Brian Holmes on Sat, 7 Mar 2009 23:00:20 -0500 (EST)

[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

Re: <nettime> Cybernetics and the Internet

Kevin Hamilton has written a great post in this thread, on 
the differences between first and second-order cybernetics 
and their socio-political consequences. Just to clarify 
before responding, it's useful to look at a chart of the 
differences between the two orders. For each theorist you 
are offered by a phrase describing the aspect of first-order 
cybernetics with which they were chiefly concerned, and then 
another phrase describing the second-order approach which 
they developed:

Author       First Order Cybernetics   Second Order Cybernetics

Von Foerster  --the cybernetics of     --the cybernetics of
                   observed systems         observing systems

Pask          --the purpose of          --the purpose of
                    a model                 a modeler

Varela        --controlled systems      --autonomous systems

Umpleby       --interaction among       --interaction between
                 variables in a  system        observer and 

Umpleby       --theories of social       --theories of the 
                       systems                      between 
ideas and society



Now I will start to reply to Kevin's post:

> I can imagine different routes to the same end of imagining there to be a
> supra-human or extra-human hand behind the mysteries of the universe. For
> some, the attribution of agency and divine order to the conditions of
> life conveniently supported a grasp for power. For others, it was a
> desperate effort at making sense of unlivable situations. For still
> others, it was a humble hope for a better world than they had inherited.
> The same could be said of the cybernetic view today.

Fabulous, that's exactly what these theories about machines 
become, a myth of the machine. To understand them you have 
to at once understand the power of the myth and disentangle 
it from the operating powers of the machine. Which seems to 
be the drift of this nettime thread...

> Before, rooms of "human calculators" (who in America were mostly those
> classed as women, disabled, or Black) worked on finite tables containing
> every conceivable combination of variables to produce, months later, a
> handy reference chart for the army gunman.  Wiener looked to these
> complex problems not as a collection of individual labor efforts that add
> up to a total solution, but as a system of possibilities that had a life
> of its own, a "self-organized" pattern that could be described and
> computed without actually doing the hard work of every algorithmic
> result.
> By describing his original problem in cybernetic terms, Wiener cut out
> some labor, and thus cost. He cut out some time, and pushed the problem
> of aiming artillery closer into the battlefield, out of the homefront.
> The gap would begin to close between the detonator and the aiming
> mechanism, until they eventually fused as one device of looking and
> destroying.

Yes, and just to pursue what you are saying, the curious thing about this
fusion is that the destructive effect, the predatory orientation, can also
be turned on labor. Consider the last 40 years of American economic
history. From the constitution of the world-scale American enterprise in
WWII, there evolved a primarily Anglo-American corporate class which by the
1970s found it in its class interests to use both robotization and the
worldwide transportation-communications network as a way of getting rid of
high-cost labor. A productive machine or an organizational diagram can be
engineered to exclude the influence of a certain category of labor, to
render their demands simply irrelevant to the functioning of the machine.
Here I do not think I am telling a myth about the powers of machines, I am
talking about corporate strategy as effected through technology. This is
one of the things that Wiener feared, he wrote about it in The Human Use of
Human Beings, though without foreseeing the role that telematic
communications would play in allowing the delocalization of industry.
Manufacturing from the 70s onward would be coordinated worldwide, and its
role as the leading sector of the economy would be replaced in the US with
reflexive, second-order operations in management, finance, entertainment
and scientific innovation. The same thing was done in the UK, and to a
lesser extent in other advanced countries (but not so much in Germany, for
example, which retained its industry). Unfortunately for those of us
interested in the positive potentials of networked communications, these
anti-labor forces were the major boosters of the computer-network myth in
the 1990s...

Once the control capacities of first-order cybernetic society had succeeded
in setting up a manufacturing system that was worldwide in scope, then the
complexity-era offshoots of second-order theories kicked in for financial
management, where the whole game is one of discrete systems observing the
behavior of other systems. According to second-order cybernetician Stuart
Umpleby, the concept of market reflexivity promoted by hedge-fund manager
George Soros is perfectly coherent with second-order formulations.  The
more I look into it, the stronger is the correlation between first and
second-order cybernetics and the two big periods of postwar economic
history, 1945-73 and 1973-2008.  For the two periods of economic history,
see Michael Hudson, Superimperialism, or David Harvey's version of
regulation-school theory, in The Condition of Postmodernity.  Amusingly
enough, Von Foerster's key send-order text, "On Constructing a Reality,"
dates precisely from 1973. The Umpleby paper on Soros is available here:

> Wiener's cybernetic description of the artillery aiming problem is itself
> the product a living, breathing, self-organizing system. It's here that
> cybernetics gets all reflexive and modernist - the analysis and
> management of feedback loops is itself a feedback loop.

