"Decentering Cooperation : Collaboration without hegemony?"
by John Duda
john*at*manifestor.org

Can cooperative production take place without the necessary accompaniment of
a shared collective understanding of the collaboration in question? In other
words, can we think of cooperation which lacks its own concept? What would a
cooperative process look like if carried out without the presence of a
community, institution, or ideology to organize and determine it?
Free Software as Anarchism?
This problematic is perhaps most obviously relevant for the ongoing and
increasingly creative attempts to reformulate political opposition to
neoliberalism and imperialism in ways which do not recreate structures of
alienation and domination within "the movement". Typically these efforts
have sought to deploy loosely affiliated networks of autonomous participants
as opposed to more hierarchical organization subsumed under a centrally
controlled platform. In seeking to support the claim that such decentralized
and radically democratic models of organization can provide, not just
structures for oppositional political practice, but replacements for the
current hierarchical systems of economic and political
organization/domination, activists have turned with interest to the
phenomenon of free and open source software development.
This theoretical and political interest in learning from free/open source
software has highlighted the latter's reliance on a decentralized
international network to support a process of production which does not
depend on wages or other forms of coercion, and which makes the product thus
produced available at no cost to anyone. [1] Yet it is surprising, given the
explicit rejection of hegemony as a political strategy on the part of those
seeking to appreciate or perhaps appropriate free software, that much less
attention has been paid to the anti-hegemonic conditions which inhere in the
free software "community" itself. [2] That is to say, the conditions which
make the phenomenon of free/open source software precisely the kind of
decentered cooperation postulated in this essay's first paragraph. And this
despite itself, or rather the despite the untiring hegemonic efforts of some
of its most active collaborators.Machines Against Hegemony
Pierre Clastres [3] famously made the case that primitive society was
structured, before the emergence of the state, in ways which sought to ward
off this very emergence. It is only through an analysis along these lines
that the more profound anti-hegemonic features of free software development
come to light. More precisely, the "free software movement" must be seen as
structured in such a way which makes the emergence of hegemony extremely
unlikely, if not impossible. This structure is not imposed on the movement
from some hegemonic position attained in the process of its
self-institution, but is rather a concrete consequence of the historical
constitution of the "movement". It has been unnecessary for participants in
the free software movement to attempt construction of a functioning
hegemonic claim in order to use this claim's legitimacy to attempt to banish
all hegemony from the movement, and where such attempts have been made, an
analysis along the lines I am proposing will reveal these attempts as both
superfluous and doomed to failure. Furthermore, such an analysis can shed
new light on the unexpected resistance which activists have encountered in
attempts to politicize the free software "community": it is not the case
merely that the free software community has decided to constitute itself as
fundamentally apolitical, but that the conditions governing the "community"
in question preclude its very formation in the first place!

The details
At least for the purposes of this provisional analysis, what one is tempted
to call the "free software movement", can be understood as a field shaped by
three distinct conceptual positions, which can only be reduced to a single
"movement" through dangerously sloppy abstraction or hopeful
idealization(one has to refuse to take seriously for the moment the various
incompatible hegemonic claims and articulations made by specific
participants in the process in question -- for example Richard Stallman's
GNU Manifesto [4] arguably includes all three rolled into one utopian
vision). It is the irreducibility of these three positions to each other
which wards off the possibility of a hegemonic mastery of this field. At the
same time, the three positions to be outlined below are to be understood
only as theoretical reconstructions of different aspects of a single
historical process, all mutually implicated by each other but without for
that being reducible to a higher synthesis. For convenience of exposition
alone, I will call them the "GNU", "efficiency", and "pirate" positions,
with the proviso that these are not to be taken as referring to any actual
actors.
