The Cybernetic Brain

by Andrew Pickering

  Figure 7.22.The Fun Palace. Source: Landau 1968, 79, fig. 56.

  Figure 7.23.The Fun Palace's cybernetic control system. Pask 1965, 3, diagram 1. (By permission of Cedric Price Fonds, Collection Centre d'Architecture/Canadian Centre for Architecture, Montréal.)

  Visitors to London will have noticed that the Fun Palace does not exist. Despite a lot of work from a lot of people, political support especially within the Labour Party, and the inclusion of such notables as Yehudi Menuhin and Lord Harewood among its trustees, the project collapsed in the second half of the 1960s, and the building was never even begun. One can cite many of the usual mundane reasons for this: the problems of finding a site, getting permissions, and, not least, raising money.62 But another problem more germane to our theme came up again and again: the sheer difficulty of saying what the Fun Palace was. Like the Dream Machine and Musicolour before it, the Fun Palace failed to fit easily into any of the accepted architectural categories. Not only did it deliberately aim to cut across the usual demarcations—combining the arts, entertainment, education, and sport in all sorts of guises, familiar and unfamiliar, including participation in what were usually taken to be spectator activities—the broader aim was to experiment: to see what might emerge from combining these opportunities in an adaptive space. This, of course, left outsiders to the project free to project their own nightmares on it, and, as Littlewood's obituary in the Guardianput it, the very phrase "Fun Palace" "evoked for councillors a vision of actors copulating in the bushes," and Littlewood's "support dissipated in a fruitless search for a site" (Ezard 2002).63

  Two thoughts before we leave the Fun Palace. The first goes back to the social basis of cybernetics. We can think once more about amateurism. I noted above that Pask's work on the Fun Palace was voluntary and unpaid, done out of interest and for fun and, no doubt, belief in the worth of the project. Here I can just add that, as Mathews (2007, 120) puts it, "like Price, Littlewood had a 'day job' and worked on the Fun Palace on the side." Again we have the sense of something welling up outside the structure of established social institutions and without support from them.

  We can also think in this connection about the relation between modern architecture and buildings like the Fun Palace. The last sentence of Landau's New Directions in British Architecture(1968, 115) reads: "So if architecture is becoming . . . anti-building . . . perhaps it shouldbe classified as not architecture . . . but this wouldsignify that it had taken a New Direction." Mary Lou Lobsinger (2000, 120) picks up the negative and describes the Fun Palace as "the quintessential anti-architectural project." We are back with the "antis"—with the Fun Palace facing contemporary architecture in much the same way as Kingsley Hall faced modern psychiatry. In neither case does the "anti" amount to pure negation. Adaptive architecture was anotherand differentapproach that crossed the terrain of established forms. If mainstream architecture aspired to permanent monuments, aesthetic and symbolic forms drenched in meaning, and fitness to some predefined function, the Fun Palace was envisaged as just a big and ephemeral rectangular box from the outside and a "kit of parts" on the inside. The heart of antiarchitecture lay in its inner dynamics and its processes of transformation in response to emergent, not given, functions—none of which existed (or, at least, were thematized) in the modern tradition. Here we have once more antiarchitecture as nomad science, sweeping in from the steppes to upset, literally, the settled lives of the city dwellers (the Walking City!), and a Situationist architecture as Heideggerian revealing—as keenly open to new ways to be—in contrast to an architecture of enframing, growing out of and reinforcing a given aesthetic and list of functions. Ontology as making a difference. No wonder that "for those who thought architecture had a visually communicative role . . . [Price's] work was anathema to everything architecture might stand for."64

  Second, we can go back to the critique of cybernetics as a science of control. Mathews's account of Price's work in the 1960s takes a strange turn just when Pask appears at the Fun Palace. Speaking of a 1964 report from the Cybernetics Subcommittee, Mathews (2007, 119, 121) picks out what he considers a "rather frightening proposal" discussed under the heading of "Determination of what is likely to induce happiness" and continues:

  This . . . should have alerted Littlewood that the Fun Palace was in danger of becoming an experiment in cybernetic behavior-modification. However, in a 1964 letter to Pask, she actually agreed with his goals, and seemed naively oblivious to the possibility that the project might become a means of social control. . . . The idea that the Fun Palace would essentially be a vast social control system was made clear in the diagram produced by Pask's Cybernetics Subcommittee, which reduced Fun Palace activities to a systematic flowchart in which human beings were treated as data [fig. 7.23 above]. . . . Today, the concept of "unmodified or modified" people would be treated with a considerable amount of caution. Yet, in the 1960s, the prevailing and naive faith in the endless benefits of science and technology was so strong that the Orwellian implications of modification went largely unnoticed.

  What can we say about this? First, this is a pristine example of the sort of critique of cybernetics that I mentioned in the opening chapters, which is why it deserves some attention. Second, the cyberneticians asked for it. They were rhetorically inept, to say the least. They went on endlessly about "control," and "modified people" in figure 7.23 sets one's teeth on edge. It invites Mathews's slide to "behavior-modification," which is a polite way to say "brainwashing." But third, of course, I think the critique is misdirected. It hinges on what I called the Big Brother sense of control—of hierarchical domination, of enframing—and nothing in Pask's work on the Fun Palace contradicts the idea that he was in the same space as Littlewood and Price (and the Situationists before them) in trying to imagine a building in which, far from being stamped by some machine, people could experiment with new and unforeseen ways to be.65 Littlewood was not being naive in agreeing with Pask's goals. The Fun Palace, from Pask's perspective, continued the lineage of Musicolour, a machine that would get bored and encourage the performer to try something new. On a different level, as I have tried to show in this chapter and throughout the book, the cybernetic ontology was one of exceedingly complex systems which necessarily escape domination and with which we have to get along—Pask's notion of "conversation"—and the Fun Palace was just another staging of that ontology. As I also said before, the control critique might be better directed here at the established architectural tradition, which in its symbolic aspect attempts, at least, to tell us what to think and feel, and in its functional guise tries to structure what we do: the factory as a place to work (not to play games, learn or have sex), the school as a place to learn, the home as the dwelling place of the nuclear family . . . This repetitious critique of cybernetics stifles its own object.66

  After the Sixties: Adaptive Architecture

  AN EVOLUTIONARY ARCHITECTURE. . . . NOT A STATIC PICTURE OF BEING, BUT A DYNAMIC PICTURE OF BECOMING AND UNFOLDING—A DIRECT ANALOGY WITH A DESCRIPTION OF THE NATURAL WORLD.

  JOHN FRAZER,AN EVOLUTIONARY ARCHITECTURE (1995, 103)

  The social history of adaptive architecture closely mirrored that of cybernetic art, reaching a zenith in the sixties with the Fun Palace and receding into the margins thereafter, but here we can glance briefly at some postsixties developments that connect to Pask.

  With the Fun Palace as the nexus, the sixties were the decade in which Pask established an enduring connection with architecture more generally. Pask (1969) described cybernetics as a theory of architecture, much as Walter and Ashby had described it as a theory of the brain. Institutionally, Pask's closest relationship was to the AA, the Architecture Association school in London. Cedric Price, who had completed his studies at the AA in 1959 (Melvin 2003), taught there part-time during the Fun Palace project and "was delighted when Gordon agreed to sit on my architectural juries" (Price 2001, 819). Thus began an association between Pask and the AA that continued for the rest of his life: "His presence and
inventions within the life of the Architectural Association are both legendary and of day to day relevance" (Price 2001, 820). Pask often spoke and gave workshops at the AA, and in the late 1980s he took up a continuing position there as an assistant tutor.67 The connection to the AA in turn proved auspicious for the propagation of Pask's cybernetics: "Of 12 successful students Pask had at Brunel University, eight were architects and six came from the Architectural Association" (Scott and Glanville 2001). Archigram's Peter Cook (2001, 571–72) speaks of "a whole generation of young architects. . . . They are, of course, the direct progeny of Gordon." To close this chapter I want to review of few examples of Paskian architecture in practice running up to the present. What these projects have in common is the Paskian idea of a dynamically evolving relation between the human and the nonhuman.

  At the level of complete structures, in 1978 Cedric Price had another attempt at designing a Fun Palace–style building that was reconfigurable in use, this time for the Gilman Paper Corporation. Again, the project was never completed, but Price hired John and Julia Frazer as computer consultants, and they constructed a working electronic model of the Generator project, as it was called.68 "It was proposed to grid the site (a clearing in a forest in Florida) with foundation pads and to provide a permanent mobile crane for moving components, allowing the users of the building to become involved in its organization. . . . We were concerned that the building would not be changed enough by its users because they would not see the potential to do so, and consequently suggested that a characteristic of intelligence, and therefore of the Generator, was that it would register its own boredom and make suggestions for its own reorganization. This is not as facetious as it may sound, as we intended the Generator to learn from the alterations made to its own organization, and coach itself to make better suggestions. Ultimately, the building itself might be better able to determine its arrangements for the users' benefit than the users themselves. This principle is now employed in environmental control systems with a learning capability" (Frazer 1995, 41). The reference to "boredom" here was an explicit evocation of the Musicolour machine.69

  Figure 7.24.Digitally controlled architectural structure. Source: Silver et al. 2001, 907.

  At a more micro level and closer to the present, a 2001 survey of projects at the Bartlett School of Architecture at University College, London, "Prototypical Applications of Cybernetic Systems in Architectural Contexts," subtitled "A Tribute to Gordon Pask," is very much in the Musicolour–Fun Palace tradition, assembling the elements for structures that can transform themselves in use (Silver et al. 2001). One project, for example, entailed the construction of a digitally controlled transformable structure—the skin of a building, say—a key element of any building that can reshape itself in use (fig. 7.24). Another project centered on communication via sounds, lights, and gestures between buildings and their users, reminiscent of the communications systems linking the robots in Pask's Colloquy of Mobiles (fig. 7.25).

  In another tribute to Pask, "The Cybernetics of Architecture," John Frazer (2001) discusses the history of a big, long-term project at the AA, in which Pask participated until his death. The morphogenesis project, as it was called, ran from 1989 to 1996 and was very complex and technically sophisticated; I will simply discuss a couple of aspects of it in general terms.70

  We have so far discussed cybernetic architecture in terms of relations between buildings and users—the former should somehow constitute an aesthetically potent environment for the latter. But one can also conceive of another axis of cybernetic incursion into architecture, this time concerning the relation between the architect and architectural design tools. The classic design tool in architecture is the drawing board—a passive object on which the architect inscribes his or her vision. The drawing board is thus not an aesthetically potent environment in Pask's terms. And much of Pask's involvement with architecture focused on changing that situation, via the development of tools that could adapt to and encourage the architect—again on the model of Musicolour. This was a topic on which he collaborated with Nicholas Negroponte at MIT in the development of what Negroponte called the Architecture Machine—a computerized system that could collaborate more or less symmetrically with the architect in designing buildings—turning crude sketches into plans, indicating problems with them, suggesting extensions, and so on.71

  Figure 7.25.Architectural communication device. Source: Silver et al. 2001, 911.

  Frazer's morphogenesis project took this idea of creating an aesthetically potent environment for design further, along at least two axes. One was to explore new ways of communicating with computers. "Our attempts to improve the software of the user-interface were paralleled by attempts to improve the hardware. The keyboard and mouse have never seemed to me well suited to manipulating models or graphics: a digitizing tablet might be closer to a drawing board, but it is rarely used that way. In any case, we were eager to get away from conventional, drawing board dependent design approaches." Around 1980 a system of cubes was developed, each with an embedded processor. These cubes could be assembled as model structures and could be read by a computer that would build up an internal representation of structures that were somehow patterned on the arrangement of cubes (Frazer 1995, 37).

