Devourer of Encyclopedias: Stanislaw Lem's "Summa Technologiae"

By David AuerbachMay 13, 2013

Devourer of Encyclopedias: Stanislaw Lem's "Summa Technologiae"

Summa Technologiae by Stanisław Lem

AT LAST WE have it in English. Summa Technologiae, originally published in Polish in 1964, is the cornerstone of Stanislaw Lem’s oeuvre, his consummate work of speculative nonfiction. Trained in medicine and biology, Lem synthesizes the current science of the day in ways far ahead of most science fiction of the time.

 His subjects, among others, include:

  • Virtual reality

  • Artificial intelligence

  • Nanotechnology and biotechnology

  • Evolutionary biology and evolutionary psychology

  • Artificial life

  • Information theory

  • Entropy and thermodynamics

  • Complexity theory, probability, and chaos

  • Population and ecological catastrophe

  • The “singularity” and “transhumanism”

Lem was one of the very few thinkers at the time to examine these burgeoning subjects in the context of both the humanities and the social sciences. Yet despite the far-ranging explorations, Lem tempers his speculations with a de-romanticized and often grim view of humanity. Indeed, part of Lem’s genius was his keen awareness that the possibilities of science and the possibilities of humanity do no more than scarcely overlap, and so our investigations must be conducted with our limited vantage point in mind.

It would be indeed unusual if it turned out that the set of orders that our mind is able to construct and accept, having as it does a deep sense of “understanding the essence of things,” matches precisely the set of all possible orders to be detected in the Universe as a whole. We should admit that this is not impossible, yet it does seem highly improbable. This way of thinking, so modest in its assessment of our abilities, is probably the only way recommended, given our lack of knowledge, because we are not aware of our limitations.

Perhaps this epistemological modesty was reinforced by Lem’s unique position in literature. Lem occupied an uneasy place in the world of letters, overlapping several fields without being claimed by any single one, enjoying a reasonable degree of popular success without reaching preeminence, at least in English-speaking countries. Even in Poland, however, he did not associate with the literary mainstream, and while he is a giant there, he remains a singular figure detached from larger trends. Lem captured the predominant attitude toward him in his satirical introduction to his collection of reviews of imaginary books, A Perfect Vacuum, which takes the form of a fake, dismissive review of A Perfect Vacuum itself: “We know that Lem has devoured encyclopedias; shake him and out come logarithms and formulas.”

 Genre science fiction had concurrently begun to touch occasionally on some of Lem’s subjects — Philip K. Dick on virtual reality, Harry Harrison and John Brunner on ecology, Cordwainer Smith on biotech, Pamela Zoline and J.G. Ballard on entropy — but Lem went far deeper than any of them. Lem was devastatingly scabrous to genre science fiction, singling out only a few authors such as the Strugatsky brothers and Dick for praise, but his assessments were still incisive: “The primary ontological elements — space and time — are Dick's instruments of torture,” he wrote in “Philip K. Dick: A Visionary Among the Charlatans.” He similarly displayed great impatience with theoreticians like Tsvetan Todorov and Jacques Derrida who elided logic and took shortcuts in pursuit of neat answers and flashy displays.

Lem’s earliest novels, never translated into English and more or less disowned by him as worthless, were optimistic tales of space travel. By the late 1950s, his work had become darker, increasingly focused on the limits of humanity’s knowledge and the limits of human society in pursuing both truth and virtue. Lem esteemed his classic novels like Solaris because “[t]hey incorporate cognitive problems in fictions that do not oversimplify the world.”

 More than any other author, Lem integrated science into science fiction, putting much “hard” science fiction to shame. But it was not enough to get facts right and respect the laws of physics: great speculative fiction needs to use creativity within the parameters of scientific and philosophical naturalism in order to explore the limits of what is metaphysically and sociologically possible. Many of Lem’s works take the form of thought experiments.

Summa Technologiae catches him at an equivocal point: the all-encompassing pessimism, reminiscent of Schopenhauer, that would characterize 1980s works such as Golem XIV, Peace on Earth, and Fiasco has yet to reach full fruition, but Lem’s focus has decisively shifted to the sociological problems of scientific knowledge. This would shortly yield one of his masterpieces, the sober and grim look at scientific inquiry, His Master’s Voice.

Whether due to the format or to its earlier date, Summa does not read as darkly. In Summa, folly is more a statistical likelihood than an object of study. Lem is concerned here more with possibility, and while he maintains a robust skepticism as to whether any of these possibilities will come to pass, and whether they should come to pass, he remains focused on his explorations rather than making moral or social assessments of them.

