Strong Constructivism: Modernity and Complexity in Science Studies and Systems Theory

Von Foerster’s theoretical position is an unusually explicit version of constructivism—a line of thought applying in various ways the idea that we invent the world that we perceive. It is important to note that von Foerster’s constructivism stems from the discourse of neocybernetic systems theory, and in this paper I will try to explain what that importance is.  Some may find the truth of his thesis easier to accept when it is presented in forms that can be dismissed as fantasy, as in any number of Escher engravings that play with perceptual paradoxes.

Figure 1. M. C. Escher, Waterfall (lithograph, 1961)

The illusion this engraving extorts from our cognitive faculties, however, is an allegory of the illusion to which we attain when we succeed in attributing sense experience solely to the exterior objects that occasion it. But, paradoxically, both levels of this illusion—the artistic or design level and the phenomenological level of perception per se—place the observer in workable relations to their environment.  Systems theory would explain that while not precisely a Möbius strip, Escher’s impossibly perpetual waterfall does exhibit the “strange loopedness” that sets cybernetic system/environment couples apart from more traditional linear and causal models of systems.  Circularity, recursivity, feedback—the re-entry of output as input—these are real processes of systematic operation that enable us to construct the real limitations of classical intuitions of system dynamics. The transition from simple input/output distinctions to complex recursive functions points us toward a crucial aspect of the constructivism associated with systems theory—namely, the paradoxicality of the boundary through which system/environment couples are organized.  For this boundary is closed and open at the same time—it both encloses the operations of the system and opens it for its internal construction of the external environment.  Without operational closure the system could not observe its environment at all, and by the same token all observations translate that which is observed into the terms of the observing system. As sociological systems theorist Niklas Luhmann puts it, the constructivist paradox is that “only non-knowing systems that can know; or, one can only see because one cannot see” (“Cognitive Program” 67).

I will call von Foerster and Luhmann’s systems-theoretical approaches to the environment strong constructivism in order to distinguish them from several other varieties of constructivism and to some extent from the controversies that have attended them.  In the field of science studies, for instance, one speaks of “social constructivism” when the content of science and technology is explicated as a construction compounded of social factors and forces. Anthropologist David J. Hess explains in Science Studies: An Advanced Introduction that “In the social studies of science and technology, the term ‘social constructivism’ is often used as a general label for studies that examine how social variables shape the pattern of choices about what research gets done, how it is done, how choices among theories are made in controversies, and the extent to which observations, laws, theories, and other knowledge claims become accepted in wider scientific communities” (34-35).  Objectors to this approach complain that when social factors alone are deemed to deconstruct the object of science, they falsify its “nature” and replace the natural object with a social cover story. However, proponents can point effectively to a procession of supposed “natural objects” that have turned out not to be natural at all—for instance, “race” and “gender,” purportedly scientific objects that have now been convincingly reobserved as social constructions. Strong constructivism would nuance this argument like so: race and gender were constructed as real substances or genetic essences because the possibility of observing their social construction occupied a blind spot in the conceptual systems available to observe them. At the same time, removing a blind spot in one location always generates another elsewhere.

In the same 1974 paper, von Foerster proceeds to a do-at-home experiment designed to show that the blind spots in our observations of environments as well as of systems are built into the cognitive apparatus at our disposal.

[Figure 2]

But unless we are guided in such a fashion to discover this blotch in our visual reception, we remain blind to our blindness, because otherwise our cognitive systems operate so as to cover over their deficits. “This blindness is not perceived at all,” von Foerster continues, “neither as something present, not as something absent.  Whatever is perceived is perceived ‘blotchless’” (43). The neurophysiology of the eye offers a paradigm for the blind spot and the occulting of that blind spot that inhabit both the structure and the operation of every observing system.

