Notes on Niklas Luhmann, the "Systems-Theoretical Approach," and the Medium of Meaning
A detour on the way to a Marxist critique of science and technology
A preface: I understand the work of Niklas Luhmann to represent the culmination of the contribution of cybernetics to social theory. While cybernetics, of course, originated in Norbert Wiener’s study of information systems, it had been subsequently advanced toward the sociological, cultural, and psychological study of the self-observation of such systems, broadly conceived, by the likes of Margaret Mead, Gregory Bateson, and Heinz von Foerster, among others. While these latter figures present remarkable theoretical developments in their own right, it is in my view Luhmann who most capaciously tries to bring the insights of “second-order cybernetics” into a new “universal sociology” in order to overcome the “crisis of theory” of his time.
Accordingly, I believe Luhmann’s work is most useful for illuminating both the great aspirations of cybernetics to become a universal science of social control, as well as its ultimate conceptual limitations. Certainly, this post is not meant to be an exhaustive critique of Luhmann’s work, and only implicitly is it even about cybernetics. Here, I only want to highlight some aspects of Luhmann’s work which can help sharpen the Marxist critique of science and technology which I have tried to develop in previous posts. After an overview of his “systems-theoretical approach,” then, I will move quickly to the importance of the “environment” and the “medium of meaning” in Luhmann’s thought, two concepts which, I believe, best illuminate the shortcomings of his overarching project, and, implicitly, reveal the means to overcome them, namely, through a restoration of the dialectical concept of mediation.
A second note: what follows is a critique still in formulation, one which will undoubtedly be better clarified (to myself) once I get to actually writing about what is at stake for a dialectical concept of mediation in the Marxist critique of science and technology. One thing that I know is clearly absent here is any review of Luhmann’s fixation on the concept of “autopoeisis,” which would have warranted several thousand more words in conjunction with the question of the ‘missing subject’ in his systems-theoretical approach (see below). I hope to get to that in a subsequent post.
Luhmann Contra Sociology
To understand Luhmann’s project, it is first useful to provide some historical context for his discipline, sociology, which he rebukes for having failed to “produce anything approaching an adequate theory of society” (1998, p. 17). Luhmann’s central complaint is that the ‘science of social facts’ cannot justify its own observations on the basis of what it observes. That is to say, sociology does not consider itself a social artifact but instead habitually adopts a view from nowhere whence its “object matter” can be abstracted into principles of social organization, or laws, which, whenever actually compared against the empirical dynamics of existing society, prove to be “no longer convincing” (ibid., p. 6). In my view, these obstacles amount to the same as those Gillian Rose diagnosed in Hegel Contra Sociology (1995), and thus the latter can provide a sketch of the problematic which Luhmann attempts to overcome and, moreover, may help us pinpoint precisely where he fails to do so.
For Rose, the founders of modern sociology, Émile Durkheim (1858-1917) and Max Weber (1864-1920), had, despite their differences, together worked to entrench a transcendental epistemological structure into the discipline. For Durkheim, society in-itself was sui generis; hence, rather straightforwardly, Durkheim retains in the very notion of society the a priori validity of social facts. For Weber, however, the analysis of society is somewhat more complicated. He did not privilege facts but values, which were the products of the ultimately arbitrary choices of a historical culture.
To mitigate the relativism which follows from this theoretical position, Weber proposed the “ideal-type” as a “conceptual construct (Gedankenbild) which is neither historical reality nor even the ‘true reality’ but instead a ‘heuristic device’” (Weber, 1949, p. 93). Rather then rooting the relationship between social values and empirical reality in an a priori social structure, then, the ideal-type denotes mere “objective possibility,” a “regulative Idea” by which the adequacy of values to their social content may be judged (Rose, 1995, p. 20). Thus, against Durkheim, Weber’s science of society attempted not to confer objectivity to social values but rather to assess through the “heuristic” of ideal-types the relationship between empirical reality and subjective belief. As Rose summarizes:
The neo-Kantian paradigm of validity and values founded two kinds of ‘socio-logy’, two logics of the social: a logic of constitutive principles for the sociology based on the priority of validity, and a logic of regulative postulates for the sociology based on the priority of values. The former identifies social reality by a critique of consciousness; the latter locates social reality within the realm of consciousness and its oppositions. (ibid., p. 21)
Identification of a shared neo-Kantian tradition between them is what is most important for Rose (and implicitly also for Luhmann): Durkheim’s structuralism and Weber’s interpretivism both postulate a “precondition and a conditioned; though their perspectives are opposite, neither can grasp the transition between spheres” (Fuller, 2017, p. 5). Neither Weber nor Durkheim can in other words justify the validity of the knowledge they produce about their object of analysis, given both impose quasi-transcendentally underivable determinations which cannot be further explicated.
