Margulis, Autopoiesis, Gaia

Bruce Clarke

Margulis, Autopoiesis, Gaia

by Bruce Clarke

07-07-2019

thread: critical ecologies

https://doi.org/10.7273/pgha-tv58

This essay was peer-reviewed.

Clarke opens with a biographical account of his own early encounters with the Gaia concept - and a skepticism he shared with the evolutionary thinker Lynn Margulis, Gaia's second author after the British scientist James Lovelock. Neither an organism nor a single cell, and not really an organic or vitalistic entity, Gaia is better understood, according to Margulis, as "an emergent property of interaction among organisms, the spherical planet on which they reside, and an energy source, the sun." The move away from metaphors and New Age vitalisms tethers Gaia theory more tightly to the sciences. Margulis’s work brings Gaia in line with the autopoietic systems theory of Humberto Maturana and Francisco Varela introduced in the 1970s.

This essay is drawn from a preliminary version of several sections from Clarke’s forthcoming book presently entitled Partial Earth: Lynn Margulis, Systems Theory, and the Evolution of Gaia (Fall 2020) and appears here by permission of the University of Minnesota Press.

How I Found Gaia

In my part of academe the general public debut of chaos theory arrived with James Gleick's Chaos: Making a New Science in 1987.¹ By the 1990s, inspired partly by chaos theory's avid interdisciplinary reception, I began in earnest to cultivate a post-tenure academic specialization in literature and science. But as I set about to reschool myself in physics, chemistry, and biology, to come up to speed on chaos and complexity theory, thermodynamics and information theory, and then cybernetics and systems theory, where Gaia was concerned, not much came to hand. And even after it had crossed my threshold, for a while I was reluctant to take it seriously. I had absorbed the usual nebulous notion that what "Gaia" named in scientific context was not quite real science but some kind of New Age notion connected to god knows what exactly. I took it to be the sort of idea that I, a recent interloper into the discourse of the sciences, in order to establish or maintain some minimal credibility, should avoid.

Around 2000 I was searching for an accessible introduction to biology for my undergraduate literature and science classes, looking for something in the vein of medical researcher Lewis Thomas's popular text The Lives of a Cell, but more recent.²The Lives of a Cell did not mention Gaia by name, but in retrospect it closely anticipated early Gaia discourse: "I have been trying to think of the earth as a kind of organism, but it is no go. I cannot think of it this way. It is too big, too complex I wondered about this. If not like an organism, what is it like, what is it most like? Then, satisfactorily for that moment, it came to me; it is most like a single cell."³ The face and figure of the Gaia hypothesis, I was to learn, would also shift about like this, appearing now as "a single cell," other times as "a kind of organism" or as a "complex" entity of some sort. As it turned out, what I did find that day was Lynn Margulis's popular exposition of her evolutionary theories in the recently released paperback edition of What is Life?, co-authored with her son, Dorion Sagan.⁴ I recollected then that The Lives of a Cell had copped many of its best riffs from Margulis's early scientific work. Popularizing her first book, Origin of Eukaryotic Cells, these "Notes of a Biology Watcher" also vetted its arguments for the pivotal role of symbiosis in cell evolution.⁵ While Thomas was writing those pieces, in fact, Margulis herself had already formed a decisive association with the independent British scientist James Lovelock, and they were well at work developing their original run of co-authored papers on the Gaia hypothesis. Now here in What is Life? was Margulis's own expansive updating of her evolutionary narrative, set forth in equally vigorous and elegant prose.

I began teaching What is Life? flanked by various works of bioscience fiction. It provided an introduction to the Gaia concept, and this was probably my first encounter with an authoritative account. However, it did not bring Gaia forward so emphatically that one had to confront it head-on. I taught this text for several years, concentrating on its main account of deep evolution while otherwise sweeping Gaia off into a neglected corner. Then the semester arrived when instead of assigning the relatively lengthy and rigorous What is Life? I went with Margulis's terse 1998 memoir, Symbiotic Planet. Its final chapter, simply titled "Gaia," retells the name-of-Gaia origin story, with a cautionary twist:

The term Gaia was suggested to Lovelock by the novelist William Golding, author of Lord of the Flies. In the early 1970s, they both lived in Bowerchalke, Wiltshire, England. Lovelock asked his neighbor whether he could replace the cumbersome phrase "a cybernetic system with homeostatic tendencies as detected by chemical anomalies in the Earth's atmosphere" with a term meaning "Earth." "I need a good four-letter word," he said. On walks around the countryside in that gorgeous part of southern England near the chalk downs, Golding suggested Gaia. . . . The name caught on all too well. ⁶

Following upon her intensive collaborations with Lovelock in the 1970s and '80s, Symbiotic Planet intimated Margulis's longstanding concern that "Gaia" as a trademark had exposed the science it covered to severe misconstructions. In an interview with Canadian science broadcaster David Suzuki, Lovelock confirmed his awareness of Margulis's mixed feelings. Regarding the name of Gaia: "Nobody, not even Lynn, liked it. She tolerated it, and was very understanding of its origins and went along with it. But her first reaction was that it wasn't a very good idea at all. It kind of brought up the idea of pagan goddesses and all and didn't fit at all with the atheistic view of science. But of course it almost instantly appealed to the New Age. And I don't say the New Age in any pejorative sense. Because in those days it was a new age."⁷ Lovelock's remarks capture my own experience prior to encountering Margulis's advocacy for the Gaia concept. However, in Symbiotic Planet one passage in particular finally made the idea of Gaia click for me.

It began with this statement—a distant echo of Lewis Thomas?—regarding the proper sense of Gaia theory: "As detailed in Jim's theory about the planetary system, Gaia is not an organism" (119). Nor was it a single cell. Margulis's negative proposition regarding Gaia proper (not "the earth" altogether) began to cut away from my misunderstanding of Gaia the things I had been vaguely worrying about, the fringe metaphysics or planetary vitalisms kept alive, so to speak, by the name of Gaia itself having "caught on all too well," and also, by Lovelock's inclination to foreground organic metaphors in describing Gaia. For her own part, in this passage Margulis rehearsed the finer points of Lovelock's developed presentation of the theory, tethering metaphors more tightly to the science, and giving her own articulation of the concept. "Gaia itself is not an organism," she continued, "directly selected among many. It is an emergent property of interaction among organisms, the spherical planet on which they reside, and an energy source, the sun" (119). So it happened that my initiation to Gaia theory did not come directly from the work of Lovelock, Gaia's primary author, but from the science writing of Margulis. Partial Earth retains that initial orientation in tracing her particular signature on the evolution of Gaia theory.

Margulis's point at that moment was that Gaia as a "living" entity cannot be reasonably submitted to standard evolutionary expectations of reproduction, random variation, survival in competition, and natural selection. Rather, she countered, Gaia is a system. Having placed the organic metaphor into this more abstract perspective, Margulis then figured Gaia's status, not as an "organism," precisely, but as a body: "Gaia, the system, emerges from ten million or more connected living species that form its incessantly active body" (119). Subtle but incisive, this exposition was my Eureka moment. If Gaia is a system, then Gaia theory is a form of systems theory. And not only that: Margulis would go on to treat Gaia theory as autopoietic systems theory and to incorporate the concept of autopoiesis into a range of her popular expositions. This book explores the concept of autopoiesis—the centerpiece of a systems discourse that first emerged in the 1970s under the title "second-order cybernetics"—in particular relation to Margulis's pronunciation of Gaia theory. By the 1990s, Lynn Margulis was coordinating the Gaia concept with the suite of autopoietic systems theories also making their paradigm-changing way against institutional and ideological headwinds.⁸

The neocybernetic line of autopoietic systems theory has developed by expanding the concept of autopoiesis beyond its origin in biological systems theory. For its inventors, the Chilean biologists Humberto Maturana and Francisco Varela, the premier instance of an autopoietic system is the living cell. Regarding the original, biotic form of the concept, living cells are autopoietic because they perform their own production. The fundamental processes of living systems are recursive. Their operations are primordially self-referring. Living systems continuously select and transform the elements they take from their environmental mediums to produce their own continuation and transformation out of their own continuing production of selective transformations. By such incrementally renovating means they maintain both their operational form and their metabolic processes, and hence, the possibility of successfully consorting and coupling with other material, biotic, and metabiotic systems and their environments. In the most effective extension of the concept of autopoiesis beyond this biotic application, Niklas Luhmann's social systems theory differentiated self-producing systems into living and nonliving, or biotic and metabiotic registers, in that, while "nonliving," psychic and social systems can emerge only from living systems. In this regard, when Margulis and Sagan write, "The biosphere as a whole is autopoietic in the sense that it maintains itself," an autopoietic conception of Gaia as a system may denote not a living system precisely, but rather, a metabiotic system—a self-generating constellation emerging from the interactions of living and nonliving elements, embodying their integrated inter-modulations.

