The Cybernetic Brain
Page 35
The Politics of Interacting Systems
LAST MONTH [SEPTEMBER 2001], THE TRAGIC EVENTS IN NEW YORK, CYBERNETICALLY INTERPRETED, LOOK QUITE DIFFERENT FROM THE INTERPRETATION SUPPLIED BY WORLD LEADERS—AND THEREFORE THE STRATEGIES NOW PURSUED ARE QUITE MISTAKEN IN CYBERNETIC EYES. . . . ATTEMPTS TO GUARD AGAINST AN INFINITE NUMBER OF INEXPLICIT THREATS DO NOT HAVE REQUISITE VARIETY.
STAFFORD BEER,"WHAT IS CYBERNETICS?" (2004 [2001], 861–62)
So far we have focused on the internal politics of the VSM—on social arrangements within a viable organization. Here, the organization's environment was conceptualized in rather amorphous terms, simply as that to which the organization needed to adapt. As we saw in the previous chapter, in the 1950s Ross Ashby was led to think more specifically about environments that themselves contained adaptive systems and thus about interacting populations of adaptive systems, including the possibility of war between them. Beer's experiences in Chile and of the subversion of the Allende regime by outside states, especially the United States, led him to reflect along similar lines from the 1970s onward. These reflections on the interrelations of distinct systems, usually conceived as nation-states, themselves warrant a short review.
Beer's basic understanding of international relations followed directly from his cybernetic ontology along lines already indicated. Nation-states are obvious examples of exceedingly complex systems, always in flux and never fully knowable. Their interaction should thus take the usual form of reciprocal vetoing or mutual accommodation, exploring, respecting, and taking account of the revealed variety of the other. Beer found little evidence for such symmetric interaction in the contemporary world, and thus, much of his analysis focused on what happens when it is absent. At the other pole from homeostatlike explorations lies the attempt to dominate and control the other, and Beer's argument was that this must fail. According to Ashby's law, only variety (on one side) can control variety (on the other). Any attempt simply to pin down and fix the other—to make it conform to some given political design— is therefore doomed to make things worse. The imposition of fixed structures simply squeezes variety into other channels and manifestations which, more or less by definition, themselves subvert any imposed order.
Beer's general analysis of macropolitics was thus, throughout his career, a pessimistic one: conventional politics is bereft of cybernetic insight and thus continually exacerbates crises at all levels. This rhetoric of crisis is a resounding refrain from his earliest writings to his last. In Beer's first book, the crisis is one of the West in general (the only instance of Cold War rhetoric that I have found in his writing) and of British industry in particular (Beer 1959, ix): "The signs are frankly bad. . . . The index of industrial production has not moved up for four years. We desperately need some radical new advance, something qualitatively different from all our other efforts, something which exploits the maturity and experience of our culture. A candidate is the science of control. Cybernetic research could be driven ahead for little enough expenditure compared with rocketry, for example. And if we do not do it, someone else will." In his later and more political writings, the crisis was often said to be one of the environment and of the conditions of life in the third world, as well as the more usual sense of political crisis: a socialist government in Chile as a crisis for the Americans and British being a prime example.42
When I first encountered this language of crisis in Beer's writing, I tended to ignore it. It seemed self-serving and dated. On the one hand, the rhetorical function of "crisis" was so obviously to motivate a need for cybernetics. On the other, we all used to talk like that in the 1960s, but, in fact, the world has not come to an end since then. As it happens, though, while I have been writing about Beer, his stories have started to seem very relevant and, indeed, prescient. Everything that has happened since those planes flew into the World Trade Center and the Pentagon speaks of an American attempt (abetted by the British) at command and control on a global scale, seeking to freeze the world, to stop its displaying any variety at all—running from endless "security" checks and imprisonment without trial to the invasions of Afghanistan and Iraq. And the aftermath of the American invasion of Iraq—what we have been taught to call "the insurgency," the killing, destruction, mayhem, and torture in the name of "democracy"—speaks vividly of the negative consequences of seeking to repress variety.
Figure 6.16.The cybernetics of crisis. Source: S. Beer, Brain of the Firm, 2nd ed. (New York: Wiley, 1981), 354, fig. 48.S.
Little more can be said here—this book is not a treatise on recent world history—but I do want to note that Beer's "cybernetics of crisis" included an analysis of how crises like the present one can arise. Again, Beer's focus was on transformative flows of information. Figure 6.16 is his basic diagram for considering such processes: the hatched area denotes a crisis affecting three different interest groups, which might be nation-states, A, B, and C. The details are less important than Beer's general analysis of the information flow from the crisis region into A ("sensory input") and the return action of Aon the crisis ("motor output"). What Beer emphasized was that such information flows necessarily impoverish variety, and that in a systematic way. His argument was that representations of crises are inevitably filtered through lowvariety conceptual models, models through which governments interpret crises to themselves and the media interpret them to the public. These models then feed into a low variety of potential actions which return to intensify the variety of the crisis along axes that are unrepresentable in the models, and so on around the loop.
