PHILOSOPHY

The Greek word sophia is ordinarily translated into English as "wisdom," and the compound philosophia, from which "philosophy" derives, is translated as "the love of wisdom." But sophia had a much wider range of application than the modern English "wisdom." Wherever intelligence can be exercised -- in practical affairs, in the mechanical arts, in business -- there is room for sophia; Homer used it to refer to the skill of a carpenter (Iliad XV, 412). Furthermore, whereas modern English draws a fairly sharp distinction between the search for wisdom and the attempt to satisfy intellectual curiosity, Herodotus used the verb philosophein in a context in which it means nothing more than the desire to find out (History I, 30). Briefly, then, philosophia etymologically connotes the love of exercising one's curiosity and intelligence rather than the love of wisdom. Although philosophers have often sought to confine the word "philosophy" within narrower boundaries, in popular usage it has never entirely lost its original breadth of meaning.

According to a tradition deriving from Heraclides Ponticus (a disciple of Plato), Pythagoras was the first to describe himself as a philosopher. Three classes of people, he is alleged to have said, attend the festal games: those who seek fame by taking part in them; those who seek gain by plying their trade; and those ("the best people") who are content to be spectators (Diogenes Laertius, De Vita et Moribus Philosophorum I, 12). Philosophers resemble the third class; spurning both fame and profit, they seek to arrive at the truth by contemplation. Pythagoras distinguished the sophia sought by the philosopher (knowledge based on contemplation) from the practical shrewdness of the businessman and the trained skills of the athlete. Whether or not these distinctions date back to the historic Pythagoras, they can certainly be found in Plato, who was much preoccupied with the question of what philosophy is and how it differs from other forms of inquiry. Some of Plato's contemporaries had thought of his master, Socrates, as a sage, some thought of him as a Sophist, and some thought of him as a cosmologist. In Plato's eyes, Socrates was none of these; he was a philosopher. But what made him different?

The Platonic conception of philosophy.

For Plato, the first characteristic of philosophical wisdom is that it can face the test of critical discussion. As is suggested in the Apology (22), this criterion at once rules out almost every type of what is ordinarily called wisdom. Neither the statesman nor the artisan nor the poet can explain why he is doing what he is doing; none of them has formulated a clear, articulate, discussible system of ideas and principles. That a man sometimes does the wise or right or beautiful thing is no evidence that he possesses philosophical wisdom; he must be able to give grounds for his action that will stand up to cross-examination.

Second, philosophy, according to Plato, makes use of a method peculiar to it, which he calls "dialectic." The exact nature of the Platonic dialectic is obscure, but this much is clear: philosophy proceeds by criticizing received opinions. Even mathematics, the most developed of the sciences, is subject to philosophical criticism. According to Plato, mathematics rests on inarticulate assumptions, and it is the philosopher's task to bring these into the open and examine them critically. Philosophy is the highest form of inquiry, just because it alone involves no presuppositions.

Third, Plato suggests, the philosopher has direct access to "true Reality," as distinct from the ordinary world of ever-changing things. That is precisely why he can offer the final criticism of received opinions: Having direct access to reality, he has no need of assumptions or guesswork. Philosophy concerns itself with the relationship between eternal and unchanging entities -- the only entities about which it is possible to have "knowledge," as distinct from mere belief or "opinion." Hence, the philosopher seeks wisdom of a very special kind -- certainty about the true nature of reality.

Fourth, to apprehend the true nature of reality is to know what everything is for. To understand the real nature of man, for example, is to know toward what ideal it is man's nature to strive. In the Phaedo (98-99), Plato suggests that the Ionian cosmologists did not possess philosophical wisdom, precisely because they had no understanding of purposes. They could not explain why, for example, Socrates did not run away from prison; to understand Socrates' behavior they would have had to take account of Socrates' ideals, as distinct from the structure of Socrates' body.

Fifth, it is on account of his knowledge of ideals that the philosopher knows how men ought to live. The Sophists professed to teach their pupils how to make immediate gains, how to win friends and influence people. This, Plato says, cannot be done. The art of making immediate gains is not a form of knowledge; it involves quick wits and rapid judgment. In such contexts the philosopher may well look like a fool. But when it is a question of understanding the general nature of man -- and, in consequence, of human society -- men must turn to philosophy. That is why the ideal ruler would be a philosopher.

Philosophy as the knowledge of ultimates.

When the Shorter Oxford Dictionary defines philosophy as "that department of knowledge which deals with ultimate reality, or with the most general causes and principles of things," it is substantially following Plato. The presumption is that science, inheriting the cosmological tradition, does not offer us a knowledge of ultimate reality; only philosophy can do this.

On the face of it, this is a very peculiar doctrine. It is scientists who tell us, insofar as the locution is a proper one at all, what "the ultimate reality" of the things around us consists in. The assertion that science can only tell us how, whereas philosophy can tell us why, things happen as they do, is in conflict with our everyday habit of turning to science precisely in order to find out why things happen. However, Plato and philosophers in his tradition (such as Louis de Raeymaker) would reply that the "general causes and principles" of the philosopher are "higher" and "more ultimate" than the causes and principles that science reveals to us.

Whether there are such higher principles, however, is itself a philosophical issue. It is obviously as improper to define philosophy in this way as it is to presume that a particular philosophical doctrine is true. Doubtless, philosophers have very often set out in search of ultimates -- sometimes ultimate explanations, sometimes ultimate foundations for knowledge, sometimes ultimate reference points for meaning. But philosophers also have sought to show that this whole project is a mistaken one. For instance, Wittgenstein is no less a philosopher in the Philosophical Investigations, where he argues that the philosopher's belief that there must be ultimate simples rests on a confusion, than he is in the Tractatus Logico-philosophicus, where he tries to show that ultimate simples must exist. The Platonic-type definition of philosophy as an attempt to discover ultimates is, in other words, at best too narrow. Philosophy can survive the abandonment of the doctrine that there are any "ultimates" in the metaphysical sense.

The philosopher and the sage.

Plato's distinction between philosophy and the pursuit of worldly success is now generally accepted. His distinction between philosophy and poetry -- or, more generally, between the philosopher and the sage -- has not won the same degree of acceptance.

The latter distinction is weakened by Plato's own concessions. In his earlier dialogues (Ion, Apology) he regularly suggests that true wisdom, as distinct from the inspired guesses of the poet-sage, must be clear and communicable. But when he professes to be describing the supreme end of philosophical inquiry ("the form of the good"), in the Republic (509), he speaks of it in the manner of a sage rather than of a philosopher. The form of the good lies "beyond all knowledge and being." It would seem that at the culminating point of philosophical inquiry, the philosopher has to abandon critical discussion and, like a sage, fall back on direct intuition.

