The Problems of Philosophy is one of the first concrete expressions of analytic philosophy, and it is, in fact, the first book of analytic philosophy whose main focus is central questions of epistemology and metaphysics, two of the main branches of philosophy. But best of all, it's a book that can be read for pleasure as well as profit by the general public, undergraduate students, graduate students, and professional philosophers. Almost alone among philosophical books of the first quarter of the twentieth century, it's read and studied today, both inside and outside the classroom.II
Bertrand Russell was forty when The Problems of Philosophy was published in 1912, but he was already a well-known British philosopher and mathematician. Educated at home until he was eighteen, he then went to Cambridge to study mathematics and, after a few years, philosophy. At first he was under the influence of the British Idealists, especially Bradley, McTaggert, and Stout, but after the turn of the century, he revolted against Idealism and became increasingly attracted to the traditional British Empiricism of Locke, Berkeley, and Hume. Empiricism in general is the view that our knowledge of what exists and what properties those things have is based upon sense experience. Little by little Russell came to embrace, and eventually to represent, an updated empiricism, an empiricism rooted in past centuries but responsive to later developments, especially those in logic. In 1912, Russell was far from a fully "updated empiricist," but The Problems of Philosophy is a first step along the path, and it's a big one.
Russell, however, is much more than a figure in the historyof philosophy. He was the second son in an aristocratic and political family, and although his parents died when he was quite young, he, like them, became a "Whig aristocrat" and frequently championed liberal causes and actively participated in political affairs. He was twice jailed for his anti-war activities, for example, and spent six months in prison during World War I for an inflammatory pamphlet he had written. He also wrote books on a number of topics of general cultural interest, including history, education, marriage, and happiness. For the ten years prior to the appearance of The Problems of Philosophy, however, most of his time and energy was devoted to work on the foundations of mathematics. The fruit of his labor was Principia Mathematica, a massive and very difficult three-volume work written in collaboration with his former teacher, Alfred North Whitehead. According to Russell and Whitehead, pure mathematics can be deduced from the principles of logic alone. At least one of Russell's discoveries during that ten-year period of intense intellectual activity, namely, his Theory of Descriptions, is utilized in The Problems of Philosophy, however the book is not technical, and no prior acquaintance with the Theory of Descriptions is needed to understand it.
Supposing that The Problems of Philosophy is the first book of epistemology and metaphysics that might be regarded as analytic, what does that mean? Actually, there are a number of questions here: (1) What is philosophy? (2) What are epistemology and metaphysics? And (3) What distinguishes analytic philosophy from scholastic philosophy, say, or existential philosophy -- that is, from other philosophies?
Russell gives us an answer to the first question. On the very first page he says that "Philosophy is merely the attempt to answer … ultimate questions, not carelessly and dogmatically, as we do in ordinary life and even in the sciences, but critically, after exploring all that makes such questions puzzling, and after realizing all the vagueness and confusion that underlie our ordinary ideas." Boiling this down a bit, philosophy is the attempt to answer ultimate questions on the basis of the most attentive and careful scrutiny and reasoning we are capable of. It's harder to define an ultimate question in a satisfactory way than it is to give an example of one -- "Does God exist?" is an ultimate question -- but generally speaking, an ultimate question is one whose answer, whatever it may be, is extremely fundamental to our beliefs about ourselves and the world. Duly organized into a whole, careful reasoning about and answers to such questions provide us with a worldview, a basic -- the most basic -- framework into which we fit ourselves and everything else. A worldview, then, is essentially a philosophy.
As for the second question: metaphysics is the branch of philosophy that critically examines (1) the most fundamental categories of things that exist -- for example, material objects, minds, events -- (2) the most fundamental constitution of things that exist -- for example, substance and attribute -- (3) the most fundamental aspects of reality -- for example, space and time -- (4) the most fundamental relations between things that exist -- for example, that between mind and body -- (5) the most fundamental properties of things that exist -- for example, the extension (occupancy of space) of material bodies, the (supposed) existence of the soul after death -- as well as (6) the very idea of existence itself. In short, metaphysics attempts to present a comprehensive, coherent, and well-argued for theory of reality, of all that is, couched in terms of the most basic categories and distinctions possible.
