Introduction to Design Cybernetics

Abstract: Cybernetics emerged in the middle of the 20th century on the basis of technological and scientific progress during the Second World War, and it influenced design theory and research. Cybernetics was initially conceived as a theoretical framework, a common language across disciplinary boundaries, but people soon found more prominent applications in goal-oriented control engineering. Since 1970, it has developed a reflective, more philosophical, and less control-oriented view called second-order cybernetics . This view acknowledges circular causality, indeterminism, subjective observers, and other concepts avoided by the natural sciences. In this way, it provides a way for self-organizing systems that negotiate their own goals in an open-ended process—in other words, design. As an introduction to design cybernetics , this chapter outlines the evolution of cybernetics from a technical engineering discipline to a design philosophy perspective.

1.1 Introduction

Cybernetics has come a long way since it was widely associated with the mechanical feedback loops that engineers build into thermostats, guided missiles and other control systems. Over the past half century, cybernetics has gone far beyond the forms of technical systems and computation, and that's why it's been stereotyped. Today, cybernetics also includes fields more closely related to humans: biology, management, social sciences, anthropology, education, therapy and, more recently, design. Cybernetics offers an abstract philosophical approach to design as a creative epistemological practice. Unlike other disciplines, design is not just an applied field for cybernetics, it is described as cybernetics in practice. The purpose of this chapter is to provide a background analysis of design cybernetics by showing how a field originally known as control engineering has become a way to explore curious, creative, and ethical human activity. Our discussion of this development begins during World War II and discusses historical similarities and philosophical connections between cybernetics and design studies.

1.2 World War II and the rise of Systems Traditions

Long before the outbreak of World War II, Norbert Wiener , known as the " father of cybernetics ," had breakthrough insights. Norbert Wiener realized that the " strong " currents of the power system and " that were transforming the sounds and sights of the Roaring Twenties " There is a clear distinction between " weak " currents. He found that in communication and control systems, currents (and radio waves) can be weakened arbitrarily, as long as they function as signals. After World War II broke out in Europe, Wiener and his collaborators Arturo Rosenblueth and Julian Bigelow had another insight: from ancient Greece From the outset, Western thought has prohibited circular causality because it has the potential to generate paradoxical conditions that traditional logic cannot resolve. However, the team realized that certain systems, such as the metabolism of higher organisms, " purposefully " respond to the effects of their own behavior in a circular causal fashion. Circular causality, while still posing a logical conundrum, can no longer be ignored in practice. Through these insights, Wiener captured the key knowledge and technological changes of his era. More than any other military conflict of the past, World War II was determined by insight into signals and causality, so it was not just a war of power and force, but communication and control war. The rise of the Nazi regime was largely due to new means of communication, albeit in a linear relationship. By installing inexpensive radio receivers in German homes, creating a one-way network of propaganda, the regime largely replaced social mutualism with a hierarchical, authoritarian structure.

Wiener tried to contribute to the war in the field of defense technology. Together with Bigelow , he sought to develop an anti-aircraft gun system that could predict the trajectory of an attacking bomber and aim its projectiles " ahead of " at the aircraft, before the fired projectiles reach their altitude hit the target. Their system showed a " positively uncanny " ability to predict the trajectory of an aircraft within seconds. However, over the required extended period, the system did not provide any operational advantage over existing methods and was never applied in combat.

At the same time, the Germans integrated internal guidance systems into aerial weapons such as the Fritz X and V-1 flying bombs. Using a gyroscope as an attitude sensor, this method was used decades ago in the Whitehead Torpedo . It is based on circular causal self-regulation and what cybernetics call "negative feedback ": once a goal has been set, and a path to that goal has been mapped out, there is a difference between motor output and sensory input. A self-correcting cyclic signal structure called a "feedback loop " is established in between to regulate its continuous motion along this path. By minimizing the deviation of the actual path from the drawn path, the so-called " error ," such a system can ensure that it will eventually reach its set goal.

As the V-1 flying bombs crossed the English Channel, they encountered defenses with negative feedback loops - automatic radar tracking stations and projectiles containing proximity fuses : an unprecedented autonomous weapon Conflict, with almost no human interference. When the V-1's successor, the faster V-2 rocket, was used to attack London, the British managed to apply error-amplifying " positive feedback" to the negative feedback control of the missile's target pursuit, allowing the double The spies reported the wrong impact point to the Germans, causing subsequent missiles to be aimed away from the populated target area, which is said to have saved many lives.