Let me close this section with three comments. First, we can note that this last discussion of the role of models in the production of crises is of a piece with Beer's general suspicion of articulated knowledge and representation. Models might be useful in performance, as in the VSM, but they can also interpose themselves between us and the world of performances, blocking relevant variety (hence the significance of the inarticulacy of Beer's algedonic meters, for example). Second, Beer died before the invasion of Iraq; the above thoughts on that are mine, not his. But, again, I am struck now not by any self-serving quality of his rhetoric, but by the prescience of his analysis. The highly simplifed story of information flows and variety reduction that I just rehearsed illuminates how global politics could have collapsed so quickly into one-bit discriminations (Beer 1993a, 33) between "us" and "them," the goodies and the baddies; how it could have been that a majority of the American population could believe there was some connection between Al Qaeda and Iraq prior to the invasion and in the existence of what we were taught to call "weapons of mass destruction"; how it is that the American public and, perhaps, their government could have expected the invaders to be greeted with flowers and kisses rather than car bombs; and (turning back to the question of controlling variety) why mayhem should have been expected instead. Of course, third, one does not have to be Stafford Beer or a cybernetician to be critical of the war on terror, a "war" in which, "allies are expected to go into battle against an abstract noun, and to assault any nation unwilling to mobilize in such folly" (S. Beer 2001, 862–63). What interests me, though, is the generality of Beer's cybernetic analysis. We all know how to generate simplistic stories of heroes and villains, and much of the political talk of the early twenty-first century takes that form. Take your pick of the goodies and baddies—Saddam Hussein and Osama bin Laden or George W. Bush and the neocons. Such reversible stories will no doubt always be with us. Beer's analysis, instead, did not focus on the particulars of any one crisis. He actually began the most extended exposition of his analysis by mentioning the British abdication crisis of 1936, arguments over Indian independence from Britain in 1946, and the Suez crisis of 1956 (Beer 1981, 352–53). His analysis did not hinge on the question of whether George W. Bush was evil or stupid; his argument was that something was and is wrong at the higher level of large-scale systems and their modes of interaction that persistently produces and intensifies rather than resolves global crises. I take the novelty
of this style of analysis to be another example of the ways in which ontology makes a difference.
Team Syntegrity
HOW SHALL WE EVER CONCEIVE
HOWEVER EXPRESS
A NEW IDEA
IF WE ARE BOUND BY THE CATEGORIZATION
THAT DELIVERED OUR PROBLEM TO US
IN THE FIRST PLACE?
STAFFORD BEER,BEYOND DISPUTE (1994B, 8)
From the time of Project Cybersyn onward, the VSM was the centerpiece of Beer's management consultancy. In parallel to the VSM, however, he also developed a rather different approach to organizations that he called "team syntegrity." This grew from the 1950s onward, "flared into considerable activity 20 years ago, and occupied me throughout 1990 in a series of five major experiments" (Beer 1994b, 4). In the 1990s also, the conduct of "syntegrations" became partly a commercial business for Beer and his friends, associates, and followers.43 Beer only published one book on syntegrity, Beyond Dispute (1994b), as distinct from three on the VSM, but he and his collaborators developed and reflected upon syntegration in considerable detail.44 I am not going to attempt to do justice to that here. My aim is to sketch out the basic form of the approach, to connect it to the cybernetic ontology, and, continuing the above discussion, to examine it as a form of micro-sub-politics.45
Put very crudely, the substance of team syntegrity was (and is) an evolving format or protocol for holding a meeting, a rather elaborate meeting called a "syntegration," and we can explore this format in stages. First, there are the connected questions of what the meeting is about and who should come to it. On the latter, Beer offered no prescriptions. The idea was that syntegration was a process focused on some topic of interest to its participants. His model for thinking about this was a group of friends who met regularly in a bar and found themselves returning to some topic, perhaps current politics, but an early example in the development of the technique involved members of the British Operational Research Society seeking to redesign the society's constitution in 1970, and the first experiment in 1990 involved a group of friends and friends of friends thinking about world governance (Beer 1994b, 9, 35). The participants were, then, characterized by their common concern and interest in whatever the syntegration was about. Beer called such a group an "infoset," and, for reasons that will become clear, the basic form of an infoset would comprise thirty people.46
But just how should the topic of such a meeting be defined? This was a matter of pressing concern for Beer, a concern that ran along much the same lines as his critique of opinion polls mentioned earlier. The usual way of structuring such a meeting would be to distribute in advance an agenda listing specific topics for discussion and action. Beer's point was that such an agenda prefigures its outcome within lines that can already be foreseen, and "anything truly novel has two minutes as Any Other Business" (Beer 1994b, 9). His idea, therefore, was that the first element of a syntegration should itself be the construction by the infoset in real time of a set of relatively specific topics for discussion. In the mature form of syntegration this entailed a fairly complicated protocol extending over some hours, but, in essence, the procedure was this: Knowing the general topic of the meeting—world governance, say—each participant was asked to write down at least one brief statement of importance (SI) relevant to the topic, aiming to encourage original discussion of some aspect of the overall focus of concern. These statements would then be publically displayed to all of the participants, who would wander around, discussing whichever SIs interested them with others, elaborating them, criticizing them, or whatever (all this, and what follows, with the aid of experienced "facilitators"). Finally, after a prescribed length of time, the participants would vote for the developed SIs they considered of most importance, and the top twelve SIs would be chosen as the focus for the remainder of the meeting (27). In this way, something like a specific agenda would be constructed, not as given in advance but as emergent itself in the process of the meeting.