In such fashion strong constructivism frames social constructivism in relation to cognitive science and philosophy. It pursues a constructivist epistemology in which the logic of knowledge and the conditions of its possibility are rearticulated through the system-environment couple.  For systems theory conceptual contingencies are as much cognitive as social, and are in any event inescapable, as in Kant’s insistence that the categories of reason process all objects of cognition according to a priori intuitions of time, space, and causality.  Hess notes as the “extreme version” of contemporary constructivism a radicalized idealism, which “amounts to more than an instrumentalist account of theories; it refers to a social idealism in which there is no material reality that constrains or structures sensory observations” (Science Studies 35).  And it is into this “nihilist” or “reality-denying” form of absolute relativism that anti-constructivists tend to blur all conceptions of constructivism.

The strong constructivism outlined by von Foerster, however, is neither a social idealism nor an anti-scientific nihilism, nor entirely a “radical constructivism” (although a group of psychotherapeutic constructivists descended from the work of Gregory Bateson, which goes by that name, has also claimed him as an inspiration).  Strong constructivism is rather, to borrow Luhmann’s characterization, a crucial refinement in the philosophical mode of epistemological constructivism for which conceptual contingencies are both phenomenological and sociological—in Luhmann’s terms, they concern the interpenetration of psychic and social systems.  The conception of the environment that informs von Foerster’s postulate that the environment as we perceive it is our invention is inextricable from the interrelated notions of system and boundary.  In this form of systems theory, environments are always strangely looped into corresponding systems.  They emerge as such only when a system demarcates itself from that environment by means of a boundary that it must then maintain in order to remain viable as a system, or in the jargon of the discipline, in order to continue the autopoiesis or self-referential self-construction of the system.

These stipulations remove strong constructivism from the realm of idealism altogether. Environments are compounded of raw reality—matter, energy, information, living beings, and social groups sufficiently loosely coupled that systems can inscribe them with their own forms. Strong constructivism accounts for both the creation and the observational recreation of that which has taken autopoietic shape in the world. What is constructed in every instance generates a complex embedding of system/environment couples. When we draw boundaries in order to indicate objects and events, our drawing of outer boundaries is itself drawn at the inner boundary that ensures the operational closure of our system of cognition. Everything known refers in the first place to the forms by which it is known, and those forms always contain more than meets the eye—as in the miraculous geometry of Escher’s Waterfall—and less: a blind spot, a place where we cannot see that we cannot see what’s there.

2. Varieties of Modernity

Why is it that we have only recently noticed that systems are inexorably coupled to the environments they construct in order to arise as systems? It is as if the environment had long occupied a cognitive blind spot from which it has just recently been shifted into view, enabling the drastic thematization of the environment that drove von Foerster’s American friends to their visionary moments to occur. And yet the emergence of this environmental awareness seems to have accompanied or occasioned a related breakdown in modern social certainty.  For instance, Jean-François Lyotard famously defined postmodernity as the historical period that witnesses the lapse of metanarratives—the inability any longer to believe the “story of salvation,” for instance, or the “story of progress.” This withering of previous cultural narratives has been variously felt to be a loss or a gain, but in any event, a profound event, worthy of its own historical distinction.

Luhmann’s discourse on modernity acknowledges this description but deflates its profundity.  For him “postmodernity” is one of many ways to observe communicative formations and their dissolution over time in the operation of modern social systems.  Luhmann withdraws the singularity of this semantics and reinscribes this description as an unavoidable effect of modern social complexity.  “The proclamation of the ‘postmodern’ has at least one virtue,” Luhmann writes in the Preface to his Observations on Modernity: “It has clarified that contemporary society has lost faith in the correctness of its self-description” (ix).  But this in itself fails to distinguish the “postmodern” from modernity per se.  Rather, Luhmann continues, “there is no métarécit because there are no external observers.  Whenever we use communication—and how could it be otherwise—we are already operating within society” (x). That is, even if we could arrive at a metaperspective entirely transcending the social systems in which we are always-already interpellated, we could not communicate that perspective to those systems. The reception of communication within given systems deletes the meta-systemic character of any message, just as the mass media co-opt any attempt to use them to indicate a state of affairs transcending the society at hand: “transgression and subversion never get ‘on the air’ without being subtly negated as they are; transformed into models, neutralized into signs, they are eviscerated of their meaning” (Jean Baudrillard, cited in Winthrop-Young and Wutz, “Translators’ Introduction” xv).