Other seminal developments of World War II occurred in the field of cryptography and cryptanalysis, with key contributions from Alan Turing in the United Kingdom and Claude Shannon in the United States. Shannon, who also worked on the aforementioned air defense challenges, developed his signal analysis work into what is known as the Mathematical Theory of Communication (sometimes referred to as Communication Theory or Information Theory ). Theory ] ). In this work, Shannon was mentored by Wiener and benefited in particular from the statistical methods Wiener developed for the air defense program.

The rise of authoritarianism in Germany, Italy and Japan prompted the establishment of the Committee for National Morale in the United States. The committee advises the president on advocacy and public morale, and develops strategies to prevent the re-emergence of authoritarianism. It explains the psychology of National Socialist in terms of the separation of emotion and reason, where emotion is magnified and reason is suppressed by mass media propaganda. This perspective will soon inspire a new medium that allows individuals the freedom to make their own, emotional and rational choices. Members of a democratic society should not be bound from above, but should be free and guided by internal values. As a result, governments find themselves confronted with the paradoxical challenge of what is mandated from the outside and must be produced from within.

After the war, the correlation of control and circular causal feedback systems did not subside. When a group of academics got together to discuss brain inhibition, Bigelow got a lot of excitement by presenting his work with Rosenblueth and Wiener on purposeful circular causal systems. This prompted a series of conferences on the subject between 1946 and 1953, with a caucus and a changing group of invited guests from various disciplines. This series of conferences, sponsored by the Josiah Macy, Jr. Foundation , is often referred to as the Macy Conferences . In addition to Wiener and his two collaborators, core participants included neurophysiologists Warren McCulloch and Walter Pitts , mathematician John von Neumann von Neumann ), physicist Heinz von Foerster , and anthropologists and National Morale Council alumni Margaret Mead and Gregory Bett Sen ( Gregory Bateson ) . Participants in the Massey conference overcame the jargon of their different subject areas to develop a common new language to explore their shared interest in circular causal feedback systems. In this way, the Massey Conference became the cradle of what is today called interdisciplinary and interdisciplinary.

Margaret Mead practiced circular causality as early as the 1920s while conducting fieldwork in the South Pacific. Mead rejected the scientific necessity of linear causation, " objectively " eliminated observation, and was probably the first American to conduct anthropological research as a participating observer. Unlike Mead, Wiener rejected the notion of the objective observer in an article published during his tenure as a visiting professor at Tsinghua University from 1935 to 1936 ( The role of the observer ) and described observation as active participation .

At a Macy's conference, the invited British cybernetician W. Ross Ashby presented his " Homeostat " - a system of four interconnected boxes Structure - when the role of the observer becomes a matter of debate. Each box has a movable indicator. When the experimenter moves one or more of these indicators away from their neutral position, the indicators on the remaining boxes will move in opposite proportions, maintaining a stable overall average that maintains glucose levels and body temperature with the organism in a similar manner. Ashby describes "homeostasis" in terms of the relationship between organisms and the environment ( Homeostat ). Ashby was frustrated when Bigelow and other attendees asked where the boundaries between organisms and the environment lie in " homeostat " because he didn't set anything between the four devices and the experimenters. specific boundaries. Two years later, he wrote:

When an organism and its environment are viewed as a single system, the dividing line between "organism" and "environment" becomes somewhat conceptual and arbitrary. ( As the organism and its environment are to be treated as a single system, the dividing line between “organism” and “environment” becomes partly conceptual, and to that extent arbitrary. )

On this point, Ashby agrees with Mead and Wiener on observer dependence. He also anticipates a shift in the way the word " system " is understood, which is at the heart of cybernetics and its broader family of systems traditions. The etymology of the term " system " points to " putting together ," which is conceptually juxtaposed with the reductionist tendencies of science to " take things apart ." After early systems theorists described " systems " in objective terms as " sets of elements standing in interrelation", cyberneticians would eventually A distinction is made when what is relevant in the observations. According to this recent view, the boundaries of the system are projected and negotiable through observational behavior, rather than observed properties.

Wiener published a seminal book called Cybernetics in 1948. Massey conference attendees adopted the book " Cybernetics " as the name of their field, and its subtitle " control and communication in the animal and the machine " became cybernetics One prominent definition of , and many others have been proposed since. The word " cybernetics " (cybernetics) is derived from the Greek adjective "κυβερνητικός", which describes the ability to manipulate, by Wiener, which is also the root of the word " government ". Despite the book's success and widespread dissemination, its suitability as a basis for a new field was ultimately questioned. Glanville argues that Wiener's publication of Cybernetics and then the more philosophical The Human Use of Human Beings was a "huge" massive tactical error . The technical and mathematical nature of Cybernetics makes cybernetics widely regarded as a technical engineering discipline. If the more philosophical The Human Use of Human Beings came out first, or, as Glanville speculates elsewhere, if Gregory Bateson Bateson ), rather than Norbert Wiener , who wrote the first book on cybernetics, then the field may today be a new field that may be more properly understood as a A realm related to living people, not a field only related to technology.

