Berkeley Foreshadowing Quine

From part VI of GJ Warnock's introduction to Berkeley's The Principles of Human Knowledge:

...he could not but be aware that the "corpuscular" theories of matter, and of light, were both too fertile to be easily rejected, and too centrally characteristic of the whole ideal of scientific understanding. But it was still not open to him to admit that such theories were, as they stood, straightforwardly true; and so, with very striking insight and ingenuity, he fell back on the distinction between the observed facts of science, and the theories devised by scientists to account for them. The aim of science, he still holds, is to reduce to "general rules" the observed phenomena; but the achievement of this aim, he now argues, is greatly facilitated by the making of appropriate suppositions. If we think of light, for example, as if it were propagated in the form of a stream of "insensible particles," then the diverse phenomena of light can be comprehended within a theory capable of expression in simple mechanical terms, and highly apt for the precise use of measurement and mathematical calculation. This is certainly useful; but, Berkeley insists, it is useful, not true. As he wrote in his tract De Motu in 1721, "to be of service to reckoning and mathematical demonstrations is one thing, to set forth the nature of things is another." Thus, Berkeley does not now object, as formerly he did, to reference to "insensible particles" and other items of supposed, unperceivable "corpuscular" machinery. He sees that such references serve a theoretical purpose, particularly in facilitating the application to physical phenomena of precise mathematical concepts and methods. But the resulting theories have the status of serviceable fictions; they are useful inventions; it cannot be objected that he leaves such theories with nothing to be true of, for in fact there is no need to suppose that they are true at all. They are theories, not facts; and the virtue of a theory consists not in truth, but in utility.