Ron Eglash (
Rensselaer Polytechnic Institute dept of STS
Talk presented at the Society for Social Studies of Science,
Vienna 2000

Cultural Cybernetics: the Mutual Construction of People and Machines


Cybernetics was coined by mathematician Norbert Wiener to describe the comparison of  information flows in artificial, natural, and social systems. These flows include communication, computation, and control. Social constructionists have approached this  powerful framework with a critical eye, finding it complicit with corporate and military interests and thus focusing on the ways in which society constructs cybernetics. Computational modelers have approached this powerful framework with an optimistic embrace, pointing to information networks as anti-authoritarian exchange media, and thus focusing on the ways in which cybernetics constructs society. Cultural cybernetics provides a methodology for the syncretic fusion of these two positions.


If we examine the changes which might characterize STS over the last few decades, we can see a decisive, well-articulated change from the “bad science” theories, which held that science and technology could lose their objectivity when influenced by society, to social constructivism, which held that both successful and unsuccessful science were the outcomes of a social process. A variety of frameworks have extended this analysis.  The least successful, relativism, merely holds that all truth is, in the end, a subjective judgement. But as Latour and Haraway have pointed out, relativist approaches attempt to silence nature. They would reduce the non-human actors in our theater of science to mere puppets or ventriloquist’s dummies. From Haraway’s point of view, this would be tragic, because the unruliness of these non-human actors makes them valuable colleagues in the fight for planetary survival. From Latour’s point of view it is self-deluding, because we spend as much time enforcing purification of the two domains, natural and social, as we do translating between them. And both authors point out that the most interesting and potent phenomena we encounter these days are typically cyborgs: hybrids between human and non-human worlds.

The move towards an STS of hybrids and cyborgs has created an academic cottage industry in boundary bashing. In this formula bad guys are in favor of boundaries, and good guys are against boundaries, so one need only produce a portrait in which boundaries or categories are permeable or mixed in order to win. But it is increasingly difficult to maintain that the mere observation of such boundary hybridity is enough to move us forward, especially given the increasing numbers of cyborgs and chimeric mixtures we see in popular culture (figure 1). If five-year olds can champion hybrids, then we need to stop leaning on them as our analytic ultimate.

Of course several authors have written critiques against the romanticizing of hybridity. Jen Croissant and David Hakken, for example, organized a 1995 AAA panel which emphasized how choices between cyborg and non-cyborg versions did not simply fall into neat ethical divisions; Hugh Gusterson (1995) described the movie “short-circuit” as the portrait of a bourgeois cyborg; and Myanna Lahsen (1997) noted that while pollution might well be described in cyborg terms, it is rarely advantageous. What these authors are pointing out is not that hybridity is a poor description, but rather the pitfalls of its utopic usage; similar to Emily Martin’s (1994) caution against “flexibility” as inherently more ethical or subversive. Neither Haraway nor Latour are guilty of utopic portraits of hybrids—indeed Haraway specifically mentions the cyborg as a figure which stands against the Garden of Eden’s utopic purity—but these caveats do help remind us that we must do more than merely highlight boundary-crossings, as if STS was a sort of mutant bird-watchers club.

Latour is no Audubon boy; his clarion call for a new constitution between the human and non-human worlds suggests intervention of history-making proportions. But his approach is also politically neutral; there is nothing in his call for a new constitution that would have prevented the infamous passages of the American constitution that legalized slavery and disenfranchised women. Latour’s strong call for a new arrangement of flows between human and non-human domains is silent on the topic of just who does the arranging.

Haraway’s suggestion for what one should do with hybridity, perhaps most succinctly put in her short essay “Situated Knowledges,” provides a stronger ethical framework. There she reminds us that rather than attempt to convert all agency into categories of the social, we should recruit non-human agency for our projects of planetary survival, just as scientists do for their projects of fact-making. But Haraway’s suggestion is taken up all too enthusiastically by Sandra Harding, who reduces the dynamic eloquence of Situated Knowledge to a static, “politically correct” formula of “Standpoint Epistemology.”  Harding’s use of this standpoint theory is so literal-minded that the puppets or ventriloquist dummies are no longer non-human agents, but rather people inhabiting specific gender race and class categories.

One reason Harding may have failed in her attempt to implement Haraway’s call for a situated knowledge is that her paradigmatic illustration depends on the sociology of Dorothy Smith. It is true that Smith’s feminist sociology is worthy of its acclaim, but to say that Smith’s caring relationship to the women she studies provides a model for how all sciences could operate is duplicitous. Of course it is possible for a social scientist to embed social concerns in her work; but that tells us nothing about how a physicist or mathematician might do the same in hers.

How might we regain the advantage of Haraway’s original insights, before they were derailed by Harding? I think it is no coincidence that Haraway chose the cyborg as her paradigmatic oppositional  figure. As much as Harding has failed in choosing Smith’s sociology  as archetype, Haraway succeeded in choosing cybernetics. In Harding’s analysis Smith was to play the role of subversive translator between human and non-human domains; the role failed here because sociology merely translated from human to human domains. Cybernetics, in contrast, is fundamentally a science of domain translation. The lingua franca of cybernetics is information; and the discipline of cybernetics was explicitly created to show how natural and artificial domains can be exchanged through this medium (figure 2).

If cybernetics indeed presaged Latour’s call for a new constitution of human and non-human hybrids, how did it fail to implement this agenda, and why did Haraway chose this unsuccessful discipline as the location for reviving the call? The same question has recently been raised in N. Katherine Hayles’ new book, How We Became Posthuman. According to Hayles, cybernetics amounted to a Pandora’s box; inviting postmodern philosophic monstrosities that were beyond the comprehension of the modernist minds that opened their box. Central to her argument is the concept of recursion. For Hayles, the feedback loops and self-referential iterations of cybernetic systems opened the door to a STS-style reflexivity that the cybernetics researchers could not tolerate. Cybernetics had to limit itself from its own radical conclusions.