In the UK, Ashby helped formalize " control " by introducing " variety " as a measure of the number of states a system might undertake. Ashby wrote in his personal journal at the time:

I want to get rid of the Shannon method of entropy and the infinitely long information averaging method; I want something that I can calculate. ( I want to get away from the Shannon method of entropies [and] averaging over infinitely long messages; I want something I can count. )

Ashby explained that traffic lights have red, yellow, and green lights, each of which has two states, on and off , which together can have eight states. In its traffic control application, however, only actual variants of the four states are used—for example, no red and green combinations are used. The difference between the potential state and the actual state is the constraint . Ashby also recognized that reliable control requires certain conditions to be satisfied in the feedback loop, which is reflected in his " Law of Requisite Variety " : the variety of controllers must be The variety of the controller must be equal or greater than the variety of the controlled .

As the world split into the " Eastern bloc " and the "Western bloc " and plunged into the Cold War, cybernetics was the tone again. Conflict is now as much an academic challenge as it is a political and military one. Academia, most notably cybernetics expert John von Neumann , developed "game theory" —a mathematical method for simulating competition among rational decision makers. Conflict, whether in the form of an arms race or full-scale violence, is now understood as positive feedback that, if left unchecked, has the potential to escalate catastrophically. In past wars, nuclear weapons and rocket technology were developed to ensure "mutual assured destruction" on both sides of the Iron Curtain. Both the Soviet Union and the United States occupied most of the resources responsible for the German rocket development research center during World War II. Now, the space race has emerged as a symbolic battleground where competition will amplify innovation rather than outright violence. Early in the race, the Soviet Union made rapid progress and launched the first Earth satellite, sending the West into a "Sputnik Shock " .

The beeping radio signal from Sputnik and its visibility at night have Americans questioning their superiority in science, technology and education. What people see as the "missile gap " and the anxiety it sparks has sparked a wave of profound reflection that has spread across the United States. After several years of "space age " optimism, American scientists and engineers have lost faith in the country's innovative edge, and policymakers and the public have begun to question the standards and methods of the American education system. The United States has fallen behind in a competition because both sides have made scientific development and technological progress for the benefit of their own people as core principles. This allows them to agree on how to disagree and provides common conditions and standards against which to measure and compare national achievements. With the United States lagging behind the achievements of the Soviet Union, President Kennedy in 1961 pledged to send humans to the moon and back within a decade. In doing so, he not only defined the race as a goal-oriented governance challenge, using a systems engineering approach, to solve it from the perspective of path goal management theory; he also drew a finish line, which is What has so far been missed in the space race. The location of this finish line, and the technical challenges it implies, redefine the race as a long-term developmental challenge skewed toward American superiority. In fact, President Kennedy has addressed "the difficulty facing every systems designer [which] is in determining the overall system specification, or ' statement of objectives' ).

The systems engineering approach (the systems approach for short) was pioneered in early ballistic missile development and was adopted by NASA's entire lunar program. As a distant relative of cybernetics in the family of systems traditions, systems engineering adheres to scientific reductionism and linear causal utilitarianism. The technical focus of systems engineering, on pre-specification, rationalization and optimization, is inherently compatible not only with the modularity of space exploration systems, such as multi-stage rockets and space station components, but also with the hierarchical organizational management structures responsible for their production and operations. For example, allowing spacecraft to be divided and subdivided into systems and subsystems such as propulsion, communications, navigation and guidance, life support, etc., this approach is based on the assumption that once each subsystem When the subsystems are integrated into the whole, the overall goal is met.

1.3 Limitations of Control, Instrumentalism, and Design Approaches

Soon after the end of World War II, technological innovation, organizational management, mission operations, and international conflict were all brought into the category of strategic governance. More and more people believe that cybernetics is not only a theory of describing control, but also a theory of purposefully exerting control. Ideological rifts have emerged around issues of ethics and instrumentalism. For example, while many cheered the coming age of industrial automation, it was Wiener who warned of the imminent demise of the average worker. The contrast between Norbert Wiener and John von Neumann , two brilliant mathematicians and founders of the new field of cybernetics, exemplifies this ethical disagreement. Wiener, who never participated in the Manhattan Project, opposed the secrecy of war research and disliked the use of nuclear weapons on civilians, while von Neumann played a leading role in the Manhattan Project, choosing Hiroshima and Nagasaki as nuclear targets Committee member and advocates a nuclear first strike policy against the Soviet Union.