Indeed, any Platonic-type metaphysics is forced to this conclusion. About the "ultimate" no further questions can be raised; if we can ask questions about it, as we can about everything else, it would cease to be ultimate. Yet it must be possible to ask questions about it, if anything definite can be said about it at all; any nontautological statement can be met with the question "Why shouldn't it be otherwise?" So the metaphysician is forced to conclude that nothing, or nothing clear, can be said about his ultimates. F. H. Bradley's "Absolute" and Wittgenstein's "totality of facts" (as Wittgenstein was well aware) share with Plato's "Form of the Good" the characteristic that they can be apprehended only "mystically." The paradox of Platonic-type metaphysics, indeed, is that what is supposed to make everything clear must itself be unintelligible, or, to give it its technical name, "transcendental."

When the critics of traditional philosophy, such as Friedrich Lange and A. J. Ayer, lay stress on its resemblance to poetry, they are identifying philosophy with transcendental metaphysics. Even so, they are mistaken; transcendental metaphysics incorporates a great deal of serious argument that must be seriously considered. But they have noticed that at its culminating point it abandons the method of critical discussion. What is true of transcendental metaphysics, however, is not true of philosophy as a whole.

In countries where the critical tradition is strong, the contrast between the sage and the philosopher is now fairly well established. However, the critical tradition is powerful in only a few parts of the world: what is commonly called "Chinese philosophy," for example, consists almost entirely of the pronouncements of sages. Neither in Russia nor in China has the tradition of critical discussion ever established itself. In France it has been greatly weakened as a result of the influence of Henri Bergson, followed by that of Martin Heidegger.

In fact, two very different forms of activity now go under the name of "philosophy": one is essentially rational and critical, with logical analysis (in a broad sense) at its heart; the other (represented by Heidegger, for example) is openly hostile to critical analysis and professes to arrive at general conclusions by a direct, essentially personal intuition. It is important to distinguish these two forms of activity from each other, even though they have sometimes been conjoined in the work of a single person. There is no value judgment involved in describing practitioners of the first kind of activity as "philosophers" and practitioners of the second as "sages." There are bad philosophers -- unimaginative, pedantic men whose criticisms are captious and devoid of understanding. There are also good sages -- men who bring us to a greater understanding of human life, even though they are neither systematic theorists nor careful analysts.

C. D. Broad draws a sharp distinction between "critical" and "speculative" philosophy. This might appear to coincide with our distinction between philosopher and sage; indeed, some of Broad's successors have suggested that, or have at least proceeded as if, critical analysis is the only sort of philosophy there is. But it is certainly not the case that philosophers in general, in our own time or in any other, have been unimaginative or unwilling to speculate. Philosophy, like science, is neither pure speculation nor pure criticism; it is speculation controlled by criticism. Recent philosophy -- as exemplified in, for example, Gilbert Ryle's Concept of Mind or Wittgenstein's Philosophical Investigations -- is quite as speculative as it ever was, although not about the "transcendental." The difference between sage and philosopher is not that the sage is imaginative and the philosopher unimaginative; it is that the philosopher submits his speculations to the discipline of close criticism.

The idea of a philosophical method.

To say that in philosophy speculation is controlled by critical discussion is not to define philosophy; the same can be said of any rational inquiry. Critical discussion takes many different forms. In science it often consists in the testing of hypotheses by experiment, while in mathematics it consists in the probing of propositions to see whether they lead to contradictions. Has philosophy a method of its own? Plato and Hegel thought so; they disagreed about its nature but agreed in calling it "dialectic." For Bergson it was intuition, for Wittgenstein the uncovering of nonsense, for Moritz Schlick clarification, and for Husserl phenomenological description. Hume, on the other hand, thought that the philosopher should imitate the methods of experimental inquiry, while Spinoza believed that the philosopher should imitate the geometrician.

The diversity of these opinions would strongly suggest that to define philosophy in terms of any particular method is to take sides in a philosophical dispute. Historically speaking, philosophers have made use of a great variety of procedures. Some of their arguments have been formal in character, reminiscent of mathematics; some of them have been attempts to overthrow hypotheses by appealing to everyday observations; sometimes they have pointed to ambiguities, obscurities, and confusions. When we look in detail at claims for the discovery of the correct method of proceeding in philosophy, we always find that they are not borne out by the writings of the philosopher who makes them. Thus, for example, Descartes's doubts are anything but universal, Hume's own arguments are not experimental in character, and Wittgenstein's Tractatus does not consist, for the most part, in the uncovering of nonsense. In other instances the method itself is suspect; this is certainly true of Hegel's dialectic or Husserl's "method of bracketing." Furthermore, the view that there is only one correct method of philosophizing always rests on philosophical doctrines that are not themselves derived by that method, such as the distinction between the given and its interpretation (phenomenology) and the distinction between ultimate simples and the complex (analysis).

In general, the methods actually used by philosophy overlap with those used outside philosophy. Some types of argument (for example, the appeal to vicious infinite regresses) may be peculiar to philosophy, but the philosopher is free to use, and does use, any type of critical discussion that promises to throw light on the problems confronting him. It would now be generally agreed that certain methods that have been advocated in the past are in fact useless: philosophers cannot arrive at the sort of conclusions which interest them by deductions from self-evident principles; philosophy cannot have the same sort of general structure as a natural science. But these negative points do not amount to a demonstration that there is a single philosophical method.

Philosophy and value.

The Platonic view that philosophy is concerned with purposes or values was revived in the nineteenth century (e.g., by Rudolf Hermann Lotze and James Ward) in response to the rise of the natural sciences. Science, it was argued, can tell us only what the world is like and how it operates, whereas philosophy can tell us what life and nature "mean," what value or purpose they have. The various branches of philosophy, according to this view, are each concerned with the "meaning" of a particular form of activity or class of thing: the philosophy of history is concerned with the meaning of history, and the philosophy of law with the meaning of law. In its most general form, philosophy elucidates the meaning of the "universe as a whole."

Once again, however, this definition of philosophy rests upon special philosophical assumptions. Some philosophers (for example, Dewey) completely reject the antithesis between facts and their value or meaning; others, even if they allow such a contrast within the limited sphere of human affairs, will certainly not grant that there is anything that can be described as the meaning of life, of history, or of the "universe as a whole," or indeed, anything answering to such descriptions as "the universe as a whole" or "the total movement of history." If it is possible to be a philosopher and at the same time to deny that the universe has a meaning, or to be a philosopher of history without admitting that history has a meaning, then philosophy certainly ought not to be defined in terms of the search for meanings.

