- Shopping Bag ( 0 items )
THE MAKING OF A SYNTHETIC BRAIN REQUIRES NOW LITTLE MORE THAN TIME AND LABOUR.... SUCH A MACHINE MIGHT BE USED IN THE DISTANT FUTURE ... TO EXPLORE REGIONS OF INTELLECTUAL SUBTLETY AND COMPLEXITY AT PRESENT BEYOND THE HUMAN POWERS.... HOW WILL IT END? I SUGGEST THAT THE SIMPLEST WAY TO FIND OUT IS TO MAKE THE THING AND SEE.
ROSS ASHBY, "DESIGN FOR A BRAIN" (1948, 382–83)
On 13 December 1948, the Daily Herald carried a front-page article entitled "The Clicking Brain Is Cleverer Than Man's," featuring a machine called the homeostat built by W. Ross Ashby. Soon the rest of the press in Britain and around the world followed suit. In the United States, an article in Time magazine, "The Thinking Machine," appeared on 24 January 1949 (p. 66), and by 8 March 1949 Ashby was holding forth on BBC radio on "imitating the brain." At much the same time, W. Grey Walter appeared on BBC television showing off a couple of small robots he had built, Elmer and Elsie, the first examples of his robot "tortoises," or, more pretentiously, of a new inorganic species, Machina speculatrix. One appeared in a family photo in Time (fig. 1.1). In 1952, Gordon Pask began work on his Musicolour machine—an electromechanical device that collaborated in obscure ways with a musician to generate a synesthetic light show. Soon he was also experimenting with quasi-biological electrochemical computers that could evolve new senses, and within a decade he was designing buildings that could reconfigure themselves in "conversation" with their users. In 1959 Stafford Beer published a book imagining an automated factory controlled by a biological computer—perhaps a colony of insects or perhaps a complex ecosystem such as a pond. By the early 1970s, he was redesigning the "nervous system" of the Chilean economy at the invitation of the socialist government of Salvador Allende.
Examples like these convey some of the flavor of the history explored in the following chapters. In this chapter and the next I want to discuss more generally what cybernetics is, or was, and why it interests me. (The tense is difficult; cybernetics as a field is alive today, but the main characters of this book are all now dead. I will tend therefore to speak of cybernetics in the past tense, as referring to a historical body of work.)
* * *
SOME PEOPLE THINK THAT CYBERNETICS IS ANOTHER WORD FOR AUTOMATION; SOME THAT IT CONCERNS EXPERIMENTS WITH RATS; SOME THAT IT IS A BRANCH OF MATHEMATICS; OTHERS THAT IT WANTS TO BUILD A COMPUTER CAPABLE OF RUNNING THE COUNTRY. MY HOPE IS THAT ... PEOPLE WILL UNDERSTAND BOTH HOW THESE WONDERFULLY DIFFERENT NOTIONS CAN BE SIMULTANEOUSLY CURRENT, AND ALSO WHY NONE OF THEM IS MUCH TO THE POINT.
STAFFORD BEER, CYBERNETICS AND MANAGEMENT (1959, VI)
TO SPEAK OF A HISTORY, ANY HISTORY, AS THOUGH THERE WAS BUT ONE SOMEHOW CANONICAL HISTORY ... IS MISLEADING.... ANY ENTITY, CULTURE OR CIVILISATION ... CARRIES INNUMERABLE, IN SOME WAYS DIFFERING, HISTORIES.
GORDON PASK, "INTERACTIONS OF ACTORS" (1992, 11)
The word "cybernetics" was coined in 1947 by the eminent American mathematician Norbert Wiener and his friends to name the kind of science they were discussing at the famous Macy conferences held between 1946 and 1953. It was derived from the Greek word kybernetes (Latin equivalent, gubernator) meaning "governor" in the sense of "steersman," so one could read "cybernetics" as "the science of steersmanship"—and this is, as it happens, a good definition as far as this book is concerned. The matter was made more interesting and complicated, however, by Wiener's 1948 book which put the word into circulation, Cybernetics; or, Control and Communication in the Animal and the Machine. There Wiener tried to tie together all sorts of more or less independent lines of scientific development: digital electronic computing (then still novel), information theory, early work on neural networks, the theory of servomechanisms and feedback systems, and work in psychology, psychiatry, decision theory, and the social sciences. There are many stories to be told of the evolution, the comings together, and the driftings apart of these threads, only a few of which have so far attracted the attention of scholars. One can almost say that everyone can have their own history of cybernetics.
In this book I do not attempt a panoptic survey of everything that could be plausibly described as cybernetic. I focus on the strand of cybernetics that interests me most, which turns out to mean the work of a largely forgotten group of British cyberneticians, active from the end of World War II almost to the present. Even to develop an overview of British cybernetics would require several books, so I focus instead on a few leading lights of the field, the ones mentioned already: Grey Walter (1910–77), Ross Ashby (1903–72), Stafford Beer (1926–2002), and Gordon Pask (1928–96), with a substantial detour through the work of Gregory Bateson and R. D. Laing. And even with this editorial principle, I have to recognize that each of my four easily warrants his own biography, which I have not attempted to write. So what follows is very much my own history of cybernetics in Britain—not a comprehensive survey, but the story of a set of scientific, technological, and social developments that speak to me for reasons I will explain and that I hope will interest others.
