Appendix 28. - (309)

As I have already observed, Locke's inexact description of abstraction was the source of his inaccuracy. Let us suppose, he says, that I see a pear tree in my garden. Comparing it with the other trees in the same garden: 1. I note something which it has in common with all the other trees to which I compare it; this common element furnishes me with the idea of the genus which is expressed, if I so wish, by the word tree; 2. I note that the tree which I observe has something in common with certain determinate trees. From this common element, which is common to one class of trees, but not to all trees, I form the idea of the species to which I give the name: 'pear tree'; 3. finally, I note that, in addition to the features common to all trees and to the features common to all pear trees, my tree has something of its own, not common with other trees. From this, I have the individual idea of that tree, which I do not name because I do not need to name it individually. Now, if we ignore the inaccuracies in this derivation of an individual idea, I maintain that, if I were to form the idea of the genus of trees by selecting the features common only to all the trees submitted to my experience (and in the case before us, to all the trees planted in my garden), this generic idea would contain nothing over and above the element common to the trees which I have examined. I could therefore apply the name tree only to that determinate number and to the actual trees I have seen. This idea would be of no use to me in designating a possible tree, or an existing tree that I had not seen, but only heard about. This, however, runs counter to the use of the word, which expresses genus. It is easy to see that, in using the word tree, I am indicating something common not to ten or twelve individual trees, nor to all existing trees, but to all the trees imaginable and, therefore, of all possible trees. It follows that in the idea of genus expressed as a common noun, there is always included the idea of a notion applicable to an infinite number of beings, that is, the idea of their possibility which transcends all the limits of experience.

Furthermore, the idea of genus does not have extension based on the number of individuals subjected to my sense experience. A person who had examined all the trees in the world, one after the other, would perhaps have an idea of the genus of trees that was more precise but certainly no more extensive than that of someone who had never set foot out of his house or left the tiny confines of his garden. Both persons would refer to the tree as a genus, assigning to the word a notion as extensive as the notion of possibility, which has no limits of any kind; both would apply this word tree to all those trees which God might create, as well as to those he has already created. The same remark may be applied to the idea of species. From this, it is clear that these ideas contain an infinitely greater extension than that which all sense experience could provide. Consequently, we have to resort to another source adequate to explain the notion of possibility, which forms part of every common idea, constitutes the common idea's great extension and does not in the least have its source in the senses.