The Cybernetic Brain

by Andrew Pickering

  56. D. Cooper (1968) reproduces some of the papers given at the conference.

  57. The "numero uno" quote is from Alan Marcuson (J. Green 1988, 209). The list of attendees is taken from Miles and Alan Marcuson (J. Green 1988, 208, 209). The Berke quotation is from Berke (1970). The argument with Carmichael seems to have been a key moment in the political history of the sixties, though people recall it differently. Sue Miles: "It was quite frightening. Stokely Carmichael started this tirade against whitey. Though one could see perfectly well why he had this point of view, it was completely unworkable. Then there was this meeting afterwards back at this house where he was staying and there was extreme bad feeling and a huge argument and split between them all. Allen [Ginsberg] was going, 'This is dreadful. We have not argued this long for everyone to start getting at each other's throats and getting divided. This is not going to get us anywhere.' " Alternatively, Alan Marcuson again: "There was this wonderful dinner at Laing's. Laing had all the superstars there for dinner and he was very into being honest and he said to Stokely Carmichael, 'The thing is, Stokely, I like black people but I could never stand their smell,' and Stokely didn't like that and left" (J. Green 1988, 209, 210).

  58. The quotations are from Trocchi's "Invisible Insurrection of a Million Minds" (1991a [1962], 178, 186) and "Sigma: A Tactical Blueprint (1991b [1962], 199). Trocchi was a key node linking several countercultural networks running from the British underground to the Beats and the French Situationist International (with situationism as central to the works just cited). For more on situationism, see Marcus (1989) and situationist international online:www.cddc .vt.edu/sionline/index.html. On Trocchi's part in the 1965 Albert Hall poetry reading, see J. Green (1988, 67–71).

  59. A great deal of the discourse of social science, at least in the United States, circles around issues of equality and inequality, power and hierarchy, but I find it hard to think of any such work that thematizes issues of enframing and revealing. The organizing problematic is usually that inequality is simply bad and that we owe it to this group or that not to oppress them. The idea is almost never that we (the oppressors) might be able to learn something positive from the others. Think, for example, of the social science discourse on race.

  Notes to Chapter 6

  1. My primary sources on Beer's biography are a CV that he provided me, dated 1 January 1998, and a biographical letter Ian Beer sent to all of Stafford's children on 25 August 2002, immediately after Stafford's death. Otherwise unidentified quotations below are from this letter, and I thank Ian Beer for permission to reproduce them. Beer was the only one of the principles of this book still alive when I began my research. I talked to him at length on the phone twice: on 23 June 1999 and 22 December 1999. Various attempts to arrange to meet stumbled geographically: Beer was in Britain when I was in North America and vice versa. I went to Toronto hoping to see him in June 2002 but Beer was terminally ill and unable to speak while I was there. I met Beer's partner, Allenna Leonard, and his daughter Vanilla during that visit. I am very grateful to them for talking to me at length at that difficult time, and for many subsequent communications which I draw on throughout this chapter. I also thank David Whittaker for an informative conversation on Beer's poetry and spirituality and his relation to Brian Eno (Woodstock, Oxon, 24 August 2004). I should mention that Liverpool John Moores University has extensive holdings of Beer's papers: "Ninety Two Boxes of Beer" is the subtitle of the catalog. I have not consulted this archive: the present chapter is very long already.

  2. The only use of Stafford's first name that I have found in print is in his dedication to his mother of his contribution to Blohm, Beer, and Suzuki (1986), which he signed "Tony."

  3. Beer (1994b, viii): "Neurophysiologists did not talk much about logic at that time . . . and logicians were completely indifferent to the ways brains work."

  4. Telephone conversation, 23 June 1999.

  5. For more on the history of OR, see Fortun and Schweber (1993), Pickering (1995a), and Mirowski (2002).

  6. Telephone conversation, 23 June 1999. There is actually a psychiatric connection here at the beginning of Beer's career. According to Beer (1989a, 11–12), "At the end of my military service, I spent a year from the autumn of 1947 to that of 1948 as an army psychologist running an experimental unit of 180 young soldiers. . . . All these men were illiterate, and all had been graded by a psychiatrist as pathological personalities. . . . I had a background in philosophy first and psychology second; the latter school had emphasized the role of the brain in mentation and of quantitative approaches in methodology. The analytical models that I now developed, the hypotheses set up and tested, were thus essentially neurophysiological in structure and statistical in operation. I had a background in the Gurkha Rifles too. What made these people, unusual as they were, tick—and be motivated and be adaptive and be happy too . . ? And how did the description of individuals carry over into the description of the whole unit, for it seemed to do so: every one of many visitors to the strange place found it quite extraordinary as an organic whole. . . . This was the start of the subsequent hypothesis that there might be invariances in the behaviour of individuals . . . and that these invariances might inform also the peer group of individuals, and even the total societary unit to which they belong."

  7. On the unpredictability of the firm's environment: "The first kind of regulation is performed in the face of perturbations introduced by the environmental economy, both of the nation and of competition in the money market. The second is performed in the face of market perturbations, which may be due to the aggressive marketing politicies of competitors, but which are fundamentally caused by the rate of technological innovation" (Beer 1981, 186–87).

  8. Noble (1986) discusses U.S. visions of the automatic factory and lays out the Marxist critique, but it is worth noting that effects of automation were viewed with less suspicion in Britain and Western Europe. In the early 1960s it was taken for granted that automation would make it possible for people to work less, but the principal concern, especially on the left, was with the so-called leisure problem: what would people do with their spare time? Would the young dissolve into deliquency while their parents sat at home watching the television all day (as in Anthony Burgess's A Clockwork Orange)?The optimistic British and European response was that this was an opportunity for a new kind of people to emerge, no longer exhausted by labor, and both Trocchi's sigma project (in the preceding chapter) and the Fun Palace project (next chapter) should be seen in relation to this—as ways to foster the emergence of new postindustrial selves (likewise the activities of the Situationist International in France).

  9. The painting also includes an R-machine operating at a lower cerebral level than the others. For simplicity I have skipped over this.

  10. "The research is under the direction of the author, but the detailed results given . . . were obtained by a project team consisting of three operational research workers: Mr. T. P. Conway, Miss H. J. Hirst and Miss M. D. Scott. This team is led by Mr. D. A. Hopkins, who is also the author's chief assistant in this field" (Beer 1962a, 212).

  11. This idea clearly goes back to Norbert Wiener's work on predictors and learning (Galison 1994), and one can easily see how it could be automated, though it had not, in fact, yet been automated by Beer's team.

  12. Beer (1959, 138–41) discusses a simple, realistic example of how this might go.

  13. For a clear statement, see Beer (1975a [1973]).

  14. The most detailed information I have been able to obtain is from D. J. Stewart (email, 8 June 2006), and I am grateful for his assistance. Stewart, another British cybernetician and friend of Walter, Beer, and Pask, had been employed at Brunel since 1965, before it was granted university status, and he was also centrally involved in the establishment of the Cybernetics Department. The fund-raising dinner was held at Claridge's on 7 June 1968, and the department began operation in that year. "Frank George was appointed Director, and the initial staff consisted of Gordon Pask part
time and me [Stewart] full time. Ross Ashby was in Illinois at the time but was expected to return and join us in 1970. In the event he never did." In 1981 "serious financial troubles hit all British universities [and] [f]rom then on further financial restrictions, together with this rather unsatisfactory structure, caused both the research and the teaching [in cybernetics] gradually to wither away."

  15. As mentioned earlier, from the 1970s onward Beer held several visiting academic appointments which made it possible for him to propagate his vision of cybernetic managagement within the system of higher education, but part-time positions are themselves marginal to the academic establishment and cannot be seen as an institutional solution to the problem of the social basis of cybernetics.

  16. In the Hitchhiker's Guide,the earth is a giant analog computer built by mouselike beings to answer the Ultimate Question. On the earth as an analog computer, see Blohm, Beer, and Suzuki (1986).

  17. For more on this robot, see www.conceptlab.com/control/. I am grateful to Ellen Fireman for first telling me about this project, and to Garnet Herz for telling me more about it when I visited Irvine in October 2005. When I first visited this website ( 21 July 2005), the project was entitled "Control and Communication in the Animal and the Machine," a direct quotation of the subtitle of Wiener's Cybernetics.The title has since changed to "Cockroach Controlled Mobile Robot," though the previous title is still listed too.

  18. The Phumox project of Andy Gracie and Brian Lee Yung Rowe does thematize adaptation, aiming at coupling an AI system into the evolution of biological systems, but it is not clear to me how far this project has got (www.aminima .net/phumoxeng.htm). In a related project, "Small Work for Robots and Insects," a neural network analyzes the song of crickets and aims at a reciprocally adaptive sonic coupling. Gracie says that "Phumox frames our interest in exploring connections between machine and nature that are outside the typical areas covered by cybernetics," but his work strikes me as directly in the line of cybernetic inheritance from Beer and Pask. I thank Guillermo Santamaria for telling me about Gracie's work.

  19. System 3 "is ideally placed to use every kind of optimizing tool in its direction of current operations, from inventory theory to mathematical programming. A dynamic, current model of the firm's internal workings must in fact emerge at this level, and offers the ideal management tool for the control of internal stability" (Beer 1981, 178).

  20. Beer had sketchily included these parasympathetic signals in his 1960 cybernetic factory essay in association with the equally sketchy R-machine. In effect, levels 1–3 of the VSM were elaborations of Beer's vision of the R-machine. As an example of what might be at stake here, think of a situation in which management calls for a sudden increase in production. Left to themselves, systems 1–3 might simply try to implement this demand even if production quality went down, the machines started breaking, and the workers went on strike.

  21. Beer contrasted his vision of higher management as a densely connected network of neurons with the traditional conception of a hierarchical pyramid, as respectively adaptive and nonadaptive (Beer 1981, 201, fig. 39; 204, fig. 40). Beer connected this to McCulloch's notion of the "redundancy of potential command" (232ff.)—the idea that control switches between structures in the brain as a function of the different situations encountered. Beer's idea was that higher management should function likewise.

  22. Beyond multiplicity, the recursive aspect of the VSM also implies a notion of scale invariance: whatever the scale of analysis, one finds the same structures: viable systems. Both of these ideas strike me as ontologically valuable (Pickering 1995). On the other hand, Beer's tidy form of recursion, with layers of viable systems neatly stacked within each other, is less cogent. My own studies have never turned up anything like this regular structure, and for this reason I am inclined to discount this aspect of the VSM as ontological theater. Beer sometimes claimed to have demonstrated logically that all viable systems have to have such a structure, but I have not been able to find a proof that I can understand. At other times, he noted that it was usefulto think of viable systems as recursive—"In order to discuss the organization of vast institutions as well as small ones, the principle of recursiveness was invoked. We should depict the organization as a set of viable systems within a set of viable systems, and so on. That decision was perhaps not a necessity; but it did offer a convenient and powerful convention for our work" (Beer 1979, 199). Recursivity can clearly be a "convenient and powerful convention" in getting to grips with complex biological and quasi-biological organizations even if it does not reflect a necessary feature of the world.

  23. This idea also runs through the development of game-playing computer programs.

  24. For an extended discussion of the system 4 model and its adaptability, see Beer (1981, 183–92).My listing of variables is an abbreviated selection from the variables that appear in Beer's fig. 36 (p. 188). As far as I know, the best-developed system 4 model was that constructed for the Chile project (below). This had the form described above, drawing upon Jay Forrester's "Systems Dynamics" approach to modelling complex systems (on the history, substance and applications of which, see Elichirigoity 1999).

  25. "These charts are, or more usually are not but could be, supported by detailed job descriptions intended to show how the whole thing works. So the charts themselves specify an anatomy of management, while the job descriptions specify its physiology" (Beer 1981, 77). Beer (79) describes these organizational charts as "arbitrary" and "frozen out of history." The rhetorical other to the VSM was always this vision of linear hierarchy of command. In recent years, of course, organizational theory and practice have moved away from that model, often, in fact, in rather cybernetic directions. See, for instance, the work of David Stark (Neff and Stark 2004; Stark 2006) on flat organizations, self-conscious concerns for adaptation, and the redundancy of potential command (mentioned in n. 21 above). Stark lists the Santa Fe Institute (chap. 4) among his affiliations.

  26. In the language of contemporary science and technology studies, Beer here addresses topics concerned with "distributed cognition"—the idea that relevant knowledge of complex organizations is spread throughout the organization rather than fully centralized. See Hutchins and Klausen (1996) and Star (1991) on automation as severing important social connections, and Hutchins (1995) for an extended treatment. Beer (1981, 109–10) also mentions the converse problem: "In fact, one of the key problems for scientists installing such systems [computers] in industry is that the connections they wishto cut are not always successfully cut." He gives an example relating to foremen and chargehands who continue to operate the old system using "little books of private information." "Surgeons have encountered a precisely similar phenomenon when performing trunk sympathectomies. . . . The surgeon does not expect the feedback circuits involved to operate any longer—but sometimes they do."

  27. This produces a Beer/cybernetics-centric account of developments within an exceptionally turbulent and eventful period of Chilean history.Medina (2006) does an excellent job of situating Beer's project within the wider frame of social, economic, and political developments in Chile, but in the present section I am principally concerned with the project as an exemplification of the VSM in action. I amgrateful to EdenMedina for discussion of her work prior to publication, and for her detailed comments on the present account of Cybersyn.

  28. This was Beer's understanding of what engagement with individual enterprises would look like, but, asMedina (personal communication, 21 September 2007) notes, the situation on the ground was rather different: "It would be more accurate to describe management as politically appointed interventors. TheOR team from the Cybersyn project rarely had any interactions with the workers at any level. In the case of the Yarkur textile mill . . . the OR scientists worked exclusively with the interventor in charge of finances. Stafford may not have realized this. Moreover, longstanding class prejudices also kept Cybersyn scientists and engineers from interacting with the rank and file."

/>   29. Medina (personal communication, 21 September 2007) points out that a government report in 1973 gave a figure of 27% for the number of enterprises connected to Cybernet. Nothing hinges here on this figure, though it suggests that Beer's account of the rate of progress of Cybernsyn might be overly optimistic.

  30. Meadows et al. (1972). For the history of Forrester's work and its part in an emerging "discourse of globality" at the heart of "planet management" see Elichirigoity (1999).

  31. Von Foerster papers, University of Illinois archives, box 1, Beer folder.

  32. Flores completed a PhD at the University of California, Berkeley, under the direction of Hubert Dreyfus, whom I thank for an illuminating conversation. Flores's management consultancy draws directly on insights from the philosophy of Martin Heidegger rather than the VSM (Rubin 1999). Traces of the ontology of unknowability and the stance of revealing are evident: "It is the third realm of Flores's taxonomy to which people should aspire: What You Don't Know You Don't Know. To live in this realm is to notice opportunities that have the power to reinvent your company, opportunities that we're normally too blind to see. In this third realm, you see without bias: You're not weighted down with information" (Rubin 1999).

  33. The quotation is from an email message to the author, 3 April 2003. I had not consciously encountered the word "metanoia" before. Besides this usage, it also turns out to have been one of R. D. Laing's favorite words for tranformative voyages through inner space (see Laing 1972, for example). The word also turns up in Pearce (2002 [1971]), a minor classic of the sixties countercultural literature. (I thank Fernando Elichirigoity for bringing this book to my attention.) The Oxford English Dictionarydefines "metanoia" as "the setting up [of] an immense new inward movement for obtaining the rule of life; a change of the inner man."