Epistemology, on the other hand, is concerned with knowledge. It critically examines (1) the nature or definition of knowledge -- for example, is knowledge the same thing as true belief? -- (2) the nature and importance of the most fundamental sources of knowledge -- for example, perception, memory, reason, and testimony -- (3) the scope or extent of knowledge -- is there moral knowledge, for example, or, more radically, knowledge of external objects (material objects) at all? -- (4) the basic structure of knowledge and justified belief -- is there foundational knowledge, for instance, upon which all other knowledge rests? -- (5) the ultimate justification for certain kinds of basic inferences -- for instance, inferences from observed cases of some connection (e.g., that between eating bread and being nourished) to unobserved cases of the same connection -- as well as (6) certain fundamental concepts connected with knowledge, such as truth, belief, and justification. In short, epistemology attempts to present a comprehensive, coherent, and well-argued for theory of knowledge, couched in the terms of the most basic concepts and distinctions possible.
And finally, what makes analytic philosophy analytic -- in other words, what distinguishes it from other kinds or forms of philosophy? Negatively, the answer is, "Not its questions or answers." Russell's questions are traditional ones, in the main inherited from Descartes and other philosophers of the modern era, and his answers are, none of them, radically new or original, though there are original distinctions and arguments in The Problems of Philosophy. Rather, what distinguishes analytic philosophy in general is its sustained effort (1) to break down -- to analyze -- and to clarify concepts, theses, arguments, and objections as much as possible, in order to ascertain exactly what is being said and argued, (2) to make distinctions which are helpful in the solution of philosophical problems, (3) to evaluate, on the basis of careful logical scrutiny, the truth or falsity of premises found in, and the nature and strength of inferences made in, philosophical arguments, and (4) to present arguments for philosophical theses that have been vetted in accordance with (1)-(3). In short, analytic philosophy is distinguished from other philosophies -- scholasticism and existentialism, say -- not by its presuppositions, subject matter, or positions, but primarily by a general methodology, or mode of approach to philosophical problems. It's continuous with previous philosophy, and especially British Empiricism but is more pronounced and self-conscious in its piecemeal approach, its emphasis on argumentation, and its insistence on clarity. These characteristics are all very evident in The Problems of Philosophy.
Although Russell tackles some metaphysical problems in The Problems of Philosophy, in the main the book is focused on epistemological problems. In the first four chapters he argues that things exist independently of mind and that we can have knowledge of them.
While he undoubtedly sees something when he looks at the table in front of him, Russell says, what he sees is "what is immediately known in sensation." This is a "sense datum," the sort of thing common to the veridical perception of tables and to dreams of tables. Sense data of qualities we're acquainted with include color (e.g., red), texture (e.g., smooth), sound (e.g., shrill), and smell (e.g., rancid). Since sense data depend on people's minds for their existence, they are subjective; and since each person experiences his or her own sense data and couldn't experience anyone else's (you can experience a red sense datum that is indistinguishable from mine but not one and the same sense datum as mine), sense data are necessarily private. The conclusion Russell draws from this is that it's not necessarily a material object, a mind-independent object, that he sees when he looks at the table. The existence of the table, a material object in public space, is problematic. Maybe it doesn't exist at all. Maybe whatever exists, or at any rate whatever can be known to exist, is mind-dependent. Russell goes on to argue, however, that even though sense-data cannot exist independently of mind, that doesn't make them mental. In other words, that sense data depend on minds for their existence doesn't mean that they are parts of minds or inherently mental in nature. It is in not noticing this point that Idealism, the philosophical theory that says that everything that exists is in some sense mental, makes its central mistake. A sensation -- an act of sensing -- is different from a sense datum -- that which is sensed. While the act of sensation is essentially mental, that doesn't mean that that which is sensed -- the sense datum -- is essentially mental. A hand may be essentially biological, but that which a hand grasps -- a book, say -- isn't necessarily essentially biological.