In the postwar period, it was believed that a variety of challenges could be purposefully managed through a scientific and systematic approach. ' Sputnik Shock ' stimulated the development of 'creativity techniques ', and soon systematic methodologies and management were applied to innovation and design. An early cybernetic preamble on creativity emerged when Wiener was invited to write a book on the philosophy of invention. In a manuscript he wrote in response (but gave up writing for other projects in 1954), Wiener explained that "truly fundamental and seminal ideas are largely fortunate and unpredictable Accident" ( the really fundamental and seminal idea is to a large extent a lucky and unpredictable accident ). He rejects the notion that invention is a rational decision:

The most critical stage of invention […] is the change in the environment of thought, which produces and is produced by a new thought. This may have untold value to society, but by the very nature of things, it is not actuarial. ( The most critical stage of invention [. . . ] is the change in intellectual climate which produces and is produced by a new idea. This may be of untold value to the community, but in the essence of things it is not subject to actuarial work. )

Wiener also likens new ideas that arise by chance to lightning. He argues that because of their contingency, conditions favorable and unfavorable to lightning and new ideas can be understood and exploited to promote or suppress them. In this view, inventions can be cultivated, but not controlled or decisively triggered. However, other precursors to design cybernetics today can be found in Wiener's work. He defines feedback as " the property of being able to adjust future conduct by past performance " , foreshadowing Simon's later description of the designer that "designing who devises courses of action aimed at changing existing situations into preferred ones , and Rittel and Webber 's approach to problems (i.e. design challenges) description of " as discrepancies between the state of affairs as it is and the state as it ought to be " .

Wiener's work also heralds the future of design-cybernetic in a more metaphorical and philosophical way. The wartime air defense system he developed with Bigelow integrated the two causal relationships described by Aristotle: causa efficiens (interpretative description: "because...") and causa finalis (Control: "For...") . These two causal relationships correspond to the descriptive agenda of the natural sciences on the one hand, and the instructive, interventional agenda of engineering and design on the other. Simon explains: " The natural sciences are concerned with how things are ... Design, on the other hand, is concerned with how things should go, by designing artifacts to achieve their goals." [. . . ] Design, on the other hand, is concerned with how things ought to be, with devising artefacts to attain goals )

If there was ever a defining moment when the cybernetic spark jumped into the realm of design, it was when design theory adopted Ashby 's concepts of variety and constraint. Oddly enough, this moment happened twice, and largely independently, when Gordon Pask in the UK and Horst Rittel in Germany both from Ash Bee finds inspiration in his work. Ritter is quick to describe the design process as " the generation of variety, and the reduction of variety " . These two operations are often referred to today simply as the " diverging " and " converging " phases of the design, as illustrated by design process models such as the double diamond model .

In the 1950s and 1960s there was a desire to " scientise " design. Buckminster Fuller announced a " World Design Science Decade " starting in 1965. The first Conference on Design Methods , held in London in 1962, launched the Design Methods movement, a decade-long academic attempt to rationalize and scientific design. The following year, Horst Rittel moved from the Ulm School of Design to UC Berkeley, where he became a leading advocate of the Design Methods movement. He later recalled:

In the beginning, outsiders from architecture, engineering and business heard about this systems approach and thought if it could handle something as complex as a NASA project, then why can't we handle something as simple as a house in the same way? ? Shouldn't we think of every building as a task-oriented design object? ( [I]n the beginning, outsiders from architecture, engineering, and business heard about the methods of the systems approach and thought that if it were possible to deal with such complicated things as the NASA programmes then why couldn't we deal with a simple thing like a house in the same way? Shouldn't we actually look at every building as a mission-oriented design object? )

Within a decade, however, the design method movement has come to its end and faced opposition from many, including some of its early proponents, who now recognize that prescriptive methodologies are antithetical to design philosophies. Until now, they considered prescriptive methodologies as antithetical to the idea of design. Jones , one of the early supporters and later opponents of the movement, explained:

Methodology should not be a fixed track to a fixed destination, but a conversation about everything that could happen. The language of dialogue must bridge the logical gap between the past and the future, but in doing so, it should not limit the variety of futures discussed or force people to choose an unfree future. ( Methodology should not be a fixed track to a fixed destination, but a conversation about everything that could be made to happen. The language of the conversation must bridge the logical gap between past and future, but in doing so it should not limit the variety of possible futures that are discussed nor should it force the choice of a future that is unfree. )

With appropriate methods and techniques, the notion that systems can be reliably predicted and controlled has also been abandoned in other fields, notably in ecology. As the design methods movement failed and was abandoned, design research took on a less instructive, more reflective stance that Ritter called " design methods of the second generation " . Perhaps not coincidentally, this transition coincided with the expansion from first-order cybernetics to second-order cybernetics, which is discussed in the next section.

