19/16 DESIGN FOR A BRAIN 



would find the system unsatisfactory. He would establish that 

 the ball, started at A, always went to A' ; and started at B it 

 always went to B'. He would find its behaviour at C difficult 

 to explain. And if he tried to clarify the situation by starting 

 the ball at C itself, he would find it went toD! He would say 

 that he could make nothing of the system ; for although each 



Figure 19/15/1. 



line of behaviour is accurately reproducible, the different lines 

 of behaviour have no relation to one another. 



This lack of relation means that they do not form a ' group '. 

 But whether the experimenter agrees with this or not, he will, 

 in practice, reject this 2- variable system and will not rest till 

 he has discovered, either for himself or by following Newton, 

 a system that is state-determined. In my theory I insist on the 

 systems being absolute because I agree with the experimenter 

 who, in his practical work, is similarly insistent. 



19/16. That the field of a system should not vary with time, 

 it is necessary and sufficient that the system be regular. The 

 proof is obvious. 



19/17. One reason why a system's absoluteness is important is 

 because the system is thereby shown to be adequately isolated 

 from other unknown and irregularly varying parameters. This 

 demonstration is obviously fundamental in the experimental 

 study of a dynamic system, for the proof of isolation comes, 



210 