But there are also positive reasons for thinking that material objects -- that is, objects that exist independent of minds, in a public space -- exist. Russell's main argument is that the existence of material objects is the best explanation of the existence and properties of our sense data, and the fact that people's sense data fit into a consistent, co-ordinate system of the world. The upshot is that although material objects cannot be immediately known, we can know that they exist, and also that our sense data are caused by them.
There is nothing particularly original in the first four chapters of The Problems. In the fifth chapter, however, new ground is broken, untrod by previous philosophers. Russell divides knowledge into knowledge of things (e.g., a red sense datum) and knowledge of truths (e.g., that the sun is 93 million miles from the Earth) and further distinguishes two kinds of knowledge of things: knowledge by acquaintance and knowledge by description. Sense data are known by acquaintance; material objects by description. All our knowledge, it turns out, "rests upon acquaintance as its foundation," and what we are acquainted with are sense data, ourselves, universals (of which more below), and, by memory and introspection, other objects of inner and outer sense (e.g., my desiring an ice cream cone). On the other hand, what we know by description -- a description is basically a phrase of the form 'the so-and-so'-- are material objects and other people's minds. More details follow, with the chapter culminating in the "fundamental principle" that "Every proposition which we can understand must be composed wholly of constituents with which we are acquainted." In essence, this is a cardinal tenet of the Classical British Empiricism of Locke, Berkeley, and especially Hume concerning meaning, but with the difference that it has been filtered through the sieve -- never explicitly mentioned by Russell -- of his Theory of Descriptions, a theory about the proper analysis of propositions that he devised in 1905 to solve certain problems in philosophical logic. That, in general, was Russell's tendency for the rest of his life: to meld Classical British Empiricism with modern developments in logic and philosophical logic. As far as the problem at hand in The Problems of Philosophy is concerned, the upshot of this is that our knowledge of material things is ultimately based upon our knowledge of sense data, and we do not and cannot directly know material things. Knowledge of material things is really knowledge of truths, with the matter out of which such truths is constructed being supplied by sense data.
The sixth chapter of The Problems is a very clear exposition and defense of Hume's problem of induction. This is the problem of justifying the principle that the future will resemble the past, or, better, that nature is uniform. The principle of induction, Russell shows, underwrites all our inferences from observed cases of connections between phenomena ('All observed As are Bs') to unobserved cases of the same ('The next A we come across will be a B' or 'All As are Bs'). If the principle is unjustified, then all such inferences are; and thus we have no more reason to think that next time we walk down the stairs, the stairs will support us, than that they won't. Both Hume and Russell think that the principle can't be justified on the basis of sense experience.
But the principle of induction isn't the only principle underwriting inferences that we habitually make and have confidence in, but that can't be proved on the basis of sense experience. In chapter seven, a number of others are identified, including principles of logic (such as modus ponens), mathematics (such as that two plus two equal four), metaphysics (such as the 'laws of thought,' as Russell interprets them), and ethics (such as that happiness is more desirable than misery). Russell claims that all such principles are known a priori, that is, known independently of sense experience. Kant's attempt to account for how we can have such a priori knowledge is criticized in chapter eight, and Russell argues that such knowledge is connected with universals, entities like whiteness and beauty (as opposed to white and beautiful things, whether material objects or sense data). According to Russell, universals are non-material, non-mental, "unchangeable, rigid, exact, [and] delightful to … all who love perfection more than life." Plato argued for the existence of universals, and in chapter nine Russell largely follows him. Like particular objects (such as sense data and material objects), universals are known by acquaintance or by description, but in both cases, through abstraction.
The basis of a priori knowledge, Russell argues in chapter ten, is our knowledge of universals. In particular, "All a priori knowledge deals exclusively with the relations of universals." The principle of induction and other fundamental principles are thus justified, but their justification is a priori, and they concern only relations of universals.
