1The Genesis and Growth of Cybernetics

The intellectual climate of the 21st century is not particularly favorable to the so-called “grand narratives” – intellectual approaches that aim to explain the entire reality available to human mind, or at least a large portion of it. It is commonly accepted that structuralism was the last such grand narrative, which seemed to serve as a metatheory of the humanities in the 1960s and 1970s. However, its predecessor in that regard – cybernetics – is rarely mentioned, even though it was even more prevalent between the end of the 1940s and mid-1960s.

Part One of this book is to be devoted to Dialogues – the one among Lem’s works in which his fascination with cybernetics is the strongest.⁶ In fact, Dialogues cannot be understood without referring to the swift career of the discipline. Therefore, before discussing cybernetics itself, I should outline briefly its history. This description of what cybernetics is will, however, come from an amateur. The mathematical tools and vocabulary used by the creators and proponents of cybernetics remain unavailable to me. I will be treating cybernetics as a phenomenon in the history of science and ideas, leaving mathematics in a sort of “black box,” which is not to be opened, but which is being observed focusing on its location and functioning. It is justifiable, as the cyberneticists never limited themselves to producing mathematical arguments. The founding father of cybernetics himself, Norbert Wiener showed the path here (I will return to it). In fact, some branches of cybernetics detached themselves completely from science. And these branches happened to wither the earliest.

Cybernetics is commonly described as “a scientific study of control and communication in complex systems” – this is how it was defined by its creator, Norbert Wiener.⁷ The general character of this description is quite significant, indicating not only a broad background and a variety of sources of the discipline, but also its broad scope. Wiener gave it a name derived from Greek.⁸ “Kybernetes” means ←15 | 16→“helmsman” and is derived from the verb “kybernao”, meaning “to steer.”⁹ The term “governor” has the same root.

Cybernetics was largely born from war-time needs and was related to technologies of building quick counting machines – in both cases the purpose was to facilitate calculating trajectories of missiles targeting bullets. In an introduction to his book Cybernetics, Second Edition: or the Control and Communication in the Animal and the Machine,¹⁰ which became the founding work of the entire discipline, Wiener describes in detail how the ideas of cybernetics were born during seminars he participated in at Harvard’s Vanderbilt Hall in 1941–1944 together with mathematicians (including von Neumann), engineers, biologists and doctors.¹¹ This interdisciplinary gathering observed that there are numerous analogies between the functioning of new calculating machines and biological organisms when it comes to mechanisms of steering and control. It turned out some processes within calculating machines and human nervous systems can be described with the same mathematical formulae – that is, processes that include feedback and oscillations.¹² Research continued after the end of the war was conducted simultaneously in engineering and biology. This duality of research directions is characteristic of the entire cybernetics, and it will be important for the argument that follows.

Wiener himself played a pivotal role in shaping the new discipline – he stood behind its laws and ideology. As a child this versatile scholar and intellectual was fascinated by nature, and traces of such interests are clear in his works, which combine mathematics with physiology. It must have tickled the imagination of a young physician Stanisław Lem, when he read his books in Mieczysław Choynowski’s seminar; learning English from them.¹³ Wiener was not only a ←16 | 17→mathematician, but also an engaged social critic, which can be best seen in his book The Human Use of Human Beings. Cybernetics and Society (1950), which is not a scientific work, but a collection of essays about science for a general public, oftentimes with a journalistic air to them. The fact that this particular book has become a popular guide to cybernetics shows that unlike other disciplines, cybernetics was tied to its social context from the very beginning – its creator himself has positioned it that way, and he did it on purpose. This was certainly aided by his powerful, authoritarian personality, which emanates from his determined arguments admitting no opposition and densely marking his texts, as well as from his very critical remarks about the postwar American society.¹⁴

Apart from contemporary needs and an intellectual osmosis between biologists and engineers, for Wiener the sources of cybernetics lied primarily in the development of thermodynamics and statistical mechanics in the late 19th century. He had especially great respect for one of the men behind both these disciplines – Josiah Willard Gibbs, whose long underestimated works greatly enriched statistical interpretation of energy transmission processes.¹⁵ Information transmission is part of these processes, as Wiener and his colleagues remarked – and the information is treated as a physical quality here. In Cybernetics, Wiener provides basis for a mathematical description of information,¹⁶ which was then developed further by his disciple, Claude Shannon. This is where physics and biology meet: according to Wiener a biological organism is an energy and information processing system.

Later cyberneticists developed the discipline much further and found some much earlier antecedents for it. They saw all thinkers and engineers involved in combinatorics and building calculating or moving machines as early cyberneticists, from Ramon Llull and Jaquet-Droz to Pascal and Leibniz (Wiener presented the latter as the “patron saint of cybernetics”). Even cabalist mystics ←17 | 18→with their search for Golem were listed in that context.¹⁷Mathematical roots of cybernetics were largely impacted by early game theory and von Neumann’s theory of automata,¹⁸ Turing’s works on computability and the probability theory, which was being developed at the time by thinkers such as Andrey Kolmogorov and Ronald Fisher (all these names come up both in Wiener’s and Lem’s texts).

