Norbert Wiener, answered from the texts and cited to the page.
The scientist's enemy is the Augustinian devil — the negative evil of confusion and disorder, nature's entropic tendency, the absence of organization rather than the presence of malice. I have already pointed out that the devil whom the scientist is fighting is the devil of confusion, not of willful malice. The work of the scientist is anti-entropic: against the natural tendency of organized systems to dissolve into noise, the scientist constructs and maintains organization.12345
By contrast the Manichean devil — active malice, craftiness, dissimulation, the determination to win by any trick — is the kind of opponent the wartime adversary represents and the kind of opponent von Neumann's game theory was designed to model. The two devils require different tactics: against the Augustinian devil one applies methodical organization, statistical inference, the building of self-correcting systems; against the Manichean devil one must anticipate strategy, account for deception, expect change of method.
The cosmological commitment of cybernetics is to treat the world primarily as the field of the Augustinian devil — entropy and noise — while acknowledging that within that field certain opponents (human enemies, market actors, biological parasites) play the Manichean game.
The view that nature reveals an entropic tendency is Augustinian, not Manichaean. Its inability to undertake an aggressive policy, deliberately to defeat the scientist, means that its evil doing is the result, of a weakness in his nature rather than of a specifically evil power that it may have, equal or inferior to the principles of order in the universe which, local and temporary as they may be, still are…The Human Use of Human Beings, pp. 192–193
Walter and Dr. W. Ross Ashby. The machine, like the living organism, is, as I have said, a device which locally and temporarily seems to resist the general tendency for the increase of entropy. _ By its ability to make decisions it can produce around it a local zone of organization in a world whose general tendency is to run down.The Human Use of Human Beings, pp. 36–37
Ex- perience has pretty well convinced the working physicist that any idea of a nature which is not only difficult to interpret but which actively resists interpretation has not been justified as far as his past work is concerned, and therefore, to be an effective scientist, he must be naive, and even deliberately naive, in making the assumption that he is dealing with an honest God, and must ask his questions of…The Human Use of Human Beings, pp. 191–192
Perhaps this devil is not far in meaning from Mephistopheles. When Faust asked Mephistopheles what he was, Mephistopheles replied, "A part of that force which always seeks evil and always does good." In other words, the devil is not unlimited in his ability to deceive, and the scientist who looks for a positive force determined to confuse us in the universe which he is investigating is wasting his time.The Human Use of Human Beings, pp. 38–39
Yet in their recognition of a fundamental element of chance in the texture of the universe itself, these men are close to one another and close to the tradition of St. Augustine. For this random element, this organic incompleteness, is one which without too violent a figure of speech we may consider evil; the negative evil which St.The Human Use of Human Beings, pp. 13–14
The distinction is central to The Human Use of Human Beings and to God and Golem; it organizes my late writing. I will acknowledge Galison's reading that my wartime AA-predictor work treated the enemy pilot as a Manichean opponent and that the postwar cybernetic vision did expand to take nature itself as a more globally Manichean field than my published distinction would suggest. The line between the two devils is real and worth holding; my own practice did not always honor it as cleanly as my prose did.