Bruno Latour has suggested another provocative version of the breakdown of modern metanarratives. In We Have Never Been Modern, Latour purports to deconstruct the metanarrative of modernity itself.  Trained in the sociology of science and technology, Latour is perhaps the most flamboyant theorist and gifted writer to emerge from the social constructivist wing of science studies. A quick summary of Latour’s historical thesis might go like this: Modernity arose from the suppression of the incoherence of premodern ideas—their mixing up of natural and cultural causalities—but this sorting operation occurred at the price of enforced separations between nature and society that have in turn now lost their coherence as well. According to Latour, by pursuing an anthropology of technoscientific practices, science studies has now unraveled the modern unraveling of premodernity. Latour would reform academic protocols and resolve the distinctly modern conflict between industrial societies and the ecological constraints of the natural world. He concludes We Have Never Been Modern with a prescription for a politics of nonmodern collectives, presenting as a way beyond the impasses of modern ontological separatisms the state of “nonmodernity,” complete with a “nonmodern Constitution” for a “parliament of things.”

Luhmann’s theory of modern social systems, in contrast, is uninflected by Latour’s brand of dialectical or perhaps Parisian drama.  What Latour perceives as modernity—an enforced purification covering over an orgy of hybrid copulations—Luhmann views as a change in habits of differentiation and a move from a less to a more complex form of differentiation.  We are modern, Luhmann would insist, due to an irreversible social evolution he describes in the following manner: Social differences in premodern societies were stratified, predetermined by a hierarchical system of proper social positions; modernity arose over time as stratified social differentiation gave way to the differentiation of social functions—for one instance, the separation of church and state. Differentiations of function (rather than of strata) now fissure society as a hierarchical totality into relatively autonomous subsystems of government, law, science, religion, education, industry, media, and so forth.  Society is now a hyper-complex system of systems within systems, occupying a built environment that is itself a network of material/mechanical systems embedded within a natural environment often unperceived until, on occasion, it twitches catastrophically.

Luhmann would neither predict where society is headed nor prescribe a direction toward which to steer it. He did, however, specify some constraints in the operation of any autopoietic system, including social systems. These systems must constantly submit their elements to dissolution in order to go on reproducing themselves, and the constant reshuffling necessary to systematic self-reproduction ensures the emergence of difference over time, system evolution.  This description goes beyond but certainly includes and valorizes social democracy in its liberal and libertarian senses as a dynamic equilibrium of social forces and a monitored interrelation of quasi-emancipated institutions. Based on such formal considerations of the abstract conditions of systemic possibility, Luhmann emphasized the paradoxicality and unpredictability that inflect the temporal evolutions of social systems.

3. Hybrids and Paradoxes

Latour’s brand of constructivist science studies exploits a strategy developed in what is called the sociology of scientific knowledge, a shift in focus from scientific theory to scientific practice, from the rarified discursive end-products of scientific enquiry to the nuts and bolts of laboratory instruments and procedures and the professional practices of assembling, interpreting, and disseminating experimental results.  At the level of experimental practice and its publication, scientists and technicians must negotiate with nonhuman actors as well as human colleagues, must construct networks that can connect the behaviors of persons, machines, organisms, material substances, forms of energy, and the protocols of informatic inscriptions.  Latour’s theory of modernity emerges from this crucible—the observation of scientific practice. According to Latour, “Science studies have forced everyone to rethink anew the role of objects in the construction of collectives, thus challenging philosophy” (55), that is, in particular, challenging previous assumptions governing the construction of scientific knowledge.  But since these assumptions have enjoyed cultural preeminence in the West since Bacon, Descartes, and Newton, Latour considers this challenge to extend to modernity altogether.

For Latour, intellectual modernity has been determined by a metanarrative that recites the epistemological charter science signed with Western society in the 17th century.

Figure 3. Purification and Translation

In We Have Never Been Modern, Latour writes:

The word “modern” designates two sets of entirely different practices which must remain distinct if they are to remain effective, but have recently begun to be confused. The first set of practices, by “translation,” creates mixtures between entirely new types of beings, hybrids of nature and culture. The second, by “purification,” creates entirely distinct ontological zones: that of human beings on the one hand; that of nonhumans on the other. . . . So long as we consider these two practices of translation and purification separately, we are truly modern. (10-11)

To cut this complexity down to size we could use an Occam’s Razor. Luhmann took a modern retooling of such a razor to his theory of social systems by redescribing it in a vocabulary derived from British mathematician George Spencer-Brown’s Laws of Form.   Spencer-Brown propounded a calculus of distinctions that systems theory has correlated with the markings made by acts of observation.

Figure 4. The Form

Systems observe by marking distinctions and crossing over them between marked and unmarked states, or between first- and second-order perspectives. For Spencer-Brown, the act of distinction, the marking of an observation, produces the following elements, which he collectively refers to as “the form”: the indication or marked state (“the inside the distinction”), that which the indication excludes or the unmarked state (“the outside of the distinction”), and the distinction itself, which both separates and unifies its inside and its outside, holding the marked and unmarked states in supplementary relation, while in addition, generating a second distinction in the very act of marking an otherwise “unmarked space.” Thus a single act of distinction always already produces another, from which an infinite series of distinctions can ramify—and conversely, a complex of systematic distinctions can often be collapsed back to a single observation.

At any given moment in the operation of an observing system, however, distinctions can function only on the inside of the forms they produce. This is the burden of operational closure, which renders the operation of a system dependent solely on its own internal states and transformations. In Spencer-Brown’s form notation, this is indicated by the location of the indication within the concave part of the distinction mark.  The reiteration of the indication refreshes but does not alter the form of the observation.  Spencer-Brown calls this the law of condensation.

Figure 5. The law of condensation

That which a system can know of its environment is necessarily translated into and maintained by the medium of its own elements: the operation of cognition in the systems-theoretical sense is both self-referential and recursive.  At any moment of its operation, an observing system can see only the inside of the distinctions it makes.  It cannot simultaneously see the unmarked state of those same distinctions—it cannot form what Luhmann denotes the “unity of the distinction,” or it can do so only in the form of paradox or logical contradiction.  Distinctions are paradoxical because they imply a coexistence between their inside and outside terms that cannot be made operational by the observing system using them at the moment of their use.  In order to do so—that is, in order to access the remainder of the information carried by the total form of distinctions, the system must cross over the initial distinction in order to mark its outer side, which crossing un-marks the previous indication.  Spencer-Brown calls this the law of cancellation.

Figure 6. The law of cancellation

Yet this clearing of the space is not an emptying: it is strong constructivism operating as the deconstruction of the initial construction, which frees the system to re-mark the mark, to maintain its re-entry as an ongoing operation, or perhaps to accept a different indication the next time around the cycle of observation.  Thus, observation can operate, but only while blinded to the paradoxicality of its own forms.  Social systems communicate through complex oscillations between the condensation and the cancellation of shared indications. However, paradoxes in the forms of observation can be resolved, that is, rendered productive, by another observer’s observing the initial play of distinctions.  This can also be diagramed as a form of cancellation, but in this case expanded to indicate that a second-order observer can take as an indication the entirety of another observer’s distinction.

[Figure 7]

A woodcut by Albrecht Dürer can clarify this distinction between first- and second-order observations.

[Figure 8]

In first-order observation—here, the artist as first-order observer of the model—all that can be seen of the form under observation is the indication, the inside of the distinction—here, the model that will occupy the entire frame of the drawing the artist is making, from which all traces of the grid through which he observes will be removed. Nor will the artist himself appear or be explicitly manifest anywhere in that drawing. In other words, in first-order observation the outside of the distinction is thrown into a cognitive blind spot. However, the form of first-order observation feeds back into itself when one can observe the operations of another observer’s observation.  In the second-order observation of first-order observations, the whole form of the distinction that produces the first indication—the “unity of the distinction” between that which is indicated and that which is not—can now be observed, precisely as Dürer has shown in this perspectival study of the shifting and multiplying of perspectives. 