1.4 First-Order to Second-Order Cybernetics

Although considered a decisive, instrumental science of control technology during the Cold War, we believe that the initiators of this field had different motivations. Margaret Mead 's work as a participatory researcher and Wiener's early recognition of the role of the observer, circular causality and non-determinability suggest that cybernetics At the time of its origins, it was more reflective than the general cognition and application of cybernetics in the mid-20th century would suggest. To illustrate this further, let's look at Figure 1.1, which shows Norbert Wiener and Palomilla , the robot he developed at MIT in the late 1940s.

Equipped with sensors, circuits and motors, Palomilla " purposefully " navigates space in relation to light sources like insects. If you're not familiar with Wiener's biography, you can see in Figure 1.1 that a Massachusetts Institute of Technology ( MIT ) mathematician is working on automating vehicles that could one day be applied to robotic vacuum cleaners, autonomous driving Cars and drones for unmanned battlefields. While it is true that some of these systems have their origins in Wiener's work, we propose a different reading of this image that we believe is more in line with Wiener's broader work and the spirit of design cybernetics. We argue that Wiener's interest in Palomilla was less in instrumental utility and more in metaphor for his own cognitive navigation, and in his 1936 essay on the role of the observer , he This effect is described as follows:

Practical mathematicians know all too well that mathematics as a living study is inductive and experimental, no matter what it looks like in a textbook. When I needed a helper function to do a definite job, I tried one by one and found that the first function was too big here and the second function was too small there, until through luck and a habit of sorts Familiar, I found a perfectly suitable function. Nine out of ten possibilities are ruled out based on a general sense of the situation before it gets to any real reasoning logic. The tenth suggestion goes some way to convince an old hand that there is something in it - it solves the difficulty at the right point, but doesn't make it easy to suspect that it's an outright mistake. Once the key is inserted into the lock and the bolt starts to show signs of turning, a perfect fit is all that's needed with a file and oil. ( The practicing mathematician knows very well that mathematics as a living investigation is inductive and experimental, whatever it may be when stuffed and mounted in text-books. When I want an auxiliary function to do a definite job, I try one after another, finding the first too big here, the second too small there, until by grace of luck and a familiarity with the habits of the species, I come on an exact fit. Nine-tenths of the possibilities are eliminated on the basis of a general feeling for the situation before it comes to a matter of any real deductive logic whatever. The tenth suggestion slips into place in a way which convinces an old hand that there is something in it – it resolves the difficulties at just the right points, but not so readily as to excite suspicions of a sheer blunder. Once the key will go into the lock, and the bolt begins to show signs of turning, it is a matter of mere filework and oil to get a perfect fit. )

Next to this passage, consider another photo of Palomilla , shown in Figure 1.2. It shows a long exposure photo of Palomilla navigating through space. The robot's glowing vacuum tubes follow a zigzag, forward-searching route, like a track in the sand, showing its path and operational logic. We believe that the optical path in Figure 1.2 has a lot to do with the motion of Palomilla in space, as Wiener described above. They are trajectories left behind by ephemeral processes, recorded for others to read, to connect with forward-looking exploration, and to engage in better metaphor and understanding for thinking about how we risk and explore the unknown. In this way, Palomilla was not only a precursor to today's Roomba vacuum cleaner, but also for understanding Roomba vacuum cleaners, and how other things and processes came into being - a " machine for showing ", cognitive reflexes, and design cybernetics pioneer.

As a "practising mathematician ", Wiener remained largely committed to the representationalist paradigm until his death in 1964. He did not live to see the self-reflection he alludes to emerge as an explicit basis for his discipline. Mainly Heinz von Foerster proposed a behavioral cybernetics that applies to itself and whose content matches its form: second-order cybernetics . Later, when asked how second-order cybernetics " came upon " him, von Forster credits Margaret Mead and her 1968 American Control Talks at the Academy of Sciences ( ASC ). In this talk, which Mead retrospectively named " Cybernetics of Cybernetics ," she asked ASC to apply cybernetic insights and techniques to its own organizations and operations.

After recognizing the role of observers and circular causality, Von Foerster explained that any description or theory must explain both observers and describers , as well as their describing ) and theorising . For those who accept it, it is a research attitude, an ethical position guided by an inherent responsibility, and ultimately an aesthetic desire. Von Foerster 's concept of second-order cybernetics ethics is based on the subjective responsibility of the self, which is defined by his/her system boundaries. Based on the view that "...freedom always exists. At every moment, I can decide who I am" ( . . . freedom always exists. At each and every moment, I can decide who I am ). Von Foerster explained that the choice of action is determined within and, therefore, the responsibility is within. Another option is extrinsically motivated action, which underlies absolutism and the rejection of personal responsibility observed in the Nuremberg trials : "I have no choice. I just obey orders!" ( I had no choice . I was merely following orders! ). Thus, Von Foerster distinguishes between ethics and morals . In this view, (linearly) directing what others think and do ("though should...", "though should not...") constitute morals , and ethics (circular) are directed against oneself ("I should...", "I shouldn't..."). Therefore, ethics cannot be made explicit, but manifested in action. To help promote necessary liberties, Von Foerster offers his Constructivist Ethical Imperative : " I shall act always so as to increase the total number of choices" to increase the total number of choices ).

For most of the 1960s, cybernetics, and in particular Von Foerster 's Biological Computer Laboratory (BCL ), benefited from the Department of Defense (DoD ) ) research funding for the development of computer technology. The Mansfield Amendment of the Defense Procurement Authorization Act of 1970 restricts the Department of Defense " with a direct and apparent relationship to a particular military function or operation" This changed when support for basic research on a specific military function or operation , ethically-oriented cybernetic experts, first of all Von Foerster himself, was not prepared to provide this support. Others, particularly in the field of artificial intelligence, responded with bold promises—often based on technological concepts that originated in cybernetics—to achieve applicability of research findings on the battlefield, and thus received generous support. Although this Pentagon policy was later loosened, its reallocation of funding strengthened AI research at MIT, Stanford and elsewhere, and is seen as leading to the closure of the Biocomputer Lab in 1974 and the 1976 Reasons for Von Foster's retirement.

From first-order cybernetics to second-order cybernetics, it is not so much a transition as an extension. Cybernetics, as a control engineering discipline, can be understood as a constrained subset of cybernetics, as a broader and more general spirit, just as Newtonian mechanics is in Einsteinian mechanics. In other systems-oriented research traditions, this broader, newer form of cybernetics is a far cry from the assumptions on which traditional empirical science is based, acknowledging " holism ", context, relational, circular causality, non- Certainty, subjective observers, and self-organization.

To illustrate the nondeterminism and observer dependence of cyclic causality, Von Foerster introduced two automata, the trivial machine ( TM ) and the non-trivial machine ( non-trivial machine ) , NTM ). Both machines are thought experiments, not proposals for technical implementations. They both have one input and one output channel, but the internal mechanism for converting the input to the output is different. TMs predictably transform inputs into corresponding outputs, so that an external observer, after a period of observation, can establish a clear relationship between possible inputs and resulting outputs, such as the “distribution” shown on the left of Figure 1.3. assignment table . A complete assignment table is a reliable model that can predict the output response of a TM to a given input, no matter how long the machine has been running. In contrast, NTMs contain methods for memorizing machine states (labeled z on the right side of Figure 1.3). This state is not only affected by each input-output transition; it also collectively determines the output of subsequent transitions.

This results in a large number of ever-changing input-output mappings. Arguably, the NTM's operational history leaves traces in the machine, which effectively turns it into a different machine, with transitions for every input and output. Von Foerster decides on the challenges of these two machines from the perspective of an outside observer who, without knowing their inner workings, must construct a mental model of their inner workings. ) - to " whiten " a "black box ", in Glanville 's words. This is easy for TMs and almost impossible for NTMs. This inability to comprehend and predict the observed system is satisfying and enjoyable because it is where magic and wonder come from. Von Foerster used the juxtaposition of his two machines to distinguish between trivial input-output systems and non-trivial input-output systems. Nontrivial systems, including humans, are equipped with memory and recurrent pathways through which the outputs of earlier operations can be re-entered as inputs to subsequent operations, affecting themselves through their interactions in unpredictable ways. The use of mechanisms to question cultural preoccupations with mechanical causality was, and still is, a cunning rhetorical device. However, to avoid misunderstanding, we must emphasize that mechanistic cybernetic metaphors of biological and social systems are nothing but metaphors. Comparing the NTM to the human mind does not imply that the mind is like a mechanism, or that the mechanism works like the human mind. This analogy is only meant to show that with knowledge of cyclic causal reentry and memory comes awareness of simple mechanisms and the uncertainty that humans encounter.

The background to von Foerster 's reference to the distinction between triviality and non-triviality is his criticism of educational institutions that treat children as trivial systems ( trivial systems ), train them to come up with reliable known answers to old questions. In this case, he also used another example of non-triviality : a schoolchild who answered the question "What is 2 times 2?" answered "Green!", so she Will be reprimanded and "trivialised" until she gives the expected answer of "4". The child's spontaneous expression of novelty, the transcendence of expected variety, captures what Wiener describes as a lightning-quick moment of creativity. While the principles at work at such moments are not explained in Von Foerster 's description of NTMs, they are explained in Conversation Theory , which is outlined in the next section.

1.5 Dialogue and Design

Jones is not the only design researcher to describe the design process as a " conversation ". Design researchers, both inside and outside design cybernetics, later recognized the cyclic structure of the design process, which they described as a " conspiracy " of " argumentative ", characterized by the " symmetry of ignorance" ignorance ), is "dialectical", is " discursive " , is " dialogue " or " dialogical ", is " negotiation ".

The cyclic structure of the design process must be divorced from the linear structure of Western logic and Shannon's Communication Theory . Gordon Pask 's Conversation Theory provides such a structure. Conversation Theory explains cognitive processes, the processes we come to understand through learning, design, and research. It is a radical constructivist theory based on circular exchanges . It does not regard " knowledge " as a storable and transferable commodity, but regards cognition and the process of cognition as a process of subjective execution. Pask's work was sometimes considered incomprehensible, but was further developed by his pupils Ranulph Glanville , Paul Pangaro and Scott , making it easier to understand.

Unlike Shannon's symbols, which travel from sender to receiver through a linear channel disturbed by noise, Conversation Theory describes a circular causal relationship between two or more conversants. Out of a desire to explain more with less (Occam's razor), Conversation Theory typically interprets and explains with only two dialogues: a subjective self ( self ) and an other ( other ). These two roles may occur in one person, who may be in dialogue with another imaginary person, or in groups of more than two, multiple individuals may play the role of one person. A key challenge in building this model of interpersonal communication is that meaning is private. Shannon's Communication Theory recognizes this challenge and explicitly excludes meaning from its concerns. Conversation Theory addresses this problem by describing a process in which interlocutors pursue mutual agreement of ideas through timely comparisons and rephrasing . This process continues until the interlocutors feel that their respective understandings are close enough for further dialogue, as if their meanings were shared, as if they were dealing with the same problem. When they do not reach such a consensus, they must seek common ground while reserving differences. This process is shown in Figure 1.4.

Many everyday conversations take place between individuals “ getting on the same page ” to align and synchronize thoughts and understanding by reducing the “ error ” of negative feedback. However, some conversations also widen differences in understanding, challenge previous perceptions, expand the diversity of other perceptions, and use “mistakes” to inspire new ideas through positive feedback. These two dialogue modes reflect how the design process " converges " and " diverges ", and how designers can both reliably meet expectations (such as in schedules, budgets, and regulations) and surprise them challenge expectations (e.g., by inventing, speculating, and challenging). A Design Self may have conversations with a wide variety of people. As shown in Figure 1.5, the other can be a person, an imaginary person, a physical model, a pen and a piece of sketch paper, or a technology.

What qualifies this contact as a dialogue is the readiness of the ego not only to influence the other, but to be influenced by the other in a circular causal fashion. For example, a designer might make a mark on a sketch paper (affecting another), then, perhaps looking at the sketch from the side, see something that wasn't meant to be expressed, and take that idea into account in the creative process (subject to another person’s influence). Fantini and Ranulph Glanville point to the importance of this openness by emphasizing the role of " listening " (which applies metaphorically to all modes of perception). Along a similar line, Von Foerster put forward his " Hermeneutic Principle ": "It's the listener, not the speaker, that determines the meaning of words" ( It's the listener, not the speaker) the speaker, who determines the meaning of an utterance ) . As a complement to this principle, he also made his Aesthetical Imperative : " If you desire to see, learn how to act " . As the Latin root of the word conversation suggests ( conversare = turn together, i.e. dance), from this perspective design is a feedback loop of influencing and being influenced, of speaking and listening, or more generally Speaking, is a feedback loop of action and understanding, with negotiable goals. We believe that this feature is a sufficient definition of design.

Scott believes that Conversation Theory is " pioneering achievement in modelling cognition as an evolutionary, self-organising process " . This description highlights a key characteristic of the conversation: it is a timely process. Dialogue thus contrasts with some of the principles of the West, especially the reasoning of the natural sciences. Feedback and dialogue run counter to formal logic, by which we evaluate declarative sentences and draw conclusions as "true" or "false." For thousands of years, Western logicians have shunned circular causality to avoid the paradoxical conditions they might create. Statements like " This is a lie. " are forbidden in traditional formal reasoning. This is because formal logic is temporal, since a statement cannot be both true and false according to the principle of eliminating intermediaries. In contrast, cybernetics acknowledges temporal processes. For example, a thermostatic heater can alternate in time on and off. What appears to be a paradox from the point of view of formal logic, is an immediate oscillation from the point of view of cybernetics. The temporal structure recognized by cybernetics can further generate desirable dynamics such as persistent self-stabilization (observed in technologically controlled systems) and spontaneous novelty (observed in dialogue). The purpose of many cybernetic feedback loops is precisely not to " conclude ", but to move on. This is one of the obstacles to doing justice to the behavioral nature of design in the rational language of academic research.

