208 DOGMATISM AND EVOLUTION 



in a variety of connections, such as would throw its own strength 

 or weakness into relief would lead to satisfactory results. But 

 in the chain of argument no such procedure is ordinarily con- 

 templated. On the contrary, each conclusion reached in the 

 course of the argument is regarded as proceeding immediately 

 from its premises; and it is upon that supposition that the rea- 

 soner advances to the later conclusions. 



But it is not only the chain of reasoning that cannot be ac- 

 counted for on the pragmatist basis. The simplest conceivable 

 argument, in which premise and conclusion are distinguished, be- 

 comes equally inexplicable; and this can be shown from an ex- 

 ample which is in constant reference by the pragmatists them- 

 selves. Let us suppose that the truth of a general hypothesis has 

 been tested in the case of a particular instance, and has been 

 found in want of correction. Here, on the basis of the hypothesis 

 under consideration, something is inferred as to the results of 

 acting in a certain way under certain circumstances; and this 

 conclusion, as compared with the observed results, is found to 

 be false. What now constitutes the validity of the inference 

 which led to the admittedly false conclusion? The whole pro- 

 cedure depends upon this point, and yet just this point is sub- 

 mitted to no practical test. To be sure it may be said that 

 similar inferences have in the past been found to be correct. 

 But, in the first place, it is probably not on the basis of such a 

 comparison that the untrue conclusion is accepted as correctly 

 derived. That is seldom a matter for reflection. And, in the 

 second place, we must observe that the pragmatist theory fails 

 equally to explain the correctness of an inference fjoni true 

 premises. In a word, the theory does not distinguish between 

 the correctness of an inference and the truth of its premises, and 

 hence virtually eliminates the former altogether. 



So far as we are aware, this result can only be avoided by an 

 interpretation of pragmatism in which its opposition to formal 

 logic is given up. It is pointed out that the acceptance of a 

 conclusion as satisfactorily derived, with the consequent passing 



