31 



relation is analyzed, as well as the atmosphere ; until it is ascer- 

 tained that only one of these constituents of the atmosphere 

 maintains life or is the cause of life, while the others are merely 

 associated. This is merely a supposed example, not true through- 

 cut, but chosen because it is familar. 



63. In the business of analysis we can specify only this rule, 

 in order to distinguish causes from mere associations (and this rule 

 must frequently be inadequate, owing to our defective means of 

 analysis), namely, to develop dependences as far as we are able; 

 to reject as causes all those things which may be separated from 

 the effect without changing its identity; and still to regard those 

 as causes which we cannot discover to be separable from the identity 

 of the effect. 



64. It is affirmed by some, and the doctrine I believe is 

 almost a fashionable one, that causation is nothing more than the 

 succession of phenomena; that we know nothing more of causa- 

 tion than that there are certain antecedents, which are regularly 

 followed by certain consequences. It is affirmed, that we cannot 

 tell why effects should succeed to causes. I do not wish to spend 

 many words upon this doctrine: to consider it then briefly: 



65. An acid and an alkali are causes, a neutral salt is their 

 effect: that is, a neutral salt succeeds to the combination of an 

 acid and an alkali. But would the neutral salt exist without the 

 acid and the alkali? To answer this question we must appeal to 

 our experience; and no one will object to this appeal, because it 

 is the best that can be proposed. Our experience tells us that the 

 neutral salt cannot exist without the acid and the alkali; hence 

 we infer, that the causes, viz. the acid and the alkali, are necessary 

 to the existence of the effect, viz. the neutral salt; and we infer 

 that the causes are necessary to the effect, simply because the 

 effect cannot exist without them : and this is agreeable to our ex- 

 perience in every example, in which the cause and the effect are 

 witnessed by the senses. 



66. Our ideas of the necessity of one thing to the existence 

 of another are derived from, or have their strongest illustration in, 

 the relation of cause and effect. Thus, we say, blood is necessary 

 to the life of a man: we mean it is necessary to the life of a 

 man, because without it the man would die. Thus food also is 

 necessary to the life of a man. If any one doubts this necessity for 

 food in order to preserve life, let him try to live without it, in 

 which he will certainly succeed if there is no necessity for it. 

 Thus, muscles are necessary to voluntary motion; thus, too, air is 

 necessary to combustion. We have no notion of a case of necessity 

 which does not respect the relation of cause and effect. If, then, 

 causes are necessary to the existence of effects, the next question 

 is, why are they necessary? and this is what certain philosophers, 

 as they call themselves, or as either in derision, or in the way of 

 civility, they might be called ; this is the question which they 

 tell us we cannot answer. Let us try 



