The Mind of David Hume: A Companion to Book I of A Treatise of Human Nature (review)

Hume Studies Volume XXIV, Number 2, November 1998, pp. 375-379 OLIVER JOHNSON. The Mind of David Hume: A Companion to Book I of A Treatise of Human Nature. Urbana & Chicago: University of Illinois Press, 1995. xi + 375 pages. ISBN 0252064569 US $24.95 paper. This work, devoted entirely to Book I of the Treatise, has two aims. The first is to provide a "post-textual" rather than a "pre-textual" interpretation; that is, one which avoids adopting "a general point of view...at the outset that molds and guides the...interpretation" but arrives "at interpretative generalizations ...only after completing its study deriving these from the text itself " (6). The second is to provide critical evaluation which respects the dictum that "a commentator should read Hume sympathetically" (14). There is much to be said for systematically going through the Treatise instread of citing the text only to support one's pre-textual interprétation. Johnson's Companion seems a bit "thin" in interpretation, but perhaps precisely because of this, it provides a useful check on one's first take on what Hume might be saying. With this said, let me focus on the second aim, since Johnson's final assessment is: most of Hume's "central arguments collapse under the weight of their internal inconsistencies" (331). Starting with a seemingly minor claim, Johnson urges that Hume has two theories of memory: the original theory in I i 3 and a revised theory in I iii 5. The original theory claims that memory ideas, unlike the ideas of imagination, (1) are more lively and (2) reproduce impressions as they were originally experienced. This theory is revised because Hume "recognizes that its requirement (2)...can never be applied in practice" (141). We cannot compare an idea with an impression that is gone forever. Thus, "to differentiate between [memory and imagination] we must instead be content with an appeal to...characteristic (1)" (141). Johnson's reading of I iii 5 is radical: For him, when Hume drops requirement (2), Humean memories cease being memories in any ordinary sense. Johnson writes: It could be argued that, although Hume concludes that the only way we can distinguish between memory ideas and imagination ideas in practice—or, as he puts it, "in their operation"—"is by their relative vivacity, in reality they differ from each other..." [by condition (2)]. Two responses can be made to this conjecture. Hume's argument gives it no support.... More important, if we cannot 376 Book Reviews appeal to the distinction in practice, it becomes gratuitous. We may repeat the difference in a verbal definition but in fact we can never put it to use to distinguish the one from the other. (145-146) What "could be argued" surely is a more sympathetic reading of Hume. The original theory requiring (1) and (2) can be taken to define genuine memory or memory„. Letting memorya be apparent or seeming memory— i.e., an idea we take to be memory„—I iii 5 plausibly asks: What consciously accessible feature makes an idea a memorya idea, that is, what makes us take an idea to be memory^? The liveliness condition (1) then answers this question . Let's grant for the argument's sake that distinguishing memorya and memory g is "gratuitous" in the sense that we can't justify a memorya idea to be memory™. Even then Johnson has given no reason to think that we can't have an idea or concept of memory- Since the problem I iii 5 raises is at worst epistemic and not semantic, we needn't saddle Hume with two incompatible theories of memory. Given his take on Humean memories, Johnson draws the startling conclusion that Hume can't appeal to any notion of constant conjunction: If we agree that, in order to recolllect past constant conjunctions, Hume must return to his original theory of memory, and if we agree that the recollection of such past constant conjunctions is a necessary condition of his having an idea of causation, we are forced to conclude that Hume is caught in a contradiction in his philosophy .... Hume must either give up his empiricist theory of ideas or abandon any hope...