If one wishes to consider the extent to which the scientific community’s image of itself corresponds to the realities of the situation, a fruitful starting point is to investigate the phenomenon of scientific change. For viewed sub specie eternitatis scientists (even physical scientists) are a fickle lot. The history of science is a tale of multifarious shiftings of allegiance from theory to theory. Newtonian mechanics had its hour of flourishing with virtual universal allegiance. Then, following a dramatic and brief period of turbulence, relativistic mechanics came to the fore and is espoused with the same universal allegiance and firm commitment on the part of the community.
Much scientific activity consists in accounting for or explaining change. This shifting of allegiances from theory to theory which will be referred to as scientific change is itself a type of change that requires explanation. But what sort of explanation? In regard to this question we face what Kuhn would describe as a pre-paradigmatic situation. Unlike the situation in contemporary physical sciences where for many areas of investigation the community of investigators are generally agreed on the form or type of explanation to be sought, we find that when we take science itself as the subject of our investigation there is no such agreement. In this case detailed putative explanations are few and far between. Instead we find only radically divergent types of explanatory sketch. The differences between the proponents of these sketches go as deep as intellectual divergences ever go, involving in this case differences concerning the objectivity of truth, the possibility of rational discourse, the nature of values, language and meaning and explanation, among others. It will be fruitful to begin by dividing models for the explanation of scientific change into two classes, one to be called rational models of scientific change, the other to be called non-rational models. For as we shall see, we shall only be justified in regarding scientific practice as the very paradigm of rationality if we can justify the claim that scientific change is rationally explicable. At this stage the division must be regarded as a tentative one drawn to assist us in focusing on the central issues in this area. In the course of this book the division will be seen to be of more than organizational significance.