162 INTRODUCTION TO PHILOSOPHY OF SCIENCE 



way the principles of mechanics, though suggested by ex- 

 perimental data, describe not these data but idealized enti- 

 ties, such as perfectly isolated bodies. 



But if the meaning and truth of these principles are not 

 determined completely by data, how are they determined? 

 Poincare's answer is definite. They are determined by 

 convenience, i.e., the principles are defined in such a way as 

 to make them most convenient. " The axioms of geometry . . . 

 are conventions; our choice among all possible conventions 

 is guided by experimental facts; but it remains free and is 

 limited only by the necessity of avoiding all contradiction. 

 Thus it is that the postulates can remain rigorously true 

 even though the experimental laws which have determined 

 their adoption are only approximative. In other words, 

 the axioms of geometry . . . are merely disguised definitions ." * 

 The same description applies to the principles of mechanics. 

 They have been drawn from experimental laws but "have, 

 so to speak, been exalted into principles to which our mind 

 attributes an absolute value." 2 



Unfortunately Poincare has not told us precisely what is 

 meant by a convention. Presumably a convenient symbol 

 would be one which is useful, and he tells us that a useful 

 science is one which enables us to foresee, 3 hence it would 

 seem legitimate to conclude that the principles of geometry 

 and mechanics are convenient because they enable us to 

 make predictions which are later verified. If this is the proper 

 interpretation of Poincare, his position reduces in this respect 

 to Hobson's. But he would disagree with Hobson in the 

 claim that principles of this kind should be called true. 

 "It is just as unreasonable to inquire whether they are true 

 or false as to ask whether the metric system is true or 

 false." 4 Definitions are not properly true or false, they are 

 only convenient. 



The exact nature of this transition from a law into a 

 principle must be exhibited if one is to understand fully 



1 Ibid., p. 65 (italics are the author's). 3 Ibid., p. 324. 



2 Ibid., p. 125. 4 Ibid., p. 124. 



