416 Prof. Forbes on the evidence for a Physical Connexion 



ration which seems to me not unimportant with reference to 

 the theory of causation treated* by the doctrine of chances, 

 though less obvious, and therefore perhaps less easily assented 

 to at first sight, than the objection already made. It aims at 

 showing that Mitchell's axiom or postulate is inconsistent with 

 any idea which we can form of a " random scattering." As 

 it is difficult to state what these words really mean, we shall 

 frame our objection so as to rest entirely upon what they ob- 

 viously do not mean. 



28. Second Objection. — To assume that " every starts as 

 likely, not hypothetically, but actually, to be in one situation as 

 another " leads to conclusions obviously at variance with the idea 

 of random or lawless distribution, and is therefore not the ex- 

 pression of that Idea. 



29. Let me first quote the words in which Mitchell states 

 the principle of his argument (which principle has certainly 

 been accepted down to this time). " Let us examine," he says, 

 " what would have been the least apparent distance of any two 

 or more stars, anywhere in the whole heavens, upon the sup- 

 position that they had been scattered by mere chance, as it 

 might happen. Now it is manifest, upon this suppositio?i, that 

 every star being as likely to be in any one situation as another, 

 the probability that any one particular star should happen to 

 be within a certain distance (as, for example, one degree) of 

 any other," and so forth, as quoted in §8. I think that it is 

 impossible to doubt that Mitchell meant to say, that the equa- 

 lity of chance referred to exists in the nature of things, in the 

 same sense exactly as we admit an actual equality of chance 

 that any one side of a well-poised die shall turn up, — as well 

 as a hypothetical equality of chance that a given side shall 

 turn up independently of our knowledge of whether the die 

 be loaded or no. In treating of our first objection, we have 

 shown that Mitchell and his followers have confounded the 

 measure of hypothetical antecedent probability of a given re- 

 sult with a probability in the nature of things, or an actual 

 probability, and have used the measure of the former (which 

 we admit may be found correctly) for the measure of the latter. 

 The Idea from which Mitchell starts is "scattering by mere 

 chance," the application of the idea to the special case is 

 61 that every star is as likely to be in any one situation as an- 

 other." The two things are assumed to be equivalent, the 

 effect of the one supposition is the same as the effect of the 

 other, and either is to be compared with the effect observed ; 

 and the probability or improbability of the Idea is to be thus 

 tested. I believe that Mitchell held, and that most of those 

 who have followed him hold this to be the true basis of the 



