Sunday 30 November 2014

Social Ecology and Soviet Cybernetics

I came across a fascinating presentation by Vladimir Kitov of the Russian Plenhanov University of Economics on the history of Russian cybernetics and the Russian computer industry. You can view his presentation here: https://njszt.hu/sites/default/files/Kitov.ppt. It's an extraordinary story. There's also an excellent paper on Soviet Cybernetics by Ben Peters at http://nevzlin.huji.ac.il/userfiles/files/47.2.peters.pdf. This is really fascinating stuff.

I'd always understood Russian cybernetics to have taken an independent direction from cybernetics in the West. Kitov (and Peters) suggests that the truth is more complex. After the Macy conferences of the late 1940s, it seems that cybernetics was seen by the Soviet hierarchy as a kind of bourgeois plot. Peter's quotes from the Russian Journal Literaturnaya gazeta  which accuses Norbert Wiener of belonging to a group of "charlatans and obscurantists, whom capitalists substitute for genuine scientists". It later carried an article: "Cybernetics—an American Pseudoscience". Those Soviet Scientists who actually read Wiener's book (which was marked 'Top Secret') took a different view. It took Stalin's death and the Khrushchev thaw for Sergei Sobolov, Anatoly Kitov and Alaksei Lyapanov (Russia's equivalent of Von Neumann) to write an article called "The main features of cybernetics" in 1955.

More importantly, the ideas of the soviet cyberneticians were particularly focused on the social and economic use of cybernetics. In 1959, Anatoly Kitov proposed to the Kremlin that a computer system was developed to manage the whole Russian economy providing real-time feedback on production. This ambitious request was rejected, although it remained a long-held dream of Kitov and other scientists: a programming language called ALGEM (a variant of ALGOL-60) was developed to assist in the realisation of this economic management system. It is fascinating to think that the project which Stafford Beer managed to realise in Chile in the early 70s was already conceived years earlier. What did Salvadore Allende know about this earlier Soviet work, I wonder?

More importantly, however, is the role of theory in these plans. Although it is relatively easy to find cursory descriptions of the ALGEM language, there is little on what the theoretical understanding for implementing a feedback system in the Russian economy would look like. Beer's intervention in Chile was based on his concept of 'viable systems'. Using the Viable System Model, Beer wasn't simply trawling data in the hope that something meaningful would come up (this is very much the approach of today's big data people). My guess is that the Soviet scientists too had theories about how their real-time economic system would work: the history of science in the Soviet Union, and the particularly social trend in scientific thinking, including Vygotsky (who had been disowned by the hierarchy - a factor which contributed to his suicide) and Leontiev, whose work on networks is only now beginning to be taken seriously by ecologists. Peters tells us that the close potential association between communism and cybernetics was not lost on American observers. When Aksel Berg published a series entitled " Cybernetics—in the Service of Communism", one american observer noted that "if any country were to achieve a completely integrated and controlled economy in which ‘cybernetic’ principles were applied to achieve various goals, the Soviet Union would
be ahead of the United States in reaching such a state." But this wasn't to be.

According to Peters, Soviet cybernetics disintegrated in a way which mirrors a similar collapse of cybernetics in the West: a process of disciplinary bifurcation helped to dilute and distort the main thrusts of what it was about. Nothing new there. But what interests me is Kitov's account of the demise of the Soviet computer industry. This came about when the Kremlin decided that their independent computer science programme was falling behind the US, and that  they might as well copy the IBM 360 architecture, consigning their existing research department to the dustbin. Kitov's analysis of this is that this might have been "one the most successful operations against the USSR by the CIA" Sadly, it was probably self-inflicted. What was lost was the diversity, flexibility and adaptability within the ecology of the Soviet scientific community.  It could no longer grow new ideas for itself; instead it began on a path of slavishly copying US corporations: a process which it has tragically continued to this day!

There's some fascinating history to trawl here. But I think there is also a warning. Globalisation, the copying of an idea from one place to the next, is not necessarily the best way that societies can make themselves adaptable. Indeed, it risks making them very brittle. The Soviet cybernetic and computer science programme effectively operated as a parallel universe to the US and UK programmes, and it was no worse for that. The rich diversity of approaches made for a fertile time where ideology and technics combined. When we look at the world today, how much flexibility are we making for ourselves? How much are we simply following each other, running after the whirling banners of global corporatism in the name of profit? What role are our universities playing in this dance? 

No comments: