Cybernetic Revolutionaries: Belated Review

So I should have done this almost a year ago, but I’m going back to the notes I made because I promised Owen Hatherley I’d eventually write it. Owen Hatherley, I hereby collect your TYR cookie and return you one unit of blog.

Eden Medina’s Cybernetic Revolutionaries is probably the definitive history of Project Cybersyn, the iconic (for some people) effort to create a real-time, decentralised planned economy in Allende’s Chile through computing. I liked the fact that the book is a Latin American history of a Latin American phenomenon, based heavily on interviews with people who were involved and on Chilean archives. This is a worthwhile counterbalance to the impact of Stafford Beer’s own telling of the story.

Show me your Cybersyn and I’ll show you your fears. Cold warriors thought the project was basically GOSPLAN with a faster refresh rate. Communists, for their part, didn’t like it much either – they assumed it was an effort to create a fascist-style corporate state, projecting the obvious point that fascism and communism have a lot in common onto the Chileans. Hippie radicals saw it as an unacceptable further extension of technocracy, instrumental rationality, and managerialism; perhaps they were projecting their own future careers. Beer himself had to suck up a lot of dog’s abuse over this, and Medina interviews one of the people who flamed him. Apparently he never realised things could go wrong in Chile.

Medina’s criticisms of Cybersyn are interesting. Her central criticism is that it baked a very specific definition of the worker into the software. This person was: a man, probably in a factory, and certainly not a peasant or even a miner. As most of what manufacturing industry there was in Chile was closely associated with mining, this set important limits for what it could achieve. This was also an important fact about Allende’s entire strategy, and the sociological assumptions that defined the politics eventually defined the code.

A lot of discussion of Cybersyn revolves around its decentralised nature. This element excited the Chileans, but Medina argues that Stafford Beer was very keen to ensure the project implemented his Viable Systems Model. Whether you consider the VSM to be hierarchical or not is a value judgment, but Beer always analogised it to the human central nervous system, spoke of the higher and lower functions, and crucially, he always drew it around a vertical axis.


Visualisation is important. So is naming things. Although Beer subscribed to the concept of embodied cognition, his most successful book is titled Brain of the Firm. Beer could have drawn the modules of the VSM from left to right; he could have called the book something else, but he didn’t.

One of the biggest problems with Cybersyn was technology adoption. For the workplaces that were meant to take part to get any benefit, they had to bother themselves to input data via the telex network, and act on the alerts that came back. As with any control system based on feedback, delay is a critical issue. Very often, because the enterprises on the system were behind with data entry, they only received predictive alerts after the events involved had happened, which was useless. Therefore they didn’t bother keeping up.

Crisis, though, caused adoption. The emergency deployment of Cybersyn to cope with the CIA-backed transport strike of 1972 actually took the system relatively close to Beer’s original design, and in a notably centralised variant. It worked well enough that take-up increased dramatically and the spice kept flowing. That said, when Allende asked for the Ops Room to be moved into the Moneda Palace, it’s possible that what he wanted wasn’t Cybersyn itself under his own hand, but rather the underlying telex network.

Most ironically of all, Cybersyn came into its own exactly for the purpose the New Left feared it would have – to use management information systems as a strike-breaking tool, or perhaps a paramilitary command and control network.

I’d strongly recommend the book, even if one thing I think it’s missing is examples of, say, a Cybersyn predictive alert message,

One comment

  1. John Styles

    My first thought is that if you haven’t read Operational Research in War and Peace by Maurice W Kirby you really should. I think it speaks to many of your interests and is an excellent book. I think I have mentioned it in passing before on your blog.

    Medina’s book is excellent. I sort of see it on some level as a debunking of the general narrative about Cybersyn etc. – or telling us what we really knew. Clearly it benefited greatly from (a) Medina going to Chile to talk to people in Spanish and (b) from her putting the male culture into perspective – is the quote ‘keeping the girl out of the control room?’ [My copy is with a friend and is boxed up as he flees the country post Brexit]
    How ‘cybernetic’ really was the use of the Telex machines vs. the lorry drivers’ strike?

    My second job was vaguely in this sort of milieu (fun fact – for pay grade reasons although I was more or less a software developer I was a ‘senior operational research analyst’ or as it appeared on print out SNR OR ANAL). The bloke in the office next to my desk in the open plan area (having been moved, I think, very very far sideways) had worked at Cybor House and worked for Tocher who took over from Beer and focused things on the less wacky word of simulation. Given this milieu one has to be skeptical about how much Beer brought this culture with him – the democracy of the upper class males. If one reads some of the documents about consultancy that Beer did that are in the archive at Liverpool John Moore University I do get the picture that rather more than one might hope / expect is what a friend calls wobbing (wise old bird) – upper class male speaks unto upper class male. The Warburtons stuff seems like this to me – talk of recursion as a way of doing the classic change management job of kicking the old man upstairs to chairmanship.

    It occurs to me in passing that Beer would get more interest in general, fitting in with the zeitgeist, if he had used the term ‘sustainable’ instead of ‘viable’ – I think sustainable is, as well as being a word likely to get him considered more widely, actually a better word for what he means.

    It is a shame that there isn’t a proper biography of Beer but I fear we are probably 10-15 years too late in terms of the age of the people one would want to talk to.

Post a comment

You may use the following HTML:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>