p.83 Ashby (1960, p. 1)... developed a model to show logically that in any strictly deterministic, "state determined," system there occurs a natural selection toward states of equilibrium... Stated concisely, "...every single-valued operation tends to select forms that are peculiarly able to resist its change-inducing action.... And when it occurs on the really grand scale... the states selected are likely... to show among their parts, a marked co-ordination tendency to make them immune to the operation" (Ashby, 1960, p. 233).

p.84 degree of coupling is a useful abstraction in understanding the stability of living systems.
 The degree of coupling, or interaction, between two systems depends on the activity of the variables which they share.

p.84 To the extent that two systems either have few variables in common or if the common variables are weak compared to other variables which influence the system, they are independent of each other. It is convenient to speak of such a situation as one of loose coupling and also to note that insofar as one system, A, is independent of another, B, we may speak of the persistence of the behavior of A in the face of the behavior of B.

p.85 A definition of loose coupling adequate for the present purpose therefore demands addition of two qualification to Klir's (1969, p. 154) set-theoretical definition of directed loose coupling as a condition in which the output variables of the driving system, A, and the input variables of the driven system, B, constitute only a subset of the effective variables:
 (1) Time must be isolated as a special variable and the above statement be taken as referring only to a given interval of time during which the coupling is judged; over the long run the influence of one subsystem on another may be stronger than it appears within any given short interval...
 (2) Provision must be made for stronger and weaker variables... the group connected by weaker variables is more loosely coupled.

p.85 the complexity of living systems means that their stability will be seen only over relatively long intervals within which one may want to look at the details of organization and also because one will ordinarily look at only part of the range of only some of the variables involved in the coupling.

p.85 Because a higher level of the living system contains components made of systems at lower levels, an operation of the higher level, e.g., an adaptation, is more complex and can go no more quickly than its component parts.

p.87 Examples: Cell and organ
 1. Hormonal regulatory systems are slower acting than those which involve strictly neural control, and are associated with stable features of physiological functioning and homeostasis... including circadian and longer rhythms...
 2. As persistence in the behavior of a system is associated with loose coupling of information into the system, behavior which involves rapid adjustments is, in contrast, associated with tight coupling. A neural example of tight coupling is provided by the motor control systems

p.87 The cerebellum, occupying a large part of the bulk of the brain, is primarily an inhibitory structure which performs a smoothing function of control of movement... The classical symptom of damage to this structure is tremor of intention, instability of voluntary, directed movements

p.87 Examples: Organism
 1. A prototypical example at the organism level of stable behavior occurring as a result of loose coupling between the behavior and feedback about its consequences is the partial reinforcement effect.

p.91-92 the prominent overall feature of a living system is its independence of momentary environmental changes, which it achieves by virtue of its self-adjusting property. A given system is stable in the face of environmental variation and even goes about increasing its organization in a universe of increasing entropy. This grand strategy of self-adjustment... may be viewed as a method by which... the system as a whole [maintains] persistent behavior in the face of variables which would otherwise disturb [it].

p.92 the degree of coupling in the system evidently represents an adaptation to environmental contingencies... the organism adapts by displaying persistent behavior which is apparently possible only because it is able to ignore, or habituate to what have turned out to be irrelevant short-term perturbations, while responding to the long-term contingency.

p.92 Other instances of loose coupling... may be classified as passively maintained. In cases of this sort, the system does not actively defend itself against the imposing variables, rather certain features of the system may be said to insulate it, giving these variables only limited access.

p.92 Any system which responds in a digital or step function manner rather than a continuous manner may be thought of as loosely coupled in this passive sense if the thresholds of response are judged significantly far apart. [JLJ - this appears to be one of Weick's 5 loose coupling criteria from his 1982 paper... in fact he seems to have read pages 87-96, thought a bit on Glassman's material, then generated his 5-item criteria as a sumamry]

p.92 An important advantage possessed by passive loose coupling is the absence of need for the difficult design which is involved in any system requiring tight coupling. However, passive loose coupling can appear... only when the imposing variables can be safely ignored.

p.94 evolution is ultimately a gradual process of accidental change, which we now observe with hindsight (Wald, 1967, pp. 70-71).

p.95 It may be noted that making judgments about adaptiveness is one part of the process whereby change is brought about through human planning rather than by accidental variation and selection.

p.95 In adult humans, comparable play may lead to invention of new adaptive combinations of subsystems.

p.96 this example also serves to illustrate the uncertainty of judging what is truly adaptive... possession of a physical feature seen as attractive by the human species may be a critically important adaptive characteristic.

p.96 Can some sort of quantitative formulation readily lead to predictability in specific cases? In accordance with the definition of loose coupling, a quantitative formulation might ask a question about the latency, probability or size of a particular response of a system at a given point in time.

p.97 in examining a specific living system we do have to deal with all or much of the complexity; the problem may... be... in enumerating the effective variables, their relative weights... and enumerating the possible responses of the system.

p.97 the real world, in the form of models and prototypes, must be used to complete the solution of problems where many variables can interact... loose coupling is a meaningful concept to apply to various systems... It provides a framework or template for abstracting the regularities in a wide variety of phenomena and provides questions with which to pursue particular examples.