Recent Issues

February 2004 Contents

Cover / In This Issue

Society News

Russell on Nuclear Deterrence

Generality and Disimilarity

The Russell Cambridge Companion

Early News Reports on Russell

Traveler’s Diary

a faithful companion

Kevin C. Klement

Review of The Cambridge Companion to Bertrand Russell, Nicholas Griffin, ed. Cambridge: Cambridge University Press, 2003. 550 pp. + xvii. $75 hardcover, $26 paperback.

We can at last release our breath: the long-awaited Russell volume in the popular Cambridge Companion series has finally arrived. It contains fifteen chapters written by well-known Russell scholars dealing with a wide array of Russelliana, along with a quite extensive introductory essay by the volume editor. It is not difficult to see what took so long. Russell’s corpus, even considering only his philosophical writings, outstrips in both breadth and volume almost all the other figures covered in the Cambridge Companion series. A further complication in Russell’s case is his characteristic habit of so frequently changing his mind even about fundamental issues. Dealing with such a vast amount of information must have required a tremendous amount of sustained collaboration. Obviously, the volume could not cover everything; but the editor and authors have done a tremendous job selectively choosing topics and themes within Russell’s philosophical work to focus on. While falling short of perfection, the result is a collection of pieces that together provide the sort of sophisticated introduction to a complex philosopher that is able to make his work accessible to relative beginners without disguising the subtlety, complexity and still controversial nature of his views.

Griffin’s introductory essay provides the requisite biographical information on Russell, along with a summary of the evolution of his philosophical views. His discussion of those views is terse, but this is understandable given that most are treated in greater length in the pieces that follow. The value of the introduction is that it provides an overall framework and chronology in which to situate the more detailed discussions that follow.

(1) The first chapter is entitled “Mathematics In and Behind Russell’s Logicism, and Its Reception,” written by Ivor Grattan-Guinness. It describes how Russell first became interested in the foundations of mathematics in the 1890s, and how his interests were transformed in 1900 and the following years by the influence of Giuseppe Peano, his associates, and others, to grow into Russell’s logicist project. It also describes the changes in Russell’s thinking brought about by the discovery of the set-theoretic paradoxes plaguing his initial formulations of logicism, his realization that his earlier proofs of an actual infinity were fallacious, and the changes to his treatment of mathematical functions with the discovery of the theory of descriptions. Grattan-Guinness also discusses the details of Russell’s collaboration with Whitehead, the writing process of Principia Mathematica, and its reception and influence among mathematicians in the decades following its initial publication.

(2) This first chapter is nicely complemented by the second chapter, entitled “Russell’s Philosophical Background,” by Griffin. Here we find discussion of Russell’s inculcation into the mindset of British (largely neo-Hegelian) idealism during his study at Cambridge, and detailed treatment of Russell’s positions during his early idealist phase. The essay immediately shows the subtlety and complexity of Russell’s philosophical thinking even during this early period, and helps counterbalance the tendency – promulgated by later Russell himself – to think of this early idealist work as simply a host of confusions engendered by rejecting relations. Russell’s positions on such matters as the nature of relations, the debate over monism and pluralism, the dependency of mathematical and geometrical truths on the mind or experience, and so on, are far more sophisticated than is generally acknowledged, as Griffin aptly demonstrates.

(3) The next piece, by Richard Cartwright, is entitled “Russell and Moore, 1898-1905.” This entry discusses the break with British idealism made by Russell and Moore in the late 1890s and their adoption of a robust realism, including commitment to propositions as mind-independent objects of belief. Russell credited Moore as leading the way in the development of this “new philosophy” (as he called it in 1903). Cartwright discusses how further investigations into the nature and make-up of propositions developed into Russell’s doctrines of philosophical logic exposited in the Principles of Mathematics, and outlines certain major features of these doctrines with regard to ontological commitment, the nature of relations, necessity and change.

(4) Michael Beaney follows with a similarly titled entry, “Russell and Frege.” Frege and Russell are together often heralded as the two primary founders of analytic philosophy, and the two primary forces behind logicism in the philosophy of mathematics and the “revolution in logic” that lead to the abandonment of Aristotelian syllogistic logic in favor of modern quantificational logic. Beaney charts Frege’s main contributions to logic and the philosophy of mathematics, such as the development of quantificational theory capable of treating multiple generality, the definitions of hereditary properties and ancestrals of relations, the analysis of equinumerosity in terms of one-one correspondence, and the resulting definition of cardinal number. He then discusses their relationship to Russell’s views, and compares and contrasts their views on the importance of relations and order, Russell’s paradox, the unity of propositions or thoughts, and the nature and purpose of philosophical analysis. Beaney also discusses their joint influence on analytic philosophy.

Disappointingly, the entry does not discuss much regarding the influence of the two philosophers upon one another (even negatively), nor does it delve into their very interesting correspondence beyond the initial letters concerning the contradiction in Frege’s logical system. In the first chapter, Grattan-Guinness had suggested that many commentators exaggerate the influence of Frege on Russell. Perhaps Beaney would agree since he does not mention a single way in which Russell’s views changed due to his reading of Frege. While it is no doubt correct that Russell did not adopt many views directly from Frege, and the most well known points of overlap between them are views they developed independently, Russell’s confrontation with Frege’s views in the years 1902-1905 lead him to rethink many of his own views on the nature of classes, functions and meaning, and while the final views Russell adopted do not coincide with Frege’s, it is unlikely they would have taken the form they did without Frege’s influence. (See, e.g., Klement 2003.)

(5) The fifth chapter bears the title “Bertrand Russell’s Logicism,” and is coauthored by Martin Godwyn and Andrew Irvine. It begins with a brief discussion of earlier logicist theorists, then sketches (what the authors take to be) Russell’s “new” type-theoretic form of logicism, which attempts to solve the contradictions plaguing Frege’s form, moves on to a discussion of Russell’s ontological commitments, or lack thereof, to such entities as numbers, propositional functions and classes, and ends with a discussion of Russell’s epistemology of mathematics. For example, while Russell thought that mathematical claims such as “2 + 2 = 4”, could, when properly analyzed, be deduced from purely logical axioms, he thought that, epistemologically, the mathematical truths were more certain, and that indeed, non-self-evident logical principles are sometimes to be justified in virtue of the epistemological status of their logical consequences. Russell therefore did not share the epistemological goals of those other logicists who hoped to secure the epistemological status of mathematics by showing it to be reducible to self-evident logical principles.

However, much of the remainder of the essay is either redundant or out of sorts with other chapters on related topics in the volume. The chapter begins with a discussion of Leibniz, Frege and Dedekind, but does not make it clear to what extent the details of Russell’s logicism were influenced by these figures, and in any case the discussion seems redundant given Grattan-Guinness’s more sophisticated look at the historical background to Russell’s logicism. The descriptions of both simple and ramified type theory are unrecognizable when compared to Russell’s actual writings, and seem to owe more to later formulations of type-theory by logicians such as Tarski and Church than to Russell’s own work. Their claim that Russell’s 1908 “Mathematical Logic as Based on the Theory of Types,” abandoned Russell’s 1905 “no-classes theory” in favor of a new approach directly contradicts Landini’s claim later in the Companion that the substitutional theory (a direct descendent of the 1905 “no-classes theory”) undergirds the logical system of that paper. Their acceptance of Quine’s criticism that Principia Mathematica’s second-order logic is based on a confusion of use and mention, and therefore, no more a reduction of mathematics to logic than a reduction of mathematics to set theory, ignores the responses made by sympathetic commentators in the past few decades (see, e.g., Sainsbury 1979, Chap. 8; Hylton 1990, pp. 217-218; Landini 1998, Chaps. 9-10, Linsky 1999, Chap. 6).

