216 INTRODUCTION TO PHILOSOPHY OF SCIENCE 



acter that all of the propositions descriptive of the data can 

 be deduced from them. But since this body of descriptive 

 propositions includes not only those which are known to 

 apply to the data but also those which constitute anticipa- 

 tions of data not yet disclosed to perception, it is readily 

 seen that, if this is to be possible, the explanatory entity as 

 obtained through the method of construction must be in- 

 creased through activities of the imagination. If the con- 

 struct is merely derived from the data it cannot be turned 

 about and used to explain the data, nor can it be used to 

 predict data not yet observed. Hence these two stages 

 represent the theory as it undergoes progressive expansion 

 through the tentative attribution to it of the desired content. 

 The distinction between the third and fourth stages is 

 relative. In the third stage one attempts to increase the con- 

 tent of the explanatory entity in such a way as to permit the 

 deduction of the propositions already known to be verified, 

 i.e., the propositions descriptive of the original data of the 

 problem. In this activity one finds it hard to avoid the feel- 

 ing that he is cheating. He is deliberately giving the ex- 

 planatory entity such content as will permit him to deduce 

 propositions which he already knows; he is making the 

 hypothesis do precisely what he wants it to do. He can be 

 sure that the explanatory entity will explain because he has 

 developed it in such a way as to make it do exactly that 

 thing. But in the fourth stage, into which the third merges 

 imperceptibly, the investigator is not proceeding in such an 

 obvious way. Here he attempts to increase the content of 

 the explanatory entity in such a way as to permit the deduc- 

 tion of propositions not yet known to be true, i.e., proposi- 

 tions descriptive of data still to be discovered. This stage 

 constitutes the essential moment in the development of the 

 theory. Only in this stage has it become a genuine theory, 

 fulfilling its predictive function. Only in this stage has it 

 become really novel, enabling one to anticipate nature. Here 

 one no longer feels that he is playing the game unfairly. On 

 the contrary he recognizes that the activity of imaginative 