  Beyond this, the morphogenesis project sought to incorporate the idea that architectural units—buildings, cities, conurbations—grow,quasi-biologically, and adapt to their environments in time.72 As we have seen, in the 1950s and early 1960s, Pask had experimented with inorganic analogue models of organic growth processes—the chemical computers—but he had moved on to mathematical experimentation on cellular automata in the later sixties, and the morphogenesis project likewise took advantage of novel mathematical structures, such as genetic algorithms and cellular automata, to simulate processes of growth, evolution, and adaptation within the computer. The architect would supply the computer with a "seed" structure for a building, say, which the machine would then evolve, taking account of coevolutionary interactions with the building's environment. At the same time, the architect could interfere with this process, in the choice of seed, by selecting certain vectors of evolution for further exploration, and so on. In this way, the computer itself became an active agent in the design process, something the architect could interact with symmetrically, sailing the tides of the algorithms without controlling them (and taking us back to Brian Eno in the previous chapter)—a beautiful exemplification of the cybernetic ontology in action. Figure 7.26 is a single example of this style of coevolutionary design, a computer simulation of how the city of Groningen might develop into the future taking account of interactions between the growing city itself, its inhabitants, and its geographic environment. The quasi-organic structure is evident. As the original caption says, the generating computer model behind it was inspired by Pask's work in the 1950s, and this was, in fact, the last student project that Pask himself supervised.

  One can thus trace out streams of Paskian cybernetic architecture ramifying from the Fun Palace toward the present, in the development of design tools as well as building structures. It remains the case that nothing on the scale of the Fun Palace has yet been built, but perhaps the sixties might be coming back: in 2002 the Royal Institution of British Architects gave the Archigram group its gold medal (Sadler 2005, 7).

  Figure 7.26.Groningen study. Source: J. H. Frazer, "The Cybernetics of Architecture: A Tribute to the Contribution of Gordon Pask," Kybernetes, 30 (2001), 641–51, p. 648.

  Figure 7.27.Gerbil architecture. Source: Nicholas Negroponte, Soft Architecture Machines, (Cambridge, MA: MIT Press, 1975), 46, fig. 1. (© 1976 Massachusetts Institute of Technology, by permission of the MIT Press.)

  _ _ _ _ _

  Let me end this chapter with one more, light-hearted, example of cybernetic architecture, an installation exhibited by Pask's collaborator, Nicholas Negroponte, at the Jewish Museum in New York from September to November 1970.73 Close inspection of figure 7.27 reveals of a mass of small cubes inha
bited by a colony of gerbils. The gerbils push the cubes around, as is their wont. At intervals, a computer scans the scene and either pushes the blocks back where they were, if they have not moved much, or aligns them to a grid in their new positions. The gerbils then go to work again, the computer does its thing once more, and thus the built environment and its inhabitants' use of it coevolve open-endedly in time in ways neither the architect, nor the computer, nor the gerbils could have foreseen—just like a Musicolour performance.

  8

  _ _ _ _ _

  SKETCHES OF ANOTHER FUTURE

  WHAT A LONG STRANGE TRIP IT'S BEEN.

  THE GRATEFUL DEAD, "TRUCKIN'" (1970)

  Writing this book has taken me to places I never expected to go. Strange worlds, new civilizations. To some I went willingly; to others, less so. I was happy to find myself in art worlds I had never imagined; I have acquired an odd interest in architecture, a field that never spoke to me before; it was fun and edifying to sit for a while at the feet of Wizard Prang. I had forgotten—and this is an important datum, I think—how literally wonderful the sixties were. On the other hand, my heart sank when I finally admitted to myself that three of my principals had written books with "brain" in the title and that I had to figure out some brain science. Likewise the realization that three of the "four founders of cybernetics" were strongly associated with psychiatry and that I needed to grapple with the history of that bleak field too. The brain and psychiatry were more daunting prospects than biofeedback and even chemical computers. Other topics just sucked me in: pursuing DAMS from a single published footnote into the labyrinths of Ashby's private journals could drive anyone mad, and the journey remains unfinished. Still, I got to visit all those places, and I am very glad I did; I thank the shades of my cyberneticians for taking me along. I am changed by the trip.