The philosopher as adviser.

The final question that arises from the Platonic description of philosophy is whether it is the philosopher's responsibility to give advice. In the Alexandrian period, Stoic, Epicurean, and Skeptic all agreed that philosophy's main objective is to show men how to achieve peace of mind. Their practical approach appealed to the Romans: Cicero defined philosophy as "the art of life." During the Renaissance the Ciceronian conception of philosophy came to be predominant, at least among ordinary cultivated men; as late as the seventeenth century John Selden wrote that "philosophy is nothing else but prudence," by which he meant that prudence is the art of life. The popular conception of the philosopher, as embodied in the phrase "taking things philosophically," indicates a similar attitude. Many of the best-known philosophers have offered practical advice that is very closely related to their general philosophical views (for example, John Locke in Some Thoughts concerning Education).

When Russell argued that philosophy should be neutral, he was not suggesting that the philosopher should be less ready to offer advice than are other people. Indeed, few philosophers have been as ready as Russell to give advice on the conduct of life. As a person unusually practiced in critical discussion, the philosopher may well have a special responsibility to do so. Furthermore, it would be very strange if, merely in virtue of his investigations, he were not sometimes in an unusually good position to offer advice. Even if the philosopher can do no more than show people that they are talking nonsense, as Wittgenstein once thought, this can be made the basis for advice on how to avoid talking nonsense. In all such cases, however, the advice rests upon, but does not constitute, the successful completion of a philosophical task. In this respect, too, the philosopher differs from the sage: not uncommonly the whole content of the sage's "wisdom" consists in advice.

Philosophy and the special sciences.

In Plato's dialogues (except in the special case of the Timaeus), the range of subjects considered is relatively small, and most of these subjects still fall within the province of philosophy. Aristotle, on the other hand, discussed almost everything. In the Middle Ages his work was read as an encyclopedia of philosophy, when he wrote about anatomy as well as when he wrote about logic or metaphysics. Aquinas was scarcely less encyclopedic than his master; Bacon, Hobbes, and Descartes all defined philosophy in encyclopedic terms, as did Leibniz and Wolff. To be sure, they did distinguish between "moral" philosophy, "natural" philosophy, "civil" or "political" philosophy, and "first" philosophy, or metaphysics. But in general they used "philosophy" as often to refer to what we now call "science" as to refer to what we now call "philosophy."

Only gradually did the "sciences" (a word that did not come into general use until the early nineteenth century) become separate from philosophy. Even now the boundaries of philosophy are by no means clear. If J. L. Austin is right in suggesting that philosophy is in process of giving birth to a new type of linguistic theory, no doubt many works which we still think of as forming part of philosophy (Plato's Cratylus or the third part of Locke's Essay, for example) will come to be thought of as stages in the history of the development of that linguistic theory. We can at least be reasonably confident that there are a good many questions which we now consider to belong to philosophy but which will some day not be so regarded -- although we are, of course, in no position to say what those questions are.

Philosophy as the science of man.

When it came to be realized that physics was an independent inquiry, philosophers, especially in Great Britain and France, turned their attention away from nature to man. Descartes's sharp distinction between mind and matter made it appear that there could be an inquiry into "the inner world" which would be wholly distinct from inquiries into "the outer world." Philosophy came to be thought of as running parallel to physics -- the science of man as contrasted with the science of nature.

This supposed parallelism reaches its extreme in Hume. Hume quite explicitly asserts that once the nonsensical ingredients are removed from philosophy, it can wholly be identified with what he calls the moral sciences, or the science of human nature. Yet in some measure he retained the view that philosophy is the "first science" by arguing that every other branch of inquiry, precisely because it is a creation of the human mind, is dependent upon the science of human nature for its foundations. In Great Britain in the nineteenth century it became the standard doctrine that philosophy consists, as Dugald Stewart put it (Encyclopaedia Britannica, 8th ed., Vol. I, p. 227), in "all those enquiries which have for their object to trace the various branches of human knowledge to their first principles in the constitution of our nature." Both J. S. Mill and Sir William Hamilton, however opposed they may have been in other respects, agreed in defining philosophy as, in Mill's words, "the scientific knowledge of man." But this is certainly an unsatisfactory definition. Hume and his successors were not able to show that such notions as causality and identity can be clarified by tracing the workings of the human mind. This mistake, like that of confusing cosmology with metaphysics, was an important one in the sense that it led to the development of psychology as a form of inquiry. However, it also had the effect of completely obscuring what we are now able to see as a distinction between philosophical and psychological inquiries. Contemporary philosophers, such as Edmund Husserl and Alexius Meinong, began by insisting that philosophy had to be sharply distinguished from psychology -- a declaration of independence facilitated by the emergence of a truly experimental psychology as distinct from what Hume had called "psychology."

Philosophy as the science of sciences.

Could philosophy entirely disintegrate into special sciences? William James thought so. He defined philosophy as "a collective name for questions which have not been answered to the satisfaction of all that have asked them," with the suggestion that once satisfactory answers were found, they would form part of a special science. In this way philosophy slowly digs its own grave.

Yet there are philosophical questions that we could scarcely contemplate as forming part of any specialized inquiry, just because they have application to any form of inquiry whatever. Aristotle thought that questions of this sort constituted what he called "the first and last science" -- the first science because it is logically presupposed by every other science, the last because in order to understand it we must in some measure have mastered other sciences. He defined it as follows: "There is a science which investigates being as being, and the attributes which belong to this in virtue of its own nature. Now this is not the same as any of the so-called special sciences, for none of these treats universally of being as being. They cut off a part of being and investigate the attribute of this part" (Metaphysics 1003al8-25).

This definition would still be accepted in some quarters (although with reservations) as a satisfactory definition of philosophy, or at least of metaphysics. However, it has its difficulties. It rests on the assumption that "being" is the highest predicate in a series of predicates such as "mammalian vertebrate living being," "vertebrate living being," "living being," "being." However, as Hume and Kant pointed out, "being" is not an attribute in the sense in which "living being" is an attribute. Nor has it attributes; pure being is indistinguishable from nothing. Only if "theory of being as being" is given a somewhat special interpretation (as often happens) can it be saved from emptiness. Thus, Samuel Alexander identifies the definition of philosophy as the theory of being with the very different definition suggested by Francis Bacon, for whom the task of "first philosophy" is to coordinate the axioms of the various specialized segments of philosophy. Herbert Spencer, in the same tradition, defines philosophy as "completely unified science." Henry Sidgwick took over and modified Spencer's definition, suggesting that the philosopher's task is to "coordinate the most important general notions and fundamental principles of the various sciences."