Moreover, already in Maturana and Varela's theory of autopoiesis, cellular life's self-referential processes produce a form of cognition. In What is Life? Margulis and Sagan name this ubiquitous self-feeling of living systems sentience. Other authors call it "sense-making," a mode of basic knowing placing any living system in sensory (and potentially, motor) relation to its own communal and material environments. In addition, the nutritive, metabolic, and excretory processes of any living system are at all times, if even infinitesimally, remaking their own niche from moment to moment. Now parlay those micro-dynamics to the worldly macrocosm. Bruno Latour has recently stated this Gaian recognition in his own idiom: "Each agency modifies its neighbors, however slightly, so as to make its own survival slightly less improbable. . . . [T]he concept of Gaia captures the distributed intentionality of all the agents, each of which modifies its surroundings for its own purposes."⁹ As I have now come to think, autopoietic Gaia taps its own modes of planetary cognition from the deep wells of these microcosmic points of biotic sensation.

So, back then, at work at once on two seemingly separate strands of autopoietic systems theory, I saw a way to bring them together. I could now construct Margulis and Sagan's phrasings in What is Life? in a fully neocybernetic sense. Margulis's biotic strand, directly responding to Maturana and Varela's original conceptuality, can join Luhmann's metabiotic strand to resolve one of Gaia theory's most persistent equivocations—the matter of whether Gaia is itself "alive." As an autopoietic system in the metabiotic register, Gaia need not be identified with the form of life per se. Rather, Gaia participates in an essential quality of individual living systems—the autopoietic form of organization, an emergent, recursive form of self-production and self-maintenance, within a metabiotic coupling of abiotic and biotic dynamics. Autopoietic Gaia arises as a "property of interactions" from the inextricable interpenetration of the biota with the seas, the skies, and the rocks, after eons of their own extra-biotic commerce with the dying generations of living forms.Materially, the Gaia of this planet emerged in a fashion unique to the peculiarities of its situation. Although Gaian processes, "other Gaias," are entirely conceivable on other planets harboring life, our Gaia is a planetary one-off. Perhaps this Gaia will produce buds in the form of space-faring closed environments launched from Earth. In the meantime, as a population of one, its autopoiesis may be the most universal property of its organization.

Gaia as Cyborg

Gaia—the blue- and green-hued, whole, living, self-sustaining, adaptive, auto-poietic earth—and the Terminators—the jelled-metal, shape-shifting, cyber-enhanced warriors fighting in the stripped terrain landscapes and extraterrestrial vacuums of a terrible future—seem at first glance to belong in incompatible universes. . . .¹⁰

The Gaia hypothesis advanced to Gaia theory in the 1990s. Lovelock's Daisy World computer models went into circulation to give Gaia some computational cred just as the non-linear recursions of chaos theory were going mainstream alongside the explosion of personal computing, email systems, and the World Wide Web. Lovelock's second book, The Ages of Gaia: A Biography of Our Living Earth, came out in 1988 and consolidated Gaia's scientific discussion. Major scientific meetings had convened by then to take up Gaia's challenge to previous verities in the life and Earth sciences. In the mid-1990s, feminist theorist and historian of science Donna J. Haraway was well positioned to step back from the consolidation of Gaia theory and the popular conceits of Gaia notions to construct her own sense of the Gaia phenomenon. She saw it critically at that moment as a tendentious and overdetermined material-semiotic entity that crossed technoscience with cultural desires.

Haraway was particularly attentive to Gaia's cybernetic origins. Her foreword to The Cyborg Handbook noted that "Gaia is the name that James Lovelock gave in 1969 to his hypothesis that the third planet from the sun, our home, is a 'complex entity involving the Earth's biosphere, atmosphere, oceans, and soil; the totality constituting a feedback or cybernetic system which seeks an optimal physical and chemical environment for life on this planet.'"¹¹ Regarding Lovelock's first book, Gaia: A New Look at Life on Earth, Haraway notes another one of the conceptual tensions or definitive equivocations that upset the placing of the hypothesis into traditional ontological categories: is Gaia's purported homeostasis a material-mechanical or a biological phenomenon? Haraway records how Lovelock's Gaia discourse renders indistinct the difference between living and non-living entities: "the whole earth was a dynamic, self-regulating, homeostatic system; the earth, with all its interwoven layers and articulated parts, from the planet's pulsating skin through its fulminating gaseous envelopes, was itself alive" (xiii). But if Gaia is conceived as a cyborg, then what's the difference? "Lovelock's earth—itself a cyborg, a complex auto-poietic system that terminally blurred the boundaries among the geological, the organic, and the technological—was the natural habitat, and the launching pad, of other cyborgs" (xiii).

Haraway's seminal round-up of cyborg formations in the later twentieth century found much to play upon in Lovelock's text. First, she noted some refraction in his lines of vision. The first line was systems-theoretical—the view informed by his professional and disciplinary location within technoscience: "Lovelock's perception was that of a systems engineer gestated in the space program and the multinational energy industry and fed on the heady brew of cybernetics in the 1950s and 1960s" (xiii). And it was certainly odd that the evocation of a planetary hypothesis called Gaia—"named after the Greek goddess who gave birth (incestuously) to the Titans" (xii)—was the brainchild of an atmospheric chemist-cum-cyberneticist and freelance NASA and Royal Dutch Shell contractor, and "not, say, the intuition of a vegetarian feminist mystic suspicious of the cold war's military-industrial complex and its patriarchal technology" (xiii). Lovelock's second line of vision was astrobiological—the view informed by NASA's pinnacle attainment of a god's eye view of Earth: "in Lovelock's prescient perspective . . . [t]he whole earth, a cybernetic organism, a cyborg, was not some freakish contraption of welded flesh and metal, worthy of a bad television program with a short run. As Lovelock realized, the cybernetic Gaia is, rather, what the earth looks like from the only vantage point from which she could be seen—from the outside, from above" (xiv).

In fact, recent realignments of Gaian vision counter the prior hegemony of the view from space. The orbital view of Earth is not the only vantage from which to observe Gaia. Bruno Latour has underlined this observational shift in commenting on Lovelock that "It was by taking 'the point of view of nowhere' that he showed that there is no 'point of view of nowhere'!"¹² However, at the moment of Gaia's inception at the end of the 1960s and for several decades thereafter, NASA-generated whole Earth imagery was completely in the ascendant with regard to Gaia's planetary imaginary. And as Haraway developed its implications, the classic view of Gaia "from the outside, from above" went well beyond Lovelock's own modest preparations for technoscientific leaps to planetary vistas. Rather, it turned Gaia into the latest episode in the epic of the human species itself arriving at the cosmic threshold heralded by the annunciation of the cyborg.

Haraway's cyborg Gaia expands well beyond a recent theory of planetary function. Gaia's scientific figure covers over its subtext as an ideological project bearing the stigmata of its cybernetic pedigree:

Gaia is not a figure of the whole earth's self-knowledge, but of her discovery, indeed, her literal constitution, in a great travel epic. . . . The people who built the semiotic and physical technology to see Gaia became the global species, in which they recognized themselves, through the concrete practices by which they built their knowledge. This species depends on an evolutionary narrative technology that builds dramatically from the first embryonic tool-weapon wielded by the primal hunter to the transformation of himself into the potent tool-weapon that seeds other worlds.To see Gaia, Man learns to position himself physically as an extraterrestrial observer looking back at his earthly womb and matrix. (xiv)

Cyborg Gaia is a potent allegorical operator, a station on the way to a distinctly virile form of the astronautical sublime rising above the womb of the world. Gaia as a cyborg would be a macrocosmic instance of the human microcosm envisioned by Manfred Clynes and Nathan Kline, who coined the term "cyborg" in 1960 "to refer to the enhanced man who could survive in extra-terrestrial environments. They imagined the cyborgian man-machine hybrid would be needed in the next great technohumanist challenge—space flight. . . . Enraptured with cybernetics, they thought of cyborgs as self-regulating man-machine systems. . . . Space-bound cyborgs were like miniaturized, self-contained Gaias" (xv).¹³ Moreover, Kline himself was engaged in psycho-pharmacological investigations "with Cold War agendas, including CIA-sponsored research on behavior control. The liberal philanthropic foundations, especially the Macy Foundation, which was so important to the configuration of cybernetics as an interdisciplinary field in the late 1940s and early 1950s, were liberally involved" (xvi). Indeed, Gaia was a sort of inspired cyborg hallucination induced by the addiction that had "technical and popular culture . . . shooting up with all things cybernetic in the 1950s and 60s in the U.S." (xvi).

Haraway's discourse is astute in its cultural references and salutary in its demolition of the popular caricature of Gaia as, in the satirical formulation given by Lynn Margulis around the same time, "an Earth goddess for a cuddly, furry human environment."¹⁴ In particular, Haraway reads Gaia as an ideological figuration of systems theory. Her text provides historical instances and contexts for the thing called cybernetics, but no gloss on the term itself. Apparently one should already know all one needs to know about the topic. Moreover, direct references to cybernetics tend to generate flippant metaphors suggesting that cybernetics is a mental agent that will addle your brains. It is a "heady brew"; its enthusiasts are "enraptured," and presumably, not in a good way; it would seem that Cold War technoscience was positively infested with cybernetics addicts "shooting up" the stuff to stave off the fever and shakes of too precipitous withdrawal from their systems-theory habits.

However, there is another, more recondite systems-theoretical term at large in this text, which Haraway does not give a hard time. Marginally glossed but largely unexplicated, it is autopoiesis. Three of the four appearances of autopoiesis in "Cyborgs and Symbionts" brandish a non-standard medial hyphen: "auto-poietic." We have already seen two of them:

Gaia—the blue- and green-hued, whole, living, self-sustaining, adaptive, auto-poietic earth . . . . Lovelock's earth—itself a cyborg, a complex auto-poietic system that terminally blurred the boundaries among the geological, the organic, and the technological. (xi, xiii)

Apart from a single instance that I will document later, Lovelock himself never applied the term autopoietic to his descriptions of Gaia. In no instance does it appear in the Lovelock texts cited in the references of "Cyborgs and Symbionts." Rather, the qualifier Lovelock consistently uses for Gaia's cybernetics is homeostatic. Regarding the text at hand, then, especially at the time of its publication in 1995, one could have wondered about the source of the term "auto-poietic" as applied to Gaia. And while Haraway does not directly state its proximate source, clearly she has extracted it from the texts of Lynn Margulis. "Cyborgs and Symbionts" introduces Margulis as "one of the formulators of the Gaia hypothesis" (xvii), but concentrates on Margulis's writings on symbiogenesis focused on the symbiotic regimes of the eukaryotic microbe Mixotricha paradoxa. An endnote thanks Margulis for having sent over the book manuscript of her then-forthcoming volume What is Life?, described in Haraway's text as "a rich exposition of the travails of the auto-poietic earth" (xvii).

"Cyborgs and Symbionts" left the meaning of autopoiesis unarticulated until the provision of a minimal bracketed gloss in the midst of a passage quoted from an earlier, published Margulis and Sagan text, Origins of Sex:

From an evolutionary point of view, the first eukaryotes were loose confederacies of bacteria that, with continuing integration, became recognizable as protists, unicellular eukaryotic cells. . . . The earliest protists were likely to have been most like bacterial communities. . . . At first each autopoietic [self-maintaining] community member replicated its DNA, divided, and remained in contact with other members in a fairly informal manner.¹⁵

This passage from Origins of Sex coordinates the concepts of autopoiesis and symbiosis within Margulis's signature theory of symbiogenesis, or the multiple endosymbiotic events in the evolutionary formation of the eukaryotic cell out of prokaryotic, or bacterial, components. What one doesn't learn in "Cyborgs and Symbionts" is that Margulis drew her presentation of autopoiesis out of crucial developments in biological systems theory that arrived in the mid-1970s. Maturana and Varela originally presented that concept as a criterion for distinguishing living systems—minimally, prokaryotic cells—as autopoietic or self-producing, as opposed to mechanical, technological or designed systems as allopoietic or "other-producing," that is, as having their maintenance-in-being outsourced to some external agency. In other words, the concept of autopoiesis deconstructed the original cybernetic splice between animals and machines, between living and non-living systems. Its virtual effect is to turn the trope of the cyborg inside out, cutting apart its two halves once more through a distinction of operation. Moreover, throughout Margulis and Sagan's What is Life?, the concept of autopoiesis is coordinated not just with the form of life of organisms but also with the particular form of the geobiological operations of the system called Gaia. And it was the Gaian application of the term autopoiesis that appears to have caught Haraway's eye in Margulis's text, which she then retailed without further comment. Autopoiesis seems to hide in plain sight in "Cyborgs and Symbionts," covering over a blind spot in its observation of systems.

As Haraway's later writings from When Species Meet to Staying with the Trouble will attest, she has continued to wrestle with the demon of autopoiesis, trying at times to get her readers to kick this pernicious cybernetic habit.¹⁶ Regarding her own dependency, however, her recent work has struck a sort of compromise formation under the symptomatic name of sympoiesis. That is a story for another occasion. However, in the context of this study, Haraway's text provides a wonderful testimony to the power as well as the problematics of the idea of autopoiesis. Additionally, Haraway's travails with the neocybernetics of autopoiesis bring out the precise conceptual twist that accounts for why this notion looks askance at some of her paradigms. Let us revisit her crucial statement from "Cyborgs and Symbionts" in this regard: "Lovelock's earth [was] itself a cyborg, a complex auto-poietic system that terminally blurred the boundaries among the geological, the organic, and the technological" (xiii).

Such "terminal blurring" has always been the battle cry of cyborg discourse. "Boundary breakdowns" and "leaky distinctions" are the very stuff of the postmodern, provocative, and critically productive "Cyborg Manifesto."¹⁷ However, along with the intellectual liberations induced by the breakdown of "boundaries" around the human, the animal, and the machine, around the material and the semiotic, the actual and the virtual, the physical and the informatic, in the subsequent critical literature such terminal blurrings have also led to significant instances of terminological haziness. The problem that autopoiesis brings into a cyborg world is precisely that it is a theory that posits boundary production for those systems that exhibit the autopoietic form of organization and operation. Thus, to call "Lovelock's earth," that is, Gaia, "a complex auto-poietic system that terminally blurred the boundaries among the geological, the organic, and the technological" is to let the technical sense of autopoiesis go by as well as to leave hanging what Lynn Margulis may have intended to convey by insisting on the characterization of Gaia as an autopoietic system. Without question, under other names of her own construction, Donna Haraway has been a profound thinker of Gaia and an early explorer of Margulis's autopoietic Gaia conception.¹⁸ Nonetheless, the cyborg description of Gaia blurs its boundaries as a matter of course. The autopoietic description of Gaia has the opposite aim. This mode of description insists on observing material and operational distinctions among "the geological, the organic, and the technological," in order to bring out a finer order of attention in the construction of their systemic couplings and compositions.

Autopoiesis in Margulis

While "autopoiesis" did not appear in the first, 1981 edition of her primary evolutionary text within her own discipline, Symbiosis in Cell Evolution, multiple entries for "autopoiesis" and "autopoietic systems" appear in its next, 1993 edition.¹⁹ Margulis now gave prominence to the concept of autopoiesis not only in regard to Gaia theory but also as running the ecological gamut from the cell to the biosphere. In contrast, the first author of the Gaia hypothesis, James Lovelock, seldom mentioned autopoiesis. Lovelock was an admirable participant in the main line of engineering cybernetics, and he continued to couch his own Gaia discourse along primarily first-order cybernetic lines. In Lovelock's main conception, Gaian self-regulation is a planetary instantiation of cybernetic homeostasis. As Margulis developed her own treatment of Gaia, it becomes the "autopoietic planet," for which homeostasis is unfolded as a neocybernetic form of recursive self-production. Bootstrapping the concept of autopoiesis to her work on symbiosis and environmental evolution, Margulis complemented and extended Lovelock's work by taking the science of Gaia on a distinctly neocybernetic path. The concept of autopoietic Gaia was a way to keep her symbiotic planet anchored to the biota. At the same time, her theorizations traced for Gaia a metabiotic course beyond strictly organic status as a post-biotic coupling that coordinates both geological and technological dynamics with the operational closure of living systems.²⁰

The texts I discuss here were begun as early as 1981 and first published between 1986 and 1991. Margulis and Sagan developed the books Microcosmos and Origins of Sex concurrently; both were first published in 1986. In December 1985, Margulis wrote Lovelock an extraordinary letter with her most vigorous speculative extension of autopoietic Gaia to date. With echoes of both Microcosmos and Origins of Sex, this letter sketches out the content of a popular article then in draft, lead-authored by Sagan, "Gaia and the Evolution of Machines," published in 1987. This text in particular indicates how the Gaia concept consistently hails its cybernetic origins and so calls upon its theorists, Lynn Margulis in particular, to consider its neocybernetic reformulation as autopoietic Gaia in relation to non-autopoietic systems, specifically, technological systems. A few years later, Margulis returns on her own to the conceptual intersection of autopoiesis and Gaia in a linked pair of articles that constitute her most concerted treatments of the autopoietic idea in the immediate vicinity of Gaia theory, with a vigorous take-down of the neo-Darwinist orthodoxy of that moment.

Microcosmos

The consolidation of Gaia theory by the early 1980s informs the entire evolutionary narrative of Microcosmos: Four Billion Years of Evolution from Our Microbial Ancestors. First of all, the concept of the microcosm is already planetary in scope. For instance, Margulis and Sagan's consistent critique of bastardized Darwinisms shifts the accent from the competition for survival of individual organisms to life as a self-reinforcing aggregation: "the view of evolution as chronic bloody competition . . . dissolves before a new view of continual cooperation, strong interaction, and mutual dependence among life forms. Life did not take over the globe by combat, but by networking."²¹ Moreover, life's planetary take-over did not have to wait for animals or plants. The early bacteria got there around three billion years ago with a wide-open evolutionary lottery driven primarily by the recombinatory dynamics of lateral gene transfers: "The result is a planet made fertile and inhabitable for larger forms of life by a communicating and cooperating worldwide superorganism of bacteria" (17). The concept of autopoiesis enters Microcosmos in the context of the origin of life. The initial statement on autopoiesis concentrates on the maintenance of structural integrity through the self-maintenance of the autopoietic identity:

To be alive, an entity must first be autopoietic—that is, it must actively maintain itself against the mischief of the world. Life responds to disturbance, using matter and energy to stay intact. An organism constantly exchanges its parts, replacing its component chemicals without ever losing its identity. This modulating, "holistic" phenomenon of autopoiesis, of active self-maintenance, is the basis of all known life. All cells react to external perturbations in order to preserve key aspects of their identity within their boundaries. (56)

As yet unremarked is the recursive form of the autopoietic operation, which circular dynamic drives the self-production of those self-maintaining mechanisms. The scare-quoted term "holistic" does not yet capture the recursive or self-referential form of autopoietic dynamics with any specificity. The manner of autopoietic recursion becomes marginally explicit only in the final chapter, "The Future Supercosm." Speculations about the possibility of taking terrestrial life successfully into extraterrestrial environments now elicit this work's most extensive rehearsal of the Gaia concept. The passage in question sketches the mechanistic paradigm that the neocybernetic view will supersede. Classical physical views of living dynamics based on the science of Descartes and Newton were linear rather than recursive. To the detriment of both popular and professional scientific ideas, this linear hangover remains the case with most of the machine cybernetics and information theory then fashionably being applied to living systems through "computer-age analogies: amino acids are a form of 'input,' RNA is 'data-processing,' and organisms are the 'output,' the 'hard copy' controlled by that 'master program,' that 'reproducing software,' the genes" (264). Autopoiesis enters this section of Microcosmos precisely to rebut such bioinformatic computationalism: "we have held to a somewhat different and more abstract view. . . . Life, a watery, carbon-based macromolecular system, is reproducing autopoiesis. The autopoietic view of life is circular" (264).

"Life . . . is reproducing autopoiesis" is an interestingly compressed but viable formulation. What is still not immediately clear from the passage itself is precisely in what way the "autopoietic view of life" is "circular." Gaia makes its entrance at this point, however, and provides the terms necessary to resolve the sense of these conceptual constructions, if just barely. It arrives with this introduction: "the freelance atmospheric chemist James Lovelock sees life best represented by a self-supporting environmental system which he calls Gaia" (265). What Lovelock's Gaia has just brought with it as an "environmental system" is the currently missing component of its proper construction. This importation of atmospheric and climatic materiality into the discussion provides an antidote to the biocentric super-organicism still at large in Margulis and Sagan's early formulations, for instance, here: "Gaia, the superorganismic system of all life on Earth, hypothetically maintains the composition of the air and the temperature of the planet's surface, regulating conditions for the continuance of life" (265). And again, in the following strictly biotic phrasing of autopoietic Gaia: "According to Lovelock's idea, which he calls the Gaia hypothesis, the biota itself, which includes Homo sapiens, is autopoietic. It recognizes, regulates, and creates conditions necessary for its own continuing survival" (266). This attribution to Lovelock is in fact a projection of Margulis and Sagan's current theorizing, in which the microcosm operates as a synecdoche for "the biota itself" as a superorganic system. However, the finer part of this construction rests in its salient observation of autopoietic Gaia as a system of planetary cognition arising from environmental recognitions.

With the following statement Margulis and Sagan interlock Gaian and autopoietic circularity, in the appropriate rhetorical form of a discursive chiasmus. Here autopoietic Gaia finds its verbal form through the depicted reciprocity of a systemic coupling of life and Earth: "On earth the environment has been made and monitored by life as much as life has been made and influenced by the environment" (265). Let us quickly compare to this hard-won Gaian construction in Microcosmos one of Margulis's own formulations of Gaia over a decade later in Symbiotic Planet. We have already noted a part of this passage. Here the biotic bias at large in Microcosmos is, if not entirely eliminated, largely corrected: "The sum of planetary life, Gaia, displays a physiology that we recognize as environmental regulation. Gaia itself is not an organism directly selected among many. It is an emergent property of interaction among organisms, the spherical planet on which they reside, and an energy source, the sun."²² This passage retains a summative biotic formulation but then grounds it in Earth processes. The rejection of an organismic description is another way of saying that Gaia, unlike any literal organism, does not reproduce and so leaves no progeny that may be more or less "fit" for natural selection. As such, this statement retains a vestige of autopoietic conceptuality even while "autopoietic Gaia" does not rise to utterance in this passage of Margulis's memoir. Rather, in this statement, life, Earth, and sun have fallen into place as the "coupled system" of Lovelock's mature idiom. A subsequent statement nicely reinforces this formulation: "Gaia is the regulated surface of the planet incessantly creating new environments and new organisms. . . . Less a single live entity than a huge set of interacting ecosystems, the Earth as Gaian regulatory physiology transcends all individual organisms" (120). Gaia's operational closure around the flows of solar radiation and terrestrial convection drive a cyclical interplay between life and Earth that is metabiotic in the final instance.

Origins of Sex

I will be glad to send you over Origins of Sex book ms (proof copy) if you'll really read it. In it I think we handle replication vs reproduction vs sex vs recombination vs autopoiesis vs mutation and evolution etc in a clear modern chemical way.
—Lynn Margulis to James Lovelock, December 7, 1985

Also published in 1986, Origins of Sex does not deserve its relative obscurity within the Margulis and Sagan oeuvre. Such oversight may be due to its sedate pacing, high level of technical detail, and lack of provocative discursive framings—in sum, its old-school editorial values under the direction of Yale University Press. In any event, its opening chapter, "What is Life? DNA, Autopoiesis, and the Reproductive Imperative," is still a highly serviceable inscription of autopoiesis as the primary process of self-production into a basic exposition of cellular and molecular biology. Varela, Maturana, and Uribe introduced the concept of autopoiesis in the 1970s as a conceptual retort to biological theory's over-emphasis on molecular genetics as the sole driver of evolution. Margulis and Sagan are at pains in this text both to honor that prior call to account for living organization and to integrate that corrective supplement into the standard presentation.

Origins of Sex makes room for autopoiesis by qualifying the function of metabolism. Autopoiesis names the principle of the imperative for continuous self-production, and "metabolism is the mechanism of autopoiesis."²³ Essentially, there are not one but two, intimately intertwined, living imperatives. In the first instance, autopoiesis must accomplish the operational continuity of a living system in its own right and at least until the arrival of an organism's reproductive capability. Then, on that basis, for life to continue beyond the finite time of that cellular or organismal self, reproduction must realize genetic continuity across generations. Meanwhile, despite anthropomorphic neo-Darwinist tales to the contrary, genes have no such desires. Macromolecules "are indifferent to existence: chemical systems have no priorities" (12). The introduction of autopoiesis into these descriptions also displaces the function of reproduction from its stereotypical presentation as the supreme expression of living beings and brings it back into relation with its inexorable preconditions. Nevertheless, at the same time, the authors render autopoiesis in its necessarily contingent relations to genetics and reproduction:

Autopoiesis occurs, then, to maintain an organism during its own life, but by itself autopoiesis does not guarantee that an organism will show genetic continuity or that the characteristics of any given organism will persist faithfully through time. The process that ensures genetic continuity is reproduction. But autopoiesis remains the primary process. On the one hand, without it the organism would not survive to reach the stage at which reproduction becomes feasible. On the other hand, autopoiesis does not depend on reproduction, at least within a single generation. (13)

The Gaia concept does not fit the topic of Origins of Sex and does not make an appearance there. Instead, Margulis and Sagan develop their account of serial endosymbiosis in the evolution of the eukaryotic cell in relation to their related theories about the evolution of sexual reproduction, which occurs well after eukaryosis has been stabilized. However, they also make some observations that bend in the direction of the higher-order ecologies for which the Gaian system could be considered the final iteration. The chapter "Meiosis and Cell Differentiation" begins, "A central thesis of this book is that the eukaryotic cell is homologous to a community of microorganisms" (170). It characterizes the "eukaryotic individual" as the systemically integrated sum of its bacterial precursors, whose separate genetic residues are not entirely bound within the nucleus but are also distributed throughout the cell as a whole:

All eukaryotic individuals must reserve, in a form capable of continued reproduction, their genetic components, the remnant bacteria in the combined form of the nucleocytoplasmic, mitochondrial, plastid, and undulipodial genomes. If we accept the cell as a microbial community, the germ plasm is equivalent to component autopoiesis: a complete set of heterologous genomes and their protein synthetic systems contained within a membranous package—not the nuclear membrane but the plasma membrane. (175-76)

Now that the discussion has shifted away from the opening chapter's rehearsal of the minimal imperatives of a basic prokaryotic being such as a bacterium to the more demanding processes needed to maintain the eukaryotic cell's integration and synthesis of its heterogeneous genetic inheritance, what arrives is an articulation of "component autopoiesis." Moreover, "We can apply the principles of community ecology directly to the development of the individual" (176). If we return Margulis and Sagan's cellular extension of community ecology back to its primary reference, then we already inhabit a theory of the ecosystem for which "component autopoiesis" drives the necessarily higher-order forms of community self-production and self-maintenance. Ecological communities have shifting but relatively stable identities. Site-specific, they endure and mature. However, even while their living components carry on their reproductive ways, insofar as they also maintain their individual autopoieses and contribute them to a group dynamic, such communities do not reproduce their organization in the form of "baby ecosystems." Rather, their continuity over time must emerge from the composite maintenance of the ecosystemic consortium. We are thus a step or two closer to a description of the operational sense of autopoietic Gaia as a self-producing but non-reproducing entity.

"Once-Autopoietic" Systems

On December 7, 1985, Lynn Margulis wrote James Lovelock a long letter I will quote at some length for its superb documentation of the intensity with which Margulis took up the issue of autopoiesis at this stage of her theoretical engagement with Gaia theory. Her letter's overriding topic is the concept of autopoiesis in relation to the fate of technology:

Machines (like hives, shells, teeth, nests, stromatolites) can be part of autopoietic systems of course . . . . Like viruses—because they do not exchange chemical components w. environment to keep form & info intact at expense of solar energy (eg, don't metabolize) they ain't autopoietic systems alone—they can't be—but they are clearly part of autopoietic systems. . .

In the main part of this passage, I take Margulis to be saying that machines are like hives or teeth because they are the non-living but post-biotic and reorganized material extrusions of living systems. From the bacteria onwards, living systems produce an organic technics that alters the composition of their environments in a manner that may become habitual if beneficial for their autopoietic and reproductive continuations. On evolutionary occasion, externalized or excreted substances (such as metabolic wastes) may be reincorporated into the body plans of individual organisms, in the manner of shells or teeth, or collectively deposited into their immediate environment in the manner of hives or stromatolites, the latter being the massive calcified inhabitations of certain bacterial communities going back to the Archean eon. Moreover, the propensity of living systems to evolve by incorporating such environmental affordances (the horizontal gene transfer of ambient plasmids, the uptake of free material elements) within their operational boundaries would be the corporeal instance of the cognitive inscription of life forms. As a reciprocal outcome of their self-reference as observing systems bringing forth their world, living systems write themselves ever further into the worlds they observe.

In similar fashion, the building out of the technosphere is not merely instrumental; its literal coming into being as a substantial material production also alters the planetary environment that absorbs its components and infrastructures into its existing, thereby transformed texture. Machines are "clearly part of autopoietic systems," then, in a view of autopoietic Gaia that considers the technosphere as a differentiated continuation of the geosphere. Machines so considered are non-autopoietic structures either produced by and within the boundaries of an autopoietic system or tightly networked with such systems as intimate environmental affordances. Technology produces a further recomposition of the environmental medium. The differentially coupled metabiotic operations of psychic and social systems effect the deposition of technological systems into the world as environmental resources, which operations are then transformed by the network of machines that emerge in their midst as metabiotic yet non-autopoietic forms of mediation.

Margulis's letter to Lovelock goes on:

Viking landers (both) are great examples. They were autopoietic from their inception ("creation", "birth"—I don't like these words but go ahead with them if you must) until ≈ 1983 when they were "turned off" by mission control. No longer part of Gaia or any other autopoietic system they are passively insulted by exposure to the disintegrating influences of the Mars surface. They are now, today examples of nonautopoietic systems even though at one time they were autopoietic by virtue of their connections with other life. Like dead animals they are "once-autopoietic" systems.

The example of Viking landers gestures directly back to Lovelock's incubation of the Gaia hypothesis as a NASA contractor during the Mariner and Viking missions. In the introduction to The Ages of Gaia, Lovelock would offer a mordant Ballardian glimpse of these abandoned machines, monuments to their own failure to fulfill NASA's exobiological dreams of Martian life: "Their mission was to find life on Mars, but the messages they returned as radio signals to the Earth returned only the chill news of its absence. . . . The two Vikings now sit there brooding silently, no longer allowed to report the news from Mars, hunched against their final destruction by the wind with its burden of abrasive dust and corrosive acid."²⁴ Whoever first composed this vision of the entropic disintegration of Earthly inceptions on the surface of Mars, here is a striking discursive revelation of the ongoing productivity of the Lovelock-Margulis collaboration. The comparison of these particular passages also underscores how avidly Margulis was bent at this moment on thinking through and with the autopoiesis concept. Her powerful locution of "once-autopoietic" still awaits its full development. However, its relevance to Gaian thought is immediately evident. For instance, we tend to dismiss the sheer materiality of Earth's crust and mantle as "non-living." Yet we might grasp the biosphere altogether more accurately, after several billion years of Gaian operations, as "once-autopoietic." The air we breathe and the rocks on and with which we build have come into being largely through the depositing of post-biotic residues processed through the self-producing operations of living systems.

"Gaia and the Evolution of Machines"

The 1987 article co-written by Dorion Sagan and Lynn Margulis, "Gaia and the Evolution of Machines," refines the autopoietic treatment of Gaia theory in relation to the technosphere sketched out in Margulis's December 1985 letter to Lovelock.²⁵ Sagan and Margulis began by deferring priority to Lovelock, as Margulis herself did not herself participate in the original gestation of Gaia's first descriptions. "The Gaia Hypothesis was invented by James Lovelock over twenty years ago to explain the tendency of the Earth's lower atmosphere to maintain for millions of years its temperature, oxygen concentration and alkalinity within rather narrow limits" (15). Nevertheless, they draw the idiom of their own description from the autopoietic language developed in Microcosmos and Origins of Sex. They underline Gaia's own systemic self-maintenance as logically prior to its maintenance of planetary variables, in that it is the "self-maintaining properties of cells, organisms, communities and ecosystems" that "can be extrapolated to the atmosphere and surface sediments of the planet Earth" (15). This implicitly autopoietic Gaia concept then adds the technosphere to the Gaian orbit: "Not only are members of the more than 10 million existing species components of the Gaian regulatory system but so are our machines. Here we argue that although not by themselves alive, like viruses and beehives, machines are capable of reproduction, mutation and evolution. That is, even though they are not autopoietic, machines do evolve" (15). These passages suggest that the Gaian matrix absorbs the technosphere within its operations, and not as so many at this moment appear to think, the other way around. Margulis's Gaia discourse in this regard turns standard notions of environmentalist separatism upside down.

Let us recall the opening passages of "Autopoiesis: The Organization of Living Systems": "reproduction and evolution are not constitutive features of the living organization . . . all biological phenomenology, including reproduction and evolution, is secondary to the establishment of this unitary organization," that is, secondary to the establishment of the autopoietic organization.²⁶ Sagan and Margulis place their current argument on the same discursive tracks. However, they repurpose Maturana and Varela's logical architecture. They endorse the autopoietic organization as the prime criterion of living systems. Then they take reproduction, genetic mutation, and evolution—the same history-bound and ontogenetic qualities that, according to the autopoietic critique, mainstream biology has misplaced as the prime criteria of life per se—and transfer them to the secondary, metabiotic realm of machines, of designed technological systems. In other words, while retaining living systems' exclusive title to autopoietic self-production, they put the mechanistic side of modern biological theory back where it belongs, on the description of machines. Machines are non-autopoietic and yet metabiotic. They do not possess autopoietic operations: they do not arise directly from the Earth, only as mediated by biotic and metabiotic systems. Nevertheless, as adjacent to them, by participating with them, and through their particular intimacy with psychic and social systems, they achieve a kind of phantom autonomy.²⁷ Under Sagan and Margulis's newly revised techno-logical scheme, Gaia becomes the metabiotic matrix within which autopoietic and non-autopoietic systems couple their distinct operations.²⁸

Later in the article, borrowing facets of the autopoietic conceptuality developed in Origins of Sex, Sagan and Margulis term the particular quality of this active matrix of coordinated operations consortial and apply it to the "community ecology" of both the biological individual and the Gaian consortium:

The consortial quality of the individual preempts the notion of independence. For example, what appears to be a single wood-eating termite is comprised of billions of microbes, a few kinds of which do the actual digesting of the cellulose of wood. Gaia is the same sort of consortial entity but she is far more complex. Consortia, associations, partnerships, symbioses, and competitions in the interaction between organisms extend to the global scale. Living and nonliving matter, self and environment are inextricably interconnected. (16)

With these Gaian formulations, Margulis and Sagan effectively remediate treatments of complex consortia such as "biospheres" and "technospheres" that lose operational differentiation. Their approach wards off the kind of analytical fade-out commonly encountered in the general run of discourse upon cyborgs, hybrids, quasi-objects, networks, and other assorted conceptual tropes of boundary transgression and ontological fusion. They demonstrate the saving utility of the concept of autopoiesis as a knife-edge for keeping operational distinctions sharp, for prying apart distinctions among, as Margulis riffed in her letter to Lovelock, "replication vs reproduction vs sex vs recombination vs autopoiesis vs mutation and evolution etc."

"Gaia and the Evolution of Machines" builds to a long passage applying the autopoiesis concept to discern the proper discontinuities of operation between biotic, abiotic, and metabiotic domains. It leverages the primal distinction between autopoietic form and reproductive history to indicate more finely the complete interdependence and mutual specification of these differential operations at the ecological scale. Still, autopoiesis takes precedence:

Although there is an ineffable continuum between the living and the nonliving, we are beginning to understand the functions and organizations that are common to living entities. Living systems, from their smallest limits as bacterial cells to their largest extent as Gaia, are autopoietic: they self-maintain. As autopoietic systems they are bounded—they retain their recognizable features even while undergoing a dynamic interchange of parts. . . . Autopoiesis is a prerequisite to reproduction . . . . Components of autopoietic systems reproduce. The reproduction of autopoietic systems depends on the autopoiesis of the components of such systems. (18)

This passage affirms and extends the idea of "component autopoiesis" introduced in Origins of Sex to account for the autopoietic form of the prokaryotic consortium that becomes the eukaryotic cell. Now they take component autopoiesis up to the Gaian instance at the zero degree of its dependence on the reproductive continuity of its autopoietic components. At the same time: "Machines reproduce. Alone, they do not self-assemble. They do not self-maintain: machines alone are insufficient parts of autopoietic systems. Despite our machineless past, however, our autopoiesis now depends on machine organization in much the same way that cells of our body depend on human organization (anatomy and physiology)" (19).These thirty-year old remarks already indicate the mutual dependence of contemporary humanity and its technosphere, but from the side of its coupling of autopoietic and non-autopoietic components.

