text: G. E. Moore, "In Defence of Common Sense" (1925)

The essay is centered around five "important" points in which Moore's philosophical views differ from those that have been taken up by some other philosophers, though he agrees with others on each one of them. Although many of the philosophical views Moore discusses here are no longer actively advocated, the general tenor of his treatment is very important. Many theorists of knowledge have, in recent years, taken a "Moorean" stance. See, for example, a paper by William Lycan. Two Web pages are devoted to this article, one from Australian National University and one from Southern Illinois University Edwardsville.

Some Knowledge Claims

The first point brings out an enunciation of the "common-sense point of view." Moore lists a number of propositions under (1) which he claims to know with certainty are true. Then he gives a single proposition which is tied directly to the list. It says that each human being has frequently known propositions like those in the list. This proposition (2) is also something Moore claims to know with certainty. That (2) is true is the first point.

The list of "truisms" is too long to reproduce here. We shall note some of the highlights.

The truism (2) is now given. It is stated twice, and both times it is very long, reflecting the length of the list just given that makes up thesis (1). He summarizes it as follows:

What (2) asserts is merely that each of us has frequently known to be true a proposition corresponding . . . to each of the propositions (1).

What makes a proposition "corresponding" is that it applies to other human beings to whom "I" and "my" refer when they use those terms. So, you have a living body that exists at the present, etc. You knew the corresponding things at various times. (This does not apply to all human beings, but it does apply to "very many" beings who have had human bodies, were born and lived for a while on earth, and had the kinds of experiences listed above.)

Against some philosophers who have held that true propositions can be partially false, Moore asserts that all the propositions in (1) and proposition (2) are "wholly true." Another disagreement is with philosophers who would regard everything in (1) and (2) as true, but give them a non-standard meaning. Moore has in mind those "phenomenalists" who would say that physical objects are not distinct from perceptions, sense impressions, or something else subjective. Moore rejects any "subtle sense" of the propositions in favor of "precisely what every reader, in reading them, will have understood me to mean."

Many philosophers do not attribute the common meaning to these propositions. When asked straight-up whether they are true, they give a qualified answer: "it depends on what you mean by . . . ." Moore finds this reaction "as profoundly mistaken as any view can be." The propositions are not ambiguous.

There is a difference between understanding what they mean and being "able to give a correct analysis" of the meaning. Moore admits that the latter is "a profoundly difficult question," to which no one knows the answer. (See the fourth point, below.) But we could not even attempt an analysis if we did not already understand the meaning. So (assuming that there is an ordinary sense), saying that he was using the propositions in the ordinary sense is "all that is required to make my meaning clear."

There are two groups of philosophers who "have really held views incompatible with (2)." The first group (Group A, metaphysical) holds that no propositions are (wholly) true in a given one of the classes based on the list. In that case, they could not be known. Whether they have held this for one class or for all, they disagree with (2), which claims knowledge of the truth of "each" of them. The second group (Group B, skeptical) only asserts that members of the classes are not known with certainty, though they grant that they may all be true.

Metaphysical Disagreement

One reason why philosophers have denied the truth of the propositions in all the classes is that they deny the reality of material things and of space. There is a sense in which some of them imply that there are material things and that space is real. Others of them, which describe experiences such as dreams and feelings, do not have this implication, though they, like the propositions of the first type, imply that time and at least one self are real. Some very hard-core philosophers have even denied the full truth of any propositions which involve present or past tense.

There is a real ambiguity what is claimed by these philosophers:

But in "the most natural and proper usage," each of these is incompatible with (2). And some philosophers have really held the views understood in this way, and hence hold a view that is not compatible with (2). Moore regards them all as wrong.

The denial of all the members of any of the classes is self-defeating. If the denial is true, then no philosopher has ever existed, in which case, no human (in Moore's sense) philosopher has ever denied it. Put another way, if there have ever been any human philosophers, then the corresponding propositions of (1) are true of all of them. If it is a fact that philosophers Moore knows and respects have denied these propositions, then he knows the propositions are true.

On the other hand, if he has no reason to believe that the propositions are true, he has even less reason to believe that the philosophers have held views incompatible with them. This is because the second claim is more specific than the first, and he would need more reason to think they held a specific view than simply that they existed and held some view. Either way, then, it is more reasonable for him to believe in the truth of the propositions than in the fact that the philosophers have denied their truth.