To begin with, there is the "GNU position" which one can most easily
identify with the term "free software", centered around the refusal to
organize (one's own) labor in a way which equates the product of this labor
(software) with property to be exchanged for remuneration, and instead gives
this product to the world at large to reproduce, reuse, redistribute, and
rework. Within this position, there are of course minor distinctions to be
made. The Free Software Foundation's GNU Public License(GPL) is essentially
a legal device which ensures that the freely released product can never
again become closed or proprietary. Less "free" licenses like the BSD
license effect the same gift of the product of labor, but make no claims on
the recipients' use of the gift, who may choose to convert the gift back
into property. One can choose to view this giving away of the product of
software development labor through the concept of "gift economy" and see
implied assumptions of reciprocity("I'll write you a webserver if you write
me a C compiler I can build it with"), or one can choose instead to see,
along with Oekonux's Stefan Merten, a new form of non-exchange; however, the
fundamental refusal to submit one's own work to capitalism's
labor=product=property equation remains the same.
But of course, given widespread corporate adoption of open source software
and sponsorship of its development, we know that such software isn't as much
of an anticapitalist project as claims emanating from the perspective of the
first position make it out to be. A second position in the field(the
"efficiency position") needs to be identified: that the availability of
source code results in an increase in the efficiency of the process of
software production. Whereas the "GNU position" treats software and source
code primarily as product, the "efficiency position" understands source
chiefly as a means of production. Before the first position was ever clearly
articulated(the launch of the GNU project [5] in 1983 would probably be the
best reference point), computer software was commonly developed in the most
intuitively productive manner possible -- with programmers freely sharing
code as much as possible(certainly a methodology made possible in part by
the fact that large amounts of development took place in academic and other
research contexts). A good example is the samizdat distribution of John
Lion's Source Code and Commentary on UNIX Level 6 among programmers in the
late 1970's, in which a printed version of the Unix kernel was photocopied
and cherished as a pedagogical instrument, and not as a functional copy of
software(though formally identical). While Unix prehistory abounds with such
stories of workers/researchers disobeying corporate and institutional
authority to share code with each other for the sake of making their lives
as programmers easier and more interesting, the end result of this ease and
interest -- the production of more and better software -- is in no way
incompatible with capitalist extraction of surplus value from this
production. Despite the misguided and only partially successful backlash
against code-sharing which started in the 1970's(comprising both acquisition
of intellectual property and its vigorous legal defense), it is now the case
that many large corporations recognize the profit to be had by encouraging
increased production through opening up source code. The Open Source
Initiative [6] explicitly claims to facilitate this recognition, whereas the
Free Software Foundation, at least in reputation, can be seen as bearing the
standard of the first more anticapitalist position.
The FSF/OSI debate or struggle is fairly public and largely acknowledged by
both sides: the incompatibilities between the "GNU" and "efficiency"
positions are on the surface, so to speak. But the situation in the field of
free and/or open source software is further complicated by the presence of a
third position, which simultaneously makes the first two positions tenable
and makes any concrete acknowledgement of this debt on the part of the first
two positions impossible. This position stems from the fact that the product
of software development, information, is axiomatically defined as the total
absence of scarcity: in other words, if one could not, given a piece of
information like a computer program, make an exact duplicate of it, it
wouldn't be a piece of information. It is the mathematical impossibility of
scarcity(at least in terms of the infinite reproducibility of each
particular product in the limited domain of software development) which
makes the "GNU" and "efficiency" positions possible at all. The "GNU"
position allows one to put one's own ethical relationship to one's own work
into a legally binding form for others, but this third position, the
"pirate" position, gives one an ethics which encourages the illegal
appropriation of the work of others without their consent and on one's own
terms, in short, piracy without reservations. One can see the conflict here
quite clearly with the FSF, which in order to claim to enforce the GPL has
to maintain the legal sanctity of copyright. When Eben Moglen, FSF head
legal counsel, writes in his "Dot-Communist Manifesto" [7] that he is for
"Abolition of all forms of private property in ideas.", one has to note that
he uses the phrase "Abolition of" rather than "Disregard for": that the
struggle against "private property" for the FSF has to take place on the
level of a hegemonic articulation through the legal apparatus of the State,
and not through the immediate transformation of everyday life into something
beyond scarcity -- whether the State chooses to allow it or not. Similarly
submerged conflicts arise when considering the "efficiency" position's
relationship to the impossibility of scarcity -- as long as we understand
these positions to be concrete historical loci of discourse and production.