This way of putting the matter, however, claims at once too little and too much for philosophy. It claims too little because the philosopher does not restrict his attention to the sciences; he asks questions about the relations between science and everyday belief, between science and religion, or, as we are now doing, between science and philosophy. It claims too much, because, although our own century has witnessed enormous advances in the coordination of the "special notions" and the "fundamental principles" of the sciences, this has resulted from, and could only result from, the work of biochemists, biophysicists, and physical chemists. Philosophers are in no position to coordinate, say, molecular structure and biological activity.

What sort of coordination, then, is left for the philosopher? One possible answer (suggested by Hegel and Croce) is that the philosopher is interested in coordinating art, economics, religion, and philosophy as forms of human activity, or modes in which the human spirit comes to an awareness of its own potentialities. This account of the philosopher's task, which equates philosophy with the theory of culture, is widely influential on the Continent. Contemporary British philosophers either frown upon such inquiries or do not consider them to lie within their competence. According to them, if these inquiries are anyone's responsibility, they fall within the province of the sociologist. But in practice, of course, the sociologist does not discuss the sort of issue raised in R. G. Collingwood's Speculum Mentis or Croce's Philosophy of the Spirit. He does not ask himself, as they did, how the theoretical and practical activities of mankind are interrelated, how religion differs from science or how both differ from philosophy. Indeed, even if these problems are not the very center of philosophy or its sole content, they may still be in some measure the philosopher's responsibility. This is a matter to which we shall return.

Philosophy as speculative cosmology.

One objection to the identification of philosophy with the study of culture is that it is unduly anthropomorphic. Philosophy's real concern, it might be said, is not particularly with the human spirit, except on the very special assumption that "nature" itself is in some sense a reification of that spirit, as the Italian idealists in fact argued. Once this assumption is rejected, it is no longer at all plausible to restrict philosophy to the realm of culture. As A. N. Whitehead and Karl Popper have maintained, everything, not only the human spirit, falls within philosophy's sphere; indeed, like science, it is a contribution to cosmology. According to a common view of the matter, this is certainly the form in which philosophy first arose: Anaximander, for example, might accurately be described as a "speculative cosmologist." But it has often been questioned whether Anaximander was a philosopher; there are good grounds for thinking of him as a scientist, or perhaps as neither philosopher nor scientist but a precursor of both philosophy and science. On the other hand, Parmenides' The Way of Truth is unmistakably philosophy.

What makes the difference? Suppose we interpret Thales as asserting that everything is made of water and Anaximenes as denying this and saying that things are made of air. These hypotheses are in important respects of the same type as the hypothesis that everything is made of electrons; the theories of Thales and Anaximenes are speculative cosmology in the same sense that Descartes's theory of vortices or Newton's theory of gravitation are speculative cosmology. If, when Parmenides denied the existence of "not-being," he simply meant to deny that there is empty space, then he was working in the same tradition.

But in fact he was doing something very different. In support of his cosmological theory, Thales might have adduced such "facts" as that when men get hot their flesh turns to water. Parmenides could produce no comparable evidence to prove that there is no such thing as not-being. His argument is of a quite different kind: "Thou canst not know that which is not (that is impossible), or utter it; for the same thing can be thought as can be." The speculative cosmology of the physicist describes the world in terms of special types of physical objects and physical processes; the speculative cosmology of the philosopher (let us rather say, following P. F. Strawson, the philosopher's "descriptive metaphysics") is expressed in terms of such logical concepts as thing, property, substance, individual, and process. These are concepts which can easily be applied to the nonphysical, should it turn out that anything is non-physical.

If Heraclitus took the view that everything is made of fire, he was a speculative cosmologist; if he took the view that everything is in process, he was a descriptive metaphysician. (Of course, he might have been both or he might not have been sure of what he was doing.) The difference between Russell's "logical atomism," which is descriptive metaphysics, and the atomism of the physicists, which is speculative cosmology, is quite apparent. Descriptive metaphysics, we might also say, is what Aristotle's definition of "first" philosophy as an inquiry into being as being reduces to, once we get rid of the notion that being is either a predicate or an attribute; it is what G. E. Moore had in mind when he said that philosophy is an attempt to arrive at "a general description of the whole universe."

If we can now distinguish with sufficient clarity between cosmological speculation and descriptive metaphysics, the fact remains that, as we can see from Aristotle's Physics, the distinction was not an easy one for men to make. Indeed, the philosophers' failure to distinguish clearly between speculative cosmology and descriptive metaphysics was one of those fruitful mistakes which, as Leonard Goddard has suggested, are largely responsible for the emergence of sciences within philosophy. It is because philosophers did not know quite what they were doing -- in what sense, for example, the concept of motion was important to them -- that they continued to work at what we now describe as problems in physics.

Philosophy as a theory of language.

What sort of inquiry, then, is descriptive metaphysics? It is easy enough to see how we can discuss the question whether everything is made of water, but how are we to discuss whether, for example, things are reducible to complexes of qualities? In his Logical Syntax of Language, Rudolf Carnap suggested that descriptive metaphysics is about the language in which we describe the world, whereas cosmology is about the world itself. Carnap admits that "thing," "property," and "relation," are not on the face of it names for elements in a language. But this, he argued, is precisely why philosophers in the past have gone astray. Although they have written as if they were talking about the world, what they have said makes sense, insofar as it does make sense, only if it is reformulated as a statement about language. For example, to ask whether things are reducible to complexes of qualities is simply to ask whether language -- or, more precisely, some particular language -- contains distinct thing-words or only conjunctions of quality-words. Quite generally, the task of philosophy is to describe the (actual or possible) language of science.

In a less formal way, the doctrine that the distinctive feature of philosophy is its concern with language has also been characteristic of twentieth-century philosophy in Great Britain. In both cases the influence of Wittgenstein has been predominant. But whereas for Carnap the philosopher is interested only in the language of science (and this is obviously too narrow a definition of the philosopher's task), in Great Britain the stress has been on "ordinary language." British philosophers have continued to work on the assumption that philosophy is as much about religion, morality, or indeed about common-sense beliefs as it is about science. But the attempt to show that philosophy is only about "the language" of these forms of inquiry gives rise to an intolerable stretching of the word "language."

The general thesis that philosophy is "really about language" may well turn out to have the same fate as the thesis that philosophy is "really about human nature." It arose in much the same way. When, in the nineteenth century, the "moral sciences" were gradually transformed into psychology and the social sciences, and even logic, in its more technical sense, was converted into a branch of mathematics, the philosopher seemed to be left with nothing to do. The distinction between "language" and "the world" has, in an important sense, replaced the older distinction between the inner world and the outer world. It provided the philosopher with an area in which to work -- one that satisfied the traditional requirement of width of scope and at the same time left "the world" to science. There was, it is true, some difficulty in explaining just how the philosopher's concern with language differed from the philologist's. But as long as philology was largely etymology, the distinction, although hard to pin down, at least had some justification. Now that philosophy has helped to create new forms of linguistics, however, the exact manner in which philosophy can be "about language" without being a scientific discipline has become more and more obscure. Once again, the attempt of the philosopher to carve out a special field for himself has been frustrated by the growth of science, and once again, that attempt assisted the development of the science which superseded it.

Philosophy as the theory of critical discussion.

"Every philosophical problem," Russell wrote in Our Knowledge of the External World, "when it is subjected to the necessary analysis and purification is found to be not really philosophical at all, or else to be, in the sense in which we are now using the word, logical." (By calling it "logical" he meant that it "arises out of the analysis of propositions.") Post-Russellian philosophers have substituted "analytic," "conceptual," or "linguistic" for Russell's "logical." The resulting definition, we have been suggesting, is at once too broad and too narrow: not all conceptual, analytic, or linguistic problems are philosophical and not all philosophical problems are conceptual, analytic, or linguistic -- that is, they cannot be solved merely by getting clearer about the way in which an expression is used, by removing an ambiguity, or by drawing attention to a temptation implicit in our ordinary ways of talking.

Russell's original definition, on the face of it, suffers from the same defect: controversies about explanation, for example, do not "arise out of the analysis of propositions." Perhaps we can modify Russell's definition by suggesting that the central problems of philosophy -- although not all problems of philosophy -- turn around the analysis of critical discussion. This is the view, which this article will now develop and defend, that the peculiarity of philosophy as a form of critical discussion lies in its being a critical discussion of critical discussion.

This, at least, is the philosopher's home ground, the area in which he must be expert. From it he can move out into a wider variety of problems, although always with reference to the way in which discussion operates. Thus, for example, most books about art are not philosophical, but a philosopher may choose to write about art. If he does so, however, he will take as his central point of reference such problems as these: how it is possible to discuss fruitfully the character and the qualities of a work; whether a work of art can itself be true or false; and whether appreciating a work of art involves a peculiar (or any) sort of knowledge. That is why a philosopher can make important contributions to the philosophy of art without being in any sense an expert on art. These are questions which arise as much in relation to a comic strip or a chocolate-box landscape as in relation to Hamlet or Guernica.

The view that we are suggesting accords with recent discussions about the nature of philosophy insofar as it regards philosophy as being a "meta-inquiry" (an inquiry about inquiry) rather than as running parallel to science and to history, differing from them only in subject matter and method. It allows that the philosopher needs to engage in investigations which are in a broad sense "linguistic," but only insofar as an understanding of how language works is essential to an understanding of discussion. It also allows that the philosopher makes use of, and is interested in, methods of analysis; but it does not see in the practice of analysis the distinguishing mark of philosophical inquiry. Since historical, legal, scientific, and moral discussions certainly have peculiarities that require independent investigation, this view leaves room for a philosophy of history, of law, of science, and of morals. However, it does not allow that the philosopher's concern is with the "meaning" of these activities.

On the face of it, no doubt, this definition also is unduly narrow. Although philosophers have devoted a good deal of attention to truth and falsity, and to meaning, evidence, and proof (all of which are involved in an analysis of discussion), they have also, for example, written at length about the theory of perception. To discuss perception is not, at first glance, to advance the understanding of discussion. But if we look at the way in which problems of perception arise for the philosopher, we see that they are intimately linked to the analysis of discussion.

In discussion we commonly regard certain types of remarks as clinching statements. At a popular level, such remarks would include "I saw it with my own eyes." Very early, however, the development of Milesian cosmology cast doubt on the reliability of sensory perception; it was generally agreed by the cosmologist that whatever the world is like, it is certainly very different from what we ordinarily take it to be. In Plato's Theaetetus, the point at issue between Protagoras and Plato was whether "perception is knowledge" or, as we might put it, whether problems can be decisively settled by an appeal to what we perceive. Plato argued that the appeal to perception is useless for this purpose, and that indeed the whole conception of a rational discussion collapses if perceptual judgments are taken as final.

In order to make his point, he was obliged to work out a relatively detailed theory of perception; in this he was imitated by his successors. As a result, many issues which were later to form part of psychology were first discussed by philosophers. But what, in general, philosophers were looking for in their epistemological writings were substitute clinching statements. Recognizing that "I saw it with my own eyes" is not decisive, they went in search of propositions (for example, "There is now a red sense datum") that would be decisive. The central points at issue in epistemology are whether "I see a table in the room" is or is not an inference; whether "I remember that happening" has a different evidential value from "I can see that happening"; and whether there are clinching statements which are not statements of perception. If we ask at what point the details of perception cease to be of importance to the philosopher, the answer is that they cease to be of importance to him when they no longer affect such questions as whether "I saw a table" is an inference. But there is no way of telling in advance what that point will be: a particular philosopher may penetrate deeply into physiology and psychology in pursuit of the answer to the problems which trouble him.

For this reason, the contrast between philosophy as "about inquiry" and science as "about the world" must not be made too sharply. Critical discussion is one way in which human beings try to come to terms with the things around them; it is only to be expected that the analysis of discussion will lead, at many points, into statements about "the world." Even if the philosopher of art can make important contributions to the philosophy of art without being an expert on art, as was suggested above, it by no means follows that he will never be led into a discussion of the qualities of a particular work of art. The peculiarity of philosophy consists only in the fact that the questions it asks about the world refer back to, and are considered with reference to, the general character of discussion.

Fields of philosophy.

