DESCRIPTIVE SCIENCE 141 



of his early life; Chapter IV might indicate how a chance 

 observation led him to formulate still another theory; and 

 so on. The total body of propositions representing his 

 conclusions in the field would be contained in his biography. 

 But their order would not be significant of their logical 

 relations to one another; the propositions in Chapter II 

 would not be logically derivable from those in Chapter I, 

 though there might be a relation of psychological suggestion 

 between them. Contrast this with the order of propositions 

 in a textbook on science. Chapter I usually contains the 

 most general principles, both of method and of subject 

 matter. Chapter II, perhaps, indicates either their most 

 immediate deductive consequences or their most apparent 

 applications to phenomena of a certain kind. Succeeding 

 chapters show progressively how more complicated theorems 

 and corollaries can be deduced by the use of more compli- 

 cated logical techniques. As in the case of a biography, the 

 total body of propositions contained in the book indicates 

 the conclusions of the scientist in the field. But their order 

 is significant of their logical relations to one another. By 

 learning where a proposition occurs in the book one would 

 be able to conclude as to its position in the logical scheme. 

 The structure of descriptive science, if it has any structure 

 at all, is that of a biography, not that of a textbook. On 

 the other hand, the structure of rational science, as will be 

 shown later, is that of a textbook, not that of a biography. 

 An alternative formulation of the same feature of descrip- 

 tive science is to say that there are no highly integrating 

 propositions. If a system of symbols is organic, its relational 

 aspects must be important. But the relational aspects of a 

 symbolic system must themselves be expressed by symbols. 

 A law, for example, expresses a relation between concepts, 

 and a highly general law states a relation between more 

 specific laws. A highly integrated symbolic scheme, there- 

 fore, would be one which contains many necessary and 

 universal laws. But this is precisely the kind of symbol 

 which is lacking in descriptive science. The correlations 