" Conversational cycles " unfold " out of control " in the manner of everyday conversation, developing in unpredictable directions, leading to unexpected ideas and new ideas. The necessary diversity is neither required nor aimed at, the conceptual grasp (diversity) of the two sides of the dialogue is different, and the dialogue itself interactively generates new diversity at some times and reduces it at other times, allowing familiar People do things they didn't know before (at least subjectively), instead of stimulating them to do known goals. While technical control systems are limited by the variety of controls available, dialogue is limitless. In conversation, diversity can (and often does) vary by interlocutor. It is itself mutable, influenced by the dialogue it shapes. Errors, differences, and misunderstandings between self and others are seen as possible sources of insight and inspiration, not necessarily corrected or avoided.

Even a digital computer, traditionally thought of as a logical machine, can precisely convert a given input into an output, and if processed in this way, become an interlocutor. This is because of the unrolling of the computational process on the timeline, allowing for cyclical human-computer interaction. Bateson observed that " The computer is only an arc of a larger circuit which always includes man and environment [...]" includes a man and an environment [. . . ] ). Glanville embodies this with a technique of digital surrealism: poets can close their eyes and make random input into a computer's text processor, which then flags the material as misspelled. The poet can then choose words and phrases she likes from the spelling checker's suggestions to interactively build the poem. What is achieved in this way is the interaction between the self and the other: " Betweenness is the source of interaction and is also its mode and its site ". In this design-cybernetics view, media is appreciated for its indeterminacy, while error, noise, and misunderstanding are valued for the imprecision we experience because "the imprecision we experience [...] may lead to [...] novelty" ( [t]he imprecision we experience [. . . ] may lead to [. . . ] novelty ." Glanville likens the relationship between this conversational design process and its results to a wheel that turns and the trail it leaves behind.

1.6 Summary: Adjusting for a New Perspective

Broadly speaking, cybernetics can be defined as the study of processes in which the states of events are adjusted with reference to the states of other events ( the study of processes in which states of affairs are adjusted with reference to other states of affairs ). In this chapter, we have illustrated how cybernetics has applied to itself since World War II, making some self-adjustments and extending it from a theory of technological control to an abstract philosophical approach to design as a creative cognitive practice.

One such adjustment is the awareness of cyclic causality (A affects B, B affects A), as it plays a role in self-regulating feedback mechanisms including thermostats and missile guidance systems. Another adjustment is the recognition that the object of observation cannot be separated from the act of observation, and the person who observes finds himself part of the subjective engagement of the object of interest. With this reorientation from an objective external perspective to an internal subjective perspective, the notion of objective reality is adjusted to that of subjectively constructed reality. Another adjustment is a shift from interest in static descriptions to dynamic processes. Cybernetics is concerned with adaptive processes, rather than producing descriptions in the form of conclusive statements, i.e. non-temporal statements expressed in terms of true and false, cybernetics is based on time. Another adjustment is from deterministic to non-deterministic. In simple mechanical systems, the outcome can be determined by previous causes. As more and more " moving parts " interact with each other in a circular causal fashion, the outcomes of actions are less likely to be predictable. A further adjustment is the realization that cybernetic principles are often recursively applied to different scales.

These adjustments constitute a departure from the ideals of natural science on the one hand, and an approximation of the designs appreciated by many designers on the other. In addition to these adjustments, there is continuity in the expansion of first-order to second-order cybernetics. Many early control engineering terms proved suitable for describing various aspects of out-of-control processes: feedback, diversity, necessary diversity, constraints, errors, and so on. We believe that cybernetics is a spirit for those closely associated with early cybernetics, as well as second-order cybernetics. In particular , the ethical ambition of design cybernetics is to increase the total number of options. The aesthetic goal of design cybernetics is pleasure, whether through magic and wonder, or through insight. Both goals apply to the design process, design results, and design studies. Design cybernetics is designed to foster rigor independent of the procedures employed and therefore does not provide a methodology. Essentially, it seeks to avoid restrictive control over others while pursuing greater choice and pleasure in the interactions of our individual and collective selves.

Compiled from: An Introduction to Design Cybernetics