A further principle of selection is also in play. Most accounts of the history of cybernetics are in the mode of a history of ideas; they concentrate on grasping the key ideas that differentiate cybernetics from other sciences. I am not uninterested in ideas, but I am interested in ideas as engaged in practice, and at the heart of this book is a series of real-world projects encompassing all sorts of strange machines and artifacts, material and social. I want to document what cybernetics looked like when people did it, rather than just thought it. That is why the opening paragraph ran from artificial brains to the Chilean economy, rather than offering an abstract discussion of the notion of "feedback" or whatever.
* * *
The choice of principals for this study makes sense sociologically inasmuch as my four cyberneticians interacted strongly with one another. Walter and Ashby were first-generation cyberneticians, active in the area that became known as cybernetics during and even before World War II, and were leading members of the first protocybernetic organization in Britain, the so-called Ratio Club, which met between 1949 and 1958 (Alan Turing was the best-known recruit). They never collaborated in research, but they knew, took account of, and commented on each other's work, though relations became strained in 1959 when Ashby briefly became Walter's boss. Beer and Pask were second-generation cyberneticians, coming onto the scene in the 1950s after the foundations of the field had been laid. They were lifelong friends, and Beer became almost the social secretary of the British branch of cybernetics, with strong personal ties not only to Walter, Ashby, and Pask and but also to Wiener and to Warren McCulloch, the guiding spirit of cybernetics in the United States. But what about the technical content of British cybernetics? Is there any unity there?
The standard origin story has it that cybernetics evolved out of the intersection of mathematics and engineering in U.S. military research in World War II, and this is certainly a good description of Wiener's trajectory (Galison 1994). But figure 1.2, a photograph taken in the early 1950s, originally appeared with the not unreasonable caption "The Four Pioneers of Cybernetics," and what I find striking is that, with Wiener as the exception, three of the four—Ashby, Walter, and McCulloch—spent much or all of their professional careers in research on the human brain, often in psychiatric milieus. We can explore the specifically psychiatric origins of cybernetics in detail in chapters 3 and 4, but for the moment it is enough to note that the distinctive object of British cybernetics was the brain, itself understood in a distinctive way. This requires some explanation now, since it is a way into all that follows.
To put it very crudely, there are two ways to think about the brain and what it does. The way that comes naturally to me is to think of the brain as an organ of knowledge. My brain contains representations, stories, memories, pictures of the world, people and things, myself in it, and so on. If I know something, I have my brain (and not my kidneys, say) to thank for it. Of course, I did not get this image of the brain from nowhere. It is certainly congenial to us academics, professional knowers, and it (or an equivalent image of mind) has been stock-in-trade for philosophy for centuries and for philosophy of science throughout the twentieth century. From the mid-1950s onward this image has been instantiated and highly elaborated in the branch of computer science concerned with artificial intelligence (AI). AI—or, at least, the approach to AI that has become known as GOFAI: good, old-fashioned AI—just is traditional philosophy of science implemented as a set of computer algorithms. The key point that needs to be grasped is that the British cyberneticians' image of the brain was not this representational one.
What else could a brain be, other than our organ of representation? This question once baffled me, but the cyberneticians (let me take the qualifier "British" for granted from now on unless needed) had a different answer. As Ashby put it in 1948, "To some, the critical test of whether a machine is or is not a 'brain' would be whether it can or cannot 'think.' But to the biologist the brain is not a thinking machine, it is an acting machine; it gets information and then it does something about it" (Ashby 1948, 379). The cyberneticians, then, conceived of the brain as an immediately embodied organ, intrinsically tied into bodily performances. And beyond that, they understood the brain's special role to be that of adaptation. The brain is what helps us to get along and come to terms with, and survive in, situations and environments we have never encountered before. Undoubtedly, knowledge helps us get along and adapt to the unknown, and we will have to come back to that, but this simple contrast (still evident in competing approaches to robotics today) is what we need for now: the cybernetic brain was not representational but performative, as I shall say, and its role in performance was adaptation.