In the larger patriarchal scenario enacted here, the male gaze sights along a phallic index, taking in a female object only to reduce it, according to the prescriptions of single-point perspective, to a shadow on a page. Systems-theoretical form analysis underwrites the paradoxical and exclusionary powers of sex and gender distinctions, but without itself privileging those distinctions. In Dürer's engraving what is seen is not just the object but also the formal mechanisms of its observation: the second-order perspective renders observable the way that the initial object is being observed. Of course, the second-order observer (for instance, the contemporary reader of patriarchal inscriptions) is always also in a first-order position, blind to or occulting something in the environment of its own system of observations (for instance, distinctions among system operations).  Here again, however, although the formal conditions of observation are paradoxical, they work—as long as one can keep shifting perspectives.

The paradoxes inherent in the subject/object distinction remain unobservable while this Constitution validates the construction of knowledge.  Latour’s second dichotomy, however, marks a second-order observation of the Modern constitution.  The first-order theatrics of Modernity—the staging of discrete objects for an audience of “objective” subjects—is laid bare with a 90° turn toward a second-order distinction that now indicates the unity of the prior distinction between nature and culture, nonhumans and humans.  Hybrids, quasi-objects, quasi-subjects, and their networks can now be seen, while the former cognition of pure subjects and mere objects is blocked or surpassed by an awareness of the conceptual machinery by which those perceptions were constructed.

[Figure 9]

Placed into a Luhmannian frame, Latour’s argument may be seen to revolve around the issues of observation and observability. “Modernizing finally made it possible to distinguish between the laws of external nature and the conventions of society,” Latour writes in the final chapter of We Have Never Been Modern, but “the double failure . . . of socialism—at stage left—and . . . of naturalism—at stage right—has made the work of purification less plausible and the contradiction more visible” (130-31).  Networks and quasi-objects can no longer be covered over by the blind spots in the operation of Modern distinctions. Rather, now “we are obliged to consider the work of purification and the work of media symmetrically” (132). With this term Latour invokes an axiom of science-studies methodology advanced by his colleague Michel Callon, “the principle of generalized symmetry”: “the anthropologist has to position himself at the median point where he can follow the attribution of both nonhuman and human properties” (96).

[Figure 10]

Translated into systems-theoretical terms, Latour and Callon’s principle of generalized symmetry entails the construction of a second-order observation allowing the unity of the distinction between nonhuman objects and human social subjects to be contemplated simultaneously.  Latour continues with an ironic inflection: “But . . . if the fruitful contradiction between the two parts—the official work of purification and the unofficial work of mediation—becomes clearly visible, won’t the Constitution cease to be effective?” (132).  Latour’s rhetorical question about the current observability of the Modern contradiction recapitulates the traditional suspicion that paradox, or at any rate the awareness of paradox, typically paralyzes the systems in which it emerges, in that its observational operations then fall into some kind of circular regress or debilitating repetition. Latour’s question can be rephrased: can the distinction that founds Modernity’s observing system be re-entered into its own indicational space without deranging the operation of the system?

Latour’s formulation of Modernity’s very efficacy as resting on the invisibility of its constitutive contradictions may be glossed by Luhmann’s explanation of the unfolding of paradox in the operation of observing systems:

If we observe such a re-entry, we see a paradox. The re-entering distinction is the same, and it is not the same.  But the paradox does not prevent the operations of the system.  On the contrary, it is the condition of their possibility because their autopoiesis requires continuing actuality with different operations, actualizing different possibilities. . . . An oscillating system can preserve the undecidability of whether something is inside or outside of a form.  It can preserve and produce itself as a form, that is, as an entity with a boundary, with an inside and an outside, and it can prevent both sides from collapsing into the other.  A self-referential system that continuously regenerates its re-entry will be, in Heinz von Foerster’s terms, a non-trivial machine. (“Paradoxy of Observing Systems” 42-43)

And in the final sections of We Have Never Been Modern we can plainly observe a strategy of oscillation as the Latourian observer shuttles back and forth between modernity and nonmodernity, ostensibly to perform a reclamation of those elements of premodernity, modernity, and postmodernity that remain worthy of representation within a Nonmodern constitution.