(6) This is followed by a chapter written by Peter Hylton entitled “The Theory of Descriptions.” This entry begins with a summary of the mechanics of Russell’s influential analysis of descriptive phrases within first-order logic, then attempts to place Russell’s 1905 discovery of this theory within the context of his developing philosophical views. Rival theories such as Frege’s distinction between sinn and bedeutung and even Russell’s own earlier theory of denoting concepts involve an indirect sort of representation according to which the thoughts or propositions we entertain, instead of containing the entities they are about, contain intermediate entities (senses or meanings) that represent the entities they are about. These theories are out of sorts with the direct realism Russell had adopted in his rejection of idealism, and according to Hylton, this is Russell’s primary motivation for adopting the theory of descriptions in their stead. Perhaps wisely, however, Hylton devotes only a paragraph’s worth of discussion to the arguments found in the infamous Gray’s Elegy passage of “On Denoting” against theories similar to the theory of denoting concepts, noting that space constraints rule out full consideration of the argumentation there. Instead, Hylton moves on to address the importance of the theory of descriptions for Russell’s philosophy after 1905, and finally discusses a number of influential objections to Russell’s theory which have surfaced since 1950. Interestingly, one lesson Hylton conveys is a warning against the traditional interpretation that Russell’s primary motivation for the theory of descriptions was the avoidance of (“Meinongian”) ontological commitment to nonexistent entities such as the round-square, the present King of France, the planet Vulcan, and so on, noting that this seems like the central motivation only in retrospect. This lesson is apparently still worthwhile, given that even other authors in Companion still focus on this aspect of the theory when presenting it (e.g., Beaney in Chapter 4, p. 162).

(7) The seventh chapter, by Gregory Landini, is entitled “Russell’s Substitutional Theory,” and deals with the highly original and interesting logical system adopted by Russell from 1905-1907 to solve the paradoxes facing logicism in which the notion of ontological substitution of one entity for another within propositions as objective complexes is taken as fundamental. Specifically, it employs a four place relation written “p/a;b!q”, which means that q results from p by substituting b for a. For example, this relation would hold when p is the proposition Socrates is wise, a is Socrates, b is Plato, and q is the proposition Plato is wise. (Here we are dealing with the substitution of the man Plato for the man Socrates within a mind-independent proposition, and not the substitution of one name for another within a sentence.) This logical system is strictly speaking type-free and employs only one style of variable – ranging over all entities whatever (including propositions) – and yet is able to proxy or do all the work required of a higher-order logic employing a simple theory of types, including providing a replacement for talk about sets or classes.

Landini sketches in some detail the origin and nature of Russell’s substitutional logic, as well the changes that it underwent as he encountered certain problems: e.g., the abandonment of quantified propositions as entities in his 1906 “On ‘Insolubilia’ and their Solution by Symbolic Logic,” as a way of resolving certain contradictions present in his initial formulations of the theory. Landini goes on to discuss that which eventually lead Russell to abandon the substitutional theory. However, against many traditional interpretations, Landini argues contentiously that certain key doctrines explicitly realized in the substitutional theory, such as the doctrine of the unrestricted variable, are maintained in a disguised form even in Principia Mathematica when one properly understands its semantics. Landini concludes that the substitutional theory is the “conceptual linchpin” connecting Russell’s work in the Principles of Mathematics with his mature logical system, and thus any proper understanding of the latter must involve an understanding of its relationship with the substitutional theory.

(8) Alasdair Urquhart follows with a contribution entitled “The Theory of Types,” which aims to summarize Russell’s type-theory, its historical roots and influence within logic, mathematics and computer science. It begins with a short discussion of Russell’s early 1903 theory of types found in Appendix B of the Principles of Mathematics and its demise, mentions briefly Russell’s intermediate non-type-theoretic solutions to the contradictions attempted from 1902-1907, and then moves on to a discussion of the more complicated ramified theory of types found in Principia Mathematica. Urquhart notes the importance of the “vicious circle principle”, stated by Russell as the principle that “whatever involves all of a collection must not be one of the collection,” in providing philosophical support for ramification. I think Urquhart perhaps gives it too large a place and applies it too sweepingly, given that for Russell the principle was not thought to be “itself the solution of the vicious-circle paradoxes, but merely the result which a theory must yield if it is to afford a solution of them” (Russell 1906, p. 205). While Russell accepts the vicious-circle principle, it is not the philosophical rationale or explanation of ramification, but a result of it.

Urquhart moves on to a summary of the technical details of ramified type-theory, but explicitly bases his exposition not on Russell’s own, but instead on later formulations of ramified type-theory given by Church, Myhill and others, explaining that the “original presentation in Principia Mathematica is both imprecise and notationally clumsy, ... [and] there is no precise presentation of the syntax of the system” (p. 295). Given that the Companion is supposed to provide a philosophical entrée to Russell’s own work, this decision is disappointing. Whitehead and Russell’s exposition of the details of their logical language is lacking when compared to modern standards, but this does not mean that an exact statement of what they had in mind would be impossible. There is unfortunately a long precedent of ignoring Russell’s own presentation of his type-theory, and an equally long precedent of attributing to him views he did not hold on the basis of later logicians’ formulations. Thankfully, in recent years there has been a movement away from the precedent. However, Urquhart ignores these attempts to understand Russell on Russell’s own terms, and neglects to mention recent findings and debates about the extent to which Principia’s formal system can be assimilated to later formulations (see e.g., Landini 1998; Chap. 10; Linsky 1999).

Urquhart’s exposition of ramified type-theory also weds that theory to precisely the sort of metaphysics of propositions Russell held prior to adopting the multiple-relations theory of judgment circa 1910. His rationale is that Russell still describes propositions as the values of propositional functions, and therefore they are required as part of the very motivation of the system. However, this is odd given that Russell’s acceptance of ramification seems to coincide chronologically almost exactly with his eschewal of a metaphysics of propositions. Again, Urquhart ignores recent attempts to clarify Russell’s seemingly-inconsistent position (see, e.g., Sainsbury 1980; Cocchiarella 1987, Chap. 5; Rodriguez-Consuegra 1989; Landini 1998, Chap. 10).

Urquhart then discusses the simplifications to ramified type-theory that were developed in the decades following Principia’s publication, especially the simple type-theories developed by Ramsey, and later, by Russell himself for the 2nd edition of Principia (1925). He lastly discusses the fate of type-theory in more recent mathematical and logical work, noting that while axiomatic set theories, based on the work of Zermelo others, are far more popular in contemporary mathematics, the ideas behind type-theory continue to play a role in inspiring certain advances in the foundations of set theory, as well as in the theoretic foundations of programming languages and study of algorithms.

(9) Next we find Paul Hager’s “Russell’s Method of Analysis,” which describes Russell’s self-conscious methodology for philosophical research. This methodology is a two phase process. In the first phase, one begins with a certain body of knowledge or set of “data”, conceived of as propositions within a certain domain of discourse which are thought to be obvious or self-evident, but somehow vague, in need of clarification or unification. The bulk of the first phase, the phase of “analysis”, consists in attempting to discover a number of logically more simple, but less self-evident, premisses or principles, employing a smaller vocabulary, in which a reconstruction of the original body of knowledge is thought to be possible. The second stage of method, the “synthetic” stage, consists in building, reconstructing or demonstrating the original body of knowledge – or at least the indispensable part of it – from the premisses and concepts discovered in the analytic phase. Mathematical examples of this methodology are easily found in Russell’s early work, and Hager goes on to describe Russell’s much later Human Knowledge as an example of this methodology applied to scientific knowledge. Hager argues that this methodology can be seen as representing the strongest continuous thread running though Russell’s philosophical work. Hager makes note of certain misunderstandings regarding the nature of analysis and its relation to language, such as the construal of analysis as having solely to do with the relationship between wholes and their parts, or thinking that analysis does not have to do with both language and the world. He argues that such misunderstandings underlie certain misconceptions about Russell’s work, most recently exemplified in Ray Monk’s well-known biography.

(10) The tenth chapter is entitled “Russell’s Neutral Monism,” written by R. E. Tully. Here we find a lengthy treatment of Russell’s consideration of neutral monism: the theory that there is only one kind of stuff making up reality, which is itself neither fundamentally mental nor physical, but out of which both mind and matter can be thought of as being formed. Tully begins with some philosophical background to Russell’s confrontation with the theory as found in the work of James and others, followed by discussion of Russell’s initial reaction and arguments against it in the early 1910s, stemming mainly from worries regarding its ability to explain fully the nature of first hand experience and its compatibility with the nature of acquaintance. Tully then discusses Russell’s gradual acceptance of the theory, at first provisionally in the late 1910s, and then explicitly in his writings in the 1920s and 1930s, as well as the maturation and changes to the doctrine in such works as An Analysis of Mind, An Outline of Philosophy and An Analysis of Matter. He goes on to describe the role the theory has, even when not mentioned by name, in later works such as An Inquiry Into Meaning and Truth and Human Knowledge.