These explorations contain the seeds of much of Lem’s subsequent fiction. The ethics of artificial life in “Non Serviam” (from A Perfect Vacuum), the evolutionary satire of Golem XIV (from Imaginary Magnitude), and the scientific malpractice of His Master’s Voice are all already present here in fully developed form, vividly and rigorously imagined.

Despite the lack of cynicism, the book is hardly idealistic. Lem reserves his harshest attacks for starry-eyed futurists who embrace technology without thinking through the often-devastating moral implications. His critique of techno-utopian Freeman Dyson’s circumsolar “Dyson spheres” is more devastating than anything from today’s neo-Luddites, precisely because he takes seriously the underlying need to deal with the issue of population, questioning the assumptions of the solutions rather than the very idea of them: “It is hard to understand why reproductive freedom would have to remain intact even if it was to lead to a total immobilization of individuals, to smashing the cultural tradition, to giving up, literally, on the Earth’s and sky’s beauty.”

Though the writing is denser than all of Lem’s fiction save Golem XIV, with which it shares much, Lem is still capable of beautiful and elegant passages. He cogently expresses the sheer metaphysical angst that besets an individual contemplating the universe, suggesting that it is this feeling more than any other that defines who we are as a species, at least at our best:

Civilization lacks knowledge that would allow it to choose a path knowingly from the many possible ones, instead of drifting in random tides of discoveries. The discoveries that contributed to its construction are still partly accidental.

As much as any of his fiction, Summa Technologiae’s intensely marries Lem’s Borgesian imagination with his philosophical and sociological rigor.

Lem does not integrate his concerns into a single unified theory; Summa is a fantasia that follows certain lines of speculative thought as far as Lem can take them. The spur, as ever for Lem, is the cognitive limitations of humanity. Lem announces early on that humanity is on the brink of becoming obsolete because it is reaching a scientific “information barrier” beyond which disorganized humans will not be able to process the amount of knowledge we are obtaining: “Science cannot transverse this barrier; it cannot absorb the avalanche of information that is moving in its direction.” That is, Lem sees that with the exponential growth of knowledge that has taken place since the Scientific Revolution, humanity is reaching a choke-point where the physical capacities of our brain, in conjunction with any and all possible societal configurations in which our brains can work together, will simply not be sufficient to (a) continue the work of scientific research, and (b) maintain a stable civilization. If we wish to exercise some sort of collective self-determination, we must accept that our current sentient forms aren’t up to the task.

The point is not to construct synthetic humanity but rather to open up a new chapter in the Book of Technology: one containing systems of any degree of complexity.... Even if man is indeed capable of anything, he surely cannot achieve it in just any way. He will eventually achieve every goal if he so desires, but he will understand before that that the price he would have to pay for achieving such a goal would reduce this goal to absurdity.

If we are limited beings, there is nothing to say that other kinds of sentient beings are similarly limited. However, we will cease to understand those other beings at the point that they wholly surpass us — or contrariwise, we will cease to be ourselves if we reach that point. The point is debatable, but Lem’s dour view is certainly plausible.

Having proposed this paradox, Lem then asks what possibilities do exist for the future. Lem starts from the framework of cybernetics. Lem was strongly influenced by cybernetics, a mid-century field of study that has since fragmented into control theory, complexity theory, artificial intelligence, and other areas. Norbert Wiener coined the term “cybernetics” to refer to “the study of control and communication in machines and living beings.” In other words, cybernetics examines structure and stability in groups of organisms: how they evolve in order to establish homeostasis, and what communication methods the organisms use to maintain homeostasis.

Loosely speaking, cybernetics is an abstracted, unified conception of evolution and artificial intelligence. Evolutionary psychology frequently falls down by trying to map complex human behavior onto reductive categories of evolutionary adaptation and fitness. Lem, like Wiener before him, does not make this mistake. Rather than looking at the outcome of evolution (this or that behavior, this or that psychology), Lem tries to understand the limits on what evolution and similar complex processes could produce, thereby avoiding the just-so stories frequently given us by Stephen Pinker and his brethren.

Though much of cybernetic theory itself has been supplanted, its issues endure, and Lem presciently extracts the perennial issues from its dated particulars. His investigations of artificial life (here termed “imitology”) and virtual reality (“phantomatics”) feel utterly up to date. Yet they are ultimately sideshows next to his recurrent obsession, which is the analysis of life, natural or artificial, in an ecological and evolutionary context, and what role technology can play in controlling it.

