THE PRINCIPLES OF SCIENCE 209 



fail, a new route to the top is sought. A new route requiring relocation 

 of one or two of the high conceptual camps allows the enterprise to 

 go on with little loss of momentum; but resistance to such relocations 

 rises swiftly with the number of them required. The situation of the 

 lower conceptual camps will be called in question only after long- 

 continued failure. The situation of the base camp of substantive prin- 

 ciples will be challenged only when every other alternative has failed. 

 Our ideas arrange themselves in a hierarchy. At the foundation 

 level occur those laws and theories we hold as of principle, while 

 at the growing edge of science appear highly tentative hypotheses 

 we are quite prepared to re-examine. These are the extremes of a 

 pecking-order that stretches all the way from certainty to doubt. This 

 conceptual hierarchy, readily understandable on historical and psy- 

 chological grounds, has also a logical basis to which Toulmin directs 

 attention. 



One cannot at the same time question the adequacy of Snell's law and 

 go on talking about refractive index. . . . Certainly every statement 

 in a science should conceivably be capable of being called in question, 

 and of being shown empirically to be unjustified; for only so can the 

 science be saved from dogmatism. But it is equally important that 

 in any particular investigation, many of these propositions should not 

 actually be called in question, for by questioning some we deprive 

 others of their very meaning. It is in this sense that the propositions of 

 an exact science form a hierarchy, and are built one upon an- 

 other; . . . 



The terms "established" and "hypothetical," as used in science, 

 need to be understood in terms of the distinction between the parts 

 of science that are actually being called in question, and those which 

 we must take for granted in order to state our working problems. 



The "established" laws and theories we take for granted define a 

 background of presupposition. As earlier noted, we need this even 

 to recognize ( as divergences from wliat we expect ) the "phenomena" 

 for which we will seek explanations. Thus, given the standard of 

 normality set by Snell's law, double refraction is unmistakably a 

 phenomenon— producing what we call, significantly enough, an "or- 

 dinary" ray and an "extraordinary" ray. For the mountain-climbing 

 expedition the location of the base camp affects decisively the chance 

 of ultimate success, by predetermining the general area in which 

 higher camps can be established. For science the standard of the 



