Contextual Contingency as a Principle of Change

Perhaps surprisingly, this indeterminacy need not have such disquieting implications for the social scientist, let alone implications of irrationality in regard to the selections of science. Recent developments in the theory of self-regulating systems (as well as in thermodynamics) suggest the opposite interpretation—that is, that such indeterminacy is a necessary prerequisite for progressive, organised adaptation, and thus for survival and reconstructive change.³⁴ In other words, the effect of indeterminacy is no longer viewed as purely disruptive as is the "noise" of information theory which prevents correct transmittance of a signal, the "errors" in the genetic code which prevent normal biological replication, or the "perturbations" in a thermodynamic system. Rather, it is seen as the sine qua поп for a progressive organisation of the system toward increasing complexity, in spite of local error or loss of information.³⁵

To expound this thesis, let us look at an example by Von Foerster as reinterpreted by Atlan.³⁶ A certain number of cubes, some of whose surfaces have been magnetised positively and some negatively, is placed in a box which is then shaken. When the box is opened, the cubes are seen to be arranged in complex, stable geometric figures which seem to have been consciously devised by an artist. However, the shaking was nothing but chance intervention, unpredictable and independent of any preceding or future arrangement of the cubes. For anyone who did not know that the cubes were magnetic, they would seem to have organised themselves in response to a chance intervention which was itself disruptive, since it destroyed the original stable, orderly pattern of the cubes.³⁷

A somewhat different example—and one which points more clearly to the levels of organisation involved—is that of biological reproduction. We know that an "error" in the transcription of the genetic code is thought to be the cause of mutations. Yet this random event on the (strictly repetitive) genetic plane can benefit the species by creating a variation which is better adapted to changing environmental conditions than the original population. The species "reorganises" itself by integrating a random mutation which has disrupted the orderly pattern of straightforward reduplication.

In the language of communication theory (which is perhaps better suited to questions of social organisation), the issue can be reformulated following Atlan (1979: 47). Suppose we have a communicative link between two subsystems A and В within a certain system. If there is no error in a message transmitted from A to B, then B will be

The Scientist as a Practical Reasoner 11

an exact copy of A and the total information of both will be identical to that of A. If the number of errors is such that the ambiguity is identical to the amount of information A transmits, this information will be lost to such a degree that we cannot even talk about transmittance. This means that the structure of В is completely independent of that of A, and that the total information of both corresponds to that of A plus that of B. To the degree that the system depends on the communicative link between these subsystems, this total independence will amount to a destruction of the global system. With regard to the amount of information of the global system, the optimum is to have a non-zero transmission of information between A and В and a certain amount of error in this transmission.³⁸

What does the claim that a certain amount of indeterminacy is constitutive for progressive self-organisation suggest in the case of science? A minimal definition of scientific development seen as directed change would assume that scientific knowledge is progressively reconstructed knowledge based upon the integration or elimination of earlier results, and that this reconstruction is a process of complexification. Complexification here means that the system is able to construct and reconstruct itself in new ways.

In common terminology, there are two correlates of this process. On the one hand, there is the ability of science to construct "new" information; that is, to produce "innovation." On the other hand, science is apparently increasingly able to construct and reconstruct itself in response to problem challenges by providing solutions to the problem,³⁹ which, I suppose, is what we mean when we talk about the success of science. Both abilities are aspects of the process of complexification, which in Shannon's sense corresponds to an increase of information.⁴⁰ But, as we have seen, without indeterminacy there could be no such increase of information. This indeterminacy seems to be nothing more than the degrees of freedom utilised by the system for a problem-absorbing reconstruction of itself. It becomes manifest in the observer's inability to specify in detail a small set of criteria or a principle of rationality according to which this reconstruction proceeds.

How does the idea of such a complexity-increasing reconstruction relate to the notion of contextual selection emphasised earlier in this section? System theory cannot conceive of self-organising systems, without assuming an environment to which a system responds.⁴¹ Deprived of this notion of context, the argument introduced here makes no sense. It is the context which orients, through the selections it promotes, the process of reconstruction and development. We have introduced the notion of context here to refer to the fabric of situated variables on which the scientists base their decisions. These variables appear as the constraints upon which the scientists hold their selections to be contingent, and as the constraints they impose through decision-translations in order to reach closure in an essentially open and expanding sequence of events. Without indeterminacy with respect to these constraints, there would be no problem of closure. And, it seems, without indeterminacy there would be no new constellations of selections.

Поиск по сайту: