This leaves one class, that of co-existence, to be examined. Here, if anywhere, must knowledge, worthy of being termed scientific, be found. This class, it will be remembered, comprehends our knowledge concerning substances. But this extends, according to Locke, "a very little way." The idea of a substance is a complex of various "simple ideas united in one subject and co-existing together." When we would know anything further concerning a substance, we only inquire what other simple ideas, besides those already united, co-exist with them. Since there is no _necessary_ connection, however, among these simple ideas, since each is, by its very simplicity, essentially distinct from every other, or, as we have already learned, since nothing is essential to an individual, we can never be sure that any idea really co-exists with others. Or, as Locke says, in physical matters we "can go no further than particular experience informs us of. . . . We can have no certain knowledge of universal truths concerning natural bodies." And again, "universal propositions of whose truth and falsehood we have certain knowledge concern not existence;" while, on the other hand, "particular affirmations are only concerning existence, declaring only the _accidental_ union or separation of ideas in things existing." This particular knowledge, it must be recalled, is, in turn, only sensitive, and thus extends not beyond the time when the sensation is had.
We are not surprised then at learning from Locke that regarding bodies "we are not capable of scientific knowledge." "Natural philosophy is not capable of being made a science;" or, as Locke elsewhere states it, knowledge regarding the nominal essence is "trifling" (Kant's analytic judgment); regarding the real essence is impossible. For example, when we say that all gold is fusible, this means either simply that fusibility is one of the ideas which we combine to get the general idea of gold, so that in making the given judgment we only expand our own notion; or it means that the "real" substance gold is always fusible. But this is a statement we have no right to make, and for two reasons: we do not know what the real substance gold is; and even if we did, we should not know that fusibility _always_ co-exists with it. The summary of the whole matter is that "general certainty is to be found only in our ideas. Whenever we go to seek it elsewhere, in experiment or observations without us, our knowledge goes not beyond particulars."