[p2p-research] Project Cybersyn

[Paul D. Fernhout]

Kevin Carson wrote:
>  Sent to you by Kevin Carson via Google Reader: Project Cybersyn via
> Marginal Revolution by Alex Tabarrok on 12/7/09
> 
> Cybersyn was a project of the socialist government of Salvador Allende
> (1970-1973) and British cybernetic visionary Stafford Beer; its goal
> was to control the Chilean economy in real-time using computers
> and "cybernetic principles." The military regime that overthrew Allende
> dropped the project and probably for this reason when the project is
> periodically rediscovered it is often written about in a romantic tone
> as a revolutionary "socialist internet," decades ahead of its time that
> was "destroyed" by the military because it was "too egalitarian" or
> because they didn't understand it.

A link for that text and related discussion:
http://www.marginalrevolution.com/marginalrevolution/2009/12/project-cybersyn.html

Three comments from there:
"""
The (mostly Californian) team working this up came to London in summer 1973, 
and wanted to talk to British Treasury economists about what they were 
planning to introduce. They talked to me. Their modelling was, in concept 
(they clearly were in the early stages of working out the detail), very good 
for the time (and their available computers would have handled the data flow 
OK). But when we got down to the content I found a materials balance set of 
ideas, with a still large private sector expected to handle things like 
retail distribution and truck transport (a vital sector in Chile). It was 
not real time - more like last week/last month for data to feed into next 
month's decisions. It was so much better than anything I had heard of in use 
in the Soviet block that I was fascinated.
   It was with real regret that I found myself suggesting that the results 
from the team's approach could be disappointing; and asking what 
improvements over the current ramshackle Chilean economy they hoped to see? 
Those hoped for enumerable improvements turned out to be mainly 
distributional. It seemed that productivity, efficiency and output gains 
would have to wait for later, improved versions of the model; versions in 
which they would somehow improve on the very crude financial incetives they 
proposed to leave in the private sector and generalise them into the state 
sector.
   As a means of incorporating incentives, I mentioned the Hungarian ideas 
of the state owning the means of production but always being willing to rent 
them out to the highest bidder. The team seemed to think that might be worth 
looking into; but it was evident that they did not realise that they were 
improving on mainstream Communist practice; let alone that they should be 
interested in heterodox Communist thought.
   For a mainstream, pretty eclectic macroeconomist, that afternoon was a 
trip through the looking glass. Nevertheless, there was more to Project 
Cybersyn ( a name the team were rather ashamed of) than a stage set.
   Posted by: David Heigham at Dec 7, 2009 12:06:57 PM
"""

and:

"""
It's funny to see the central role that contempt plays in MR's rhetoric. 
(Note that that's hardly a ringing defense of Cybersyn or Beer: it was silly 
and he was a wanker.) This site and its maintainers are wedded -- very 
wedded -- to beliefs that are trvial modifications of Beer's beliefs and 
Cybersyn's pillars. Why not compare the computational power behind the moon 
landings to an iPhone and 'prove' that they were impossible too? Yeah, what 
a crazy idea that WU Telex machines or other retrospectively 
primitive-seeming communications techniques might play a vital role in 
large-scale economies! &c., &c.
   Posted by: m3t00 at Dec 7, 2009 5:27:24 PM
"""

and:

"""
m3t00,
   Your last sentence was a silly bait-and-switch: using Telex machines to 
improve communication and coordination -- between humans -- was never 
controversial. So let's stick to the topic of computers. The financial 
crisis of the last few years has demonstrated just how sharply our best 
efforts at mathematical modeling of the economy can diverge from reality, 
even with today's most modern computer technology let alone 1973's much more 
primitive kind.
   By contrast, plain old Newtonian physics for orbital calculations is far 
more straightforward. Before computers, astronomers accurately calculated 
those by hand, albeit not at the near-real-time speed needed for lunar missions.
   Posted by: anonymous at Dec 7, 2009 7:55:24 PM
"""

What the last poster misses is that using Telex machines in the concept of a 
communist command-and-control economy was revolutionary, because it changed 
the nature of the entire thing from purely hierarchical to more of a 
cybernetic hierarchy-meshwork balance (Manuel de Landa 
http://www.t0.or.at/delanda/meshwork.htm ). So, there was a move from "five 
year plans" to monthly, weekly, or even daily plans.

One example that Alex Tabarrok (who wrote the article)
ignores is how this system was used successfully to deal with the 
consequences of strikes created by the right wing.

 From the New York Times in 2008, hardly a bastion of communist thought:
   "Before ’73 Coup, Chile Tried to Find the Right Software for Socialism "
   http://www.nytimes.com/2008/03/28/world/americas/28cybersyn.html
"""
Cybersyn’s turning point came in October 1972, when a strike by truckers and 
retailers nearly paralyzed the economy. The interconnected telex machines, 
exchanging 2,000 messages a day, were a potent instrument, enabling the 
government to identify and organize alternative transportation resources 
that kept the economy moving.
   The strike dragged on for nearly a month. While it weakened Mr. Allende’s 
Popular Unity party, the government survived, and Cybersyn was praised for 
playing a major role. “From that point on the communications center became 
part of whatever was happening,” Mr. Espejo said.
"""

See, that is what was revolutionary about Cybersyn compared to Soviet style 
central planning. It was a hybrid of central planning and responding to 
daily needs in the way the market does.

Would it have needed to change as it grew? Surely. But the idea itself was 
very threatening ideologically to capitalism and the "free market". Still, 
if the USA had not helped take it out, maybe the USSR might have done so 
later? :-) It's success would also have disproved long term central planning 
as an ideal, too. :-)

I'm sure if Google had been taken out by the US military in 1999, a similar 
article would have been written. I'll try, just for fun:
   http://www.google.com/corporate/history.html
"""
(Making up an alternative past history written from the future. :-)
As a footnote in history of government-approved search engines, "Google" was 
the name of a small company created more than twenty years ago, in 1998, as 
the starry-eyed impracticable vision of two graduate students at Stanford 
University. Using a computer built partially from LEGO bricks:
   "Google Founders built Server Casing with LEGO Bricks"
   http://www.i4u.com/article14320.html
and operating from a garage, these wild-eyed dreamers had a vision of 
stealing all information on the planet and supplying free access to it. 
While there was no real risk that such people could have succeeded, even 
with US$100,000 worth of personal backing from someone at an established 
computer company (a trivial amount of money in those days for a startup), 
because of the vast computation and electrical demands of such an enterprise 
(clearly impossible to meet within a garage if you just do a little math on 
even just the electrical demand from several servers), clearly these 
individuals involved did not understand free market processes and how all 
information would need to be sold in order to promote the creation of more 
information. On April 1, 1999, the garage that this small company in was 
raided by the US FBI in conjunction with representatives of RIAA, the MPAA, 
and the SPA, and the server was confiscated. This was also justified because 
the FBI was able to perform successful search queries on explosives 
manufacture, and such an information service could have been used for more 
terrorists like Timothy McVeigh to learn enough to bomb Federal Office 
Buildings. Google co-founders, Larry Page and Sergey Brin, are still 
awaiting trial in Guantanamo Bay (as of this writing in 2019) for 
un-American activities as enemy combatants as threats to the state order. 
Through enhanced interrogation techniques, it was discovered that Page and 
Brin also had plans to steal the world's books, and more fancifully, set up 
an independent society on Mars. While it is laughable to think that such 
people could have indexed all the world's proprietary information with a 
server that has less power than one of today's highly secure "trusted 
computing" cell phones, none-the-less the concern was that such people had 
no respect for private intellectual property and the free market that 
creates it, and so these two info-terrorists would no doubt have helped 
physical terrorists like Timothy McVeigh or others to perform a variety of 
anti-social acts. Likewise, such a system could potentially have made copies 
of proprietary news articles and slightly harmed our vibrant mainstream 
newspaper industry, or worse, created synthetic newspapers with biased views 
compared to the unbiased reporting that professional journalists and 
publishers are well celebrated for (like with the Pulitzer Prize). The US 
government stopped the project and probably for this reason when the project 
is periodically rediscovered it is often written about in a romantic tone as 
a revolutionary "socialist search engine," decades ahead of its time that 
was "destroyed" by the US government because it was "too egalitarian" or 
because they didn't understand it. Hopefully this article will show why that 
is foolishness.
   This footnote in history is brought to you by Mickysearch, the search 
engine that gives you the appropriate information you need, when you need 
it, and is offering a discount this week, sign up for one year of searching 
(maximum 10 queries per month, all with completely safe high-quality US 
government approved results), and the cost is only US$1000 a month instead 
of the usual US$1500 a month, a cost savings of more than US$50 per search.)
"""

Now, does that not sound a bit like the rhetoric in the original article? :-)

--Paul Fernhout
http://www.pdfernhout.net/
http://www.beyondajoblessrecovery.org/