Given a set of thirty participants and twelve SIs, what happens next? In a short but complicated process, participants are each assigned to a pair of SIs, respecting, as much as possible, their preferences. Then the process of syntegration proper begins, and things get complicated to explain. How do you organize the discussion of twelve topics by thirty people? A completely unstructured agora-like situation is imaginable, but experience dictates that it would get nowhere. One might try to structure the meeting by, say, ranking individuals or topics in terms of priority, but this would return to Beer's critique of agendas, one step down the line. Inspired by Buckminster Fuller's geodesic domes (Beer 1994b, 12–14), the solution that Beer arrived at was to structure discussions in the form of a geometric figure, the icosahedron (fig. 6.17).47
Figure 6.17.The syntegration icosahedron. Source: S. Beer, Beyond Dispute: The Invention of Team Syntegrity (New York: Wiley, 1994), 338, fig. S6.2.
An icosahedron has thirty edges and twelve vertices, and hence the appearance of these numbers above. Each of the twelve topics is assigned to one of the twelve vertices of an imaginary icosahedron; each participant is imagined to be placed on an edge and engages in discussions of the two topics assigned to the vertices at the end of his or her edge. In turn, this implies that each participant is a member of two discussion groups of five people, each associated with the five edges that meet at any vertex (plus some additional niceties, including the participation of "critics" from disconnected edges, which I will not go into). These groups then meet repeatedly (three or more times) over a period of days for discussions that take off from SIs at their vertex, adding to, refining, and elaborating these statements in the course of their interactions. These discussions cannot all take place at once—one cannot be a member of two groups discussing two topics simultaneously—so participants alternate in time between their topics. And, according to Beer, the effect of this is that discussions reverberate around the icosahedron. On the first occasion, the discussion of any topic has a sui generis quality defined by the interaction of the five people concerned. But by the second iteration, their positions have each been inflected by different discussions at the other end of their edges, themselves inflected by discussions at one further remove. And by the third iteration these inflections are traveling around the geometrically closed figure, and there is the possibility that an earlier contribution returns "to hit its progenitors in the back of the neck" (Beer 1994b, 13). This is what Beer meant by reverberation: ideas travel around the icosahedron in all directions, being transformed and becoming progressively less the property of any individual and more that of the infoset as a whole. At the end of the process, each vertex has arrived at a final statement of importance (FSI), and these FSIs are the collective product of the syntegration (Beer 1994b, 32–33).
Thus, in outline, the form of the syntegration process, and to put a little flesh on what the products of such a process can look like, we can look briefly at a syntegration held in Toronto in late 1990 (Beer 1994b, chap. 6). "The group who came together were recruited mainly by word of mouth. . . . Thus the Infoset was assembled in an unusually arbitrary way: we may call it such a unity only because of its members all being drawn to the heading on the poster: 'What Kind of Future do You Want?' " (87). The first three of the SIs constructed at the start of the syntegration were: 'God is a verb not a noun,' 'Each child spontaneously desires to develop responsibilities commensurate with its abilities,' and 'Censorship is a personal issue.' " In Beer's précis, the first three of the FSIs, the products of the syntegration, were (97–98)
1. Local Empowerment: the need to push decision making downwards, especially in the case of abolishing nuclear war.
2. Law and Government: the move from ownership to stewardship, control to guardianship, competition to cooperation, winners and losers to winners alone.
3. How to Make World Peace: sovereign individuals acknowledge and accept the responsibility of a (human) world social contract, towards environmental protection, security, and evolution of the planet.
This and similar experiences in other syntegrations led Beer to remark that "amongst many others I have often claimed that in planning it is the process not the product that counts" (Beer 1994b, 97), and Beyond Dispute documents in various ways the fact the participants in syntegrations generally found them enjoyable and productive. The phrase "consciousness-raising" comes to mind, and we will see below that such phrases had a very literal meaning for Beer—his idea was that a genuine group consciousness could arise from the reverberations of syntegration. Let me close this section, however, with some general reflections on the syntegrity approach to decision making, with the critique of Beer's VSM in mind—"a topic of which" Beer declared himself in 1990 to be "heartily sick" (Beer 1990b, 124).