The interdependence between contemporary human life and machine reproduction, over and above the reproduction of Gaia's living components, presses the purview of autopoiesis beyond its biotic base. Taking the premises of the machine issue when framed by the concept of autopoiesis to their logical conclusions, Sagan and Margulis independently retrace the metabiotic course of that concept's neocybernetic development toward the self-producing operations of technological society and the social reproduction of communications. Moreover, "The reproduction of technological societies and their components is part of the autopoiesis of the biosphere" (19). And,

From a biospheric view, machines are one of DNA's latest strategies for autopoiesis and expansion. The classification of machines as non-autopoietic and nonliving does not negate the fact that they reproduce, and reproduce with mutation, as avidly as viruses. Like beehives, termite mounds, coral reefs, and other products of the activity of life, machines—if indirectly through DNA and RNA—make more of themselves. Through us they make other machines. (19)

In what may well be an oblique satire of Richard Dawkins's selfish-gene concept, this passage completes the transfer of neo-Darwinism's cherished biological priorities from the blind dispersion of bodily phenotypes to the designed and directed evolution of machines. The Gaia concept dismissed by Dawkins now presides over "DNA's latest strategies for autopoiesis." Genetic determinism implodes within a creative and combinatory metabiotic biosphere now in the process of sending out mechanical spores to other planets.In this vision of a multiply-coupled autopoietic Gaia, humans are variously entrained parts of the technosphere, but that network or grid is itself a non-autopoietic part of this Earth's biosphere and as such takes part in its incessant modification, not just of the geosphere, but also, incrementally, of the cosmic environment:

The Viking Lander on the surface of Mars does not maintain its own structure or actively preserve its boundaries. Alone, lacking communication, it is no longer autopoietic. But from 1975 to 1982, when all of its communication with the Earth was halted, even the Viking Lander was part of an autopoietic system. Machines, by themselves on Mars, are not autopoietic. Machines tended by their workers form part of the autopoietic systems of their makers. (19)

The autopoietic Gaia discourse Lynn Margulis in particular developed through neocybernetic systems theory anticipated the discourse of the Anthropocene technosphere by several decades. A Gaian discourse of the technosphere indicates the need to maintain the biospherical bona fides of machine beings. Margulis and Sagan fastened upon a mode of systems description founded on the self-produced membrane-bounded operational closure of living systems, the cellular organization, in the midst of those systems' ultimate capacities in the fullness of evolutionary time to arrive at higher-order autopoietic consortia of pre-evolved components. Gaia itself exhibits these improbable but evolutionarily successful metabiotic couplings of living ecologies, geological formations, and technological systems across its planetary interface; mutual feedbacks of living and non-living processes that continuously remix the system; and deeply interfolded differential effects of Earth and life processes, including minds and societies. While its own autopoiesis emerges in particular from the integration of those subsystems bearing biotic organizations, Gaia's manifold responsiveness to its own planetary situation goes beyond the biotic occasion and propagates as well from both its abiotic and metabiotic components. Gaia has always partaken of the air, the rocks, and the oceans. In the era called the Anthropocene it also wraps itself globally around technological processes and productions, the processes of combustion necessary to run them, and the waste heat and cast-off substances they vent. A strange brew indeed. Just how lethal or not it becomes we will surely find out.

Replacing Neo-Darwinism

The big trouble in biology is directly related to the big trouble in our social structure and its priorities. This is a big subject. —Lynn Margulis, "Big Trouble in Biology"

Sagan and Margulis conclude "Gaia and the Evolution of Machines" in a relatively detached, fatalistic, and futuristic manner. Speaking out of Ronald Reagan's America, the authors note in one of their final remarks that "Neutron bombs, capable of destroying our human bodies, are expected to leave the more durable machine systems intact. The future of our machines, provided they can remain part of the autopoietic biosphere, is less bleak than that of ourselves" (21). A few years later, Margulis returns on her own to the conceptual intersection of autopoiesis and Gaia in a pair of articles that constitute her most concerted treatments of the autopoietic idea in the immediate vicinity of Gaia theory. Around 1989-90 Margulis composes twin essays at the nexus of Gaia and autopoiesis connected by a fair amount of textual overlap, a shared sense of cultural distress, and a dramatic increase in institutional animus. One of them is delivered within the biological academy: Margulis first presents "Kingdom Animalia: The Zoological Malaise from a Microbial Perspective" to a plenary session on "Emerging Systems: Molecules, Genes and Cells" at the centennial meeting of the American Society of Zoologists.²⁹ The other article returns to the extra-disciplinary and putatively avant-gardist venue of John Brockman's Reality Club trade publications: "Big Trouble in Biology: Physiological Autopoiesis versus Mechanistic Neo-Darwinism" launches an extraordinary diatribe against the biological establishment in particular and big science in general.³⁰

"Kingdom Animalia"

In "Kingdom Animalia: The Zoological Malaise from a Microbial Perspective,"Margulis informs an audience of dedicated zoologists celebrating the 100th anniversary of their professional association why their views on animal biology participate in an intellectual disorder rooted in the overextension of insular and philosophically misguided concerns to the rest of life and the planet in general. In the original published version, this combative tone is immediately established by a remarkable "Synopsis" that begins with two images of bodily trauma: "Pain and cognitive dissonance abound amongst biologists: the plant-animal, botany-zoology wound has nearly healed and the new gash—revealed by department budget reorganizations—is 'molecular' vs. 'organismic' biology. Here I contend resolution of these tensions within zoology requires that an autopoietic-gaian view replace a mechanical-neodarwinian perspective."³¹

"Kingdom Anmalia" begins with a preliminary rehearsal of how to break out of the battered institutional and conceptual shackles of the "the plant/animal dichotomy" once and for all and to see animals from her announced "microbial perspective," as "embedded in the context of their microbial predecessors. They are not 'superior,' or 'higher' forms of life to be contrasted with the 'lower' animals and 'higher' plants. Rather, animals are peculiar, if familiar, descendants of coevolved microbial communities" (862). This classic Lynn Margulis move prepares one to carry their evolutionary vision back to the Archaean scene before the emergence of species specificities, at which point later differentiations among animals, plants, and fungi dissolve into the microcosmic commons from which all eukaryotic cell forms first arose through the symbiogenetic mergers of the ur-bacteria. The autopoietic form of the living is rooted here with the origin of life altogether, as Gaia first arises in the midst of an entirely prokaryotic world. From this microbial vantage, Margulis now shifts the discussion directly to autopoietic Gaia as representing the non-zoocentric worldview that must supersede modern biology's prostration before the humanistic gods of philosophical mechanism in the doctrinal commitments of neo-Darwinism. Margulis's current treatment of autopoiesis is worth drawing out at some length, because here she provides her professional audience with an unusually high level of physiological and biochemical detail. This finer grain of presentation derives in part from the recently completed doctoral dissertation of her student Gail Fleischacker, to whose work she makes repeated references in her autopoietic writings from this period.³²

What becomes particularly clear in this presentation is how the original theory of autopoiesis is rooted in physiological biochemistry rather than upon the neo-Darwinist fetish of molecular genetics. This orientation is patent in the original conceptuality put forth by Varela, Maturana, and Uribe in 1974: "Consider for example the case of a cell: it is a network of chemical reactions."³³ We also recall that her letter to Lovelock of December, 1985, discussing the forthcoming publication of Origins of Sex, noted her satisfaction that "we handle replication vs reproduction vs sex vs recombination vs autopoiesis vs mutation and evolution etc in a clear modern chemical way" (my italics). Margulis unfolds her autopoietic perspective now with increased biochemical detail. The section of "Kingdom Animalia" titled "Autopoietic Gaia to replace neodarwinian mechanics" begins:

Autopoiesis, a term invented by Maturana and Varela (1980) and elaborated by other authors (Fleischaker, 1988) refers to the living nature of material systems. Well within the materialist view that recognizes the physical-chemical composition of organisms, autopoiesis refers to the self-making and self-maintaining properties of living systems relative to their dead counterparts. Autopoietic, unlike mechanical, systems produce and maintain their own boundaries (plasma membranes, skin, exoskeltons, bark, etc.). Autopoietic systems incessantly modulate their ionic composition and macromolecular sequences (i.e., amino acid and nucleotide residues in their proteins and nucleic acids). (865-66)

This autopoietic materialism is reinforced by the concreteness of its physiological examples (the "plasma membranes" of cells, the "bark" of plants) and by the biochemical specifics related to the "incessant modulation" of intracellular processes. As Margulis now shifts to a definition of Gaia, she carries over the systems inflection of cellular chemistry to Gaia's own "autopoietic" regulation of planetary chemistry: "Gaia is defined as the large self-maintaining, self-producing system extending within about 20 kilometers of the surface of the Earth. The Gaia hypothesis states: the surface sediments and troposphere of the Earth are actively regulated by the biota (the sum of the live organisms) with respect to the chemical composition of the reactive elements (e.g., H, C, N, O, S), acidity (e.g., H+,OH-, CO3--, HCO3-), the oxidation-reduction state and the temperature" (866).

