Some recent and historical writers in the philosophy of science have concerned themselves with a certain problem which seems to occupy, at least in the minds of those who have written about it, a position of peculiar importance. Whether the problem is really as significant as its authors maintain need not be decided here; certainly many writers in this area have either neglected it or made only vague allusions to it. It can best be described as the problem of the (...) development of science, and it is usually solved in terms of a certain theory that science passes in its growth, either naturally or under the control of conscious logical techniques, through various “stages” or “phases.” For some strange reason these levels are usually exactly three in number. One would like to believe that this number is not a magic one, but empirically demanded by the actual way in which the sciences develop. But the fact that the descriptions of these stages in the writings of the various authors appear to agree only in their triplicity, and not in content, would incline one to believe otherwise. There is, in fact, the slight suspicion that the number three has been preferred to the number two simply because a triad may be ordered in a way in which a couple may not, and to the number four or any larger number merely because of the preference for simplicity over complexity. (shrink)