Operational Dialectics: Gotthard Günther’s Cybernetic Systems

31 October 2023

Hegel did not write a philosophy of technology. However, there have been many attempts to find possible intersections between Hegelian dialectics and the philosophy of technology. Today, we should look back to Hegel’s ideas of self-reflection, the organism and causality when considering self-organizing systems. But this relationship has already been explored in the past, especially in the context of the history of cybernetics. Inspired by Hegelian dialectics, the cybernetician and philosopher Gotthard Günther, proposed to rethink the metaphysical assumptions on which bivalent logics—the logics according to which every sentence must either be true or false—are based upon. Günther’s questioning was inspired by his collaboration with the Biological Computer Laboratory (BCL) headed by cybernetician Heinz von Foerster. 

Günther’s reflections on the metaphysics and logics of cybernetic systems are related to the idea that self-organized systems are both effect and cause of themselves, an idea that can be found in Kant’s reflections on teleology and life in his Critique of Judgment, but also further developed in Hegel’s discussion of the concept of life in his Science of Logic. In this text, I will focus on Günthers critique of bivalent logics and how it is related to his thoughts about self-organizing systems, such as living systems. I think that considering these ideas today can be helpful in questioning the kind of assumptions underlying many contemporary debates about the causality of the living, the emergence of life, but also how we think about our knowledge of these phenomena when we study them with computational systems.

  1. Contextualizing Günther’s Work 

Gotthard Günther (1900-1984) was a philosopher who worked primarily in the field of logic. In 1933 he published a dissertation on Hegel with the title “Basic Features of a New Theory of Thinking in Hegel’s Logic.” These considerations on Hegel’s logic would accompany Günther into his later work, especially when he proposed the theory for which he became known: operational dialectics, i.e. the formalization of the logic of self-referential systems. In the 1950s, Günther emigrated to the United States, where he came into contact with some of the most important figures in the early days of cybernetics. For example, Günther was well acquainted with Warren McCulloch, the neurophysiologist and cyberneticist known for his contributions to the thinking of neural networks, which laid the groundwork for later developments in artificial intelligence. Günther’s collaboration with McCulloch is well known because he dedicated to McCulloch his three-volume work Contributions to an Operational Dialectics, which contained essays on topics ranging from a critique of Aristotelian logic to the metaphysics of cybernetics, published from 1937 to 1979. Günther’s relationship with McCulloch brought him into contact with the cybernetician Heinz von Foerster, and in 1961 he was invited to contribute to the interdisciplinary research of the Biological Computer Lab (BCL) in the Department of Electrical Engineering at the University of Illinois, where he worked until 1965. The essays in Contributions to an Operational Dialectics and his book The consciousness of the machine (1963) should therefore be read in the context of this collaboration.

The BCL focused on studying living systems in order to operationalize their structure and functions in computational systems. Although cybernetics is less recognized as a field today than it was in the 60s and 70s, we can recognize some of the contributions of the BCL in fields such as today’s network intelligence, artificial life, artificial networks, and systems theory. The BCL had a special focus on the analysis of the self-organization of living systems in the context of thermodynamics. The core of Günther’s critique of bivalent logic is the study of the logical structure of self-organizing systems, its formalization and operationalization. Günther’s research dealt with questions surrounding von Foerster’s notion of noise from information (a fundamental notion in the context of explaining the emergence of complex systems) and the question of the role of subjectivity in modeling self-organizing systems (a central question of second-order cybernetics). This background should be read when dealing with Günther’s metaphysics of the notion of information and his investment in the question of subjectivity in self-referential systems.

  1. Against Dualist Metaphysics 

What is the metaphysical status of information? How do reflexive systems—that is, systems that are the result of a process in which content and form coindice—come out of irreflexive states of being—that is, processes that are determined from outside themselves? Where does the organization of self-organized systems emerge from? These questions are central to Günther’s critique of classical metaphysics and, according to him, these are the metaphysical questions opened up by second order cybernetics. Inspired by the questions of the Biological Computer Laboratory, Günther sought to understand what it means for computers to be modeled on the forms of organization of the living. His thoughts on what information is led him to reconsider how we think about that which is not self-conscious, and what is the concept of matter we assume when we think about nature.