For Luhmann, such “epistemological obstacles” irreparably mystify the science of society, once again, leading sociologists to oppose their conceptual constructions to the content of their analysis. Hence, what Luhmann seeks to provide, first of all, is a justification for the conditions of possibility of sociological observation which necessarily originate within the object the sociologist is observing—precisely, then, a justification for the “transition between spheres” of social theory and social fact.
Most remarkably, this leads Luhmann not to any repudiation of the “critique of métarécit” of his anti-foundationalist contemporaries but a concession to its consequences (Lyotard, 1984 [1979]). That is to say, Luhmann’s “systems-theoretical approach” starts from the vantage point that the validation of knowledge, objectivity, and truth is indeed troubled by the contingency of their historical appearance. Thus, Luhmann does not avoid the “crisis of theory” but seeks to wade directly through it, embracing the contention that any descriptive claim about society cannot be rendered independent from but is irreparably a part of society—thus, sociological claims are necessarily produced and always open to transformation, that is, to history.
And here is where Luhmann makes his first move against neo-Kantianism: the possibility for the sociological observation of society can therefore not be deduced transcendentally. His “systems-theoretical approach” turns away from “critique”—in the sense of a deduction of the a priori forms which organize the content of thought—and toward an immanent study of “the function of such descriptions for the operations of systems” (Nassehi, 2005, p. 179). Put differently, sociological description is only justified for Luhmann insofar as it is not independent from but a self-differentiation of the social system it describes, a differentiation which does not simply “cut reality into two parts”—which would merely repeat the division between noumenon and phenomenon—but is the “correlative to the operation of observation, which introduces this distinction (as well as others) into reality” (1995a, p. 178). Sociology attains its perspective, hence its possibility of observation, within the indication of the difference between itself and the social system—an indication which is nothing other than the self-differentiation of that social system realized in the process of observing itself.
It is worth dwelling on this notion of indication and its relationship to the self-differentiation of system, as it is the most important relationship in Luhmann’s systems-theoretical approach. Luhmann borrows this correlation from the “calculus of indications” developed by the British mathematician George Spencer-Brown in Laws of Form (1979). For Spencer-Brown, no distinction is made without also being indicated as a distinction, and, moreover, its indication is made always from the interiority of that distinction. Put another way, a distinction can indicate what it is not, but it can only do so in reference to itself. This is what Luhmann calls the “guiding difference” which marks the possibility of a system against its “environment,” that is, the “unmarked state” indicated in the “self-referential closure” of the self-differentiation of a system:
[…] we can conceive of system differentiation as a replication, within a system, of the difference between a system and its environment. Differentiation is thus understood as a reflexive and recursive form of system building. It repeats the same mechanism, using it to amplify its own results. In differentiated systems, as a result, we find two kinds of environment: the external environment common to all subsystems and a separate internal environment for each subsystem. This conception implies that each subsystem reconstructs and, in a sense, is the whole system in the special form of a difference between the subsystem and its environment. Differentiation thus reproduces the system in itself, multiplying specializes versions of the original system’s identity by splitting it into a number of internal systems and affiliated environments. (Luhmann, 1982, p. 230-231)
This, to be clear, is not to say that the “affiliated environments” cannot change in themselves or that they have no effect on a system but rather that, from the perspective of system, these changes are “adapting effects, which [always] appear in conjunction with the system itself” (Nassehi, 2005, p. 180). Systems are thus not determined by the environment against which they differentiate themselves but rather “procure causality for themselves,” such that they can no longer be “‘causally explained’ […] They presuppose themselves as the production of their self-production” (Luhmann, 1995a, p. 41).
Thus, in Luhmann’s account, to suggest that the environment (or contingency) contains the possibility of system leaves the analysis of society at the stage of the social system reacting “to an unclear picture of itself” (ibid., p. 28). At this stage of observation, the ‘in-itself’ is, in a word, a bad abstraction. It is only with greater self-clarification that the environment can be seen to be “neither ontologically nor analytically more important” than system or vice versa: “both are what they are only in reference to each other” (ibid., p. 177).