As a preliminary definition, then, we can regard cybernetics as a postwar science of the adaptive brain, and the question then becomes: What did cybernetics look like in practice? Just how did the cyberneticians attack the adaptive brain? The answer is, in the first instance, by building electromechanical devices that were themselves adaptive and which could thus be understood as perspicuous and suggestive models for understanding the brain itself. The simplest such model was the servomechanism—an engineering device that reacts to fluctuations in its environment in such a way as to cancel them out. A domestic thermostat is a servomechanism; so was the nineteenth-century steam-engine "governor" which led Wiener to the word "cybernetics." Working with servomechanisms in the war was, in fact, what led Wiener into the field he subsequently named. Walter's robot tortoises and Ashby's homeostat were more striking and original examples of adaptive mechanisms, and they were at the forefront of "brain science" in the late 1940s and throughout the 1950s. A phrase of Warren McCulloch's comes to mind. Speaking of another British protocybernetician, the experimental psychologist Kenneth Craik, McCulloch remarked that Craik always wanted to understand "the go of it"—meaning, to grasp the specific mechanical or quasi-mechanical connections that linked inputs and outputs in complex systems like the brain. Cybernetic devices like tortoises and homeostats aimed precisely to illuminate the go of the adaptive brain.
There is something strange and striking about adaptive mechanisms. Most of the examples of engineering that come to mind are not adaptive. Bridges and buildings, lathes and power presses, cars, televisions, computers, are all designed to be indifferent to their environment, to withstand fluctuations, not to adapt to them. The best bridge is one that just stands there, whatever the weather. Cybernetic devices, in contrast, explicitly aimed to be sensitive and responsive to changes in the world around them, and this endowed them with a disconcerting, quasi-magical, disturbingly lifelike quality. Wiener himself was well aware of this, and his writings are dotted with references to the Sorcerer's Apprentice (who casts a magical spell that sets matter in motion and cannot be undone) and the Golem of Prague (magically animated clay). Walter likewise spoke of "the totems of primitive man" and invoked the figure of Frankenstein's monster (1953, 113, 115). This sense of mystery and transgression has always attached to cybernetics, and accounts, I think, for much of its glamour—the spell it casts over people, including myself.
* * *
I need to say more about cybernetics, the brain, and psychiatry. The early cybernetics of Walter and Ashby directly concerned the brain as an anatomical organ. The tortoise and the homeostat were intended as electromechanical models of the physiological brain, normal and pathological, with the latter providing a direct link to the brutal approaches to psychiatry that were dominant from the 1930s to the 1950s, chemical and electrical shock therapies and lobotomy. In the 1950s and 1960s, however, a different form of cybernetic psychiatry emerged, often, though somewhat misleadingly, labeled "anti-psychiatry" for its opposition to violent interventions in mental illness (and, indeed, for its opposition to the concept of mental illness). I associate this latter form of cybernetic psychiatry with the work of the expatriate Englishman Gregory Bateson (1904–80) and, in the 1960s, with the radical therapeutic experiments of the Scottish psychiatrist R. D. Laing (1927–89).
Unlike my four principals, Bateson and Laing are relatively well known to scholars, the subject of several book-length studies, so I will not discuss their work here to the same depth as the others. But I include a chapter on them for three reasons. First, because Bateson's approach to psychiatry exemplifies a move in cybernetics beyond a concern with the physiological brain and toward something less biologically specified. If Walter and Ashby focused on the adaptive brain, Bateson was concerned with something less precise and less structured, the adaptive subject or self, and how that could be disrupted by what he called double binds. Laing, from this perspective, played out what Batesonian psychiatry might look like in practice. Second, simply to emphasize that cybernetics was not forever irrevocably locked into the world of electroshock. And third, continuing that line of thought, because there is an important sense in which Bateson and Laing were more cybernetic than Walter and Ashby. Laing's psychiatry took seriously, as Walter and Ashby's did not, the idea that we are all adaptive systems, psychiatrists and schizophrenics alike. I am interested to follow the practical and institutional ramifications of this move here.
These features of Bateson and Laing's work—looking beyond the biological brain, and an extension of cybernetics into the field of the self and social relations—move us to another theme of this book, namely, the multiplicity of cybernetics, its protean quality. I began by defining cybernetics as the science of the adaptive brain, but even the earliest manifestations of cybernetics ran in several directions. Tortoises and homeostats could be understood as "brain science" in the sense of trying to explicate the functioning of the normal brain as a complex adaptive system—a holistic counterpoint to reductive neurophysiology, say. At the same time, as I just mentioned, tortoises and homeostats could also simulate the abnormal, pathological brain—madness—and hence stand as a contribution to psychiatry. Furthermore, these cybernetic devices did not have to be seen in relation to the brain at all, but could also be seen as things in themselves. Walter's tortoises, for example, were foundational to approaches to robotics that are very influential today—the situated robotics that I associate with the work of Rodney Brooks, and extremely interesting related work in biologically inspired robotics. From a different angle again, although Ashby's work from the 1930s onward has to be understood as attempting to shed light on the brain, by the 1950s he had begun to see his cybernetics as a general theory, applicable to all sorts of complex systems besides the brain: adaptive autopilots, the British economy, the evolution of species.
Excerpted from THE CYBERNETIC BRAIN by Andrew Pickering Copyright © 2010 by THE UNIVERSITY OF CHICAGO. Excerpted by permission of THE UNIVERSITY OF CHICAGO PRESS. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.