Continuity and Change in the Development of Russell's Philosophy (review)

150 JOURNAL OF THE HISTORY OF PHILOSOPHY 35:1 JANUARY a997 rank among souls," there isn't a health proper to the soul, but only a health proper to types of soul. Furthermore, ill health in some souls is tolerated, even extolled, for the sake of slowing the slide into mediocrity. CHRISTINE SWANTON UniversityofAuckland Paul J. Hagar. Continuityand Changein theDevelopmentofRussell'sPhilosophy.Dordrecht: Kluwer, 1995. Pp. xiii + 195. Board, $97.oo. The main theme of this book is that "there is a lot more continuity in Russell's philosophy than has usually been acknowledged" (xii), which the author explains in term of Russell's method of analysis and the role of relations in the analyses given by Russell throughout his career. The author divides his discussion into two parts, the first on analysis and relations, and the second on Russell's theories of space and time as the "key" to the changes in Russell's philosophy. He also divides Russell's development into four phases: (1) a neo-Hegelian phase (up to x899), (2) a Platonist phase (1899-1913), (3) an empiricist phase (1913-18), and (4) a modified empiricist phase (from 1919 on). Russell's method of analysis in both mathematics and philosophy "is essentially that of logical analysis, followed by logical synthesis," as Russell himself said repeatedly. The first stage of this method, logical analysis, is sometimes considered all there is to the method. The author is critical of this view--which he calls "the narrow interpretation of analysis" (23)--because it erroneously claims that the point of Russell's method is to end up with trivial, self-evident logical truths in the case of mathematics, or with self-evident premises about sense data in the case of empirical knowledge. The author observes that, for Russell, the first stage never really ends, i.e., that analysis is unlikely ever to be final, and the premises one ends up with at any given point are frequently less self-evident than the data with which one begins. The "broad interpretation" of Russell's method recognizes the importance of the second stage, logical synthesis, as an essential part of the method. This method of analysis and synthesis is what Russell also called "philosophical analysis." The author gives a systematic account of the method in terms of analysis diagrams, with different boxes for unanalyzed propositions and inferred commonsense objects, other boxes for the premises, basic particulars, still others for the properties and relations that result by logical analysis from the first stage, and finally boxes for the propositions and logical constructions that result by logical synthesis from the second stage. In Part I of the book, the author explains how Russell's method of analysis was applied in each of the four phases of the development of philosophy and how relations are involved in those analyses. In his neo-Hegelian phase, e.g., Russell's project was to analyze the nature of the different sciences and the dialectical relations between them, a project that led to his description of "paradoxes of relativity" and of "parts and wholes." Both types of "paradox" were based on the Hegelian doctrine of internal relations, which Russell abandoned in his Platonist phase and replaced by a doctrine of external relations. The author notes how three sorts of relations--many-one relations BOOK REVIEWS 151 as the basis of descriptive functions, one-one relations as the basis of Russell's construction of cardinals, and serial relations as the basis, e.g., of progressions and the natural numbers--were particularly important in this phase. The author is also critical of how Russell's views in this phase have been misunderstood by Imre Lakatos and others. The empiricist phase, the author notes, was "marked by a retreat from the realism of subsistent being" that was characterisitic of the Platonist phase (49). It is also noted that Russell's construction in this phase of a neo-Leibnizian three-dimensional world of perspectives of sensibilia, some but not all of which are perceived by subjects, depends essentially on spatial and temporal relations. The modified empiricist phase, finally, is marked by the "the inescapable necessity of postulating inferred (or nonexperienced) entities"--namely, events--and...