Bertrand Russell’s The Problems of Philosophy is an introduction to some of the central issues in metaphysics and epistemology. The work is still regarded as one of the best introductions to philosophy, and it is also historically significant as one of the first book-length examples of analytic philosophy, the type of philosophy founded by Russell, his Cambridge colleague G. E. Moore, and German mathematician Gottlob Frege.
Analytic philosophy is distinguished by its emphasis on clarity and the attempt to establish its conclusions by the strongest rational means possible. While the book deals with a number of philosophical issues, much of it is centered on the problem of the external world and the problem of a priori knowledge—that is, knowledge that in principle can be gained through thought alone, independent of empirical experience.
Russell begins by asking whether there is any knowledge that is so certain that no reasonable person can doubt it. While it may seem obvious that there is such knowledge, Russell shows that it is not easy to arrive at an adequate answer to this question. Take for example an ordinary table. It is usually taken for granted that sensory experience reveals what the table is like: that it is brown, smooth, rectangular, and so forth. This seeming obviousness, however, ignores, for example, that the table may not look uniformly brown. As a perceiver changes location, the color of the table will appear to change. The same is true for all of the properties of the table apprehended through sensory experience: They may seem stable, but they are in fact alterable based on the situation and attitude of a perceiver. While this fact is usually ignored in everyday life, it is vitally important to a philosopher trying to determine whether sensory experience yields genuine knowledge of the external world. Physical objects can appear to have incompatible properties: A table that appears to be smooth to the naked eye may appear to be rough when viewed under a microscope. However, Russell believes, physical objects cannot really have incompatible properties; thus, it seems to him that perception can reveal only how things appear and not how they really are.
Based on considerations such as these, many philosophers have adopted attitudes of either skepticism (the real world is unknowable) or idealism (the world is essentially mental). Russell argues that drawing either a skeptical or an idealist conclusion is not mandatory. His first step in this argument is to introduce the concept of sense data, or the things that are immediately known in sensation. Colors, sounds, smells, and so forth are examples of sense data. Russell is careful to distinguish sense data from sensation, which is the mental act of being aware of sense data. Without this distinction, there is a tendency to think that the only things that can be known or can exist are mental, and this tendency leads to either skepticism or idealism.
Russell next asks about the relation between sense data and mind-independent physical objects. While skeptical arguments may demonstrate that the existence of physical objects can be denied without contradiction, Russell argues that there are good reasons for thinking that such objects do exist. His argument begins with the fact that the patterns of sense data experienced by the mind are not totally chaotic but are, rather, relatively coherent. This fact requires an explanation, and, according to Russell, the best explanation is that the sense data are caused by the interaction of physical objects and a person’s sense organs. This interaction ultimately causes the person to be aware of a particular pattern of sense data. Thus, since...
(This entire section contains 1615 words.)
positing the existence of physical objects provides the best explanation for the patterns of sense data that are experienced by humans, it is reasonable to believe that there is an objective world of mind-independent physical objects.
Russell’s important distinction between knowledge by acquaintance and knowledge by description can be used to explain the difference between the sorts of knowledge one has of sense data and physical objects. Sense data are known directly, immediately, and noninferentially; this type of knowledge Russell calls knowledge by acquaintance. Physical objects, on the other hand, are known by description. For example, a table is the cause of this particular cluster of sense data. Knowledge by acquaintance is extremely important for Russell’s epistemology. He suggests that any proposition that a person can understand must be entirely composed of elements with which that person is acquainted. (That is, each element of a proposition must correspond to some sense datum or data that the person has personally experienced.)
Russell maintains that one may have knowledge of both things and truths. While some knowledge of truths is empirical, based on sense experience, knowledge of other truths is a priori. For example, knowledge of the necessarily true principles of logic and mathematics could not be based on experience because experience can only reveal how things are at a given moment, not how things must be.
Russell’s solution to the problem of a priori knowledge involves the introduction of universals. A traditional metaphysical distinction exists between universals and particulars. Particulars are unique individual things. Thus, a specific red tomato would be an example of a particular. Particulars have properties that they can share with other particulars. For example, the red tomato shares the property of redness with a red cherry and the property of tomatoness with other tomatoes. The property red or redness is a universal—that is, a general, shareable feature exemplified by particular things. Russell’s conception of universals is similar to Plato’s. According to both philosophers, universals are abstract (not existing in space or time) and nonmental entities that are apprehended in pure thought. Russell maintains that universals are known by acquaintance.
Sense data and universals should not be confused. Seeing a red tomato involves the sensation of a red sense datum, which is a particular, but the sense datum exemplifies the universal redness, which can be shared by other particulars and which is known by reason, not the senses. One of Russell’s innovations is that he emphasizes not only that there are monadic universals, such as redness or justice, but also that there are relational universals, such as those expressed by the phrases “north of” or “greater than.”
Russell’s argument that there are universals is based on a reductio ad absurdum: Rather than argue directly that universals exist, Russell attempts to show that the assumption that there are no universals implies an absurdity and must therefore be false. For there to be no universals, there would instead be only particular things with properties that merely resemble the properties of other particular things. If one assumes that this is the case, one must assume that two red things, a and b, do not share universal redness but only resemble each other. Similarly, two green things, c and d, would again only resemble each other without reference to a universal greenness. However, consider the relation of resemblance that holds between a and b and the one that holds between c and d. The assumption that everything is particular implies that these instances of resemblance are themselves particular and so only resemble each other. That is, there is no universal known as resemblance either, so the relationship between a and b is similar to, but not exactly the same as, the relationship between c and d. Thus, an infinite regress is being generated. The only way of blocking the regress is to acknowledge that the instances of resemblance are the same—that is, that the relation of resemblance is a universal. Thus, there must be at least one universal, the relation of resemblance. Having admitted the existence of one such universal, there is no reason not to admit the existence of others.
Believing that he has made the case that there are universals, Russell is now in a position to offer his solution to the problem of how a priori knowledge is possible. The basic idea is that a priori knowledge is explained in terms of the apprehension of relations between universals. Take the a priori proposition that nothing is entire red and entirely green at the same time. This seems necessarily true and knowable independently of experience; there is no need to examine all colored surfaces in the world to make sure that none are both red and green. By just reflecting on the universals redness and greenness, a philosopher may recognize the proposition as self-evidently true.
The Problems of Philosophy concludes with Russell’s reflections on the limits of philosophical knowledge and the value of philosophy. Many philosophers have thought that a priori metaphysical reasoning can establish truths about the nature of reality as a whole. They have asserted, for example, that reality is essentially spiritual. Russell, as a founder of the analytic tradition in philosophy, has a more modest view regarding the scope of philosophical knowledge. This modesty is due to the higher standards of clarity and proof that one typically finds among analytic philosophers. Russell is quite skeptical of the claim that human reason, unaided by empirical investigation, can determine the ultimate nature or final truth about the world.
Regarding the value of philosophy, Russell asserts that this value does not lie in the definite knowledge that philosophy produces but rather in the effect it has on those who study it. Those who do not study philosophy, he believes, are imprisoned by the unexamined prejudices of their society. The study of philosophy allows one to examine critically these beliefs, to retain those that survive critical scrutiny, and to reject those that do not. Philosophy for Russell thus liberates those who study it.