There are a number of passages of the essay that are somewhat unclear, and parts, especially when discussing Russell’s earlier views, in which Russell’s doctrines are misleadingly stated. To focus on a single example, on p. 348, Tully suggests that Russell’s multiple relations theory of judgment was a reaction against a doctrine according to which “propositions are entities occupying an intermediate position between the minds and facts,” a doctrine “associated with Meinong.” In fact, neither Russell nor Meinong ever held such a view. Russell’s early view of propositions did not make them out as being intermediates between the mind and facts, and indeed, on that theory, facts and true propositions were identified. (This point is aptly made in the Companion itself by other contributors, e.g., by Griffin on pp. 27-28, by Cartwright on pp. 110-111, by Landini on pp. 253-255, etc.). The advance of the multiple relations theory was not that it avoided, as Tully suggests, “treating propositions as objects in their own right separate from facts.” Instead, it was that it allowed not treating propositions as singular objects at all. There are similar difficulties elsewhere in the essay; but such small difficulties – given the aim of Tully’s essay – are perhaps forgivable. However, more problematically, nowhere does Tully offer the non-specialist a simple overall statement of Russell’s neutral monism, nor a simple explanation of how Russell or others believed that either physical objects or minds should be conceived on this position. (For this the reader has to wait until Grayling’s contribution later in the Companion, pp. 461-463.) Tully mainly concerns himself with details of the theory, problems within it, or changes to it without giving a simple description of the overall theory.

(11) Next we find a chapter called “The Metaphysics of Logical Atomism,” written by Bernard Linsky. Linsky discusses in general Russell’s characterization of philosophy as an “atomism,” arguing that this should primarily be understood as commitment to analysis as a method coupled with a rejection of idealistic monism, rather than a pretense to have discovered the genuine metaphysical “atoms” making up the world of facts, or even the belief that such a discovery is possible. Linsky also discusses the epistemological aspects of Russell’s logical atomism, his notion of logical construction, as well as a number of related questions regarding the nature of Russell’s metaphysical views on propositions, propositional functions, universals, extensionality, atomic facts and the relationship between logical constructions and eliminative metaphysics, not all of which can be discussed in detail here. I will restrict my comments to two relatively small points. First, Linsky oddly claims that Russell introduces the name “logical atomism” in his 1918 lecture series The Philosophy of Logical Atomism, whereas in actuality, that phrase first occurred in Russell’s writings at least as early as the 1911 “Analytic Realism” paper (see Russell 1992, p. 135). Secondly, Linsky seems to assume that giving a nominalistic reading of Russell’s use of higher-order propositional function variables in his logic would amount to ascribing to Russell a nominalism about universals. However, these two issues are unrelated. On my own interpretation of Russell, he became a nominalist about “propositional functions” as early as 1905, but was never throughout the period in question a nominalist about universals. At least prior to his having been influenced by Wittgenstein, Russell never equated in his mind the propositional function “x^ is red” with the universal of redness – as Linsky knows full well (see Linsky 1999, chap. 2) – and so a realism about the later would not entail a realism about the former. However, I cannot fully elaborate this point here.

(12) William Demopoulos’s contribution, “Russell’s Structuralism and the Absolute Description of the World,” appears next. Demopoulos sketches Russell’s “structuralism”, i.e., his view that perception alone provides us directly at most with knowledge of structural features of the physical world – a view Russell held explicitly from 1919 through 1948, and perhaps implicitly as early as 1912. Demopoulos discusses its relation to Russell’s theories about propositional understanding, and how these lead him to consideration of difficulties regarding the proper interpretation of scientific theories, as well as Russell’s solution taken from the standpoint of the program of logical construction. Demopoulos also discusses certain questionable assumptions within Russell’s position. For example, he sketches Russell’s subjectivist treatment of color vocabulary according to which color predicates such as “yellow” or “blue” are to be understand as first and foremost representing qualities of subjective percepts or sensations, upon which our understanding of these predicates as applied to external surfaces is thought to be derivative. Demopoulos contrasts this with a “relativist” view, according to which while it is admitted that our initial understanding of such predicates is given in terms of perceptual criteria, with the advancement or our scientific understanding of color, this understanding is replaced by an “absolute form” of description that abstracts away from the particularities of our perceptual systems. The pre-theoretic and post-theoretic understandings can nevertheless be co-extensional. Demopoulos sketches certain other difficulties with Russell’s position, and while he does not find Russell’s position to be incoherent, he suggests that his rival position accommodates much of Russell’s insights while ending up as less revisionary with regard to our ordinary discourse and conception of the physical world.

(13) The thirteenth chapter is written by Thomas Baldwin and has the title “From Knowledge by Acquaintance to Knowledge by Description.” Baldwin charts over 35 years’ worth of the development of Russell’s epistemology, beginning with 1912’s Problems of Philosophy and the distinction between knowledge by acquaintance and knowledge by description. He then proceeds to discuss the changes to Russell’s conception of a priori knowledge first made explicit in the 1918 Philosophy of Logical Atomism lectures brought on by his rejection of logical objects due to the influence of Wittgenstein, and his movement towards a more linguistic notion of analyticity and a prioricity. Baldwin continues on to discuss the more radical changes to Russell’s epistemology from 1921’s Analysis of Mind, when Russell abandoned his former understanding of acquaintance as a relation between the mind and non-mental objects in line with his newly adopted neutral monism. Baldwin also discusses how Russell’s epistemological work during this period anticipates later discussions in the theory of knowledge such as the debate between internalism and externalism, as well as the causal and reliabilist theories of knowledge. Baldwin continues on to consider further changes to Russell’s epistemological doctrines in An Outline of Philosophy (1928), and An Inquiry into Meaning and Truth (1940), finally concluding with a discussion of the role causation plays in Russell’s final epistemology in Human Knowledge: Its Scope and Limits (1948). In particular, Baldwin discusses what Russell calls “weakly a priori” truths such as the principle of induction. Unlike standard a priori truths, our belief in these principles cannot be justified by reason alone; however, our belief in them is at least amenable to a sort of causal explanation that shows it to have a kind of validity based on the fact that it reliably leads to other true beliefs.

(14) The penultimate chapter, “Russell, Experience and the Roots of Science,” contributed by A. C. Grayling, sketches Russell’s long-running project of attempting to explicate the relationship between sense experience and scientific knowledge. Grayling argues that it should be understood quite differently from the traditional Cartesian project of attempting to justify scientific claims on the basis of experience. Russell’s task was rather to clarify how the objects of the sensible world and of scientific discourse relate to the data of immediate experience. He first discusses Russell’s approach to the issue in Problems of Philosophy and works of that period, in which Russell conceived the problem as having to do with how we are able, beginning only with our direct acquaintance with sense-data, to achieve “knowledge by description” of the objects of the external world. He then proceeds to sketch how Russell reconceived the project after initially accepting neutral monism, when he abandoned both the distinction between the act of sensing and what is sensed, and the distinction between sense-data or sensations and objects themselves. Baldwin then discusses Russell’s later return to an inferential view about our knowledge of physical objects in The Analysis of Matter, and finally Russell’s naturalistic epistemology in Human Knowledge. The chapter overlaps heavily in theme and substance several previous chapters in the Companion (specifically, those by Hager, Tully, Demopoulos and Baldwin), but Grayling does an admirable job tying together the various themes discussed by others into a unified account of the development of Russell’s philosophy from the 1910s through the 1940s.

(15) In the final chapter, “Bertrand Russell: Moral Philosopher or Unphilosophical Moralist?”, Charles Pigden switches gears and examines Russell’s contributions to moral philosophy. Pigden outlines six phases in the development of Russell’s ethical theorizing, challenging the views of many that Russell’s writings in this area were mostly derivative by highlighting significant points of originality, including influence on Moore’s ethical work, as well as anticipations of both Mackie’s error theory and the emotivism of Ayer and Stevenson. Even more contentiously, Pigden argues against Russell’s own evaluation of his popular writings on political and moral themes as being unphilosophical, noting by way of example that Russell’s call for world government involved a number of philosophically interesting convictions and arguments.