This was the way Bateson and Mead saw it. In a famous interview with
Stewart Brand, they represented the difference between themselves and the
"engineers" as being the inclusion of themselves in the diagram of the
overall system. In addition to their own names they included that of
Wiener: (see p. 6 of the PDF)

But I would say the diagram is somewhat misleading. When Wiener worked as
an engineer, he did not only work on what was happening in the engineered
object, but on the dynamics of the entire environment in which that object
was supposed to operate. This not only means identifying the very specific
environmental factors to which the machine should respond, but also
building the receptor functions to get the information about those factors,
as well as the effector assemblages to alter those exterior factors by
altering the interior state of the machine (causing it to fire a missile,
or explode or whatever). In doing all that, however, I really don't think
Wiener thought much about how his personal agency was going to change the
battle space!

Yet as soon as he stepped back from working directly with engineers, that's
what he did. He produced very far-reaching reflections about how the new
generation of machines was going to change society, particularly through
widespread unemployment. Still there is a difference in the styles of
thought: Mead and Bateson were concerned with the real-time patterns of
cultural change, as they unfold repetitively and dynamically among
manifolds of speaking subjects. The reflexivity is immediate, and rather
than being measured by standardized units, it is "assessed" - at various
levels of consciousness - by the people involved, according to their
knowledge and whatever they are able to make of it. The theorists also
attempt to assess the impacts of their own interventions. This interaction
matrix can no longer be formalized in toto as a problem of control
engineering, because there are too many variables and their interactions
are too complex; and Bateson, for his part, went on to try to figure out a
completely ecological view where individuals and groups would try to find
the secrets of harmonious "coevolution." Mead, as you indicate, stayed
closer to the US elite's way of seeing things, which I will try to describe
a little more further down. Perhaps to understand Wiener's specific insight
into all this, one would have to study the period when he made hearing
aids, and how he conceptualized what he was doing. 'Cause that is one of
the things he did in order to avoid working with the military version of

> But the "cybernetics of cybernetics" was equally applied to a less
> individualistic sphere - that of management and administration. Macy
> Conference alum Margaret Mead framed it this way in her address to the
> inaugural meeting of the American Society of Cybernetics in 1968....

The text by Mead is very interesting for a lot of reasons.  One is that you
can clearly see where the problem of "systems observing systems" actually
came from. It was the problem of two nuclear powers, both with very strong
control systems, but not always constituted on the same bases. Game theory
had been invented to parse out how the control systems would react to moves
by each other, but then there arose a theoretical problem: what if their
underlying rationality were different, what if they were making
calculations based on different assumptions, or on a differently
constituted social order? Maybe even a more efficient one? Fear arose in
the early 1960s that the widespread Russian interest in cybernetics might
give them a tremendous advantage in industrial planning. That's why the
Cybernetics Society was founded. See the history here (I realize you know
it, Kevin, but this is for everyone):

Thus the perception of a "cybernetics gap" between the US and the USSR is
what led to Mead's prescription for how to run the (Cybernetics) society:

"Why can't we look at this society systematically, as a system with certain
requirements, certain possibilities of growth, and certain constraints, in
a world which is making demands? If this society is to pay attention to the
way cybernetics is developing in other countries, especially in the Soviet
Union and in other countries from the Eastern European block, what are the
devices for adequate cross-national and cross-ideological communication? Do
we have the necessary techniques? When are we likely to need either death
or transformation?"

Some of the American policy establishment did learn to think systems-
theoretically; one of the most interesting forerunners of that whole
generation is Karl W. Deutsch (of Czech origin, despite the name).
Influenced by Wiener and by Talcott Parsons, he wrote an explicitly
cybernetic book, The Nerves of Government. But he's more famous for his
work in the field of international relations, on the notion of 
"security communities." The first book on this subject is Political
Community and the North Atlantic Area (1957). It's about groups of states
which have managed to eliminate violence as a way of relating to each
other. Deutsch went on in the 70s to edit a book called PROBLEMS OF WORLD
of what Guattari says about "Integrated World Capitalism" would undoubtedly
resonate with Deutsch's cybernetically informed notions of the different
kinds of technical and cultural integration required to form a "security
community." From what I do know about Deutsch I would say that all this
was done, exactly as Kevin indicates, out of desperation on the one hand,
and a humble desire for a better world, on the other.

> Now back to my original question - what is gained, and by whom, in
> framing a problem this way?  We need a second-order example to compare to
> Wiener's artillery computing, so let's move forward to the present-day,
> 35 years after Von Foerster's lab folded, after the US military lost
> interest in funding "blue-sky" research with no battle-ready results....

> Wulf, an outspoken critic of current patent and intellectual property
> law, anticipates a "coming age of mass customization," in which low-wage
> labor is replaced by a "knowledge-intensive kind of manufacturing." His
> criticisms of patent law, antitrust legislation, and even drug-testing
> protocols are not that they are wrongly-based.  Rather, they don't change
> fast enough. For Wulf, we don't need different laws, we need a different
> system within which those laws are developed, changed, managed. Citing
> Thomas L. Freidman, Wulf draws a boundary around a new domain of concerns
> that need to be better linked into one dynamic, changing system. This
> domain includes "intellectual property law, tax codes, patent procedures,
> export controls, immigration regulations."
> The picture conjured here by this much-lauded leader is one strikingly
> reminiscent of Mead's "cybernetics of cybernetics," or even of Wiener's
> artillery firing data... The target keeps moving, so we need to stop
> computing the distances and just assume that the target's life can be
> anticipated, merged with that of the gunner.