Certainly one can conceive of a utopian society in which freely shared code
results in wonderful levels of software productivity and in which
intellectual property has also been abolished. But considering software
production as it exists today, no government, corporation, or university(in
short no employer) can with any degree of legitimacy claim that you have the
right to steal as much information as you want(even if some kinds of
"stealing" let you program better and faster). At the same time that
production is increased because of an absence of scarcity, scarcity has to
be continually manufactured (elsewhere) to insure that someone profits from
this production. Only by abstracting away the real working conditions of
software production can the "efficiency" and "pirate" positions be made
compatible.
Putting it all together
Despite the fact that the three positions above cannot be reconciled in a
single functioning hegemonic articulation, it nevertheless did not
consequentially transpire that free/open software development never took
place. In fact, it might be argued that it is precisely because hegemony is
"warded off" by the interactions between the very basic structures of the
field that free software development has been so successful. I have referred
to "positions" within the field of free/open source software; perhaps the
term "attractors" might better illustrate the productive results of the
contradictions between the positions. Like a three body problem in physics,
the system is chaotic, with a variety of stable solutions and many unstable
ones. These solutions would correspond to the diverse and unpredictable
sites of potential collaboration which spring up in the spaces between the
three attractors. Because these collaborative spaces are inherently
unstable(i.e. not convertible through word or deed into a hegemony over the
system), all collaboration taking place within them happens on a sort of
neutral territory, where for example anti-capitalists can cooperate with the
NSA on the same Linux distribution.
To return to the political, I do not wish to suggest that hegemonic claims
should be entirely abandoned: even in decentralized networks, "minimal"
hegemonic articulations like the hallmarks of People's Global Action [8]
serve a useful purpose. And any political practice may, following Laclau and
Mouffe, involve some provisional claim to hegemony. Neither do I wish to
suggest that anti-hegemony be erected as the hegemonic claim, as is done in
capitalist articulations of libertarianism. But if one considers the example
of free/open source software, one has to admit that exploiting unresolved
contradictions might in some cases be more productive than insisting on
theoretical and political purity. [9] So not only should one seek to seek to
institute anti-hierarchical, non-hegemonic structures in which to live,
work, and create(on the basis of the ethical superiority of such
structures), but to recognize situations and configurations which by their
own logic prevent the emergence of a realizable hegemonic claim, and learn
to explore, intervene in, and exploit the productive resources such
situations can offer.
[1] see "The Free Software Movement -- Anarchism in Action"
http://info.interactivist.net/article.pl?sid=03/12/31/070202 for a
representative example
[2] The concept of hegemony which is operative here owes much to that
developed by Ernesto Laclau and Chantal Mouffe in Hegemony and Socialist
Strategy
[3] Society Against the State Also extremely relevant is Deleuze and
Guatarri's discussion of Clastres in the "Treatise on Nomadology" chapter of
A Thousand Plateaus
[4] http://www.gnu.org/gnu/manifesto.html
[5] http://www.gnu.org/gnu/initial-announcement.html
[6] http://www.opensource.org/
[7] http://info.interactivist.net/article.pl?sid=03/10/12/2237257
[8] http://www.nadir.org/nadir/initiativ/agp/en/pgainfos/about.htm#hallmarks
[9] One might be able to use the idea of a non-hegemonizable configuration
to construct a ethical-political rule for when "collaboration" with one's
enemies is desirable. If hegemony is conceivable, say in the possible
absorption of an activist antiglobalization struggles by fascist ideology
espousing nationalism and covert anti-semitism, than one has a reason for
refusing to cooperate with the right that does not apply in the case of the
same antiglobalization struggle advocating the use of Linux, even though it
is developed with support from multinational corporations and from
government security services.
http://creativecommons.org/licenses/by-nc-sa/1.0/
<http://creativecommons.org/licenses/by-nc-sa/1.0/>
This work is licensed under a Creative Commons License
<http://creativecommons.org/licenses/by-nc-sa/1.0/> .
_______________________________________________
http://freecooperation.org
conference, Buffalo, April 24-25, 2004
list archive: http://molodiez.org/ocs/mailinglist/archive.html