Finally, it should not escape mention that the volume also contains an up-to-date and extensive 36 page bibliography, with separate listings of Russell’s own book-length works, prominent articles, collections, as well most of the important monographs and articles in the secondary literature. The bibliography of course is not comprehensive – remember that the extensive bibliography of Russell’s own writings published by Kenneth Blackwell and Harry Ruja in the mid-90s was itself a three volume affair! The bibliography also contains some minor mistakes: for example, in the listing of Russell’s philosophical articles, those that actually appear in Volume 4 of the Collected Papers of Bertrand Russell are all erroneously listed as appearing in Volume 3. Nevertheless, the bibliography provides an invaluable resource for anyone wishing to pursue further research on any aspect of Russell’s philosophy covered in the Companion.

In summation, the Companion consists of four essays addressing Russell’s logic and philosophy of mathematics, three essays primary concerned with Russell’s philosophical background and interactions with other philosophers, three essays concerned with Russell’s metaphysics and theory of meaning, four essays addressing Russell’s epistemology, philosophy of science and theory of philosophical methodology, and one essay dealing with Russell’s ethics. If Pigden is right that much of Russell’s writings concerning political, social and moral affairs constitute philosophy, then much of Russell’s philosophy is not covered in the Companion, from his early writings on German Social Democracy to his later writings on nuclear warfare and disarmament. No doubt, these omissions will disappoint certain die-hard Russell fans. However, I think by and large the choices regarding coverage were wise. The titles in the Cambridge Companion series are aimed primarily at working philosophers and philosophy students. The topics chosen are those that are most likely to be of use to that audience. While in a perfect world, a “companion” volume to Bertrand Russell would have covered all of Russell’s work, in reality, this would have doubled its size and price and left it without a single identifiable market.

Even if largely restricted to works easily and uncontroversially “philosophical” in the mainstream sense, the Companion’s coverage is by no means limited to the “usual suspects”. While Griffin apologizes in the introduction that Russell’s later philosophy is given “relatively sketchy treatment” (p. 46), in fact by comparison to other treatments of Russell’s philosophy, the Companion contains a number of chapters that contain serious engagement with Russell’s philosophical writings from the 1920s, 1930s and 1940s. Even among Russell’s earlier philosophical career, the Companion covers areas of Russell’s thought that are not widely known, such as the works of his idealist period and the substitutional theory.

If I were to give any criticism of the coverage of the anthology, it would be a small complaint about the lack of a single piece tracking the development of Russell’s thinking about truth, perhaps in connection with his views on representation and judgment. While these topics are covered in a piecemeal fashion in various selections, a single exposition of the changes in Russell’s views would have served to reconcile some otherwise contradictory-seeming statements found in chapters dealing with different phases of Russell’s thought – and indeed, would also have shed light on those few instances in which the statements made by the authors are in fact at odds with one another. Room for this might have come from eliminating one of, or amalgamating, either the two chapters on Russell’s logicism or the two chapters on Russell’s epistemology.

With regard to quality, most of the entries are both well-written and show an excellent grasp both of Russell’s writings and their historical situation. Certainly, some of the chapters fare better than others in this regard. I have noted some minor difficulties in my discussion above, and with one or two chapters there are some more systemic difficulties which space limitations preclude me from elaborating upon here. However, such problems are far outweighed by the strengths of the Companion as a whole. Moreover, while there are some disagreements and even direct contradictions between the various authors on certain points – some of which I’ve noted above – I do not take this to be a fatal flaw of the Companion. While sometimes the disagreements are straightforwardly due to a misreading by one of the authors which could be cleared up by consulting the primary texts, more often they reveal the sort of disagreements about interpretation that are inevitable when engaging with a highly original and productive philosopher such as Russell. A good introduction to a philosopher need not and should not hide the fact that there remains serious contention about certain aspects of his work. Instead it should highlight the unresolved disputes in a way that invites the interested reader to investigate them for her or himself. This is the spirit of many of the more controversial passages in several of the chapters, though there are a few occasions in which a contentious point is made without attention being drawn to it.