I will stay on this course of chemical interest for the moment. For alongside the polemical content, which will only intensify in "Big Trouble in Biology," Margulis carries the theory of autopoiesis farther up the phylogenetic line from the usual idealized autopoietic cell of Maturana and Varela. This physiological specificity makes the ultimate destination of autopoietic Gaia that much more plausible. The following passage offers several striking examples of salutary concreteness in presenting a "chemically self-conscious, autopoietic point of view" and also provides a blunt reminder of how the "autopoietic imperative" mixes life and death together in the chemical matters of Gaia's persistence over the eons.

Why do Pacific salmon swim upstream to die in the area where they themselves spawned? A neodarwinian uses military or economic terms, tending towards an explanation in terms of "reproductive strategies," of offspring outcompeting others with fewer genes in common. In the autopoietic point-of-view, attention is paid to the chemical components of the fish. For example, that the dead bodies of the upstream adult salmon provide phosphorus for the diatoms that, during the next season, serve as food for salmon fry.

Another example: Why do small quantities (less than or equal to 0.5 ml inocula) of certain bacteria added to fresh growth solution not grow whereas larger ones (greater than or equal to 1.0 ml) grow well? The observation that death of the organisms comprising nearly the entire inoculum provides conditions for growth of the few remaining bacteria is described as pure "altruism" (and thus rejected) by neodarwinians. From the chemically self-conscious, autopoietic point of view it is sufficient to recognize that component lipids and other compounds shed by a large inoculum provide sufficient ambient conditions, probably including food, for the initial growth of at least a few of the bacteria in pure culture. At least ten orders of birds contain species in which parents or nestmates eat their offspring. Cell death, tissue resorption and cannibalism are common means for the autopoietic imperative of replacement of molecular components. (868)

The coevolution of meiotic sexuality and programmed death specific to kingdom Animalia would appear to motivate this particular emphasis on the place of mortality in a zoologically fleshed-out autopoietic conception. As an old man himself, the Irish poet William Butler Yeats had his aged bard sing about such worldly events, from which the singer sought his soul's escape by "Sailing to Byzantium":

. . . The young

In one another's arms, birds in the trees,

—Those dying generations—at their song,

The salmon-falls, the mackerel-crowded seas,

Fish, flesh, or fowl, commend all summer long

Whatever is begotten, born, and dies. . . .

Instead of yearning after release from bodily conditions, Margulis the autopoietic materialist places this zoological interest in metazoan sex and death against the Gaian backdrop of deep evolutionary time extending six times farther back than the origin of animals per se. During all that time it has been Gaia that has enjoyed a form of autopoietic immortality, a deathlessness of incessant operation: "All organisms are part of a single continuous bounded autopoietic system that has never been breached since the origins of life in the Hadean or Archean eon. While portions of the system (cells, individuals, populations, species) are always losing autopoietic properties, the entire system itself persists. Death must co-occur with life. Failure to retain autopoietic properties is death—and death by loss of components, desiccation, disintegration, and atrophy is intrinsic to the continuity of life" (879). Moreover, by insisting on an autopoietic Gaia theory, Margulis develops all of this conceptual imagery as a neocybernetic antidote to the zoological malaise of neo-Darwinian genetic parochialism and planetary small-mindedness. However, Margulis reserves her most stringent animus against mainstream neo-Darwinism for a sister paper to this one, aimed at the wider if still elite intellectual audience for the self-styled Reality Club.

"Big Trouble in Biology"

What are we to do with such a vehement critique of the author's own scientific discipline and academic institutions? Should we write off her tirade as the idiosyncratic outcome of strictly personal stresses? Is her published rancor a professionally foolhardy rejoinder to the usual accumulation of workplace or disciplinary resentments? Does this essay vent the justifiable anger of a female genius who has finally had it up to here with the daily indignities of a masculinist academic culture within a patriarchal society? However, even if there were some truth to each of these rationales, settling for any of them would be a patronizing response denying wider validity to her extraordinarily deep and specific arguments. And what if she is right? What if her critique of the philosophical aberrations of mainstream Anglo-American science was correct in 1990 and is still largely correct thirty years later? Perhaps Margulis's uniquely refined scientific awareness of planetary dynamics enabled her to see farther and sooner into the global crisis that is crashing down all around us now. What if her arduously acquired access to alternative ways of scientific seeing and knowing gave her sufficient perspective and courage to expose the deformity of some of modern Western culture's most misguided and destructive verities, no matter how ceremoniously they have been wrapped in the robes of science?

The primary destructive verity or "big trouble" under Margulis's extra-disciplinary attack is what she considers the indoctrination of biologists in particular with the conviction that the aim of their science is to reduce the phenomena of life to non-living mechanisms, that "life is a mechanical system fully describable by physics and chemistry. Biology, in this reductionist view, is a subfield of chemistry and physics."³⁴ As Margulis will detail in a later iteration of the Reality Club publications, biologists in general suffer from a bad case of "P.E.," that is, "physics envy . . . a syndrome in which scientists in other disciplines yearn for the mathematically explicit models of physics."³⁵ In biology proper, the reigning disciplinary expression of such "physicomathematics envy" (214) is the set of neo-Darwinist doctrines that encourages her colleagues and students to ditch field work and intimacy with the physiologies of embodied living beings for mathematical and computational approaches to molecular genetics and population dynamics. "Hence biologists receive Guggenheim Fellowships for calculations of the evolutionary basis of altruism or quantification of parental investment in male children, while the tropical forests are destroyed at the rate of hundreds of acres per day" (213). It is not just a matter of academics preferring ivory-tower offices with supercomputer interfaces to strenuous confrontations with real-time environmental depredations. It is also that, falling in line with such bloodless dogmas, biologists have been giving away the store, ceding the knowledge of life to people who really have no idea what life is, yet are "their supposed superiors: physicists, chemists, and mathematicians" (215). And because, "like monasteries of the Middle Ages, today's universities and professional societies guard their knowledge" (213), her biological colleagues may not even know that there actually are respectable alternative ways of conceiving their disciplinary objects and doing their science. The trouble is that they have been too carefully guarded against heretical doctrines that open out onto alternative cosmologies. Too few biologists, Margulis laments, know that "a life-centered alternative worldview" even exists, "called 'autopoiesis,' which rejects the concept of a mechanical universe knowable by an objective observer" (214).

However, as the auto mechanic said to the owner of the broken-down car, "there's your trouble!" Scientists are not supposed to challenge the popular notion of their unique possession of the truth about nature. Proper scientific mechanists guarantee to deliver the kinds of ontological finalities that their funders prefer to anticipate—well, eventually, once a complete data set—say, the "human genome"—has been assembled and completely computed upon, so please keep the funds coming. The theory of autopoiesis comes with no such 100,000-mile guarantees. It promises only a universe in unpredictable flux on the edge of whatever it takes to maintain living self-productions contingent upon the integrity of operational closures. Moreover, autopoietic theory is wedded to constructivist epistemologies that have withdrawn any refuge in positivism from mainstream science's conceit of objectivity. Margulis has read Autopoiesis and Cognition well and has laid out the ramifications of the theory. An autopoietic universe is not an absolute totality or a grand unified anything. As it comes into being out of a pre-living world, an autopoietic universe does not supersede but supplements the physicochemical cosmos with processes of life and cognition that are possible only for embodied living beings, among which scientists may be placed, incapable of bodiless omnipresence and absolute knowledge, however Pythagorean or Platonic their visions may be.

In its primer on autopoiesis (considered strictly in its original biotic sense), "Big Trouble in Biology" follows up an earlier table in "Kingdom Animalia" giving "Criteria of Autopoiesis" with an expanded table breaking down "Properties of Autopoietic Systems." Six properties—identity, unitary operation, self-boundedness, self-maintenance/ circularity, external supply of component raw materials, external supply of energy—are unfolded with regard to aspects such as "Boundary structure produced by system" and, uniquely in the literature of autopoiesis at this moment, "Examples of Biochemical/Metabolic Correlates" (216). Margulis fleshes out biotic autopoietic materiality with regard to such factors as nucleic acids, fatty acids, multienzyme-mediated networks, lipogenesis, polymerization, and so forth. In the following passage, she works through an autopoietic framing of metabolism, placing the familiar workings of cell biology and physiology into the encompassing logical framework of autopoietic organization and self-production:

Autopoietic systems metabolize, whereas nonautopoietic systems do not. Proteins, viruses, plasmids, and genes are all components of live material. When contained within the boundaries of animal, plant, or other cells, they may be required to sustain cells or organisms and their autopoietic behavior; yet proteins, viruses, plasmids, and genes, intrinsically incapable of metabolism, are never autopoietic in isolation. Metabolism includes gas and liquid exchange (breathing, eating, and excreting, for instance); it is the detectable manifestation of autopoiesis. Autopoiesis determines physiology and hence is the imperative of all live matter. Autopoietic entities, that is, all live beings, must metabolize. These material exchanges are the sine qua non of the autopoietic system, whatever its identity. (217)