I too have often argued in favor of a strong relation between recursion in cybernetics and reflexivity in STS (cf Eglash 1992, 1997). But I find Hayles’ portrait of the cyberneticians themselves, starting with her analysis of the Macy conferences, to be a poorly fabricated target. She begins by telling us that recursion was such a strange concept to them that they had no term to describe it. In fact, the Macy conferences were all about reflexive systems; the original title was “Circular causal feedback mechanisms.” Wiener coined the term cybernetics as a way to provide a more succinct description for a discipline which would focus on such phenomena. Nor were the social implications lost on Wiener, who was quite devoted to anti-authoritarian political critique. And Hayles can hardly accuse Wiener’s colleagues, Gregory Bateson and Margaret Mead, of reflexive ignorance; in fact Mead’s 1962 lecture to the newly formed Society was titled “The Cybernetics of Cybernetics.”

Contrary to Hayles claim that the early cybernetics movement lost its radical potential due to lack of reflexive insight, I think the failure of cybernetics to fulfill its promise of monsters is in part due to the internal political division between the opposing camps of Wiener and John von Neumann during the 1950s. While Wiener made efforts such as his famous letter to the president of the UAW, warning him of the rapidly approaching threat of automation, and calling for collaboration between working class interests and the new information sciences, von Neumann secured elite access to what would soon be called “the military-industrial establishment.” It was von Neumann’s centralized computing camp who won the power of mainstream support, while Wiener’s decentralists were marginalized as cranks or mystics.

By lumping together the oppostional Wiener with von Neumann’s elitists, Hayles erases a heroic history of ethical scientists who put social justice above their own careers. But she is right in putting the blame on humanism. The essential claim of Wiener and his followers – that political progress could be achieved from the resonance between decentralized computing and decentralized social authority – was fundamentally flawed. Flexibility is not inherently subversive; the chaos, complexity and fractal nonlinearity of Wiener’s neural nets are just as easily put to the service of postmodernity’s global capitalism as they are to any counter-hegemonic resistance.

What does remain valuable, however, is the methodology that Wiener and others developed to achieve translation between these different domains of the social and the natural. The value of such translation is clearly seen by global capitalism–for example by the genetics industry, where information media offer Latourian exchanges between DNA and HMOs—but that shouldn’t stop us from asking what manner of translations might better serve the purposes of planetary survival and social justice? How might the lingua franca of information be deployed by those of us who oppose the claims for universal, transparent communication, and champion the recognition that all representations are partial, provisional, and social? I refer to this vision as “cultural cybernetics”, summarized by the following three starting points:

1) Epistemological egalitarianism between Science and Society:
E.O. Wilson’s recent book, “Consiliance,” masks genetic reductionism under a façade of mutual understanding (Caporael 1999). Similarly, Hayles’ insistence that psychoanalytic truths trump cybernetics is at best an analog to the U.S. Supreme court’s Dred Scott ruling, maintaining racial segregation under the motto, “separate but equal.” Under this analogy, Cultural cybernetics should offer the STS equivalent of racial pollution; encouraging the crime of miscegenation between technoscience and society.

2) Multiple Objectivity:
We typically assume that subjective views, while unable to be falsified outside their local, are multiple, and that objective views, while having the advantage of falsifiability, are singular. Thus objectivity threatens us by its authoritarian rule of one over many, while subjectivity seduces us to sacrifice testability for multiplicity. We need knowledge that is both falsifiable and multiple. Cultural cybernetics should allow translations between locals, keeping Haraway’s Situated Knowledge from becoming mere relativism, while still empowering contestations for truth.

3) Participant Simulation:
Anthropologists have long championed participant observation; cultural cybernetics should provide its postmodern equivalent.
At the core of both cybernetics in particular and science in general lies the goal of modeling. Simulation – whether in the virtual worlds of the internet or the hyperreality of everyday postmodern life—expresses this goal in the media of information technology. Participant simulation should be both loyal to its modernist roots in the democratic ethics of participation, and take advantage of its postmodernist critique of authenticity and realism.

Some of you are probably familiar with my work on participant simulation from the African Fractals book; let me conclude with an example of participant simulation from more recent research on Native American culture. The Shoshoni-Bannock tribe originally inhabited the Cache valley in northern Utah. Pushed out by environmental degradation due to cattle ranching, they are now located on a reservation in nearby Ft. Hall, Idaho, and have purchased some of the surrounding ranch land for return to its natural state. Shoshoni biology teacher Ed Galindo has pioneered innovative techniques for restoration of the local ecosystem, most importantly re-introduction of salmon. Working with various community members--Mr. Galindo and other local teachers, students, and tribal culture representatives--we are creating an agent-based simulation that can model natural and social features of the Shoshoni environment in past, present and future. The simulation promises innovative approaches to historical anthropology in its abilities to run "counter-factuals" -- to see what history might have been like if certain events or processes had been different. It offers new approaches to health education; examining the prevalence of disease with variations in different nutritional practices and other behaviors. And it provides the opportunity to translate the indigneous knowledge systems -- which included botany, zoology, astronomy, and number and geometric patterns -- into the framework of contemporary science education.

In conclusion, cultural cybernetics offers a new outlook for STS, one in which science and society can engage in a more egalitarian interbreeding, where multiple objectivity provides for falsifiable local knowledge, and where participant simulation can support more beneficial flows of information across mental and material, natural and social, and mundane and spiritual domains.