It should perhaps be noted that the Companion is not – nor do I think the authors intended it to be – an anthology containing new and cutting edge research. Indeed, there is remarkable overlap between it and previous writings by the same authors. The chapters by Griffin and Landini are largely just summaries of their respective books (Griffin 1991, Landini 1998), and the information contained in the chapters by Grattan-Guinness, Hylton, Hager, Linsky and Pigden overlap heavily with their previous writings (see Hager 1994, Hylton 1990, Linsky 1999, Grattan-Guiness 2000, Pigden 1999). Specialists already familiar with these authors’ works won’t find anything remarkably new here, but it is certainly convenient and useful to have a single source-book bringing all the recent secondary literature together in a summary form. As it reads on the back cover of the book, the Companion aims to be a “conspectus of recent developments in the interpretation of Russell,” and in this regard it certainly fulfills its aim.

Yet in the end, it is not specialists for whom the collection will be most useful. Currently, there is nothing to compare to it in providing an accessible but comprehensive introduction to Russell’s philosophy for advanced students, particularly, intelligent undergraduates and graduate students capable of doing work on Russell at a high level. To be sure, there are some introductions on the market, but most are usually too short or too unsophisticated to give students a sense of the nuances and detailed rigor of Russell’s philosophy. Most of the writings in the Companion are pitched at a level that make them accessible to someone with a solid background in analytic philosophy and only minimal exposure to Russell’s own writings. Some of the contributions are pitched higher than others (e.g., those by Grattan-Guinness, Landini, Urquhart and Demopoulos), but in general these are precisely those dealing with topics that would likely only be tackled by relatively more advanced students and specialists. I can speak from first hand experience from having taught a graduate seminar on Russell’s philosophy in the most recent term; the Companion had appeared just in time for me to recommend it to my students. Their feedback was nearly uniformly positive, and this, perhaps more any anything else, speaks to the quality of Griffin’s anthology.

Department of Philosophy
University of Massachusetts
Amherst, Massachusetts

Works Cited

Cocchiarella, Nino. 1987. Logical Studies in Early Analytic Philosophy. Columbus: Ohio State University Press.

Grattan-Guinness, Ivor. 2000. The Search for Mathematical Roots, 1870-1940: Logics, Set Theories and the Foundations of Mathematics from Cantor through Russell to Gödel. Princeton: Princeton University Press.

Griffin, Nicholas. 1991. Russell’s Idealist Apprenticeship. Oxford: Clarendon.

Hager, Paul. 1994. Continuity and Change in the Development of Russell’s Philosophy. Dordrecht: Kluwer.

Hylton, Peter. 1990. Russell, Idealism and the Emergence of Analytical Philosophy. Oxford: Clarendon.

Klement, Kevin. 2003. “Russell’s 1903-05 Anticipation of the Lambda Calculus,” History and Philosophy of Logic, vol. 24, pp. 15-37.

Landini, Gregory. 1998. Russell’s Hidden Substitutional Theory. Oxford: Oxford University Press.

Linsky, Bernard. 1999. Russell’s Metaphysical Logic. Stanford: CSLI Publications.

Pigden, Charles, ed. 1999. Russell on Ethics. London: Routledge.

Rodriguez-Consuegra, Francisco. 1989. “Russell’s Theory of Types 1901-1910: Its Complex Origins in the Unpublished Manuscripts.” History and Philosophy of Logic, vol. 10, pp. 131-164.

Russell, Bertrand. 1906. “On ‘Insolubilia’ and their Solution By Symbolic Logic,” in Essays in Analysis, ed. Douglas Lackey. London: Allen and Unwin, 1973. Originally published as “Les paradoxes de la logique,” Revue de métaphysique et de la morale 14 (1906): 627-50.

——. 1992. The Collected Papers of Bertrand Russell. Vol. 6. Logical and Philosophical Papers, 1909-1913. Edited by John G. Slater. London: Allen and Unwin.

——. 1994. The Collected Papers of Bertrand Russell. Vol. 4. Foundations of Logic 1903-05. Edited by Alasdair Urquhart. New York: Routledge.

Sainsbury, R.M. 1979. Russell. London: Routledge.

——. 1980. “Russell on Constructions and Fictions.” Theoria, vol. 46, pp. 19-36.

© 2004 Kevin Klement.