This difficulty was demonstrated in the work of the empiricist philosophers George Berkeley and David Hume. Their initial premise was that it is not possible for the human mind, which knows the world only through its ideas, to compare an idea with anything except another idea—that is, with another one of the mind’s mental states. This is, of course, a straightforward requirement of empiricism, the philosophy of experience that bases all knowledge on the deliverances of the senses and thus on the ideas that are thereby produced in the mind. On the other hand, the conception of the “external” world, which Descartes and Locke had advanced as the philosophical basis for the new physics, presupposed the possibility of comparing, and thus distinguishing between, an idea within the mind and the external object the idea is supposed to represent. The irony here is that, for most of those who subscribed to it, the way of ideas had served mainly as a way of pulling high-flying abstractions down to earth by putting them to the test of sense experience. It was easy to forget that what the human senses deliver is a modification of a mental state, which is itself a mental state—an idea, rather than (as might be instinctively assumed) something that is unambiguously “out there” in the world.

Once it is acknowledged that the way of ideas applies to every bit of knowledge humans claim to have, the principle can be interpreted in a variety of ways. Some interpretations, in the manner of Hume, are highly skeptical—humans have no access to a world of stable, perduring objects—while others, following Berkeley, are ambitiously metaphysical—the world itself is made of ideas. In either case, the conception of a reality that lurks behind sensible experiences has to be given up.

What is perhaps even more significant is the impact that this line of inquiry can have on the premises of the way of ideas itself. In his A Treatise of Human Nature (1739–40), Hume argued that he was unable to find any sensible idea—his word was impression—of a “self” or “mind” in which ideas were supposed to be received. He concluded that not only things in the world but also minds were only loose collections of impressions and their fainter copies, for which he reserved the term ideas. Although he understood very well that he was really undermining the entire notion of mind, his line of thought had other consequences of which he does not appear to have been fully aware. One of these is that, if each human being is indeed locked within a circle of impressions and ideas, the reasoning by which these come to be referred to in this way—that is, as mental contents—must itself be suspect. Because humans can think only in terms of the ideas that are supposed to be the products of actions involving things outside the mind, no human being can be in a position to claim any knowledge of the process by which ideas are produced in the mind—that would require familiarity with something that is not an idea. All that remains is a collection of qualities. Although the qualities themselves are neither mental nor material, they can be combined to form objects that may be either the one or the other. It may be said, then, that in Hume’s thought (and in much of the work of empiricist philosophers who followed him), the concept of the self or mind has been dispersed into just these atomic units, which supposedly combine and recombine to form a self and a world.

It is nevertheless difficult to see how many characteristic functions of human life can be understood in these terms. How, for example, would it be possible to explain action unless one is prepared to assume that units of this kind can have intentions? And what can be said about each person’s relation to other human beings if their minds, too, have to be analyzed as collections of such units? Is one such collection supposed to be able to divine the presence of another? It is not surprising that Hume himself acknowledged that it was impossible to live by these conclusions and that, upon quitting his philosophical speculations, his ordinary beliefs in selves and in an external world resumed their usual power.