[Figure 11]

The very distinction between the modern and the nonmodern itself is viewed with an eye to the paradoxical productivity of oscillating across this newly constituted semantic boundary. What, then, is the outside of this second-order distinction whose inside or indication is “hybrids and networks,” the quasi-objects constructed and stabilized by nonmodern observations?  Invoking Spencer-Brown’s law of cancellation, the environment of the nonmodern system is as yet an unmarked state that Latour can only re-mark within or re-enter into the system constructed by the nonmodern Constitution as the “non” in “nonmodernity.” After all, there’s not much new about hybrids and networks other than the newly acquired form of their observability, but where this new distinction will lead the operation of the system is anyone’s guess.  I think that Luhmann would have been receptive to this mode of theoretical communication, but would also have insisted that Latour’s nonmodernity remains a description of modernity in its present states of paradoxicality and hypercomplexity.

6. An Environment that Remains Unknown

By way of conclusion, I will glance at Luhmann’s late essay “The Ecology of Ignorance,” a meditation precisely on the vertiginous quality of the “unmarked state” that lies beyond our present systematic horizons. This essay focuses directly on the ecological uncertainty generated by our awareness of the possible catastrophes that may have been set into motion by twentieth-century systems of military and industrial production. Those systems, predicated to a great degree on the defense of capitalist and socialist societies from each other, have actually contributed to the collapse of the metanarratives of salvation and progress that previously drove or guided both sides of the old geopolitical divide.

If we have lost conviction in the possibility of articulating a coherent self-description of hypercomplex modern social systems, however, perhaps we have also gained some useful knowledge about our epistemological limitations. No matter how you cut them, social systems are embedded in environments more complex than anything attainable within those systems. Ignorance is the flip side of knowledge. How can we know what we can’t know? We are ignorant of the extent of our ignorance. When all moral or ethical prohibitions on knowledge are lifted and we know everything we can know, what is it we still can’t know? Luhmann suggests: We still can’t know the full complexity of our environment. What will the environment do that we will not have anticipated? Luhmann calls this “the paradox of warning: a warning, if successful, prevents us from determining whether what we were warned of would have occurred at all” (Observations on Modernity 76). Present indications remain ignorant of the knowledge whether what we now expect of the future will in fact come to pass. No matter how powerful or predictive, science will have to remain a drastic simplification of the real, and governmental surveillance, unfortunately, will always work better on the inside than on the outside of the surveyed system. The point is not: Why bother with knowledge? Obviously we care to know what we can know, and that will continue to drive us toward knowledge. But we need to know that we don’t and can’t know everything there is to know, and we need to act as if we knew this. One of the many terrors of terrorism as a message from without our accustomed channels of communication is that it forces us to ask for more intelligence from our social authorities than they can ever give us—complete knowledge of the environment beyond our own borders.

Perhaps both our strategic and architectural politics and could use a dose of systems-theoretical ethics. At the end of “On Constructing a Reality,” written toward the end of the Vietnam conflict and in the midst of the Cold War, Heinz von Foerster stated his own “ethical imperative: Act always so as to increase the number of choices.” This is also an information-theoretical way of saying: do not foreclose unnecessarily the options a system possesses for continuing its constructions of meaning and thus its behaviors toward its own environment. If ethics concerns the operations of systems, aesthetics concerns the role of observations in guiding those operations. Von Foerster concludes his article with a second, “aesthetical imperative” that feeds back into its ethical partner: “If you desire to see, learn how to act.”

How should one act?  Despite Luhmann's own reticence in saying so, the proper strategy for hypercomplex modernity is to shift into systems-theoretical mode intellectually as well as technologically, to reform classical dyads such as part/whole divisions and subject/object dichotomies into system/environment couples, and to rethink—as strange loops rather than iron curtains or chain-link fences, as paradoxical rather than simple partitions—the borders that autopoietic system/environment couples must use in order to operate.  What we might ask of current design philosophy, then, is some assistance in the task of refashioning our guiding distinctions so that they reflect the complex embeddedness of systems within systems and all systems’ codependency with environments that must always remain in crucial ways out of sight and out of mind.  What forms the construction of such environments will take are also anyone’s guess. However, one thing at least now seems clear: to design an environment is to construct a way of seeing the world.

Works Cited