Intuitive knowledge, truth, and the nature of knowledge itself are the focus of the next three chapters. By intuitive knowledge, Russell means knowledge of self-evident truths, propositions either incapable of proof (because they are so basic) or just clearly true, even if capable of proof. Many propositions known a priori are self-evidently true, but so are some propositions known on the basis of sense experience, that is, known empirically. Some propositions concerning sense data and some concerning memory are self-evident, for example. Truth itself is "some form of correspondence between belief and fact," but belief isn't "a relation of the mind to a single object, which [is] what is believed." Believing is a multi-termed relation. If John believes that Tom hit Bill, then John is the subject of the belief, and Tom, Bill, and hitting are its objects. Belief ties all these terms together in a certain order. If Tom believes that John hit Bill, the terms of the belief are the same, but the belief is a different one, because the order of the terms is different.
As for knowledge: it's certainly not true belief. First, lucky guesses which happen to be true aren't knowledge. Consider someone who believes, on the basis of no solid evidence, that Seabiscuit will win the fifth race; and suppose that Seabiscuit does win that race. Still, that person doesn't know that Seabiscuit will win the race. Second, true beliefs validly deduced from false beliefs aren't necessarily knowledge. Imagine that someone believes that the United States has had a Republican President but validly deduced that belief from the false belief that Clinton was a President of the United States and a Republican. Again, such a true belief wouldn't count as knowledge. This last counter-example, incidentally, a modification of one of Russell's own, also shows that knowledge can't be defined as justified true belief. Over fifty years after Russell wrote, the fact that justified true belief isn't necessarily knowledge would be re-discovered and give rise to repeated efforts to solve what is known as "the Gettier Problem." Solving the Gettier Problem means formulating an adequate definition of knowledge that avoids such counter-examples. No solution to the Gettier Problem has met with widespread acceptance among philosophers.
In the last two chapters, Russell discusses philosophy itself, and in particular, the scope of philosophical knowledge and the value of philosophy. Philosophy can't prove the "dogmas of religion, the essential rationality of the universe, the illusoriness of matter, [or] the unreality of evil," Russell thinks, and to try "to prescribe to the universe by means of a priori principles," as many philosophers have done, is a mistake. "Knowledge as to what exists [is] limited to what we can learn from experience." Essentially, what Russell is rejecting in such remarks is a conception of philosophy that erects grand and emotionally satisfying systems, such as (as he interprets it) Hegel's, on a foundation consisting of nothing more than a few a priori building blocks.
But on a more positive note, there is "the value of philosophy and why it ought to be studied." In as eloquent a statement as can be found anywhere in the history of the philosophy, Russell argues that "it is exclusively among the goods of the mind that the value of philosophy is to be found; and only those that are not indifferent to these goods can be persuaded that the study of philosophy is not a waste of time." The "knowledge [which philosophy] aims at," he tells us, "is the kind of knowledge which gives unity and system to the body of the sciences, and the kind which results from a critical examination of the grounds of our convictions, prejudices, and beliefs." Without philosophy, we go through life "imprisoned in the prejudices derived from common sense." Philosophy liberates us from such prejudices by suggesting "many possibilities which enlarge our thoughts and free them from the tyranny of custom," by removing "the somewhat arrogant dogmatism of those who have never traveled into the region of liberating doubt," by "keeping alive our sense of wonder [in] showing [us] familiar things in an unfamiliar aspect," and by making us aware of "the greatness of the objects. . .it contemplates, and [thus making possible]. . .freedom from narrow and personal aims." This last point is an especially important one for Russell. As we normally view the world, it's simply a means or an obstacle to the attainment of personal and private ends. Philosophy enables us to escape this feverish pursuit of petty, self-centered interests. It enables us to view the world "as God might see [it], without a here and now, without hopes and fears, without the trammels of customary beliefs and traditional prejudices, calmly, dispassionately, in the sole and exclusive desire [for] knowledge -- knowledge as impersonal, as purely contemplative, as it is possible for man to attain."
Plato, Aristotle, the Stoics, and Spinoza all would have smiled in agreement with these remarks.
Hye-Kyung Kim is part of the Department of Philosophy at the University of Wisconsin at Green Bay.