According to Günther, classical metaphysics assumes the existence of only two positions in the process of knowledge: that of the subject of knowledge and that of the object of which knowledge is produced. This epistemic dualism corresponds to a dualistic metaphysics, visible in dichotomies such as that between mind and matter, or more precisely, subject and object. Since information is neither fully material nor fully immaterial, the question arises whether this dualistic metaphysics is sufficient to explain it. Information requires the relationship between more than one system because it creates a feedback between at least two systems. Instead of a subject (of judgment) and an object (what the judgment is about), that could be brought together in correspondence with each other, information works as the mediation between two reflective systems. A reflective system is, however, not to be confused with what we would understand as the process of thinking in minded subjects, but it is rather the capacity for self-organization. Thus, in cybernetic systems we are not dealing with two metaphysical sides—the subject and the object—but with a more complex structure in which our thinking about the system must be taken into account. It is in this sense that Günther speaks of three “proto-components” of metaphysics, instead of two positions of knowledge: 

1) The transcendent object 

2) Information

3) Introscendent consciousness 

The three proto-elements are Günther’s explanation for how we refer to self-referential systems, that is, for the kind of feedback loops found in cybernetics. On the one hand, the notion of subjectivity names self-relationality and self-organizing systems, on the other hand, it names self-relationality as self-awareness, which is the one Günther names introscendent. The first one refers to self-referentiality as such, the second one to self-consciousness. This would imply that beings which are not self-aware are reflective in the sense that they produce and organize information. A living being is a reflective system, but not necessarily a self-conscious one. 

These two forms of subjectivity are Günther’s way of framing the question of the role of the modeler in the modeling process, a question at the heart of second-order cybernetics; or how we model self-organizing systems, that is, how computers modeled after self-organizing systems work. Reflective subjectivity remains, however, with a certain degree of objectivity. Self-referential systems are the reference of our reflection on them, so we transform them into the content of our reflection. As the content of our reflection, this form of subjectivity is always referred to by another. It is the content of another reflection. Günther calls this form of objective subjectivity ‘You’ in opposition to the second form of self-reflection, which in resonance with transcendental idealism he identifies as the position of the ‘I.’

This form of objective subjectivity is then distinguished from introscendent subjectivity where self-relationality allows not only for self-organization but also for awareness of the self. Here, Günther takes a transcendental stance regarding reflection, according to which, if there are self-reflective beings as humans are, reflection must be a structure that is found in everything that is a condition of possibility for the existence of self-reflective beings. Thus, matter must have a certain degree of reflection or the capacity for self-organization, which in his reading of the debate between idealism and materialism is recognized by dialectical materialism. Matter is not an inert, undifferentiated mass, but a dynamic organization. This dynamic organization in matter is what Günther calls information.

If matter is always in a sense self-organized, what is the metaphysical status of our relation to self-organized systems? Moreover, how do we deal with computational systems that are modeled on the structure of our thought? Günther maps out this metaphysical discussion with the concepts of specified and underspecified systems. Specified systems are those that achieve complete reflexivity, that is, the system is completely self-relational or absolute reflective. Underspecified systems, on the other hand, are those of which some aspects remain inaccessible to reflection. Here Günther says that humans remain unspecified systems because some aspects of our experience of the world remain inaccessible to reflection. But this is precisely the condition of possibility of self-reflection. In the case of systems modeled on the structure of self-consciousness, they appear to us as specified systems because we always refer to them as the content of our reflection or as objective subjectivity. Reflection is an operation in which that which is thought is always transformed into another, or a ‘You.’ For this reason, Günther argues that machines always remain a You, or a system that is reflected by us, but is not capable of self-reflection. He argues, in this spirit, that it is not possible to create a self-conscious machine: 

The you—and only the you!—is that subjectivity which is repeatable and reconstructable in the machine. The repeatability in the material construction means that here the living introscendent reflection has been shut down and that now in its place objective, physical processes, which can be primordially reflectively conceived, take over the role of the soul. In this sense, the machine in fact confirms the truth of the Marxist thesis for that reflective subject which can be objectively apprehended because it can never be identical with the subjective subject—i.e., the unattainable introscendent of the ego.

Günther’s critique of dualist metaphysics is a critique of the simplistic position that the objects of our reference are never themselves reflective, that is, that they are not self-organized. For him, this would be problematic because then, their organization would depend on our cognition of them, on how we conceptualize them. Günther, on the other hand, grants self-organization to processes in nature. However, he still distinguishes between self-organization and self-consciousness. 