The second point Moore makes against these philosophers is that they are inconsistent. At the same time (so to speak) that they deny that material things and space are not real, they write as if other people exist, using "we" in the sense Moore has used it, i.e., to refer to human beings who have had bodies and lived on earth. The philosophers are members of the class of human beings only if those beings have frequently known propositions corresponding to (1). So the deviant views of the philosophers are inconsistent with propositions they knew to be true, which is a strange thing.

The third point against the first type of philosophers is that one of their pet forms of argument cannot be right. They often claim that to affirm the kind of propositions that are in the list implies a contradiction. But, according to Moore, they cannot be right about this. Because those propositions are true, they cannot imply a contradiction.

The second claim, that the philosophers are inconsistent in the way they present their views, does not mean that the views themselves are inconsistent. Moore admits the possibility of the non-reality of time, space, material things, and selves. But the truth of the propositions in the list is enough for him to conclude that these things are real, despite the possibility of their unreality. This is the classic "Moorean strategy" in dealing with such possibilities.

Skeptical Disagreement

The second type of philosopher only denies that we know that the propositions on the list are true. Moore claims that such a view is self-contradictory. These philosophers deny that we know for certain only some of the propositions on the list, those asserting the existence of material things or of other selves. They do not deny that they have had various experiences at different times.

On the other hand, they will allow that they believe the propositions about material things or other selves. Although they may be "beliefs of common sense" which are widely shared, they have denied that the propositions are known for certain. But to make the claim that they are widely-shared beliefs is to fall into a trap. For it is to assert that there are other human beings who have these beliefs, which entails the propositions on the list.

There is a real inconsistency here, unlike with the philosophers of the first type. Assertions about human knowledge presupposes "the existence of many human beings." The metaphysical philosophers do not presuppose this in their thesis, though other things they do presuppose the existence of many human beings. The skeptical philosophers presuppose that it is certainly true that they exist, and this cannot be unless some member of the human race (themselves) knows "the very things which that member is declaring that no human being has ever known."

(Note, however, that that a skeptic need make no knowledge claim while holding that it appears to him that there are no human beings who know what is in the list. This is the Pyrrhonian strategy. Or, a skeptic might claim that if there are human beings, then they are ignorant of these things.)

The fact that claiming knowledge of our ignorance of all the propositions in the list is inconsistent does not imply that there is a contradiction in the denial of them all. Moore's position is that he knows the listed propositions and knows that others know corresponding propositions. But he may not have the initial knowledge. "Isn't it possible that I merely believe them? Or know them to be highly probable?"

The answer is that it seems to him that he knows them with certainty. The knowledge need not be direct, as knowledge of many of the propositions depends on evidence, some of which he does not know. But that is not a good reason to doubt that he does know these things. We know them, we know we have evidence for them, but "we do not know how we know them."

The Common-Sense View of the World

If Moore's position is to be given a name that has been used in classifying philosophical views, he would call it the "common-sense view of the world." He takes that view to be wholly true. He also thinks he has established that all other philosophers have held this view, so that their only difference from him is that they hold other views inconsistent with it. The common-sense view should not be disparaged without absurdity.

Physical Facts and Mental Facts

The second point, which is "next in importance" has to do with the relation between physical facts and mental facts. He rejects the thesis that all physical facts are either logically or causally dependent on some mental fact. (Theses of this kind are generally called "idealist."

Rejecting that thesis is not the same as asserting that there are mind-independent physical facts (though Moore indeed thinks that there are). He is only holding that there is no good reason to accept the idealist thesis.

Moore professes to use the term "physical fact" in an ordinary way. He gives as an example "That the mantelpiece is at present nearer to this body than that bookcase is." Other physical facts are "like this" (and other facts) in a respect that "I cannot define." None of the examples he has given are such there is reason to suppose that they are dependent on mental facts.

The term "mental fact" is trickier, though he claims that he is using it in a natural way. There are three different kinds of possible mental fact. He sure that there are mental facts of the first kind but does not claim certainty about the existence of the other possible kinds of facts.

Facts of type (a) are those that resemble in a certain respect "I am conscious now" and "I am seeing something now." They resemble "I am conscious now" in that they are about a particular individual at a particular time, asserting that that individual at that time is conscious." "I am seeing something now": entails "I am conscious now," adding a specific way of being conscious.

Possible facts of type (b)are harder to understand, as they are based on an analysis of facts of the first type. The idea is that being an experience is a property that an event at a time has. So when one is conscious, one is related to that experience.

