Free As In Beer: Cybernetic Science Fictions

For the past year or so, Jem Axelrod and I have been working on a research project focusing on the history of computing. Last month, we presented the first product of that collaboration at the 2009 Pacific Ancient and Modern Languages Conference in San Francisco.

The paper was titled Free As In Beer: Cybernetic Science Fictions. It told the story of Project Cybersyn, an early 70s socialist pseudo-internet built by British cyberneticist Stafford Beer in Chile. We explored how Beer’s writing, infographics, and industrial design worked together to create a science fictional narrative of omniscience and ominpotence for Salvador Allende’s socialist government.

Recently, Jem and I put together a video version of this paper and put together a site for the larger project, which we’re calling, Computer Science Fictions. The site is pretty basic for now, but we’ll add to it as the research project progresses.

In working on the paper I’ve had a number of interesting conversations with various outside advisors including Clay Shirky who suggested thinking about the parallels with the East German Stasi, Tom Igoe who helped me work through some of the actual circuit diagrams in the Project Cybersyn documentation, and Matt Webb from Berg London who’s in the midst of his own Cybernetics research project (much of Webb’s thinking on the subject is beautifully summarized in his recent Web Directions South Keynote: Escalante). I just wanted to take a moment to publicly thank these three as I haven’t had the chance to do so elsewhere.

We’ll be continuing this project along many different avenues going forward including exploring the relationship between AI research and the early video game industry. For now, without further ado, the video:

Free As In Beer: Cybernetic Science Fictions from Greg Borenstein on Vimeo.

This entry was posted in Opinion. Bookmark the permalink.

0 Responses to Free As In Beer: Cybernetic Science Fictions

  1. Nick Green says:

    Hi Greg,
    I should comment that Cybersyn was intended as a distributed management system driven by performance statistics from all sectors of Chile.
    Distributing the data produced a rational basis for the development of better methods shared by all. There was nothing sinister in that in fact quite the reverse. Beer’s principles should be applied more widely.
    The technology was the best available at the time and it’s remarkable so much was achieved in so little time.
    Nick Green FCybS
    Metaphorum.org
    Cybernetics Society
    cybsoc.org

  2. Greg,
    Your interpretation of The Chilean project and my understanding of it are so different that I hardly know where to begin.
    I knew Stafford Beer very well, have written a book that is an introduction to the main ideas used in the Chilean project, and also know several of the Chilean scientists and managers who were involved in the project.
    Beer himself would be horrified at the idea that the project was intended to provide centralized control. Perhaps the element of cybernetics that is most important for management is called the law of requisite variety, which (to paraphrase) states that the control achievable by any regulator is limited by the ratio of the complexity of that regulator compared to the complexity of the thing to be controlled. This law shows that the entire idea of centralized control is an impossible myth. No conceivable centralized group of managers can approach the complexity of the real-world of factories, etc. and therefore can never hope to reasonable manage that world in a centralized fashion.
    On top of this, Beer’s own political ideas were diametrically opposed to government centralized control.
    Beer’s management cybernetics theories / models were much concerned with developing and protecting local autonomy. The Chilean project was intended as an aid to decision making. The reason the control room had no connection to the outside world is that it was simply intended to allow senior management to understand what was actually going on in the nation in the hope that this understanding would allow better decision making. And there were all sorts of protections built in to support local autonomy at the shop level, at the factory level, and so on.
    I am not sure what you were trying to do with this work, but it is a serious misinterpretation of management cybernetics generally and the Chilean project in particular.
    Barry Clemson

  3. Hi Nick. Thanks for your thoughtful comment.
    While I agree that Beer frequently discussed the idea of using Cybersyn to distribute information for use in management decisions at all levels of the hierarchy (especially in his sketches fo Cyberfolk), I think the decisions about what parts of the project actually got built are very telling about the priorities that emerged between beer and the Allende government in collaboration.
    The Ops Room — and the fantasy of omniscience it embodied — fed any centralized government’s desire for perfect knowledge and hence received the lion’s share of the attention and resources and so logically became the most real. The more distributed parts of the project (from the local factory-level ops rooms to Cyberfolk itself) never really caught on much with the project’s backers and so remained mostly at the level of design concept.
    As Latour points out in his book about the French Aramis transit project, ideas such as this move gradually and inconsistently from idea to reality, frequently losing ground as often as gaining it. And no idea survives contact with real life in tact, each one is transformed and translated as it goes through this process.
    Finally, I think it’s important to distinguish between the notions of hierarchical management and distributed authority. Beer’s vision is profoundly hierarchal. While each factory and regional coordinator was supposed to use Cybersyn to coordinate with the layer below them in the hierarchy, the only thing they ever passed up to the higher levels was information, not opinion. This was also not a “peer-to-peer” system. Cybersyn did not help the factories communicate with each other, only report their statistics upward
    In our current era of an peer-to-peer internet that emerged from a different cultural and technological context it’s important to keep in mind that not all networks have this property.

  4. Greg,
    Your interpretation of The Chilean project and my understanding of it are so different that I hardly know where to begin.
    I knew Stafford Beer very well, have written a book that is an introduction to the main ideas used in the Chilean project, and also know several of the Chilean scientists and managers who were involved in the project.
    Beer himself would be horrified at the idea that the project was intended to provide centralized control. Perhaps the element of cybernetics that is most important for management is called the law of requisite variety, which (to paraphrase) states that the control achievable by any regulator is limited by the ratio of the complexity of that regulator compared to the complexity of the thing to be controlled. This law shows that the entire idea of centralized control is an impossible myth. No conceivable centralized group of managers can approach the complexity of the real-world of factories, etc. and therefore can never hope to reasonably manage that world in a centralized fashion.
    On top of this, Beer’s own political ideas and values were diametrically opposed to any sort of centralized control.
    Beer’s management cybernetics theories / models were much concerned with developing and protecting local autonomy. The Chilean project was intended as an aid to decision making. The reason the control room had no connection to the outside world is that it was simply intended to allow senior management to understand what was actually going on in the nation in the hope that this understanding would allow better decision making. And there were all sorts of protections built in to support local autonomy at the shop level, at the factory level, and so on.
    I am not sure what you were trying to do with this work, but it is a serious misinterpretation of management cybernetics generally and the Chilean project in particular.
    Barry Clemson

  5. Nick Green says:

    I have to say railing against hierarchy is like railing against the telephone directory or the notion of filter or choice. How else can you disambiguate responsibility? Also note in Beer’s scheme policies with clear criteria for change not people fill the hierarchy. As to the web that’s still client-server not peer-to-peer yet (as in, say, file sharing).
    As for information flows recall all the local/global system 4s (for Development and change) are connected to the outside world.
    We could talk further on Skype if you wish and I see Barry Clemson has made some excellent points for you to consider.
    Best
    N.

  6. Nick, I’m not railing against hierarchy at all. Many of your — and Barry’s — comments assume that I’m somehow opposed to Beer’s ideas on an ideological level. Broadly speaking, this isn’t the case at all. What our paper tries to do is track how Beer’s ideas shifted and evolved, frequently in ways he didn’t fully control or understand, as they left the realm of utopian vision and rubbed up against the real world with all its texture.
    I actually think evaluating any utopian project for success or failure is a fool’s errand’s. Unless the rapture or the heat death of the universe comes sooner than we’re all expecting, the real world isn’t going anywhere. All utopian projects fail. It’s in their nature.
    What’s fascinating for me about utopian projects is that they reveal the limits of our conventional thinking about the world. Maybe something daring had never been done because it had simply been “unthinkable” and when someone sets out to actually do it, it turns out to be totally possible. Other times, those conventional limits turn out to be there for good reasons and the result ends up extremely complex. Either way, we end up seeing our previous conventional world view in a new light.
    Beer’s project in Chile was never given a chance to become real enough to succeed or fail on its own merits. The tragedy of the Pinochet coup cut off that possibility. What we tried hard to point out in our talk was how the project shifted and changed in response to the needs of Beer’s client, the Allende government. The fantasy of omniscience into which the Ops Room evolved, emerged from the combination of Beer’s ideas and the needs (aesthetic, emotional, political) of Allende and his circle
    Barry — Musn’t Beer have been horrified about much of how the project turned out? By the end, many of his more populist ideas (Cyberfolk, much of the local infrastructure) had fallen by the wayside and the Ops Room had grown in importance. None of that is to mention the obvious and complete tragedy of the regime’s end. Cybersyn was not simple and clean cybernetic theory, but an actual political reality with compromises and flaws. While I agree with you about much of what you say about Beer’s intentions, I think the real world results were a lot more complicated.

  7. Roger Harnden says:

    Interesting, Greg.
    I too new Stafford well, and have become something of an expert on his ideas and also on the more linguistically-focused ‘second order cybernetics’.
    An interesting presentation with unusual and good examples.
    I think your response is well considered and apposite. The insight of myself and others who are ‘fans’ of Stafford, tends to be coloured by Stafford’s own ‘story’ of the events, rather than informed by the pragmatic (and frequently messy) realities.
    You do refer to this in passing in the presentation, but perhaps it would be more respectful if you felt able to slightly lessen the “cyborg” tone of the story. As for the model Beer went on to refine into what he called the Viable System Model (VSM), it encapsulates lessons he learnt from the painful Chilean experiment. As you probably know, thereafter he severed his links with big business and more rigorously attempted to communicate his insights into distributed planning, and the need for modern societies to feel able and willing to rely on more horizontal forms of communication and control, thereby lessening more authoritative vertical ones.
    But, leaving this to one side, you have put together a very interesting account

  8. Roger,
    Thanks for the comment. I can definitely appreciate how hard it is to have enough distance on people and events you were close to in order to be able to listen to other people’s comments on them with an open mind.
    I think that part of why our video has received such an emotional response from many people who knew Beer himself (or were fans of his) has to do with the unusual context in which it’s being distributed. As I mentioned in my post, this paper was originally given to a panel at the Pacific Ancient and Modern Languages Association conference. This is an intimate setting populated by academics The paper was, of course, delivered in person by us and a discussion ensued afterwards between the other speakers (three other papers were presented as part of the same panel) and the small audience in the room.
    Being excited about our work, Jem and I decided to produce this video in order to share it more widely online. Doing so was an experiment. We’d never seen an academic paper presented in this manner online and wanted to find out what would happen. Because of this novelty, I think it was also difficult for people viewing the presentation online to reproduce anything like the context in which the paper was created and intended. Many people view videos on the internet as impersonal “media” that makes some implicit claim towards authority rather than as simply a part of a larger discourse the is multi-way and ever contingent. I think our relatively professional (if I do say so myself) production values probably contributed to this formal officious tone. And getting it picked up by a site as widely read as BoingBoing probably helped as well.
    Hence, people with a stake in Beer or Cybersyn or cybernetics more generally tended to watch the video with a hyper-critical eye for being misrepresented. They felt that we were producing a public document that would represent their private community. And so they feel the need to defend themselves and their group, understandably. Where we, on the other hand, were focused on communicating our interpretive ideas to other people with a scholarly (or other) interest in Cybersyn and the relation between technology and culture more generally. In that context we were initially very surprised by the strength and emotional quality of much of the reaction.
    Anyway, thank you for your kind comment and for greeting our experiment with patience and interest.

  9. Dear Greg,
    As Stafford’s partner, my bias with respect to his character and values is a given, and I will leave that defense to others.
    I am a reader of science fiction and fantasy literature, though, and know that authority and reactions to is is a consistent theme although authority in those contexts has more in common with the Mafia and the Red Queen from Alice than with the complex demands of contemporary governance which in any case are not well served by autocratic control.
    The point about the Operations Room was that it was never supposed to be omniscient, never mind omnipotent. The information it received was filtered (each entity at each level reported around ten values) so that the decision makers were not flooded with data but could see more clearly where to allocate their limited resources of time, materials and money. If all the indices were operating within expected values, no action was called for and each entity had a set amount of time within which to bring an index back into the expected range before it became visible at the next level. The computers were used to apply statistical filters to provide decision makers with the best available information, not to take decisions for them. The graphics were designed to make the information easier for managers to absorb and understand.
    In fact this approach is the exact opposite of the massive data bases and data mining that is possible, although not necessarily useful, with today’s computer power. It was not a pseudo or proto internet. It was wringing the maximum out of the technical capacity at hand and doing it in as innovative and aesthetically pleasing a format as possible.
    It should also be pointed out that Stafford and the technical team weren’t the people who were to use the facilities. They were for the responsible ministers and managers at each level who are in charge in any country with a sizable portfolio of state-owned industry. That situation was by no means confined to Chile; it was common among the small countries in Latin America who weren’t large enough markets to attract private investment.
    The utopian ideal was that the United States would permit an elected Marxist government to survive. Cybersyn didn’t ‘doom’ Allende’s government – it may only have led to a larger amount of violence due to helping it survive the earlier attempts to bring it down. If you check out the hearings chaired by US Senator Frank Church on the Chilean situation you would find that Kissinger wanted a far off country to use for a destabilization experiment as the Bay of Pigs had gone wrong.
    If you are looking for a near contemporary (1977) published account from someone who was there, go to Hermann Schwember’s article ‘Cybernetics in government: experiments with new tools for management in Chile’ in “Concepts and Tools in Computer Based Policy Analysis” Vol 1. edited by H. Bossel.
    What is unfortunate about your presentation is that it adds to the misinformation circulated at the time to discredit the Allende government and everyone associated with it. Your statement that Stafford ‘appropriated’ notation from science and engineering suggests that using such to help a government manage its affairs is illegitimate. Such assertions add to the public’s difficulty in distinguishing evidence based deductions from partisan opinion.
    Allenna Leonard

  10. Greg,
    Writing as yet another good friend of Stafford Beer (he still has a lot of good friends around the planet) I must say that the man I knew for 20 years bears no resemblance to the character depicted in your film.
    The tone of the piece does suggest a very sinister intention by both Beer and Allende. Soon after the military coup there was a good deal of propaganda from the new Pinochet regime attempting to justify the brutal overthrow of a democratically elected government. They claimed evidence had been found to demonstrate Allende was planning a communist style dictatorship (this was known as ‘Plan Z’). Over 36 years later not one scrap of this so-called evidence has ever come to light, but once again your film seems to resurrect this absurd fiction by implcation.
    Beer was always impressed with Allende’s deep concern that the government and the technology should be there to serve the will of the people. In fact just recently Allende was voted the greatest Chilean in history. (This was part of a national TV vote based on the pioneering example here in the UK, when Churchill achieved the equivalent success.) This indicates that perhaps a younger generation can now see from an historical perspective that Allende really was attempting something daringly innovative in the face of conservative forces at home and abroad. Let us not forgot that there is plenty of information available to show that the Nixon administration, with Kissinger in charge of foreign policy, seriously sabotaged the Chilean economy. The CIA received 10 million dollars to assist with this, including the funding of strikes. In other words the Allende government was doomed from the moment it took power.
    It’s also worth noting that no dictatorship has ever attempted to use Stafford Beer’s ideas to increase its control over people. In essence Beer’s work is not about hierarchy (though it may give that superficial impression on the page); it concerns autonomy and self-organization recursively structured. ‘Control’ in cybernetics is about intrinsic control where a system maintains its own regulation and dynamic adaptability (with a capacity for learning) through feedback loops.
    My worry about your film is that it sends out completely the wrong signals as an online resource for many people who may not know anything about Project Cybersyn, disseminating misinformation across the internet.
    For anyone who is interested I have recently edited a wide-ranging anthology of Stafford Beer’s writings: ‘Think Before you Think: Social Complexity and Knowledge of Knowing’ (Wavestone Press 2009). Here you will find several papers concerning Chile. But the selection also shows much more of Beer’s overall transdisciplinary philosophy as well as his moving poems relating to the drama of Chile. It should be clear for anyone reading these writings that they come from a highly ethical, compassionate and courageous man.
    David Whittaker

Leave a Reply to Greg Borenstein Cancel reply

Your email address will not be published. Required fields are marked *