International Encyclopedia of Systems and Cybernetics

Description is necessarily done by an observer, (or a group of them).

Awareness of "something there out" implies some change "here in". In von FOERSTER's words: "Terminal representations (descriptions) made by an organism are manifest in its movements; consequently the logical structure of descriptions arises from the logical structure of movements" and this is the case for "the two fundamental aspects of the logical structure of descriptions, namely their sense (affirmation or negation) and their truth value (true or false)". As a result: "The information associated with a description depends on an observer's ability to draw inferences from this description" (1981, p.262-3).

Synthetically, any description is by somebody, in this somebody's terms. von FOERSTER closes the circle, by saying: "The environment contains no information: the environment is as it is" (Ibid).

Thus, as stated by A. GOUDSMIT: "… description takes place through the interactive behavior itself: it is the activity of specifying distinctions in the environment" (1989, p.172).

Interactive behavior supposes that the system (observer) is equipped to somehow perceive the inputs that will lead to the description. An interesting example are the sunburns by infrared or ultraviolet radiations. When we receive them, we are still unaware of the burns in the making. Since the effect is delayed, we will be able to describe the sunburns only later on. And, as we cannot directly perceive many electro-magnetic radiations, we are able to describe them only through their perceived delayed effects. (two centuries ago, nobody was even aware of their existence).

R. LILIENFELD observed that: "It is possible that "explanation" is a form of knowledge that is both secondary and defective and that another, and neglected, form of knowledge must precede it. This neglected form … would be something called description" (1979, p.6).

H.H. PATTEE, on the other hand, states: "Description is not simple". He sees this as a result "… from our inability to reduce the number of elements and the number of rules that appear essential for the writing and reading of any structure" (1977, p.261).

He offers various possible explanations of this inconvenient situation: "First, there is the requirement in formal systems for explicit representation of every change of the system, whereas in physical systems the laws of motion involving real space, time and energy can take care of many changes without needing description…

"Second, the linguistic rules of formal descriptions can produce only discrete serial operations rather than parallel continuous operations. Construction, on the other hand, can often be performed most efficiently with parallel and coordinated dynamics.

"Third, interest in formal programs and machines usually depend on mathematicians' predisposition for generality or universality which greatly complicates their expressions" (Ibid).

PATTEE went even farther and observed: "… the more you describe the measuring device, the less effectively it measures or describes the system" (Ibid., p.262).

He also stated: "… what we measure is determined by what we choose to ignore about the dynamical details of a system" (1972, p.36). This is quite coincident with HEISENBERG's indeterminacy relation.

We are thus not only restricted in our perception of reality by sensorial limitations, but altogether in our description of it by our modeling languages, formal or just common. In L. LÖFGREN's words: "To notice that a describing system is influencing its descriptions does indeed suggest the idea that what can be objectively describe is limited" (1978, p.244).

This is in some way a startling generalization of HEISENBERG's indeterminacy! It also implies a significant limitation to the power of the observer.

PATTEE's conclusion: "However, until the essential epistemological relation between descriptions and dynamics is faced as a fundamental problem, I doubt if we can claim a useful theory of complex system behavior" (Ibid.,p.265).