Luhmann Pro Kant
Even as Luhmann therefore succeeds in rationalizing the necessity of contingency (qua environment) in the self-differentiation of system—a decidedly Hegelian formulation (see also Kjaer [2006])—ultimately, he will conclude that there is an ‘environment of environments’, as it were, which serves as the absolute “precondition” to the system-environment differentiation. In a rather revealing passage in Social Systems, Luhmann writes that the rationality once presumed to inhere in the scientific object was abandoned “by transcendental philosophy” and replaced, “in correlation with the inclusion of self-reference in the ‘subject,’ by the hypothesis that reality is unknowable ‘in itself’” (1995a, p. 101). The “re-objectivation of self-referential systems,” i.e. Luhmann’s “systems-theoretical approach,” does not
falsify this thesis, but rather generalizes it: every self-referential system has only the environmental contact it itself makes possible, and no environment 'in itself.' But this 'itself makes possible' is not possible in a structureless, arbitrary, and chaotic environment, because within such an environment it is impossible to carry out 'internally' satisfactory proofs of worth and, from the perspective of evolution, to acquire permanence. With this, one does not return to the postulate of a corresponding rationality or a lawfulness in nature; but knowledge in particular and system behavior in general presuppose structured and in sufficient measure graspable complexity. (1995a, p. 101)
Thus, even in Luhmann’s sophisticated account of social systems, we are ultimately forced to appeal to the unintelligible chasm between system-environment and the pure contingency of the thing-in-itself. Backgrounded and unable to be explained is that which potentiates this “evolutionary” relationship between system and environment. Incapable of being explicated, the space between projection and thing projected remains, in the last instance, “dark and void” (Lukács, 1971, p. 119).
But there is another limit Luhmann runs into with the “special case” of “meaning systems,” namely, social and psychic systems (1995a, p. 102). Luhmann had long stressed that meaningful systems are uniquely capable of self-observation. This is because the categories of being which comprise its “frame” subsist in the “medium of meaning,” which is “the same on both sides of the frame, on its inside and on its outside” (1995b, p. 41). Meaning, in other words, bears not only what is “framed” as being but all manner of paradox, non-sense, and logical interdiction in excess to that frame: “every possible use of this medium called ‘meaning’ will itself reproduce meaning, and even an attempt to cross the boundary of meaning into an unmarked space will be a meaningful operation” (1995b, p. 41). Thus, this “evolutionary universal” of meaning, as Luhmann remarks elsewhere, is an “unnegatable category” precisely because its negation, non-meaning, also has meaning. It is “a category devoid of difference. In the strictest sense, its sublation would be ‘annihilation’—and that could only be the matter of an unimaginable instance” (1995a, p. 62).
Not only does this medium allow self-observation by subsisting on both sides of the frame, then, but precisely because of this topology, meaning can always “re-enter” itself, guaranteeing that whatever is actual is always open to future possibility—“continuing actuality with different operations, actualizing different possibilities” (1995b, p. 42). Indeed, the development of psychic and social systems is “based” on the possibility of re-entry (ibid., p. 42). Meaning, Luhmann writes, is coextensive with the “form of the world” overlapping the “difference between system and environment” (1995a, p. 61). Situated at the interstice of the point of actuality and the “horizon of possibilities,” meaning thus serves to “redifferentiate differences among open potentialities: to grasp them, to standardize them, to schematize them, and to acquire informational value from the ensuing actualization” (ibid., p. 75).
It is only in this way that the medium of meaning can subvert the distinctions it had previously indicated. That is, it is through this torus-like topology that both sides of meaningful distinctions can “re-enter the space they distinguish, turning up in their own forms, thus capable of developing some kind of contact with themselves,” in this way slipping between “what it is indicating” and “what it uses to make indications” (Schiltz, 2004, pp. 17-18). Once again, as Michael Schiltz insists, this possibility of re-entrance, of the medium of meaning referencing itself, is the condition of possibility of the self-development of all meaningful systems, as well as the guarantee of their incompleteness: “Meaning as our phenomenology of this world can only be partial, as the difference between form/medium can only be actualized as a form” (2009, p. 173).
The Missing Subject in the “Systems-Theoretical Approach”
But here, precisely what is curious to me in Luhmann’s account of the self-development of meaningful systems is how this substratum of meaning is completely cleansed of its subjective conditions of possibility. Indeed, the subjective process of signification appears to be ruled out from the very beginning: Subjects do not constitute a system but are merely illusory means by which meaningful systems cover up the underlying “paradox” of self-reference: put differently, subjects are “mechanisms which release social systems from the danger of losing themselves in their own self-reference” (Nassehi, 2005, p. 182). As we have seen above, the ultimate precondition of pure contingency also serves this role; but it remains ‘in itself’, unintelligible to system; hence, Luhmann’s recourse to the “undifferentiated” and “unnegatable” medium of meaning in the justification of self-observation.
What does Luhmann lose in excising subjects from the “medium of meaning,” in turn positing the latter as not product but precondition of the evolution of social (and psychic) systems? First, it impoverishes the nature of social action. Luhmann, in other words, succumbs to the same condition Georg Lukács diagnoses in context of the reification of bourgeois thought, where action is rendered a purely contemplative reflection of the ‘objective’ system of laws within which it insulates itself. As reality becomes rationalized in systems theory, ‘action’—rendered idle with respect to system transformation—comes to concern only those actions validated by system, such that the more action approximates itself to this reality, “the more the subject will be transformed into a receptive organ ready to pounce on opportunities created by the system of laws and his ‘activity’ will narrow itself down to the adoption of a vantage point from which these laws function in his best interests” (1971 [1923], p. 130).
Moreover, without a subject, or else with the subject posited merely as an illusory “mechanism,” society remains opaque to itself. That is to say, the removal of the subject from the object of society results in the unknowability of ourselves, both as concrete subjects of social experience and as moral agents capable of transforming that ‘social system’ to which we are subject. In turn, the opacity of the subject in the “systems-theoretical approach” makes completely unintelligible the course of real historical development. As István Mészáros remarks,
The only way to make dialectically intelligible the course of historical development is by adopting as the theoretically necessary point of departure the dynamic transformations of objectively existing need and necessity […] with reference to the progressive self-constitution and potentially emancipatory self-mediation of the human agency. That is to say, by accounting in historical theory for structurally meaningful change through the intervention of the actual human subject of history not as fictitiously inflated ‘sovereign maker’ of historical change—to the arbitrary exclusion of the immense weighty objective conditions, found by the social individuals at their point of arrival, and in a partially modified form left to the next generation both to live with and to modify—but as a vital and genuinely active part (and only a part, no matter how important) of the overall process.
In this sense the challenging problem of dialectical intelligibility in historical theory requires focusing on the actual process of the human being becoming the subject of history in a properly defined sense. (2011, p. 303)
In the Grundrisse, Marx critiques how the bourgeois system of thought obfuscates its own historical genesis: “new forces of production and relations of production do not develop out of nothing, nor drop from the sky, nor from the womb of the self-positing Idea.” The mystique of the “completed bourgeois system” is precisely that it nevertheless gives the appearance of ‘objectively’ presupposing itself—“everything posited is thus also a presupposition”—a system of totality subordinating “all elements of society to itself […] The process of becoming this totality forms a moment of its process, of its development” (1973 [1857], p. 278). As itself the culmination of what cybernetics brings to bear on social theory, Luhmann’s “systems-theoretical approach” is, in my view, the apotheosis of this tendency in bourgeois thought.
Conclusion: Toward a Notion of Dialectical Mediation
In History and Class Consciousness, Lukács remaks that, in its attempt to unversalize itself, bourgeois rationality extricates itself from its own givenness, such that the “ultimate substance of knowledge” becomes indeterminable. Reason remains, in other words, dependent upon the relation of its own contents to contingency it cannot rationalize: “Thus, the attempt to universalise rationalism necessarily issues in the demand for a system but, at the same time, as soon as one reflects upon the conditions in which a universal system is possible, i.e. as soon as the question of the system is consciously posed, it is seen that such a demand is incapable of fulfilment” (Lukács, 1971 [1971[1923], p. 117). The issue, as I see it, is that every proposal for a universal system, including Luhmann’s “universal sociology,” depends upon an actually given content which cannot be derived from its own principles—the pure contingency beyond the “guiding difference” of system and environment; the missing subject in the “medium of meaning.” For Lukács, this means that either the content of the system of thought in question must be assumed to be non-existent in spite of its immediate sensuous experience, or else one must concede “that actuality, content, matter reaches right into the form, the structures of the forms and their interrelations and thus into the structure of the system itself” (ibid., p. 118). While the former is simply to ignore the problem of contingency, the latter abandons the notion of thought as self-systematizing and hence turns ourselves blind to the objective necessity of its determinations.
The notion of labor as “sensuous productive activity” grounding all production, even the “most complex and mediated intellectual production,” is how Lukács (and Marx) resolves this antinomy (Mészáros, 2011, p. 36). At the ground of the “social system”—in Lukács’ terms, social being—is, in other words the “dynamic material/intellectual telos of labor: both as self-production and as the production of the conditions of emancipatory social transformation in the direction of the ‘realm of freedom’” (ibid., p. 36). Telos is here thus not “theological” or “naturalized” but immanently social, concerning the “way in which the human being—this unique ‘self-mediating being of nature’—creates and develops itself through its purposeful productive activity” (ibid., p. 70).
The uniqueness of labor is that it “always causally determined but the causes are posited by the subject, and are thus transformed” immanently in the process of labor (Browne, 1990, p. 205). Thus, precisely by changing its own objective determinations, labor changes itself. This is the dialectical mediation missing in Luhmann’s “medium of meaning” which, despite his best efforts, exists as some underivable, quasi-transcendental apriority to the transformation of social systems. He therefore takes as immediate and given what is historically mediated by labor—a metabolic relation between society and nature which is not “foisted on to the objects from outside,” nor attached as a “value-judgment,” but the “manifestation of their authentic objective structure” (Lukács, 1971 [1923], p. 162).
By contrast, in the absence of the subject and the “one-sidedness” of the operation of indication, what we are left with in Luhmann’s “systems-theroetical approach” is a profoundly conservative reflective judgment. Sociology thus remains, even in Luhmann, mired in neo-Kantianism, failing to advance conceptually from this hypostatized notion of the “medium of meaning” toward a dialectical notion of mediation. While my post on Hans-Dieter Bahr should already give hint at the stakes of dialectical mediation for a Marxist critique of technology, I will concretize these connections in a subsequent post.
Works Cited
Browne, Paul. 1990. “Lukács’ later ontology.” Science & Society 54 (2), pp. 193-218. https://www.jstor.org/stable/40403069
Kjaer, Poul. 2006. “Systems in Context: On the Outcome of the Habermas/Luhmann-Debate.” Ancilla Iuris: 66-77.
Lee, Daniel. 2000. “The Society of Society: The Grand Finale of Niklas Luhmann.” Sociological Theory 18(2), pp. 320-330. https://doi.org/10.1111/0735-2751.00102
Luhmann, Niklas. 1982. The Differentiation of Society. Translated by Stephen Holmes and Charles Larmore. Columbia UP, New York.
Luhmann, Niklas. 1995a. Social Systems. Translated by John Bednarz, Jr., with Dirk Baecker. Stanford UP, Stanfrod.
Luhmann, Niklas. 1995b. “The paradoxy of observing systems.” Cultural Critique 31, The Politics of Systems and Environments, Part II, pp. 37-55.
Luhmann, Niklas. 1998. Theory of Society, Volume I. Translated by Rhodes Barrett. Stanford UP, Stanford.
Lukács, Georg. 1971 (1923). History and Class Consciousness: Studies in Marxist Dialectics. Translated by Rodney Livingstone. MIT Press, Cambridge.
Lyotard, Jean-François. (1984) 1979. The Postmodern Condition: A Report on Knowledge. Translated by Geoff Bennington and Brian Massumi. University of Minnesota Press, Minneapolis.
Marx, Karl. 1973 (1857). Grundrisse. Translated by Martin Nicolaus. Vintage Books, New York.
Mészáros, István. 2011. Social Structure and Forms of Consciousness, Volume II: The Dialectic of Structure and History. Monthly Review Press, New York
Nassehi, Armin. 2005. “Organizations as decision machines: Niklas Luhmann’s Theory of Organized Social Systems.” The Sociological Review 53(1), pp. 178-191. https://doi.org/10.1111/j.1467-954X.2005.00549.x
Rose, Gillian. 1995. Hegel Contra Sociology. Athlone, London and Atlantic Highlands.
Schlitz, Michael. 2003. “Form and medium: A mathematical reconstruction.” Image & Narrative 6, pp.
Schlitz, Michael. 2009. “Space is the Place: The Laws of Form and Social Systems.” In: Emergece and Embodiment. Edited by Bruce Clarke, Mark N.B. Hansen, Barbara Herrnstein Smith, and E. Roy Weintraub, pp. 157-178. https://doi.org/10.1177/0725513607072452
Spencer-Brown, George. 1979. Laws of Form. E.P. Dutton, New York.
Weber, Max. 1949. The Methodology of the Social Sciences. Translated and edited by Edward A. Shils and Henry A. Finch. The Free Press, Glencoe, IL.