There are important parallelisms between the main concepts philosophers use in analyzing a discussion and the concepts they use in descriptive metaphysics. If, for example, the philosopher embarks upon a discussion of explanation, he soon finds himself confronted by the question whether there is a peculiar kind of explanation describable as "causal" explanation, and if so, what its peculiarities are. Again, in analyzing discussion it is natural to draw a distinction between what a person is talking about and what he says about it; such descriptive-metaphysics distinctions as those between substance and attribute and between objects of acquaintance and objects of description arise naturally out of, and are only discussible in relation to, the analysis of discussion. In Theaetetus, Plato argued, against Heraclitus, that unless there are unchanging entities, there can be no such thing as discussion. Wittgenstein maintained in the Tractatus that the world must consist of simples, because otherwise none of our statements could be intelligible. Without begging the question whether descriptive metaphysics is simply reducible to a set of assertions about the elements in a fruitful discussion, we can say at least that the two are related to each other in a very intimate way, so much so that to regard the central problems of philosophy as turning on the analysis of discussion does not exclude descriptive metaphysics from philosophy.

Nor does it even exclude, at the outset, transcendental metaphysics. For the transcendental metaphysician commonly sets out to show that the existence of transcendental entities is implicit in the practice of critical discussion. He tries to show, for example, that there must be a single subject which all our statements are about, or else they will turn out not to be about anything; that in our practice of seeking explanations, there is implicit the assumption that there is an ultimate explanation, something whose nature is such that it must exist; that it is nonsense to say that ethical controversies about what is right can be settled by an appeal to ends, unless there is an ultimate end, and so on. We might go on to reject transcendental metaphysics on the ground that in fact it professes to, but cannot possibly, solve problems about discussion by appealing to the undiscussible. But our definition does not rule out transcendental metaphysics a priori.

Of all forms of critical discussion, the most developed is certainly science. It is not surprising, therefore, that philosophers have taken a quite special interest in the structure of scientific discussion. At the other extreme, ethical, aesthetic, and political discussions are notoriously unsatisfactory. The philosopher interests himself in such topics as the good, the beautiful, and the public interest, just because the mechanism for discussing differences of opinion about them strikes him as being inadequate. Such discussions, however, are so confused that the philosopher may find himself obliged to look in more detail at the whole character of ethics, of aesthetics, and of political theory. He can discuss science on the presupposition that everybody knows what a scientific problem is and would agree that such and such scientific statements are accepted as being true. He cannot with the same confidence set about discovering the characteristics of a successful ethical, aesthetic, or political discussion; he has to consider which, if any, statements in these fields are true and what the subject, as a whole, is attempting to do. In fact, insofar as ethics is concerned, the subject still lies almost entirely in the hands of philosophers. But except for the fact that the nature of moral discussion is particularly difficult to analyze, on the face of it there is no reason why philosophers should take any special interest in rectitude or goodness, and many philosophers have not done so.

Legal discussion falls, in some respects, between aesthetic discussion and scientific discussion. Within a legal system, there is fairly general agreement about what is and what is not "a good discussion"; what counts as evidence; and under what circumstances somebody's conduct can be held to explain why a subsequent course of action occurred. At the same time, however, legal criteria may be difficult to reconcile with our more general views about, for example, causal explanations of conduct. The philosopher will need to look very carefully at the workings of a legal system before he can pronounce on such points. Insofar as legal systems form part of "the world," he has to examine "the world" in order to understand their peculiarities. Indeed, it is impossible to talk fruitfully about any form of discussion without considering to some degree its nature as a mode of life and its general aims and character. This is one reason that investigations along the general lines of "philosophies of the spirit" -- although without their idealistic presuppositions -- are of more than marginal interest to the philosopher.

Many philosophers have taken a special interest in such social institutions as education and censorship. This is not surprising. Education itself is, or can be, a training in critical discussion. Furthermore, the discussion of education and the attempt to settle controversies about the organization of educational institutions not uncommonly involve a confusion of theoretical, practical, and moral issues, which a philosopher may properly attempt to disentangle. Similarly, defenses of censorship often rest on a doctrine of "sacred truths" or of undiscussible issues, which are naturally of importance to the analyst of discussion.

The philosophical problems which seem most remote from the analysis of discussion are those which turn around what are often called the great issues -- God, freedom, and immortality; mind and nature. But consider, for example, the difference between a neurophysiologist talking about mind and body and a philosopher talking about the same topic. The neurophysiologist tries to find out which cells in the brain are involved in particular forms of mental activity. The philosopher is concerned with quite different questions, such as whether everyday explanations of human behavior in terms of reasons, motives, and intentions are or are not compatible with explanations of human behavior in physiological terms. It is the meshing of two different types of explanations that presents him with his problem.

Similarly, God is important to the philosopher in three ways -- first, insofar as it has been argued that the proposition "God exists" is both existential and indubitable; second, insofar as "God" is an explanatory concept and the use of God as an explanation involves peculiar difficulties; third, insofar as the question can be raised whether God's existence and nature are discussible at all. "God" is a metaphysical concept of such historical importance in Western culture that the philosopher may well pay more than ordinary attention to it. But God is not intrinsically of greater philosophical interest than, let us say, Bradley's Absolute or the Cartesian cogito.

Similarly, in the case of freedom, the question is whether our habit of discussing human behavior in a special kind of way (regarding as a real issue whether somebody is deserving of praise or blame) is compatible with the supposition that there are sufficient and necessary conditions for a person to act as he does -- that is, with the possibility of explaining his behavior in the ordinary scientific fashion. As for immortality, there have often been doubts about whether this is really a question for philosophy. But insofar as it is, the question is whether it is possible to identify the being who is said to live after death with the living being by any of the ordinary means used in identification -- that is, the means by which we determine whether we are both talking about the same person. Philosophy has only slowly discovered how such issues can be discussed, as distinct from merely being pronounced upon. This process of discovery has gone hand in hand with philosophy's creation of new fields of inquiry, which have brought philosophers face to face with the problem of how their investigations differ from the investigations of the scientist on the one side and the pronouncements of the sage on the other.

Philosophical problems are so interconnected that one could perhaps take some group of topics other than those relating to discussion and show how they can be expanded to fall within the province of traditional philosophy. But there are certain advantages in treating the analysis of discussion as central. For one thing, there can be no doubt that explaining, criticizing, and claiming to know are activities that actually go on, in a sense in which, for example, there can be doubt whether there are ultimate principles. Yet to define philosophy in terms of discussion does not rule out ab initio the search for ultimate principles.

