Published online by Cambridge University Press: 03 July 2024
Cybernetics has fallen prey to snobs and journalists, who, in dealing with it, tend to mix myth with science. In this study we shall try to sift out the chaff, which a regrettable sensationalism has needlessly mixed with the good grain. Concentrating our attention on the rational bases and some of the lines of force of this new field of study, we shall try to eliminate the element of fable, but we shall not prohibit ourselves from opening windows on any perspectives that seem reasonable.
The notion of “information,” cornerstone of cybernetics, sheds light on the theory of knowledge. What is science, indeed, but an interpretation, which tries to be objective, of the flood of signals with which the universe submerges us? A Protean notion, it permits us to discover a certain unity among the most disparate phenomena. But it is a dangerous notion, too, because of the multiplicity of senses one may give to it.
1. This term was invented—or rather, reinvented, for Plato and Ampere had already used it—by Norbert Wiener in 1947, as a tribute to Clerk Maxwell.
2. An electric or electronic current passes, or does not pass, through a conductor or a tube.
3. Certain intellectuals will not permit the "disorder" of their papers to be regulated, because they know how to find their way through it; what appears to the layman to be dis order is, for them, order.
4. Sometimes, but much more rarely, with other authors: "ectropy."
5. Information may be measured in units defined as above, or in thermodynamic units of entropy. Each unit ofinformation is equivalent to an entropy equal to K In 2—that is, about 10-16 thermodynamic units.
6. "Principe de Néguentropie pour l'Information," in Louis de Broglie, Physicien et Penseur (Paris, Michel, 1953), p. 368.
See also the same author's article in the Journal of Applied Physics for May, 1954.
7. Sometimes the term "bit" is used—an abbreviation of "binary digits."
8. Telecommunications networks use a special method of transmitting messages. Signals circulate as modulations (amplitude, frequency, etc.) of a sinusoidal current.
9. Attempts have even been made—without great practical success so far, it must be said— to consider the propagation of such a slow geological phenomenon as the formation of a mountain chain as a kind of transmission of information.
10. For example, to find in three weighings—without the use of weights or tares—a counterfeit coin mixed with twelve good ones.
11. In Foundations of the Theory of Signs (Chicago, University of Chicago Press, 1938), p. 30.
12. The word "circuit" is obviously borrowed from electrical terminology. But it re mains valid when one is speaking of the nervous system or of social arrangements.
13. It should be clearly understood that all the words in this paragraph which are placed between quotation marks are to be taken only as images.
14. Walter B. Cannon is responsible both for defining the notion and applying the term.
15. We should, however, cite Emile Borel in one field, and Bascal and Edgar Allan Poe (apropos of games) in another; and also Buffon, Condorcet, Poisson, and Cournot in connec tion with the probabilistic study of judicial decisions, electoral laws, commercial competition, etc. As early as 1928, von Neumann demonstrated a theorem into which enters the notion of "ruse."
16. A semi-lattice is an ordered group—that is to say, a collection of elements in which a (partial) relation of order can be defined—in which there exists an operation by which a third element can be associated with any two elements.
17. In some of these problems, the penetration of biology by the exact sciences is illus trated by the complementary character of the specialists who collaborate in the same research. Thus it was a mathematician, Norbert Wiener, who first advanced the hypothesis that if the organism uses negative feedbacks we should expect to find pathological troubles with oscillating structures, and who, applying Van der Pol's calculations, predicted their forms. It was a physiologist, Rosenblueth, who discovered them experimentally; they are the clonic spasms of stretched muscles or the ataxias of the cerebellum. Cooperation of the same type associates mathematician W. Pitts with the neuro-physiologist W. MacCullough. At times, a single scientist combines fields of competence that are usually separated. W. R. Ashby, a psychiatrist by profession, handles figures with talent. All those who have met Norbert Wiener have been astounded by the versatility of his culture.
18. These comparisons have been pursued by Warren MacCullough, who established them on probabilistic bases.
19. According to MacKay (1950), the sensation of the passage of time may be nothing more than the appreciation of the greater or smaller flow of information that penetrates our consciousness in practically continuous fashion. As François showed in 1927, this sensation is linked with the temperature of the body; and, according to Hoagland, the latter regulates the rate of oxidation of certain glucides in the brain.
20. "Transformation groups" are mathematical notions which play an important role in many natural phenomena.
21. We do not assume in any way, here, a substantial existence of the consciousness. If we suppose that consciousness is only an "illusion" which should not be taken into considera tion in building an objective psychology—as certain of the behaviorists believe—then this appearance nevertheless requires an explanation. Optical illusions have a real existence, as illusions, just as do erroneous beliefs.