Hume on the Abstract Idea of Existence: Comments on Cummins' "Hume on the Idea of Existence"

Hume on the Abstract Idea of Existence: Comments on Cummins' "Hume on the Idea of Existence"1 Fred Wilson Hume'sviews on theconceptofexistence: thisisone ofthemore obscure parts of Hume's philosophy. Professor Cummins has done a valuable service simply by trying to unravel some ofthe puzzles; it is still more valuable for shedding as much light as it does on the issues. There are nonetheless problems with the interpretation that he develops, and I would like to bring out some ofthese. I will address three questions: (a)What, more precisely, is the (abstract) idea ofexistence? (b)What is the notion of"necessity" as Hume would use it in connection with the notion that something exists "necessarily," or, contrarily, "contingently"? (c)Does it follow from the fact that whatever thing we conceive we conceive as existent that every thing necessarily exists? In order toanswerthese questions somethingwillhave tobe said about Hume's doctrine ofabstractideas. Butonall theseissues, Humeis often not so much inadequate as incomplete; and I shall suggest that a numberofpoints canbeilluminatedbylookingatthe historical context that Hume himselfcould take for granted. The upshot will not be that Hume's views are unproblematic, but only that they are perhaps less problematic than Professor Cummins makes out. I. Abstract Ideas In many respects, Hume's doctrine of ideas is fairly traditional. We think in terms ofideas. In particular, our capacity to think generally, and to use general terms, is a matter ofhaving abstract ideas. In this, Hume follows the views ofhis predecessors such as Locke, Descartes and the Port Royal logicians. Hume also adopts the principle that what is conceivable is possible. Again, he follows the same tradition ofLocke, Volume XVII Number 2 167 FRED WILSON Descartes and the Port Royal logicians. Where Hume differs from his predecessors is in his account of what precisely an abstract is. For Hume's predecessors, one forms an abstract idea by separation. Thus, for Locke, forming ideas by abstraction consists in "separating them from all other Ideas that accompany them in their real existence."2 This sortofseparation is the mechanism bywhich "the Mind makes the particular Ideas, received from particular Objects, to become general" (Essay, 2.11.9). The Port Royal logicians make the same point: abstraction occurs with respect to "choses ... composées" when one "les considérant par parties, et comme par les diverses faces qu'elles peuvent recevoir." In abstraction what is thus known is considered separately from the whole; either one considers "les parties séparément" in the case of"parties intégrantes," or one "peut séparer les choses en divers modes," or, finally, "quand une même chose ayant divers attributs, on pense à l'un sans penser à l'autre."3 The former occurs when, for example, we separate in thought the parts ofthe body. Thisis the simplestform ofabstraction. Berkeleyhas noproblems with this notion: "Iwill notdenyI can abstract, ifthatmay properlybe called abstraction, which extends only to the conceiving separately such objects, as is possible may really exist or be actually perceived asunder."4 Neither does Hume have problems with this notion; it is the basis of his—and Berkeley's—view that ideas are either simple or complex and that some of the complex ideas are a result of our combining simple parts in ways that are not found in reality, as, for example, in our idea ofthe New Jerusalem.5 It is the other two sorts ofabstraction that are more problematic. Consider the first of these two forms of abstraction. We begin with simple ideas ofsensation, for example, white (Essay, 2.3.1). One forms the abstract idea of white by separating the property white from the other qualities with which it is conjoined in the white particulars, for example, snow or milk (Essay, 2.21.73), that are given in sense experience. The resultingideais distinct from other ideas, for example, blue or heat (Essay, 2.12.1). The example given by Nicole and Arnauld is thatofthe geometers who explore the idea ofthree-dimensional space by considering first length alone, then area, and finally volumes or solids (which is the same, for these Cartesians as "corps," body) (Logique, 1.5). This is one form ofseparation. By this means one forms abstract...