14/5 AN INTRODUCTION TO CYBERNETICS 



A third method for getting a regulator made is to assemble it in 

 hardware, as a mechanic makes a water bath. Again selection is 

 necessary: components have to be selected (distinguished) from other 

 possible objects, and the mode of assembling and coupling has to be 

 selected from the other, incorrect, modes. The quantity of selection 

 used can be measured ; and any dispute about the measurement can 

 be resolved by the method of S.13/11 (final paragraph). 



It follows from S.13/18 that if the final regulator can be arrived 

 at by stages (the whole selection occurring in stages) the possibility 

 exists that the provision of a small regulator at the first stage may 

 lead to the final establishment of a much bigger regulator (i.e. one of 

 larger capacity) so that the process shows amplification. 



This is the sense in which "amplifying" regulation is to be under- 

 stood. The law of Requisite Variety, hke the law of Conservation 

 of Energy, absolutely prohibits any direct and simple magnification 

 but it does not prohibit supplementation. 



14/5. Let us consider some examples which will actually show such 

 ampUfication of regulation. 



Suppose the disturbances are fluctuations in the mains' voltage, 

 which come to an apparatus owned by Q at the rate of hundreds a 

 second, and threaten to disturb it. Assume that the variety per 

 second provided by these disturbances far exceeds his capacity as a 

 channel, so it is impossible for him to regulate against them by direct 

 personal action. However, he has available a manufacturer's 

 catalogue, which shows three items : 



1 : Television set, 

 2: Mains stabihser, 

 3 : Frequency changer. 



Assume that it is within his capacity for him to make a suitable 

 selection of one from three; if now he performs the appropriate 

 selection, the end result will be that the mains' supply to his apparatus 

 will become stabilised. Thus his three possible primary selections 

 can be put into correspondence with three outcomes, one of which 

 is "mains' voltage stabilised". 



The latter regulation (over, say, a year) involves far more selection 

 than of one from three; so over the whole transaction an undoubted 

 amplification has occurred. 



In this example the supplementation is so obvious, and his 

 dependence on the manufacturer's power as a designer so blatant, 

 that the reader may be tempted to dismiss this "amplification" as 



268 



