Nature published a special issue last week for Alan Turing’s centenary; were he alive, he would be 100. Turing is widely regarded as the father of the computer. His contributions to code-breaking were key to cracking the Enigma code, and helped win World War II for the Allies. His untimely death at the age of 41 was a sad reminder of the power of prejudice over gratitude.
Turing’s work was fundamental to the field of cybernetics, which was as short-lived and as influential as he was. Although Turing himself never published on the subject, it drew heavily on code analysis of the kind he did during the war. Cybernetics existed because early computing existed, and all at once the passage of messages became at least as interesting in their content. This is especially germane if you think about computer science in terms of its code-breaking background; Neal Stephenson’s Cryptonomicon tells this story, among others, beautifully.
Cybernetics made sense of information transfer by simplifying down to a system that could transduce input to output. It had a lot in common with thermodynamics, the study of heat transfer, which had solidified over the previous century or so and which also takes an input/system/output view of the world.
This ground rule was elaborated as different systems were studied in cybernetic terms. One of my own favorite elaborations (it’s delightfully silly!) was the study of second-order cybernetics, or understanding how information is transferred when we try to understand information transfer. (In the picture above, the lower diagram shows second-order cybernetics; Wiener, Bateson and Mead are three of the founders of the field).
The field was at its strongest during the Macy conferences, a series of academic conferences between 1946 and 1953 dedicated to both developing and spreading the ideas of cybernetics. You could make a cybernetic study of the conferences themselves; although we have the schedules, and writings from academics that reflected on the ideas shared there, but no proceedings were ever published, so we’ll never know exactly what was said. (So much has changed in science communication; today conferences are liveblogged up one side and down the other, like the SciOnline unconference series.)
Cybernetics was holistic at a very reductionist time; it played a major role in coming to understand biological systems in terms of their emergent properties. Its findings took two directions: toward the development of computers as more-and-more complicated information processing machines, and toward understanding the brain, which remains the most complicated information processor we know of.
Conventional wisdom says that the movement lost momentum after the conferences ended. Its progenitors were from different fields, and they never made a real interdisciplinary mark despite their ambition to found a unified study of everything. However, it threw a long shadow across academia; cybernetics often turns up in astrofuturist poetry of the sixties, and modern scholars of systems and complexity point to cybernetics as their progenitor field.
I think of the 1940s and 1950s with a weird nostalgia, as a golden age in American academic science, and cybernetics plays a big part in why I think that way. After all, American science was in no way socially or ethically ideal in that period; scientists developed the bomb, stole Henrietta Lacks’s cells, carried out the Tuskegee study, all the while excluding from their ranks many people who would have been great scientists. But strictly in the realm of ideas, it was a time of very great discovery: the structure of DNA, the most basic signaling that controls a cell, the rules behind genetics and virology; not to mention the great strides in physics and chemistry that were made at the same time. Because discoveries made back then are fundamental to science education now, and experiments were both elegant and easy to explain, there’s an illusion that information was more manageable. But then I’m reminded it wasn’t; scholars back then founded a whole new field strictly for understanding information flow.