Page:The complete works of Henry George vol. 1.djvu/64

 labor and capital as taking separate parts in the work of production and separate shares in the distribution of its proceeds; when they speak of wages as drawn from capital, or as depending upon the ratio between labor and capital, or in any of the ways in which the term is generally used by them. In all these cases the term capital is used in its commonly understood sense, as that portion of wealth which its owners do not propose to use directly for their own gratification, but for the purpose of obtaining more wealth. In short, by political economists, in everything except their definitions and first principles, as well as by the world at large, "that part of a man's stock," to use the words of Adam Smith, "which he expects to afford him revenue is called his capital." This is the only sense in which the term capital expresses any fixed idea—the only sense in which we can with any clearness separate it from wealth and contrast it with labor. For, if we must consider as capital everything which supplies the laborer with food, clothing, shelter, etc., then to find a laborer who is not a capitalist we shall be forced to hunt up an absolutely naked man, destitute even of a sharpened stick, or of a burrow in the ground—a situation in which, save as the result of exceptional circumstances, human beings have never yet been found.

It seems to me that the variance and inexactitude in these definitions arise from the fact that the idea of what capital is has been deduced from a preconceived idea of how capital assists production. Instead of determining what capital is, and then observing what capital does, the functions of capital have first been assumed, and then a definition of capital made which includes all things which do or may perform those functions. Let us reverse this process, and, adopting the natural order, ascertain what the thing is before settling what it does. All we are trying to do, all that it is necessary to do, is to fix, as it were, the metes and bounds of a term that in