Again, while the definition of philosophy as descriptive metaphysics has its attractions, it is certainly a matter of controversy -- philosophical controversy -- whether there can be such a subject, and how it can be developed. Nor, beginning at this point, is it at all easy to understand how there can be such subjects as "the philosophy of law" or "the philosophy of history," except insofar as a particular theory of law or history raises metaphysical issues. One can see why, for example, philosophers should be interested in what is meant by saying that "good is a nonnatural property," but not in why they should carry their interest in ethics beyond this point. In contrast, as Hare has pointed out, to begin from the nature of ethical discussion is to be led into the very heart of ethics.

Description, prescription, and rational reconstruction.

If we say that philosophy is concerned with types of inquiry, we still have a problem to face: Does it merely describe what goes on in these forms of inquiry? It was stated above that aesthetic discussions were unsatisfactory, and it was presumed that scientific discussions were satisfactory. On the other hand, a good many philosophers (for example, James Ward) would not accept this account of the matter. Scientific discussions, they would argue, are profoundly unsatisfactory unless the question at issue is a purely technological one. The scientist never actually describes the concrete things he purports to be describing: his descriptions are wholly abstract, whereas the things he is describing are concrete. Many of these philosophers would add that aesthetic discussions do not suffer from this defect. Indeed, at any but the abstract technological level, scientific discussion itself is about whether to choose one theory or another; and in the end, this comes down to an aesthetic choice, since it is a mere pretense that argument, whether in the form of calculation or in the form of observation and experiment, can decide the issue.

When faced with such an opposition, how is the philosopher to proceed? One suggestion is that he should not allow himself to call one kind of discussion satisfactory and another unsatisfactory; it is his job, simply, to describe how aestheticians argue, how theologians argue, what considerations they accept, and what forms of reasoning they regard as decisive. This way of looking at the matter, which may be christened "descriptionist," is a natural reaction against the "prescriptionism" of classical rationalism. Classical rationalism begins from the notion of real proof or real explanation or real knowledge. Sometimes it derives this ideal concept from some existing form of inquiry (for example, proof from axiomatic geometry or explanation from physics), and then proceeds to complain that there are no proofs (that is, axiomatic deductions) in philosophy and no explanations (that is, statements of necessary and sufficient conditions) in history. In even more extreme instances (in the case of F. H. Bradley, for example), classical rationalism begins from some supposed "requirement of the intellect" and then asserts that not even mathematics really proves and not even physics really explains -- much as for Russell not even our best friends are "objects of acquaintance" and not even "Socrates" is really a proper name.

It is easy to understand why, in reaction to this attitude, philosophers should begin to argue that the philosophers' concern is to describe how, let us say, historians actually reason, without attempting to impose on historians a concept of good reasoning derived from "ideal considerations." Yet even if he wishes to, the philosopher cannot restrict himself to mere description. Suppose, for example, the philosopher of science attempts to "describe what the scientist does." Then he has first to determine what he is going to count as science. Is he going to include Marxism and Freudian psychology or will he, with Karl Popper, exclude them from science? To say that he will let the scientists "determine what is to count as science" is to be involved in a vicious circle, for how is he to decide who are the scientists? Hence, a "pure descriptionism" could never even get started.

But this does not mean that we are forced back to pre-scriptionism. There is a third possible approach, "rational reconstruction." To employ the method of "rational reconstruction" is neither simply to prescribe nor simply to describe. A satisfactory reconstruction of explanation, for example, will almost certainly not apply to everything which has previously gone under that name. It may well reject, for example, explanations in terms of the guiding hand of Providence, not on the ground that they are false but because they are not the sort of thing which ought to be reckoned as an explanation; they do not enable us to account for the fact that things happen in one way rather than in another.

So far, rational reconstruction is prescriptionistic, but not arbitrarily so. The rational reconstructor will try to show that there is no way of distinguishing between explaining and other forms of human activity (for example, reasserting), unless some of what have ordinarily been called explanations are excluded as pseudo explanations. Quite similarly, the test of whether a "rational reconstruction" of philosophy is a reasonable one is whether it provides us with a method of demarcating philosophy from other forms of inquiry, even if in order to do so it is obliged to exclude much that ordinarily passes as philosophy. The rational reconstructor is not so much prescribing as drawing attention to a difference.

Furthermore, he is a descriptionist to the extent that he will expect his theory of explanation, for example, to apply to a wide range of what have ordinarily been called explanations. He bases his reconstruction, not on a priori views about what kind of explanation would be fully satisfactory to the intellect, but rather on what counts as an explanation in developed forms of inquiry. His rational reconstruction of philosophy may exclude Kierkegaard but must apply, at least in large measure, to the work of Plato, Aristotle, Descartes, Hume, and Kant.

Obviously, there can be disagreement about whether a reconstruction is rational. Some people will think it so intuitively obvious that Kierkegaard and Kant are both engaged in the same sort of activity that they will reject as prescriptionist any rational reconstruction which has the effect that Kierkegaard is not a philosopher; or they will think it so obvious that Marxism explains what happens in history that they will reject out of hand any theory of explanation which denies that the materialist interpretation of history can serve as an explanation. Controversy on this point is all the more difficult because words like explanation, philosophy, science, and knowledge are used eulogistically. But this is only to say that agreement in philosophy is difficult to secure -- which is scarcely news.

The variety of philosophical tasks.

The analysis of discussion raises problems of such proportions that difficulties arising out of an attempt to solve them, rather than the original problems themselves, can come to be the sole preoccupation of philosophers. To take an instance, many philosophers have greatly interested themselves in the peculiarities of mathematical discussion. A philosopher may well restrict his attention almost entirely to the analysis of that particular variety of discussion, asking what conditions have to be fulfilled by an adequate proof, what role is played by axioms and by definitions, what exactly mathematical propositions tell us, and so forth. In the course of such a critical discussion of mathematics particular suggestions are made -- for example, that mathematical propositions have the peculiarity of being synthetic a priori or of being tautologies. A philosopher may choose to concentrate almost entirely upon the question whether, and how, it is possible to distinguish between synthetic a priori propositions and analytic propositions.

What he says in the course of his attempt to settle this question may provoke further controversy about whether, for example, analytic propositions are "really about language." In the end the philosopher may be engaged in highly technical controversies which seem to have very little to do with the theory of discussion. But the point of such controversies, the reason that they are something more than technical exercises, is the light they throw on the connection and distinction between mathematical and scientific reasoning.

Similarly, a philosopher may concentrate his attention upon the problem of "counterfactual conditionals." But in this case, too, the point at issue is how a person who utters such a counterfactual conditional as "If Caesar had not crossed the Rubicon, the Republic would still have fallen" can, by that means, make a contribution to discussion. If this question is forgotten, controversy about counterfactual conditionals can easily collapse into scholastic pedantry.

