7/10 THE ULTRASTABLE SYSTEM 



7/8. What has been deduced so far in this Chapter is necessary. 

 That is to say, any system that has essential variables with given 

 limits, and that adapts by the process of testing various behaviours 

 by how each affects ultimately the essential variables, must have 

 a second feedback formally identical (isomorphic) with that 

 described here. This deduction holds equally for brains living 

 and mechanical. 



To be quite clear in this matter, let us consider the alternative. 

 Suppose some new species, or some new mechanical brain, were 

 found to change from the non-adapted to the adapted condition 

 (S. 5/7), doing this consistently even when confronted with quite 

 new situations ; and suppose that, in spite of what was said above, 

 investigation showed conclusively that there was no second feed- 

 back of the type described — what would we say ? 



There seem to be only two possibilities. We must either invoke 

 a hitherto unknown channel (in spite of the investigations), as 

 one was invoked after the demonstration by Hertz (S. 4/13); 

 or we must be willing to accept as natural that the system S 

 should go to correct values without being given an appropriate 

 input. This second possibility would be accepted by no one, 

 for the situation would be like asking an examiner to accept as 

 natural a candidate who gives the correct answers without being 

 given the questions ! If this possibility is rejected, we are left 

 only with the possibility that the second channel, in some form 

 or other, must be there. 



The implications of double feedback 



7/9. We may now usefully consider the relation between adaptive 

 behaviour and mechanism from the opposite point of view. So 

 far in this chapter we have taken the facts of adaptive behaviour 

 as given and have deduced something of the underlying mech- 

 anism. We will now take the mechanism and ask: Given such a 

 mechanism, in whatever material" form, will it necessarily show 

 adaptive behaviour ? The answer to this question will occupy 

 the remainder of this chapter and the next. 



7/10. Let us get the basic assumptions clear for a completely 

 new start, assuming from here to the end of the chapter only 

 what is stated explicitly. 



85 