I have not been able to form a very clear idea about William Wulf, as I
have not found many of his texts online, but his recent editorial in
Science does say this: "To prosper, we need an international process that
can, time after time, fundamentally rethink the elements of our innovation
ecology." Which is indeed a second-order, systems-theoretical kind of
statement. The aim is to remodel a whole discursive universe: the
"innovation ecology." And this is exactly the kind of governance approach
"that I am trying to talk about. But I would still make a
"first/second-order distinction. A Wiener-type machine is always aimed at
"changing some singular factor in its environment: that is its purpose, the
"end-goal, the 
"teleology." The second-order systems approach tries to 
change the parameters of the entire envoronment and not just one factor
within it. At the extreme limit, that means changing the parameters of the
world economy, or even of world society. It's delirious, but it can be
done. It was done by Paul Volcker, back in the late 70s, who changed the
monetary order by raising interest rates, forcing a halt to inflation in
the US and weeding out small businesses, while at the same time making US
T-Bills into the basic tool for attracting worldwide savings and propping
up a non-productive economy. It was then done again by Clinton and Bush 2,
whose governments transformed the rules of finance capitalism and made it
possible, for a short time, to turn the American stock and then housing
markets into the strange attractors of worldwide savings. Peter Gowan's
recent article in the New Left Review, as well as his book Global Gamble,
shows how this was carried out: interesting reading. One of the important
things about this kind of strategy, insufficiently discussed by Gowan, is
that all the actors know very well that they are just another part of a
system which by definition goes far beyond them: therefore they need to
play on the structure of beliefs that helps other people determine what
they do, in order to be sure that they will spontaneously do some of the
things that you want. This is why American power (and to a lesser extent,
English power in the previous period) depends so much on very reasonable
notions of democratic legitimacy and the extension of fundamental freedoms.
These are major motivators for systemic changes that will ultimately only
benefit a very few.

My belief is that the so-called "Wall-Street-Treasury complex" is only
imaginable after the second-order, systems theoretical proposals had made
their impact on a broad range of elite actors. And yet I do agree that the
second-order proposals in many ways flow directly from the first.  Probably
an analysis of the way game theory evolved would show the exact
progression. I bought the recent history of the Rand Corporation (Soldiers
of Reason) which, I imagine, would help to make that analysis. But I never
got through it....

> Second-order cybernetics seems to be alive and well, if even its
> current adherents are ignorant to its history. The questions that
> remain for me around cybernetics are around these issues of rhetoric,
> where cybernetic language and representation assumes a particular end
> for the world, a particular basis for action.

The point of my text on Guattari was to argue that the "particular basis of
action" that has been assumed in order to bring the world to its current
sorry state also has its effects on all of us who participate in this
world. Those effects are often predatory, whether we are treated as targets
from a first-order perspective, or whether our self-organization is
accepted in the second-order mode, and even mythified, in order to entice
us into certain environments and modes of thought. At the same time, the
rules of our interactions and the content of our beliefs are subtly
modified so that our apparent freedom produces the desired effects.

My question is double. On the one hand, how to avoid or derail the
multiplicity of new and even more effective targeting systems that have
been installed in the past twenty years? And on the other, how to react to
changes in the coordinates of an entire system, which builds most of the
environments in which we live? To change the world, we need to experiment
in consciously changing the microsystems in which we participate. Only a
capacity to experiment with systemic change can produce it. We in so-called
"civil society" need (and have needed for a long time) to change it from
below. That's why I wrote the text Swarmachine.  Self-organization in
society does not happen spontaneously and involuntarily, the way water
changes into gas. That is the myth that was propagated for the last twenty
years, serving to distract our attention from other realities. Who decides
to change the basic parameters (the laws, the technological toolkits, the
beliefs) according to which social reflexivity operates? Intellectuals need
to join with artists and social movements, to try to create the logics,
myths and concrete territories in which counter-systemic efforts can cohere
and gain consistency. Because achieving a new "ecology" of social relations
and relations with the natural world is always the major question, which as
the Wulf example shows is being asked again right now. I assume that Kevin
meant this example to stand in for the far larger set of proposals which
will utimately determine what the US and Britain try to do in the next year
or two, now that their former world-orienting strategy has failed. I share
Kevin's apprehensions about the answer. As Guattari pointed out, the matter
of who speaks, and of what kind of language they use, is the key problem in
present-day society.

best, Brian

#  distributed via <nettime>: no commercial use without permission
#  <nettime>  is a moderated mailing list for net criticism,
#  collaborative text filtering and cultural politics of the nets
#  more info:
#  archive: contact: