17. On Thoughtsticker from the early 1980s to the present, see Pangaro (2001). Pangaro is an American who moved to Britain to work with Pask. He records that, having been funded by the British Social Science Research Council, the project was later continued with support from the Admiralty. He also comments on the improvised nature of the computer hardware that typified System Research: "This THOUGHTSTICKER used display hardware that was obsolete at Negroponte's lab [see below] and had been shipped to System Research for dynamic displays of a Paskian nature. An incessant software bug, which Pask contended was a 'feature,' led to extraneous lines in these displays, but did little to discourage the imagination that, with decent funding, something really amazing could be done here" (Pangaro 2001, 794). Pangaro himself, with Pask as consultant and adviser, established a consultancy to build a more up-to-date version of the machine with a hypertext interface. This fell victim to a software platform battle in the mid-1980s, though Pangaro continues to imagine some version of Thoughtsticker as a more active and intelligent Web browser than any currently available. The affinity between Pask's interface techniques and hypertext was noted at the time by Ted Nelson (1987), quoted as "Ted Nelson on Gordon Pask," www2.venus.co.uk/gordonpask/clib.html.
18. Philosophers of science will recognize an affinity between Pask's entailment meshes and Mary Hesse's (1966, 1974) network theory of knowledge. Pask argued that entailment meshes close in on themselves (for more on this, including the idea that the topology of closure is a torus, see B. Scott 2001b), which is obviously a useful approximation in the construction of teaching machines, but surely wrong in general.
19. Other recollections and historical accounts of Pask's educational work at Brunel and the OU include Thomas and Harri-Augstein (1993) and Laurillard (2001). The most successful pedagogical machine in the Paskian tradition up to the present appears to be a system called Byzantium, developed in the late 1990s by a team from a consortium of British universities including Bernard Scott (Patel, Scott, and Kinshuk 2001).
20. Leaving the realm of machines, one could explore the cybernetics of education further by following the lead of another of Pask's Brunel students, Ranulph Glanville. Glanville (2002a, 2002b) takes seriously the idea of children as exceedingly complex systems with their own dynamics and connects it to the idea that education should be understood as a constructive fostering of that dynamics rather than any simple transmission of information. This idea returns to the work of Friedrich Froebel (1782–1852) who, according to Glanville, invented the metaphor of the child as a growing plant—a very cybernetic image. Concretely, Glanville points to Froebel kindergartens,Montessori schools, and a form of studio education common in teaching design as examples of cybernetic pedagogic institutions. (Grey Walter was also keen on Montessori schools: Walter 1953, 269.) I would be tempted to add the Anti-University of London (discussed in chap. 5) and Joan Littlewood's Fun Palace from the same period (discussed below) to the list, with Kingsley Hall as a sister institution in the realm of psychiatry. As usual, we can note that the cybernetic approach entails a symmetrizing shift away from the familiar teacher-student hierarchy, and that Glanville's examples of his preferred forms of pedagogy remain marginal to the educational establishment.
21. If we wanted to follow this line of thought further, we could note that Pask is seen is one of the founders of second-order cybernetics and a significant contributor to Maturana and Varela's "autopoiesis" tradition and recall that Francisco Varela was scientific adviser to the Dalai Lama.
22. Pask's publication list begins with eighteen papers written between 1957 and the publication of An Approach to Cyberneticsin 1961. Ten are about teaching machines; the other eight are about cybernetic controllers such as chemical computers.
23. On genuinely biological computers, Pask (1960b, 258) says that he and Beer "have examined models, where currency is food supply and unicellulars like paramecium are active elements, sufficiently to show that such colonial organization may be coupled to a real process."
24. As usual, Pask was himself not very forthcoming on the practical details. Pask (1958, 166–67) sketches out a means for monitoring the development of threads in some region using four ancillary electrodes. Two electrodes emit current periodically (presumably to inhibit growth of threads from them) and the others register some response. The trick would be to look for changes in response correlated with sounds in the required range (as detected by a filter attached to the microphone). Positive correlations could then be encouraged by injecting more current into the assemblage as they occur.
25. See also the distinction between a cybernetic "natural history" approach and the traditional scientific method in Pask (1960b).
26. For further development of Pask's ideas on epistemology in relation to adaptive and self-organizing systems see the work of Peter Cariani, available at his website: www.cariani.com. I am very grateful to Cariani for discussions about Pask's chemical computers, and responsibility for inadequacies in my account rest with me. Isabelle Stengers's (1997) critical philosophy of science is an attempt to push science in the direction of a cybernetic and "risky" method. Looking in a different direction, the "grounded theory" approach to socialscientific research foregrounds the kind of dense engagement with the object that Pask took to characterize cybernetics. Genuinely "risky" research conducted under the aegis of grounded theory thus constitutes yet another of example of ontology in action.
27. Pask (1961, plate IV, facing p. 65) reproduces images of the evolution of an "activity surge in 2-dimensional cell array of neurone-like units, simulated on a computer by R. L. Beurle." On the later simulations discussed below: "The programs . . . were written by D. J. Feldman in the machine language of the ICT 1202 computer and the programs were run on this machine. (A good deal of reprocessing of cards was needed to augment the limited storage capacity of this machine.)" Pask (1969a, 106).
28. See, for example, the online journal, first published in 1998, the Journal of Arti-ficial Societies and Social Simulation,jasss.soc.surrey.ac.uk/JASSS.html.
29. Pask was not alone in the 1960s in his interest in cellular automata as models for the biological and social sciences. He cited the work of five other individuals and groups (Pask 1969a, 102). In chapter 4 I discussed the way in which aspects of Ashby's cybernetics were elaborated in the work of Christopher Alexander, Stuart Kauffman, and StephenWolframin later developments centering on cellular automata. In a 1969 paper on architecture, Pask included Alexander in a list of people with whom he had personal contact (1969b, 496). The hand calculations were presented by Pask in talks and publications in the early 1960s. Pask (1962) discusses models of slime molds, which makes a bridge to the work on self-organization by Ilya Prigogine (and thus back to Stengers): see Prigogine and Stengers (1984) and Pickering (2005a).
30. I thank Peter Asaro for bringing this work to my attention.
31. I continue to suspect that Pask was the model for one of the BBC's Dr. Who's (Patrick Troughton—the perky little one who played the penny whistle), but the only documented connection to Dr. Who that I can find is this (Moore 2001, 770): "One evening [in the 1970s]. prior to our meeting [of the Cybernetics Society, of which Pask was Chairman], I recognised at the bar Tom Baker, famous for his performances on television as Dr Who, all round scientist and 'Time Lord.' I invited him to come upstairs and join our monthly meeting. Alas he declined saying that he was only an actor and did not understand such high-level science."
32. I thank Sharon Irish for many conversations on what follows and its wider context, and for access to much relevant literature, especially on architecture.
33. For a fuller account of Littlewood's role in what follows, see Littlewood (1994).
34. Pask (1964b, 10): "The structural organisation of a dramatic presentation suitable for this system closely resembles the branching programmes used extensively in teaching machines." Pask's diagram 2 (1964b, 9) shows a monitoring device which displays possible lines of plot development as a network of interconnected
branches, and ends with a skeletal example of the possible forms of such a play (28–30). One of Pask's suggestion was that in rehearsal actors could work back from various nodes in the action, thus developing the overall network of possible trajectories for the play and the metainformation which would be provided to the audience at the branch points (29–30). I thank Paul Pangaro for a copy of this proposal.
35. The Guardianobituary also recalled that "the company's exceptional flair for improvisation and rewriting—Behan's script was chaotic—drew full houses" (Ezard 2002).
36. Pask's cybernetic theater was by no means unique in the sixties in its interactive and emergent aspects: one thinks of "happenings" and performance art. My suggestion is that we can see the latter, too, as exemplifications of the cybernetic ontology in action, somewhat differently staged from Pask's project. I thank Ranulph Glanville for pointing out to me that in 2000 Jeffrey Archer wrote a West End play about his own trial for perjury in which the audience voted on Archer's innocence or guilt at the end (see Wikipedia on Archer). This is a heavily watered down version of what Pask had in mind.
37. For a historical overview of the British art scene in the 1960s, see Stephens and Stout (2004).
38. The exhibition had a relatively small budget of only £20,000; Reichardt was employed part time at the ICA at a salary of £30 per week; artists were not paid to exhibit their work; and the ICA did not produce a catalog for the exhibition (Reichardt, personal communication, 21 February 2002). Instead, a special issue of Studio Internationalwas published to coincide with the exhibition: Reichardt (1968a).
39. Robin-McKinnon Wood and Margaret Masterman (1968) contributed a short poem that had been generated by a computer at the Cambridge Language Research Institute: a bug in a language analysis program had randomized the output to produce something resembling a Burroughs style cut-up. The only roboticist to appear in histories of the British underground, Bruce Lacey, also contributed to the exhibition. As discussed in Lacey (1968), his exhibits at the ICA were interestingly cybernetic, and able to interact with their environment (Owl,1967), with humans (Rosa Bosom (R.O.S.A.—Radio Operated Simulated Actress),1965), and with other robots (Mate,1967). Nuttall (1968, 125) refers to Lacey's "magnificent hominoids, sick, urinating, stuttering machines constructed of the debris of the century, always with pointed socialist/pacifist overtones but with a profound sense of anger, disgust and gaiety that goes far beyond any political standpoint."
40. Pask (1968, 35) acknowledges "Maurice Hyams in cooperation with System Research Ltd" as "patron of the project."
41. Speaking of their undergraduate days in the early 1950s, Harry Moore (2001, 769) recalled that "Grey Walter's experiments on 'artificial neurones' and tortoise models in his book The Living Brain. . . also provided Gordon with additional stimulation."
42. As Gere (2002, 95) points out, this "computer" is visible in the background of fig. 7.17. Pask (1971, 98) creditsMark Dowson for constructing the electronics, Tony Watts for the electromechanical construction, Yolanda Sonnabend (an eminent theater designer) for the design of the female robots, and himself for the male design and the overall setup.
43. One can also think about the Colloquy from a different angle. The idea of robots mating in public seems to have escaped explicit discussion from 1968 to the present, but it is hard not to see the Colloquy as some sort of reflection on the "permissive society."
44. Zeidner et al. (2001) recalls Zeidner's first meeting with Pask and is a good place to begin an inquiry into this strand of Pask's work. "He infused our research with new concepts and paradigms for understanding decision making in complex, poorly-structured systems. He also introduced us to the use of formal theory and logic to better understand and predict interpersonal or personmachine communications" (984–85). Pask organized four conferences for Zeidner's Army Research Institute in Richmond (close to his home and System Research) on decision making in complex systems, in 1975, 1976, 1978, and 1983. The aim of the first was "to elicit a fair picture of the state of the art in decision making in Europe; the status of decision oriented disciplines; and ongoing or contemplated lines of research" (Pask 1976c, i). One can get a feeling for the overall problematic from the first paper at the meeting, "SIMTOS: A Review of Recent Developments," by J. Baker, the supervisory project director of the Organisations and Systems Research Laboratory of the Army Research Institute in Washington (Baker 1976). SIMTOS was a computerized battle simulation in which military personnel conducted operations against a simulated enemy. "Decision making" for the Army Research Institute thus referred to decisions made in the flow of battle, though neither the academics at the meeting nor Pask referred to this context. The second conference (Pask 1978) was devoted to the problematic of training "decision makers." I amgrateful to Joseph Zeidner for providing me with copies of the proceedings of these conferences. There is more on this work in Pask and Curran (1982, 130–31): "Conference discussions stimulated research into a different type of decision aid, a sort of on-going model which System Research has since updated. This model was an evolving computer system called TDS, or Team Decision System." This system featured a space battle scenario, in which "TDS can aid decision making by allowing the commander to interrogate the system; by giving information; by presenting the tactics currently in use; by doing calculations. But it also has an extra ingredient: there are real emergencies which parallel the unexpected emergencies of real life. . . . However much the commanders interact with TDS as a decision making aid, they have no chance to deal with emergencies, simply because these happen too fast. . . . TDS takes over and makes the choice on the basis of what it has learned about the strategies commanders have chosen in the past." Pask also carried out research for the U.S. Air Force in the periods 1961–65 and 1975–77. The first report on that work was "Research on the Design of Adaptive Training Systems with a Capability for Selecting and Altering Criteria for Adaptation" (Pask et al. 1965).
45. Simon Penny is an interactive artist and theorist based at the University of California, Irvine. I thank him for enlightening discussions on the past and present of his field. Penny (1999) discusses the writings of sculptor Jack Burnham (1968) on the history and future of sculpture. In Beyond Modern SculptureBurnham devotes the first half of the book to "sculpture as object," meaning the modern tradition, the past. The second half is on "sculpture as system" running from "sculpture and automata" up to "robot and cyborg art," machine art, and the future. Cybernetics is central to Burnham's discussion of the last of these, beginning with the tortoise, the homeostat, and the maze-running mice built by Claude Shannon and continuing with Pask's Eucrates (204, 337) (the book was written before the Colloquy was exhibited).
46. Several of these points are made by Penny (2008), who also alerted me to the difficulty of keeping these machines going in an art museum (the ScienceMuseum in London has a motionless tortoise displayed in a glass case). The threat to the distinctive identity of the artist was explicit at the Cybernetic Serendipityexhibition: "Two aspects of this whole project are particularly significant. The first is that at no point was it clear to any of the visitors . . . which of the various drawings, objects and machines were made by artists and which were made by engineers; or, whether the photographic blow-ups of texts mounted on the walls were the work of poets or scientists. There was nothing intrinsic in the works themselves to provide information as to who made them. Among the contributors . . . were forty-three composers, artists and poets, and eightyseven engineers, doctors, computer systems designers and philosophers. The second significant fact is that whereas new media inevitably contribute to the changing forms of the arts, it is unprecedented that a new tool should bring in its wake new people to become involved in creative activity. . . . Graphic plotters, cathode-ray tube displays and teleprinters have enabled engineers, and others, who would never even have thought of putting pen to paper, to make images for the sheer pleasure of seeing them materialize" (Reichardt 1971, 11). Artists might well have been concerned for their alread
y tenuous social status.
47. A growing if ambiguous fascination with technology is visible in the history of British art in the sixties (Stephens and Stout 2004). One indicator of the cultural centrality achieved by machine art is that Cybernetic Serendipitywas opened by AnthonyWedgewood-Benn, then minister of technology, with Lord Snowdon (Anthony Armstrong-Jones, photographer) and Princess Margaret also in evidence (Reichardt, interview, 21 February 2003).
48. We can see this as a return to "normal" in the art world, reinforced in the late sixties and early seventies by a critique of science and technology withinthe counterculture. The critique grew out of disgust at the Vietnam War and the complicity of scientists and engineers with the military-industrial complex. Sadler (2005) discusses this in relation to adaptive architecture (below).
49. In 1972, Jasia Reichardt tried to put together another project fusing art and science, but was unable to find support for it: "Every proposal was dismissed." Later she sought to develop a project called "Fantasia Mathematica" on the arts and mathematics for the BBC but was again unable to attract a sufficient budget. She told me that no one in Britain was interested in archival papers on Cybernetic Serendipity,and that the papers are now in Japan. Pask led a discussion at "Event 1" of the Computer Art Society, 29–30 March 1969; "Event 2" never happened (Reichardt, interview, 21 February 2002).
50. For examples of Penny's work, see ace.uci.edu/penny/. Penny knew Pask personally; Hertz was working in Penny'sMasters Program in Arts, Computation, and Engineering at the University of California, Irvine, when I met him (24 October 2005).
51. Gere (2002, 110) gives a different list of names, including Roy Ascott (below), David Medalla ("who refers to himself as a 'hylozoist,' a philosopher of the pre-Socratic Ionian school devoted to the belief that all matter and life are inseparable"; Burnham 1968, 345) and the ubiquitous Stelarc. For a recent illtempered critique of interactive art fromthe mainstream as represented by the New York Times,see Boxer (2005).
The Cybernetic Brain Page 61