xxxrx.] MENTAL INTELLIGENCE. 157 



axioms of mathematics; whereas it is not certain, but of thin f^s 

 only highly probable. It is highly probable that the certain 

 sun will rise to-morrow ; it is so probable that we are ^^"t only 

 right m acting as if it were an absolute certainty ; but it 

 is not so certain that the suu will rise to-morrow as it is 

 that he rose to-day ; and it cannot be regarded as even 

 highly probable that the course of nature will continue to 

 go on for a thousand millions of years longer.^ I reply, 

 that all this is true ; nevertheless, it is as certain as the t>ut it is 

 axioms of mathematics, that the course of nature will go ^o be 

 on until it is interrupted by some cause not now in opera- constaiit 



•i-i o 1 unless ni- 



tron. The law of causation, which is a part of the law of temiptwl. 



the uniformity of the oixler of things,^ is only that the 



same antecedents will continue to be followed by the same 



consequents ; but it cannot guarantee that the antecedents 



will continue to be the same. Our knowledge of things 



is but finite, while our ignorance is infinite ; and we must 



consequently regard all known lines of causation as being 



liable to be cut through by unknown ones. 



Of course the truth that all causes in the natural world 



act regularly, so that their action is in its nature capable of 



being predicted,^ is not by any means self-evident ; on the 



contrary, it is a discovery due to the progress of science. 



But the belief that every effect must have a cause is, I ^" '^^^^ 



•' sense tlio 



think, universal among men, and among animals which law of 



^ I believe that the attempt to find a numerical value for the proba- 

 hilitj' of the sun rising to-morrow is one of the idlest of all amusements. 

 At the best, it is an attenqjt to arrive at a conclusion for which there are 

 DO possible data. But I am inclined to agree with Mr. Venn, the author 

 of " The Logic of Chance," that in such a case the expression " numerical 

 value of the probability " has no meaning at all. 



^ The uniformity of the order of things includes uniformities both of 

 co-existence and of succession ; it is only those of succession that belong 

 to causation. By uniformities of co-existence I mean, that the same sub- 

 stance wDl always have the same properties ; as, that ice, at a temperature 

 below freezing, will always be hard and brittle. By uniformities of succes- 

 sion I mean, that the same causes will always have the same effects ; as, 

 that heat will melt ice. These two classes of uniformities run into each 

 other, as the instances just mentioned show. I doubt whether there is 

 any really fundamental distinction between them. 



' This truth is what Mill, in his I-ogic, means by the law of universal 

 iMusatioii. 