This obsession is rooted in three central themes which occupied Lem his entire life:

1         The limits of human cognition.
2         The nature, function, and restrictions of the evolutionary process.
3         Probability’s metaphysical role.

These themes are interlinked. Evolution, through brute-force stochastic processes, has produced such finite, limited creatures as us and is responsible for our cognitive limits. As sentient creatures, we intentionally attempt to exert control over and influence evolution, but we are limited in our agency and in our ability to grasp the structures of life and the universe. We may fail intellectually, socially, or simply physically.

Our ambitions so often outpace our efforts that humanity’s actions appear to be primarily folly in the macro context in which Lem is examining them. (For example: politics, economics, warfare, reproduction, even technology itself.) Hence the ever-darkening satirical viewpoint that Lem took in his fiction. For Lem, communism and capitalism are delusional twin faiths: communism, that we can collectively and centrally control chance and causality; capitalism, that chance and causality will intrinsically prove benevolent and productive for us.

Lem reflects this nondeterministic fatalism back onto science itself. The scientific process, he says, is by no means insulated from our lack of self-determination. Science just happens to be the best of many bad options.

Civilization lacks knowledge that would allow it to choose a path knowingly from the many possible ones, instead of drifting in random tides of discoveries.

From this comes the irony of the title, a play on Aquinas’s Summa Theologiae, our assured knowledge of God replaced by our uncertain advances into physics and technology. Having gained sentience, we are the only knowing gods around, though mostly ignorant and powerless.

To understand the cosmos, evolution, and even ourselves, one should start by examining how chance works. Sentience stands in opposition to probability, the most daunting of the three themes. In the nineteenth century, evolution and thermodynamics showed that the processes of the universe are the result not of clockwork laws but of complex and chaotic processes, and as such are governed, at least from our vantage, by the laws of chance and probability. The theories of Bolzmann, Gibbs, Maxwell, and ultimately Einstein disposed of the dream of the ordered, clockwork universe. Wiener’s cybernetics treated biological, social, and mechanical systems as thermodynamic systems, intending to establish what processes existed to preserve homeostasis against the universal, disordering forces of entropy. Lem was profoundly influenced by this basic model: the forces of probability, chance, and chaos loom heavily in his work as fundamentally conditioning all systematic activity, human or not.

This theory was quite far from the Chomskyan model of cognition that shortly became dominant in linguistics and artificial intelligence. The Chomskyan model was deterministic, structural, and considerably easier to implement. The cybernetic model was feedback-based, evolutionary, and algorithmically opaque. Alas, it is the latter that more closely mirrors life.

The idea is not new. The rule of Fortune goes back to Heraclitus, the Stoics, and Seneca and Lucan, but the modern age brought with it such an emphasis on determinism and agency that Lem achieved something novel in reintroducing metaphysical chance into literature (half-assed efforts like Pynchon’s “Entropy” notwithstanding). Into this probability-governed world comes the concerted, intent-driven efforts of sentient beings. And they are mostly futile, for at the macro level, society seems to be governed more by those laws of chance and complexity than by rationality.

Olaf Stapledon, another of Lem’s great influences, had treated the interaction of society and chaos in his classic novel of future history Last and First Men (1930). Stapledon similarly took it as a given that human beings in their current forms would be supplanted by other creatures that would call themselves “men.” Some would fly, some would be inorganic robots, others would simply be beyond our current understanding. Stapledon, however, chronicled these changes on a primarily sociopolitical level: the changes in beings occur as a result of some kind of upheaval or long-term social change. Lem shifts the focus to the limits of our understanding.

Lem’s shift owes mainly to intervening scientific discoveries, particularly theories of computation such as Alan Turing’s, and the discovery of DNA. These provided Lem with the tools to analogize two devices that, against the general statistical law of entropy, produce increasing rather than decreasing organization: human thought and biological evolution. If evolution was charted in a form of code, and a simple Turing machine could be shown to be equivalent to the most complex computational machines imaginable, then evolution and computation should not be treated as incommensurable processes, but as divergent models of a single type of development. DNA and the Turing Machine were fundamentally alike.