With this reservation, however, it is quite natural that there should be a great many philosophers with interests of a highly technical nature that find expression in short studies on technical points rather than in large-scale constructions. At the other end of the spectrum, there are philosophers whose main interest is in the interrelation of such forms of human activity as religion, science, and art. These philosophers run the serious risk of woolliness, emptiness, or arbitrariness, unless they take as their central point of reference the structure of religious, scientific, and aesthetic discussion. There is room within philosophy for an immense variety of types of investigation, some very minute and some highly generalized. Nor is it a matter of any importance if a philosopher, in trying to solve a particular problem, passes outside the boundaries of philosophy altogether. In the end, it is problems, not the divisions between subjects, that are crucial.

Alexander, Samuel, Space, Time and Deity. London, 1920.

Austin, J. L., "Ifs and Cans," in Philosophical Papers. Oxford, 1961. Pp. 179-180.

Ayer, A. J., Language, Truth and Logic. London, 1936.

Bacon, Francis, De Dignitate et Augmentis Scientiarum. London, 1623. Especially Book III, Ch. 1.

Bergson, Henri, "L'Intuition philosophique," in La Pensee et le mouvant. Geneva, 1934. Translated by Mabelle L. Andison as The Creative Mind. New York, 1946.

Bradley, F. H., Appearance and Reality. London, 1893.

Broad, C. D., "Speculative and Critical Philosophy," in J. H. Muirhead, ed., Contemporary British Philosophy, First Series. London, 1924.

Carnap, Rudolf, Logische Syntax der Sprache. Vienna, 1934. Translated by Amethe Smeaton as Logical Syntax of Language. London, 1937.

Cicero, Marcus Tullius, De Finibus ("Concerning Ends"), translated by H. Rackham. Loeb Classical Library, No. 40. Cambridge, Mass., 1914. III, 2, 4.

Collingwood, R. G., Speculum Mentis. Oxford, 1924.

Croce, Benedetto, Filosofia come scienza dello spirito, 5 vols. Bari, Italy, 1909-1917. Translated by D. Ainslie as Philosophy of the Spirit. London, 1909- 1921.

Descartes, Rene, Philosophical Works, 2 vols., translated by E. S. Haldane and G. R. T. Ross, Cambridge, 1911. Vol. I includes Meditationes de Prima Philosophiae (1641) and the Preface to the French translation of Principia Philosophiae (1650).

Dewey, John, Reconstruction in Philosophy. New York, 1920.

Ducasse, C. J., Philosophy as a Science. New York, 1941. Contains the best general discussion of the word "philosophy,"

Goddard, Leonard, Philosophical Thinking. Armidale, New South Wales, 1962. Inaugural lecture.

Hamilton, William, Lectures on Metaphysics and Logic, 4 vols. Edinburgh, 1859-1860.

Hare, R. M., Freedom and Reason. Oxford, 1963.

Hegel, G. W. F., Encyclopadie der philosophischen Wissenschaft. Heidelberg, 1817. Part III. Translated by William Wallace as Hegel's Philosophy of Mind. Oxford, 1894.

Heidegger, Martin, Was ist das -- die Philosophie? Pfullingen, 1956. Translated by W. Kluback and J. T. Wilde as What Is Philosophy? London, 1958.

Hobbes, Thomas, De Corpore. London, 1655. Translated as Concerning Body. London, 1656.

Hume, David, A Treatise of Human Nature. London, 1739.

Husserl, Edmund, Logische Untersuchungen, 2 vols. Halle, 1900-1901.

Husserl, Edmund, Ideen zu einer reinen Phanomenologie und phänomenologischen Philosophie, Vol. I. Halle, 1913. Translated by W. R. Boyce Gibson as Ideas. London, 1931.

James, William, Some Problems in Philosophy. New York, 1911.

Lange, F. A., Die Geschichte der Materialismus, 2d ed. Leipzig, 1873-1875. Translated by E. C. Thomas as History of Materialism. London, 1877.

Leibniz, G. W., De Vita Beata ("Concerning the Blessed Life"), first published in Opera Philosophica, J. E. Erdmann, ed. Berlin, 1840.

Lotze, (Rudolf) Hermann, Metaphysik. Leipzig, 1841. Translated and edited by Bernard Bosanquet as Metaphysic, 2 vols. Oxford, 1884.

Meinong, Alexius, Untersuchungen zur Gegenstandstheorie und Psychologie. Leipzig, 1904.

Mill, J. S., Auguste Comte and Positivism. London, 1865.

Moore, G. E., Some Main Problems of Philosophy. London, 1953.

Passmore, J. A., Philosophical Reasoning. London and New York, 1961.

Passmore, J. A., "The Place of Argument in Metaphysics," in W. E. Kennick and Morris Lazerowitz, eds., Metaphysics. Englewood Cliffs, N.J., 1966.

Popper, K. R., Logik der Forschung. Vienna, 1935. Revised English version, The Logic of Scientific Discovery. London, 1959.

Popper, K. R., Conjectures and Refutations. New York, 1962.

Raeymaker, Louis de, Introduction a la philosophie. Louvain, 1938.

Russell, Bertrand, Our Knowledge of the External World. London, 1914.

Schlick, Moritz, Gesammelte Aufsätze. Vienna, 1938.

Sidgwick, Henry, Philosophy, Its Scope and Relations. London, 1902.

Spencer, Herbert, First Principles. London, 1862.

Spinoza, Benedict, Ethica, in Opera Posthuma. Amsterdam, 1677. Translated by W. H. White and A. H. Stirling as Ethics. Oxford, 1927.

Strawson, P. F., Individuals. London, 1959.

Ueberweg, Friedrich, Grundriss der Geschichte der Philosophic, 3 vols. Berlin, 1862-1866. Translated by G. S. Morris as The History of Philosophy. London, 1871. Contains a discussion of the word "philosophy."

Walsh, W. H., Metaphysics. London, 1963. Contains one of the best general discussions of the word "philosophy."

Ward, James, Naturalism and Agnosticism. London, 1899.

Whitehead, A. N„ Process and Reality. Cambridge, 1929.

Wittgenstein, Ludwig, Tractatus Logico-philosophicus, translated by C. K. Ogden. London, 1922. Corrected translation by D. F. Pears and B. F. McGuinness. London, 1961.

Wittgenstein, Ludwig, Philosophical Investigations, translated by G. E. M. Anscombe. Oxford, 1953.

Wolff, Christian, Philosophia Rationales. Frankfurt, 1728.