Hume's System: An Examination of the First Book of His Treatise (review)

BOOK REVIEWS 615 At the end of his book Yolton points to the work yet to be done on the reception of the rest of Locke's work in France and of his famous suggestion elsewhere. Similarly, it may be that more work will yield an interpretation answering the questions raised here. If so, rich and abundant material for the interpretation is already to be found in this book, and we should be grateful to its author for compiling it. THOMAS M. LENNON University of Western Ontario David Pears. Hume's System: An Examination of the First Book of His Treatise. Oxford: Oxford University Press, 199o. Pp. ix + 2o4. Cloth, $49.95. Paper, $x4.95. The framework of Pears's critical reading of the First Book of Hume's Treatise is a "balanced interpretation" which seeks to take into account both Hume's empiricism and his naturalism, which Pears wishes to correlate with a theory of meaning and a theory of truth, respectively. In spite of some differences, there are thus certain parallels in Hume's theory with modern linguistic philosophy which, according to Pears, allow Hume's work to be read across this grid. Hume is understood as wishing to solve, through his scientifically grounded theory of mind, fundamental philosophical problems such as those concerning causation, personal identity, and perception; so that the second part of Pears's book will test Hume's position against its success in resolving these issues, and thereby determine concretely whether I-Iume is a sceptic or a naturalist , even if a cautious one. In the first part of his book, Pears advances a closely argued critique of Hume's theory of mind, based primarily upon two philosophical observations: that the mind is active in the bestowing of meaning upon ideas, so that thinking is essentially intentional (as opposed to Hume's treatment of the mind as wholly passive); and that certain ideas, such as those of memory, cannot be understood apart from the ascription of a referential or contextual character to the ideas themselves (as opposed to Hume's atomism). Particular ideas, for example, can take on more general signification-according to Hume--insofar as they become annexed to certain words which bring to mind, on the appropriate occasion, the relevant set of particular ideas. But Hume neglects the active role of the mind in this process, or rather exploits it without acknowledging it, since we ourselves must select the different lines of resemblance involved in recalling one train of ideas rather than another. Meaning thus rests upon an intentional activity, since we ourselves give to the general ideas their role in the selective process. Also, Pears focuses upon the "shade of blue" counterexample to illuminate Hume's treatment of each simple idea as separable from every other, so that simple ideas bear no intrinsic reference to one another. This neglect of "lateral connections" among ideas is symptomatic, for Pears, of Hume's atomism. Memory-judgments, for example, are not "composed" of atomic ideas but offer a definite reference to the past which cannot be accounted for by the combination of isolated ideas. Pears shows, further, that the difference between memory and imagination lies in the way the idea is meant, not in some specifiable quality of the "idea." The same failure to recognize the active role of the mind characterizes Hume's attempt to account for the belief accompa- 616 JOURNAL OF THE HISTORY OF PHILOSOPHY 30:4 OCTOBER 1992 9 nying an idea through a quality of the idea, namely its liveliness. One is immediately aware of belief; it is not something added, as a special "property," to the idea. In his examination of the applications of Hume's theory in Part II, Pears shows why, for Hume, resemblance and continguity cannot produce belief, and then proceeds to a critique of Hume's explanation of inference based upon the association of ideas. There is nothing in association alone which will distinguish between merely bringing ideas into the mind and producing belief. Constant conjunction turns out to be a merely descriptive notion, having nothing in it which can account for the production of belief. With regard to the question of Hume's...