With this in mind, we can now see why computational systems need an explanation. Computational systems are modeled on the forms of organization found in nature, specifically living things and our brains. However, they always remain a reference system for us. This is where the second-order relationship to these systems appears. Systems that are not self-reflective require a second level in which we think about our relation to them. But computational systems always remain the object of our reflection because they are fully specified by us, while self-conscious beings always remain unspecified for themselves. Self-conscious beings are always more than what they can describe. Thus, although computers are modeled after us, they always remain objects for us.

  1. Bivalent Logics and Poly-Contextual Systems 

Günther’s critique of dualistic metaphysics inspired what would become his proposal for dealing with self-reflective systems. Here his debt to Hegel is evident, as he seeks to develop a logic that can formalize the different levels necessary to localize our thinking about self-organized systems. This model should also help to formalize the structure of the living in a form that can be operationalized for computers.

Bivalent logic is the logic corresponding to the dualist metaphysics. For, as we have seen, in bivalent logic thinking is transformed into an operation with two positions: that of the knowing subject and that of the object to be known. The totality of logical space can only be defined disjunctively either from an absolute position of the subject or from an absolute position of the object: “All our scientific terms—as they develop on this Aristotelian ontological basis—retain a semantic ambiguity. They can, in their totality, either be taken as describing the Universe as the absolute Object or as the absolute Subject.” These two positions then correspond to the two available values (true and false) in logic. The idea is that bivalent logics only can reflect on objects if they are not dynamic. 

In Cybernetic Ontology and Transjunctional Operators, the question is how to account for the emergence of an ordered system from non-informed elements. This question is central to Heinz von Foerster’s notion of order out of noise. To bivalent logic, Günther opposes an operational dialectic. Operational dialectic gives rules for the localization of subjectivity or the position of the observer in knowledge. This proposal depends on a different conception of negation than that of bivalent logic. Bivalence logic has only one logical operator: negation. Negation divides the logical space into two positions for truth values. This means that we can only assert something as either x or -x. Thus, for any proposition, we can only find two spaces in which to locate it, either true or false. Operational dialectics would rather be considered as a multivalent logic, which requires not only to judge the content of a proposition, but also to locate the position in which the proposition is made. This is done by introducing multiple logical operators, which Günther wanted to introduce with a graphical notation system. This would allow Günther to introduce logical operators that are created by the system itself rather than imposed on it, which means to think of the operationality of self-referential systems.

In Life as Poly-contexturality, Günther distinguishes between context and contexture. Context is our everyday concept of the frame of reference that gives meaning to something. Contexture, on the other hand, is the logical domain of a two-valued structure whose scope is determined by the principle of the excluded third, that is, by what is not included in the frame. As we have seen, in the case of bivalent logic there is only one logical operator, so that, according to Günther, the contexture in which negation operates is only one. The two positions for the values produced by negation in bivalent logic correspond to the same contexture, insofar as it is from the context of being that what is negated can be defined. If x is not the case, then it cannot be that x is the case. The logical space is then divided into two parts: that of which is (the case) and that which is not (the case). In Aristotelian ontology, the contexture is only one—mono-contextuality— or there is only one framework upon which to decide what is the case and what is not. In Aristotelian logic it is not possible to generate other value positions, let alone other systems for ordering those values: “The mono-contextual ontology offered no place for the observer of the world or the thinking subject because it would have been absurd to suppose that the cognizing subject belonged to the context of the cognizable.” To explain the complexity of life, it is necessary to consider multiple levels or systems of reference. Therefore, Günther opposes the mono-contextuality of classical logic with a poly-contextuality.

Tables III and VII, from Life as Poly-Contextuality, display respectively a simpler and a more complex contexture.

According to Günther, Hegel proposes precisely a poly-contextual structure of reality. That means that speculative dialectics opens up the possibility for subjectivity to become the object of its own reflection, that is, one can think about oneself. Hegel suggests thinking about the position of the subject when it is the content of its own reflection. This idea becomes helpful for developing poly-contextuality. That is, the idea that when we think about ourselves, subjectivity is divided into two: one side that performs the activity of thinking and the other side that is us as the content of our thinking. This allows us to think of different positions of subject and object, since we subjects also become objects. Thus, in systems that operate poly-contextually, subjectivity is multiplied at all levels of organization. If, as we have seen, subjectivity denotes the capacity for reflection, and this refers only to self-organization, then all self-organizing systems are already endowed with subjectivity. Living systems contain different levels of organization and at each of them there is a subjectivity or an organizing principle. This organizing principle is introduced as a logical operator in the formalized version of Günther’s dialectical logic. 