The bottom line of Margulis's autopoietic critique is that the mechanistic worldview sees no fundamental distinction between living and non-living entities. Lovelock's own ultimate orientation is to a form of cybernetics that countenances this mechanistic orientation: "The only difference between non–living and living systems is in the scale of their intricacy, a distinction which fades all the time as the complexity and capacity of automated systems continue to evolve.Whether we have artificial intelligence now or must wait a little longer is open to debate."³⁶ For Margulis, autopoietic Gaia is the major form of her protest against this mindset even as it resides within her long-time collaborator. I would elaborate her vision as follows. For "the physics-centered philosophy of mechanism and its runt offspring Neo-Darwinism" (271), the universe is essentially a dead place where cosmic energies move in and out of material phase, within which astral dynamics the odd epiphenomena of living beings may emerge from time to time, as apparently they have done here on Earth, and then come and go, creating momentary bursts of noise within the universal silence. Meanwhile, the mechanistic worldview finds a place for "intelligence" within its universe, somewhere or another, artificial or otherwise, ideally constituted as an informatic patterning with no particular location and rising above the contingencies of living bodies. In our own moment, the aggravated symptoms of this still-mainstreamed orientation are everywhere. At the popular level, we can see it in computer-generated cinematic superheroes with death-defying superpowers battling intelligent robotic nemeses amidst mindless explosions of super-cool hardware raining down debris on a passive, forgotten environment. At the specialized level of corporatized research and development, we can see it in the ever-trendy anticipation of the arrival of the Singularity—the moment at which machine intelligence leaves its human cradle behind—the sooner the better, just as soon as we can get our machines to get this AI thing completely figured out. At that point, autopoiesis and cognition, organic life and natural mind, for all their self-organization of a worldly technics of Gaian construction, will be headed for extinction, left behind by the "higher" evolution of our informatic replacements as they depart for, as Neil Young prophesied in "After the Goldrush," a new home in the sun.

Perhaps you would agree that these scenarios are juvenile and self-defeating. They are the manifest repercussions of specific scientific and technological thought collectives granted ideological dominance in modern culture. The concept of the thought collective was propounded by the Polish microbiologist Ludwik Fleck, and "Big Trouble in Biology" registers the arrival of Fleck's sociology of science within Margulis's polemical armory. Margulis introduces Fleck's work midway through this article with a harrowing biographical sketch of his survival of the Holocaust, spared, like the chemist Primo Levi, due to his utility to the Nazis and pressed into service in Auschwitz and Buchenwald manufacturing typhus vaccine for German troops. According to Margulis's account, Fleck sent bogus vaccine to the front while holding back "the real vaccine, in exceedingly short supply, to protect himself, his family, and friends. Surrounded by lives in daily danger, Fleck paid close attention to how easily scientists and technicians mentally imbibe the prevalent 'common myth'" (222). The implication is that even in the halls of science, professional ideals will defer to the demands of personal survival and succumb to overriding ideological forces. However, Fleck developed his sociological theory of scientific facts prior to these dire experiences. He originally published Genesis and Development of a Scientific Fact in 1935.³⁷ Brought into English translation in 1979, Fleck's sociology of scientific thought collectives anticipated by over a generation the rise of the sociology of scientific knowledge in the 1970s and 1980s, for instance, in a work such as Latour and Woolgar's Laboratory Life: The Construction of Scientific Facts, also published in 1979.³⁸ Margulis offers this synopsis of Fleck's sociological self-analysis:

The theory claims that all "scientific facts" are merely consensuses among socially interacting "card-carrying" scientists. . . . "[T]he fact" is a product of a complex social process beginning with individual observation or measurement and terminating with the integration of a stylized "true statement" into the knowledge of the society at large. . . . [C]ertain words and phrases become banners for the immediate identification of scientific friend or foe. . . . [S]ocial activities . . . cement into cohesive groups otherwise unruly scientists and technicians. . . . "[T]hought-collectives"—are then recognizable. . . . [T]he thought-collective achieves the status of "professional tribe," as do today's Neo-Darwinists, whose members are bound together by many ties, including those of common scientific language. (223)

Using Fleck's theory of thought collectives to identify and characterize the biological tribe against which she will wage mental fight, Margulis bootstraps an astonishing sociological and ideological critique of her own field. "Why do members of the Neo-Darwinist social group dominate the biological scientific activities in U.S. and other English-speaking academic institutions? Probably there are many reasons, but a Fleckian one is that the Neo-Darwinist mechanistic, nonautopoietic worldview is entirely consistent with the major myths of our dominant civilization" (225-26). At this point Margulis makes the first of several references to William Irwin Thompson's 1981 mythopoetic analysis of cultural evolution, The Time Falling Bodies Take to Light.³⁹ The residual apocalypticism in Thompson's rehabilitation of Neolithic matriarchy and account of past cultural upheavals leading to the current showdown between our "materialistic civilization that is concerned almost exclusively with technology, power and wealth" (cited in Margulis, "Big Trouble" 226) elicit some of the edgiest, most politically trenchant commentary anywhere in her writings:

A world philosophy based on the recognition of the autopoietic and nonmechanical nature of life must upset the believers in the fundamental myths of our technological civilization. In the world of the Native American, humanity belongs to the earth; in the world of the money machines, the earth belongs to humanity. In the autopoietic framework, everything is observed by an embedded observer; in the mechanical world, the observer is objective and stands apart from the observed. (227)

I suspect that Neo-Darwinists, upon observing physiology and contemplating autopoiesis, suffer cognitive malaise. Their mathematized formulations systematically ignore physiology, metabolism, and biological diversity; they fail to describe the incessant, responsive, reciprocal effects of life embedded in environment. Suffering philosophical distress, physics-worshiping Neo-Darwinists must reject autopoiesis and its attendant life-centered biology with the same zeal with which the Spanish true church, guarded by its Inquisitors, rejected the mescal- and peyote-eating religions of the Native Americans. (228)

Gaia arrives in "Big Trouble in Biology" as the destination of these meditations on embeddedness. For the theory of autopoiesis, the self-reference of cognition is axiomatic: the observer is always already embedded in their observation, whether or not they observe this to be the case or are aware of their existential and epistemological status in this regard. For its part, as Margulis telegraphs the matter, life itself is "embedded in environment." Following out this autopoietic logic, living (autopoietic) systems are ineluctably in cognitive relation to the environments from which they emerge. Living beings are environmentally embedded observing systems, period. The recursive self-constitution of living systems feeds back into the incessant recursive re-constitution and thus evolution of their environments. This is autopoietic Gaia. The passage into the autopoietic worldview is strongly mediated by the very kind of immersive experience of planetary connection that the bloodless biology against which Margulis inveighs denies to its disciples, who thus become its unsuspecting victims:

Who are the victims of these latter-day religious wars for the souls of the biological science practitioners? Primarily graduate students, young investigators, and teachers, in whom direct observations of life and experience in the field often foster an expansive autopoietic attitude. The study of physiology and immersion, especially in tropical nature, tends to lead students to a perception that the living planetary surface behaves as a whole (the biosphere, the place where life exists on the Earth). Yet the Academy guards, using Neo-Darwinism as an inquisitory tool, superimpose a gigantic super-structure of mechanism and hierarchy that protects the throbbing biosphere from being directly sensed by these new scientists—people most in need of sensing it. The dispensers of the funds for scientific research and education and other opportunity makers herd the best minds and bodies into sterile laboratories and white-walled university cloisters to be catechized with dogmatic nonsense to such an extent that many doctoral graduates in the biological sciences cannot distinguish a nucleic acid solution from a cell suspension, a sedimentary from an igneous rock, a kelp from a cyanobacterium, or rye from ergot. (228-29)

Margulis's plea on behalf of her students and colleagues is one expression of the ethical imperative that drives "Big Trouble in Biology." An exposition of the Gaia hypothesis proper arrives directly after this quintessentially Gaian evocation of the need for initiation into immersive experiences leading to "a perception that the living planetary surface behaves as a whole." In largely the same words of "Kingdom Animalia," here the late rehearsal of Gaia provides one last hook on which to hang neo-Darwinism in effigy. Margulis's ecological fatalism reappears as well, anticipating that our mechanistic civilization is beyond self-reform. Its transformation will have to be apocalyptic, some form of global convulsion: "among academic biologists inside the convent walls, Neo-Darwinist reductionism will prevail until the suddenness of a new planetary culture replaces the technological civilization to which Thompson refers. Only after the new civilization binds us consciously to our nonhuman planetmates, especially the truly productive green ones, can the physiology of autopoietic visionaries replace the mechanics of the Neo-Darwinists inside the academic cloister" (229). The neocybernetics of Gaia rises in these sweeping indictments to a consummate expression of the international systems counterculture that nurtured the concept of autopoiesis as a philosophical as well as biological discourse.

Cite this Essay:

Clarke, Bruce. “Margulis, Autopoiesis, Gaia”, Electronic Book Review, July 7, 2019, https://doi.org/10.7273/pgha-tv58.

Readers wishing to respond to an essay in ebr may send ripostes or short glosses to the journal’s Managing Editor, Anna Nacher.

PDF