Ross Ashby's General Theory of Adaptive Systems (Umpleby, 2004, 2008)

p.2 The second type of more general theory is a more abstractly worded theory. The theories of Ross Ashby are examples. However, these theories still require the knowledge in more specialized fields in order to operationalize them and put them to use. For example, Ashby spoke about the need for requisite variety in a regulator... In game theory variety is expressed in possible moves.

p.3 [Ashby] Cybernetics treats not things but ways of behaving. It does not ask "what is this thing?" but "what does it do?"... It is thus essentially functional and behavioristic... The truths of cybernetics are not conditional on their being derived from some other branch of science. Cybernetics has its own foundations (Ashby 1956, 1). [JLJ - Italics added to match Ashby's use in original]

p.3 As I read Ashby's books I imagined my own examples in fields of interest to me... [JLJ - so did I] Hesitancy to exercise imagination [JLJ - no problem for me...] may be an obstacle to appreciating the relevance and importance of Ashby's work.

p.3 Ashby was concerned... with organized complexity [JLJ - so am I]... His approach to studying organized complexity was unusual... Ashby chose to look for constraints or interaction rules which reduce the maximum possible variety to the variety actually observed. Laws... are examples of constraints, which reduce variety from what can be imagined to what is observed.

p.3-4 Ashby's... interdisciplinary theories are more general or abstract than the theories in most disciplines... However, theories at a more general level are neither sufficient nor necessary. A more generally worded theory is not sufficient because "domain-specific knowledge," which is obtained from discipline-based theories, is still needed in order to apply the theory in practice.

p.4 It is remarkable that Ashby was able to formulate theories that work for so many domains. Discipline-based theories do not. One can take the formal structure and operationalize it in many fields. Ashby's general theories then become a tool for developing more specific, operationalizable theories in specific disciplines.

p.4 What is observed, he called the "machine." For Ashby, "the system" is an internal conception of "the machine." A "system" is a set of variables selected by an observer... because he defines a system as a set of variables selected by an observer, his work is quite compatible with second order cybernetics [the idea that the observer should be included within the domain of science].

p.4 Ashby divides all possible outcomes into the goal subset and the non-goal subset. The task of the regulator is to act in the presence of disturbances so that all outcomes lie within the goal subset.

[JLJ - perhaps this is asking more than we really need. We can never know the future with certainty in a complex environment - but we don't need to. We can instead use the regulator to guide our consequential exploration efforts in developing scenarios. We make sure that the possible future that is critical strategically, as it emerges, consistently lies in the goal subset, with margin for safety and the unforeseen. The future that is not as critical strategically - perhaps involving our our opponent disengaging from the struggle for key squares - we can spend less time investigating or even ignore, when quickly constructed diagnostic tests show evidence of sustainability (for our position).

Think of an animal moving in the woods. It does not "calculate" the future, it positions itself for possible futures. It listens for signs of being stalked, knowing that it can rely on speed (and other forms of resilience, such as climbing a tree) to flee from an attacker.

Our goal then (in playing in the positional style in a strategic game) is to guide the reflex-driven development of diagnostic tests of adaptability, not to determine, once and for all, "certainty" about future states of our system. We are satisfied when our well-crafted scenarios - diagnostic tests of adaptability - show evidence of stretching and adapting behavior, and our strategic challenge lines show good fallback positions for possible action if needed.]

p.4-5 An error-controlled regulator can be very simple, for example a thermostat. A cause-controlled regulator requires a model of how the machine will react to disturbance. One consequence of Ashby's view of regulation is the Conant and Ashby theorem, "every good regulator of a system must be a model of that system." (Conant and Ashby 1970). Von Forester once said that Ashby told him this was the idea he was looking for when he began his explorations in cybernetics.

[JLJ - every good regulator of a system must be able to generate hypotheses about how the system might act - these hypotheses are then constructed into scenarios by some higher-level cognition. We can regulate only when we can imagine how the present and future forces might interact - perhaps jointly - to shape the evolution of the system.]

p.5 For Ashby learning involved the adoption of a pattern of behavior that is compatible with survival... As a psychiatrist and director of a psychiatric hospital, Ashby was primarily interested in the problem of adaptation.

[JLJ - yes, including the type of selection which involves choosing which of the possible futures that might emerge demand further attention, in determining whether the positions further down the road are sustainable, and which possible futures do not - at least now - need attention.]

p.6 Ashby... asked the question, "can a mechanical chess player outplay its designer?" He answered the question by saying that a machine could outplay its designer, if it were able to learn from its environment.

[JLJ - a machine can outplay its designer if the designer (the programmer) is able to write a sequence of steps that better reduces the complexity in the environment, and which produce better attention-guiding diagnostic tests of adaptive capacity. It is that simple. A machine cannot outplay its designer in the game of tic-tac-toe (noughts and crosses), because the game is simple.]

p.6-7 When confronted with a complex situation, there are only two choices - increase the variety in the regulator... or reduce the variety in the system being regulated. [JLJ - hire a consultant, run, find someone to blame, address the real problem, perform research, ask someone smarter than you, come up with a trick that works - usually, construct scenarios of how things might play out and determine what critical skills or information is necessary to survive, change jobs, change careers, laugh, cry, laugh and cry, ...]

p.7 By choosing a more conceptual strategy, rather than a more direct and immediate strategy, it becomes possible to regulate a very large system... The law of requisite variety says that variety must be controlled, if successful regulation is to be achieved, but variety need not be controlled directly. If one is clever in creating conceptualizations and organizational structures, the amount of variety that can be controlled can be very large.

p.8 As a transdisciplinary field cybernetics serves as a catalyst for further developments in many fields. That is the role that cybernetics and general systems theory have played until now.

p.8-9 Despite the fact that more general theories are more valuable because they explain more phenomena with fewer statements, Ashby’s theories have not received as much attention as they deserve. The reason no doubt lies in the traditions in universities that enforce narrow specialization. However, as knowledge grows and an integrated understanding is needed to cope with the problems of a global society, probably increased attention will be paid to more general theories. When that day comes, Ashby’s work will receive renewed attention and acclaim.