The third type of possible mental facts, (c), has to do with "timeless experiences, which might or might not be the experiences of some individual." Moore does not see how there can be such things, but he can't rule it out, and they would be mental facts. A fourth type would be facts about the first three types of possible mental facts.

Moore reiterates that he finds no reason to believe that all physical facts (those which resemble his examples) depend on mental facts as he defines them. Berkeley's idealism holds that there is a fact about mental facts which is logically entailed by all physical facts. People who believe the universe was created by God hold that all physical facts are caused by some mental facts. Moore claims there is no good reason to believe either of these.

The Existence of God

Moore differs from all those philosophers who have held that there is good reason to believe that God exists and created material things. He also thinks there is no good reason to believe that human beings will continue to exist after death. He does not say why he thinks so.

The Analysis of Material Things

Although Moore claims to know many things about the physical world with certainty, he is not so confident about how statements about the physical world should be analyzed.

The claim that "Material things exist" can be analyzed only if simpler statements such as "I am now perceiving a human hand" can be analyzed. But knowledge of the latter statement is possible only because it is deduced from statements such as "I am perceiving this," and "This is a human hand."

Moore breaks these statements into two components. First "there is always some sense-datum about which the proposition i question is a proposition." Secondly, there is the thing itself, the hand, which is not a sense-datum. The sense-datum is just what you see (or feel, etc.), and Moore claims to know that there are such things.

But it is problematic how these sense-data relate to the hand. The data do not include all the parts of the hand, for example. What is perceived is only part of the surface of the hand, a part that is supposed to be representative of the hand. Moore can only be certain about this aspect of the analysis of "This is a human hand." He is not certain about what it is he is knowing when he knows that "This is part of the surface of a human hand."

Moore finds only three types of answer to the question of what he knows when he knows about sense-data. The first type claims that he knows is that the sense-datum is part of the surface of the human hand. In that case, he directly perceives part of the hand. This may well be true, and if it is, then we must abandon the popular view that "our sense-data always really have the qualities which they sensibly appear to us to have." Someone else looking through a microscope would see the surface quite differently, yet it would be the same surface.

A more serious objection is a traditional one (found in Hume, for example), that when we have "double vision," there are actually two sense-data of the surface. It is not the case that both can be the surface of the hand. Given this conclusion, the sense-datum looks not to be a direct perception, but only a representative perception of the hand.

This leads to the second view, that the sense-datum is "of" the surface of the hand, and it is related by some relation R to the hand itself. Moore thinks that this relation cannot be analyzed. This view is subject to "very grave objections," which again are traditional. How can we possibly know of any of our sense-data that they bear the relation R to some object? Even if we know this, how do we know that they represent the very qualities the object has.

The final view is that of John Stuart Mill, who called material things "the permanent possibilities of sensation." What we know, when we know that this is part of the surface of a human hand, is a wide range of hypothetical facts. For example, if I had changed the position of my head, I would have had sense-data which in broad outline are a mirror-image of the sense-data I have. More generally, "If these conditions had been fulfilled, I should have been perceiving a sense-datum intrinsically related to this sense-datum in this way."

Moore finds three "very grave" objections to this view, which at least show that it is not known with certainty to be true. The first is that the conditions listed in the hypothetical seem to have to be about material things, as in our example. But this is what is to be analyzed.

The second objection is hard to understand, as it is very complex and stated as part of a single enormous sentence. We want to say that we know that the sense-datum we actually have of the hand is of the same object as the hypothetical sense-data would be of. This is not equivalent to knowing that there is a relation that ties the actual and hypothetical sense-data together. Knowledge of the first sort does not imply knowledge of the second sort. But, Moore seems to assume, we would need to know that relation in order to give a proper Millean analysis of material things.

The final objection is that the way in which a body has a certain shape, for example, would be extremely complicated and quite different from the way in which "our sense-data sensibly appear to us." Looking round seems very simple from the standpoint of our sense-data.

The Analysis of the Self

The fifth point is that a similar problem arises for "There are and have been many Selves." This is certainly true, but its analysis is doubtful (though many philosophers hold the converse--that its analysis is known but its truth is not). The problem lies in explaining the relation of my sense-data to me. (David Hume discussed this problem notoriously in Book I, Part IV, Chapter VI of the Treatise of Human Nature.) There are many possible analyses and none is nearly certain. The problems are especially acute with respect to the relation of past sense-data to me.

[ Course Home Page | Menu of Lectures ]