International Encyclopedia of Systems and Cybernetics

"A box to which inputs are observed to lead and from which outputs are observed to emerge" (St. BEER, 1968, p.293).

The "black box" model has been - according to ASHBY - introduced by J.C. MAXWELL to "justify the building of functioning descriptions (in his case equations) that accounted for the observed behavior of some phenomenon when the workings of the phenomenon were not clearly visible" (R. GLANVILLE, 1982, p.1)

And "Briefly, a black box can be characterized as:

a) being believed to be distinct

b) having observable (and relatable) inputs and outputs

c) being black (that is, opaque to the observer") (Ibid. , p.1)

Of course, any black box is black for some specific observer, i.e. becomes a construct of the observer through his peculiar way of observation.

GLANVILLE, who discusses widely the black box model, adds that, after experimenting with the box, "what the observer has is a functional description that has worked in the past. That it will continue to do so is a pure article of faith: The black box regularity is an assumption" (Ibid, p.2).

However, when repeated observations produce similar results, this assumption is reinforced and "it is said that the black box has become white. The whiteness (as WIENER, 1948, asserts) is not a property of the box, but of the observer's interaction with it" (Ibid, p.4).

Viewed possibly from outside (by another observer), the interaction between the black box and the observer is, in fact, between two black boxes, and the whole system is thus a higher level black box.

In order to reach some consensus about a (tentative) conclusion we should thus clearly state our own observation level.

The black box can be either a concrete experimental artifact, or a conceptual model.

R. ESPEJO writes: "The black-box construct is a shorthand for the real world, that is, for the transformations actually taking place in it"(1988, p.140).

It however should never be forgotten that different observers may interpret the same black-box behavior in different ways (W. KARGL, 1991 , p.577). The internal organization of this box is at the start of the experiment totally unknown. It must be conjectured by making hypotheses about the links between inputs and outputs and modifying these hypotheses as the experience is furthered.

St. BEER comments: "The reason why we contemplate a box having such odd properties, is that the more familiar box in which something is known about the internal connectivity, has its variety (already) constrained… Even if an arrangement having fairly high variety were chosen, the potential variety of all the other arrangements which could have been built in is suppressed. But a black box is assumed to be able to take on any internal arrangement of input- output connectivity at all; it can therefore proliferate maximal variety" (Ibid.).

→Grey box.