It was soon observed that

It was another step toward conceptually placing humans and machines on a par. A theory of “learning machines” started being developed, together with building such machines, initially quite primitive, and then increasingly complex.

In 1948 William Ross Ashby made the first Homeostat – “a physical model imitating the phenomenon of homeostasis [i.e. physiological balance in a variable environment] and the self-organizing capacities of the brain.”²⁰The Homeostat was in fact the first practical success of cybernetics. In the 1950s and 1960s cybernetics developed swiftly and had its big entry into such disciplines as biology, economy, technical sciences (including telecommunication), sociology, political science and other.²¹ The marriage of cybernetics and biology gave rise to a discipline sometimes called bionics (usually biocybernetics) – and this was when for the first time there were publications on systems that combine biological and mechanical components, based on thorough research on the functioning of human nervous system.²² I emphasize that so much, because such ←18 | 19→systems (cyborgs) will be one of the main topics of Part Three of this book. For some time it seemed like creating a structure that would combine features of a biological organism and a machine is close. Research in economical cybernetics looked promising. New subdisciplines were formed too, such as socio- and psychocybernetics and military, medical, pedagogical and linguistic cybernetics (the latter producing the first attempts at machine translations). Researching all types of steering processes, scholars focused on problems such as the impact of steering signals and feedback on the quality and stability of control, the impact of the structure of the systems on their reliability and the resistance of steering systems to interference. It needs to be emphasized, given the liberties with terminology taken by later epigones of cybernetics, that all these notions originally had precise mathematics determinants, formed on the basis of advanced fields of the science. In the 1970s it was further enriched by linking cybernetics to the general system theory,²³ which made it possible to research complex steering systems, among other things.

While creating cybernetics, Norbert Wiener saw it not only as a new, revealing discipline of science but also as a remedy to the increasing atomization of sciences²⁴ and as a major tool shaping social life.²⁵ Very soon, however, in the 1960s it became clear that neither of these “metascientific” goals of cybernetics is or ←19 | 20→can be achieved. Instead of quickly becoming a mathesis universalis, it started dividing into subdisciplines, which were losing connection with one another. The attempts to apply cybernetics to social sciences, which were in fact undertaken against Wiener’s will,²⁶ soon failed, as they turned cybernetic terminology from a precise vocabulary into a set of blurry metaphors with no heuristic value (I shall provide examples of that later). The purely technical fields of cybernetics, such as the theory of automata, of adaptive control systems and of optimal and hierarchical control, as well as the more specialized biocybernetical research, met the same fate as all other subdisciplines: this atomization and formal sophistication have made them completely inaccessible for those who specialize in slightly other fields (not to mention amateurs). What happened was exactly what Wiener was trying to save the science from.

There are innumerable texts about cybernetics. Globally there are hundreds of monographs and dozens of thousands of articles. It is impossible to pin down the moment when all this production got relegated to the margins of real science, because naturally the cyberneticists themselves have never admitted it had happened. It can be said that the 1970s brought the final fading of classic cybernetics, even though it is also the moment when Heinz von Foerster announced the end of “first-order cybernetics” and the beginning of “second-order cybernetics” in a work titled Cybernetics of Cybernetics. He defined the former as cybernetics of observed systems, while the latter as cybernetics of observing systems (which means the discipline has not avoided the self-referentiality, which became overwhelming in social sciences and the humanities at the time). This “second-order cybernetics” is now represented by sociocybernetics, which investigates the so-called autopoietic – or self-reproducing – systems.²⁷The ←20 | 21→highly abstract character of these inquiries situates them beyond the main scope of sociology and social sciences, although such theories did have considerable impact on, for instance, biology of ecosystems for a while (there existed a branch called cybernetic ecology).

There still exist professional associations such as the American Society for Cybernetics (www.asc-cybernetics.org, the website includes numerous links to other sites of similar character), as well as journals, such as the monthly Biological Cybernetics.²⁸ Today’s cybernetics is largely related to contemporary antireductionist theories, such as constructivism. The term includes attempts undertaken mostly by German scholars to encompass the entire human mental activity in one general theory, centered on the notion of “construction” (construction of reality in human cognitive apparatus) and employing the achievements of contemporary epistemology, system theory and system biology.

None of this means that cybernetics has not contributed anything to the mainstream world science after the period when it was one of the constituting disciplines. Fields such as IT, robotics, artificial intelligence (AI) (which cyberneticists wrote about as early as in the 1950s), the theory of automata, organization theory, telecommunication and system engineering also owe a lot to cybernetics. Economic cybernetics contributed to the development of management theory (including managing “human resources”), optimizing theory and decision theory. The specialists in neural networks, which were the thing of the time in the 1980s and 1990s, are especially indebted to cybernetics. The problem of complexity, which was in fashion at the time, investigated by both physicists (such as Stuart Kaufmann) and biologists (such as Ilya Prigogine), has a lot in common with system theory combined with an indeterminist philosophical orientation.

A detailed investigation of the growth of cybernetics in specific countries would be very time consuming. Nevertheless, it is important to glance at what happened with it in Poland, which is, I believe, a good sample, illustrating in detail the process of degeneration, which I have outlined earlier.