CHAPTER 17 



Serial Adaptation 



17/1. We have now reached a stage where we must distinguish 

 more clearly between the organism and its environment, for the 

 concept of the l multistable ' system clearly refers primarily to 

 the nervous system. From now on we shall develop the theme 

 that the nervous system is approximately multistable, and that 

 it is joined to, or interacts with, an environment. But before 

 discussing the events in the nervous system we must be clear on 

 what we mean by an ' environment '. So far we have left the 

 meaning very open : now we want to know what we mean exactly. 

 The question occurs in its most urgent form to the designer of a 

 4 mechanical brain ', for if he has designed this successfully he 

 still has to decide with what it shall interact : having made a 

 model of the brain, he must confront it with a model of the 

 environment. What model could represent the environment 

 adequately in principle ? 



It seems clear that we can, in general, put no limit to what 

 may confront the organism. The last century's discoveries have 

 warned us that the universe may be inexhaustible in surprises, so 

 we should not attempt to define the environment by some formula 

 such as ' that which obeys the law of conservation of energy ', 

 for the formula may be obsolete before it is in print. In general, 

 therefore, the nature of the environment must be left entirely 

 open. 



On the other hand, we may obtain a partial definition of some 

 practical use by noticing that the living organism on this earth 

 adapts not to the whole universe but to some part of it. It is 

 often not unlike the homeostat, adapting to a unit or two within 

 its immediate cognisance and ignoring the remainder of the world 

 around it. Yet, given a particular organism, especially if human, 

 we cannot with certainty point to a single variable in the universe 

 and say ' this variable will never affect this organism '. This 



179 