Now, insofar as the logical operator follows the principle of the excluded third, each contexture negates another, thereby producing a determinate negation, that is, a negation that depends on that which it negates to be determined. Poly-contextuality proposes to increase determinate negation by focusing on the intersections of more than one contexture. In this way, the intersections between contextures or levels of organization become new positions that can be occupied by new truth values. This replication of new contextures can be seen at different levels of reflection, and according to Günther, levels of reflection are what create complexity in self-organizing systems: “For it is a property of all systems of reflection that—insofar as they produce new structures of reflection at all—these must inevitably assume a higher degree of complexity than the system which produced them.” With poly-contextuality Günther seeks to explain how the complexity of the living operates and what kind of logic is necessary to understand that complexity.

Although Günther takes on Hegel’s concept of reflection for his theory of complex and self-organizing systems he also criticized him. While Hegel thinks of the machine only in the context of mechanical labor, Günther explores machines that have consciousness as their model. For Hegel, machines remain bound to the logic of the mechanical and never achieve the self-referentiality that appears only in the organic. Complex systems are for Günther systems that are both the effect and cause of themselves, in other words, they operate internally, as Kant in the Critique of Judgment or Hegel in his Science of Logic have proposed to think of the living.

I think that when we think about self-organization today, it is worth looking at Günther’s ideas about what kind of metaphysics it requires. This would help us to reflect on the kind of assumptions that underlie many debates about the causality of the living, the emergence of life, and how we think about these phenomena when we study them with computational systems. What is striking, however, is that Günther has focused more on Hegel’s ideas about consciousness and less on his reflections on life. This sometimes leads Günther to make some rather anthropocentric claims, which I think should be critically revised in an engagement with Günther’s work. Nor should one overlook the historical and political context in which these developments in cybernetics took place, and the involvement of their protagonists in Germany during the Nazi government. How this context influenced the development of cybernetics would, however, would be the subject of a more extensive study of Günther’s work and the Biological Computer Lab which is to be undertaken. 

References

Günther; Gotthard. Das Bewußtsein der Maschinen. Eine Metaphysik der Kybernetik. (Baden Baden: Agis Verlag, 1963)

_____________. Beiträge zur Grundlegend einer operationsfähigen Dialektik, Band I (Hamburg: Meiner; 1976). 

_____________. Beiträge zur Grundlegend einer operationsfähigen Dialektik, Band II (Hamburg: Meiner; 1976). 

Hegel, G.W.F. Wissenschaft der Logik, Band I (WS5). Suhrkamp: Frankfurt am Main, Edition 2017.

_____________. Enzyklopädie der philosophischen Wissenschaften (WS10), Band III. Suhrkamp: Frankfurt am Main, Edition 2016.

Hörl; Erich. Die offene Maschine. Heidegger, Günther und Simondon über die technologische Bedingung. MLN, Volume 123, Number 3, April 2008 (German Issue), pp. 632-655. 

Köpf, David. Mit dem Weltgeist rechnen. In Baecker, Dirk  (Ed.). Schlüsselwerke der Systemtheorie. Wiesbaden: Verlag für Sozialwissenschaften; 2005, pp. 225 – 242.

Lemmens, P. C. Gotthard Günther. (2013). Roudboud Repository. Online, Retrieved on 29/07/23. 

Magerski, Christine. Gehlen, Arnold Karl Franz (1904–1976). Bloomsbury Encyclopedia of Philosophers. London: Bloomsbury Publishing, 2020.

Müller, Albert. Eine kurze Geschichte des BCL. Heinz von Foerster und das Biological Computer Laboratory“. In: Österreichische Zeitschrift für Geschichte, 11(2000), 9–30.

Yang, Ou-Sork. Hegels Technikverständnis und Versöhnungsinteresse zwischen Natur und Geist. Würzburg: Ergon-Verl., 1998. 

Leave a Reply

Your email address will not be published. Required fields are marked *

Comments