9/13 ULTRASTABILITY IN THE ORGANISM 



thing special, in the sense that the fully three-dimensional fields 

 (e.g. that of Figure 3/5/2) cannot be reduced to the simple form. 

 Thus certain behaviours, though they do not permit the deduc- 

 tion that they are due to something that is a step-mechanism, 

 may none the less permit the deduction that they are produced 

 by some entity with the special property that its behaviour can 

 be reduced to step-function form; and the latter is a meaningful 

 statement, for not all entities produce behaviour that can be so 

 reduced. 



9/13. What of the nervous system ? If the variables in S (of 

 Figure 7/5/1) were to vary as full-functions, the observer would 

 see only one very complex system moving by a complex continuous 

 trajectory to an eventual equilibrium. Often, however, he 

 observes that the organism ' makes a trial ', i.e. produces some 

 recognisable form of behaviour and then persists in this way of 

 behaving for some appreciable time. Then the trial is aban- 

 doned, a new way of behaving occurs, and this too is persisted in 

 for an appreciable time; and so on. 



When this happens, the observer may justly claim that the 

 system is showing less than its full complexity; for, through the 

 duration of the trial, the fact that it is persisting as this particular 

 form of trial means that some redundancy is occurring. The 

 redundancy is similar to that of the sequence of letters that goes, 

 e.g. 



FFFFFLLLLLLLTTTTTJJJJJJJ 



rather than with full variety: 



EJYMSNASGCGHLAAPEYPJVRQJ 



Within each trial, if it shows a characteristic way of behaving, 

 there will be defined a field ; and these fields will follow one another 

 in a discrete succession, as in Figure 7/23/1. When this occurs, 

 if the observer is willing to assume that the changes of field are 

 occurring in a state-determined whole, he may legitimately deduce 

 that the parameters responsible for the changes from field to field 

 are of such a nature, and so joined to the main variables, that 

 they may be represented by step-functions (for full-functions 

 could not give the discrete movement from distinct trial to distinct 

 trial). That they may be so represented is a meaningful restriction 

 on their nature. 



If now we couple this deduction with what has been called 



129 