Yet DNA has succeeded in creating sentient creatures, whose creation processes are very different from those of any other known material process. We utilize intention and meaning. And we are at a point where we are taking up the work previously done by evolution, for better or for worse. Lem is keenly conscious of the paradox that our powers are at once superior to evolution’s in their directness, yet inferior to evolution’s powers in their reach and scope, for, as Lem says, “One creates philosophers while the other only creates philosophies.”

Thus we reach the main thesis of Summa Technologiae: human society is becoming increasingly large and complex at an exponential rate, but our capacity to regulate our society is not keeping up, making homeostasis increasingly difficult to preserve (if indeed it ever has been established). The technologies by which we might preserve homeostasis will not — cannot — be fully controllable by us, for to be controllable, the technologies would have to be so impotent as to be useless in assisting us. They too will have to be probabilistic and chance-governed, yet influenced by our intentions.

The deepest and most difficult chapter, “The Creation of Worlds,” deals with evolution and the possibilities for creating intelligence through genetic algorithms. The analogies here between technology and evolution become considerably deeper and more complicated, but the basic idea is that (if I may paraphrase what I’ve written elsewhere) if computers are stupid, then so is evolution, and consequently they share more with each other than either does with us.

Lem’s conceit that if one were to develop a program that generates self-reproducing organisms which would be incrementally capable of accommodating themselves to their environment — that is, functionally understanding it — then that program would end up looking a fair bit like evolution. It would push out increasingly complex machines/organisms that would increasingly grasp the nature of their world and of themselves. Would such a process really be analogous to evolution? Lem presses on forcefully without analyzing the details of where the points of difference might be, but it is a breathtaking analogy, high-level brain candy. He then speculates on what a semantics for evolution would look like: survival is truth, extinction is falsehood.

This is all rather grim stuff, for we turn out to be waste products of a self-reproducing genetic code trying to survive through clumsy brute-force efforts, presaging Robert Trivers and Richard Dawkins’s sociobiology work in the 1970s. Lem would later summarize some of his points here through the mouthpiece of the supercomputer in Golem XIV, which along with “The Creation of Worlds” probably represents the pinnacle of Lem’s intellectual achievement:

Evolution is forced to use death, since it cannot go on without it; it is lavish with death in order to perfect successive species, for death is its creational proofreader. Thus it is an author publishing ever more magnificent works in which typography — the code — is merely its indispensable instrument. However, according to what your molecular biologists are now saying, Evolution is not so much the author as a publisher who continually cancels works, having developed a liking for the typographic arts!

Not that things are hopeless. Lem holds out the possibility that we may be able to wrest control of such evolutionary processes and redirect them to serve our purposeful ends rather than the ends of propagating genetic code above all. Lem would later retract this hope and decide that humanity simply wasn’t up to the task, but the final chapter of Summa, “A Lampoon of Evolution,” at least speculates on the chances for us to dethrone evolution from its vulture’s perch. He ridicules evolution’s inefficiency, ruthless violence, and lack of foresight, and in these regards we have indeed been able to do better, albeit in vastly smaller contexts. The paradox remains, though, that it was evolution that created us, and its scope of achievements remains unfathomably greater than ours. Do we even have a chance of beating it? We are helpless idiots competing with omnipotent, undirected chaos. (This, in fact, is the same competition portrayed by Laszlo Krasznahorkai in his fiction.)

Certainly the construction of complex systems has gone far beyond what once seemed feasible. Abstraction and modularity in software have proven surprisingly robust, at least within carefully controlled settings. This has been made possible by the immense boom in computing power that has come to be known as Moore’s Law, stating loosely that processing power doubles every two years. This absurd prediction has held up, and consequently computers are over ten million times faster today than they were at the time of Summa Technologiae. Moore’s Law makes efficiency considerably less important than one would think: within a few years, your optimizations will cease to be a chokepoint.

In 1981 Lem spoke of sentient self-programming computers a million times faster than ENIAC. Lem’s prediction of the date was exactly right: 2000. He knew his stuff. (In a follow-up essay from 1991, “Thirty Years Later,” Lem flaunted the accuracy of his virtual reality predictions to Leszek Kolakowski, who had critically trashed Summa on publication.)

On the other hand, Lem predicted sentient machines quickly to follow such increases in speed. Computer intelligence hasn’t kept up. My laptop is a million times faster than ENIAC, but little smarter. So it has been with cybernetics. Creation of genuinely viable artificial intelligence has proven exceedingly difficult if not impossible. Today’s code lacks any of the self-organizing and evolutionary abilities about which Wiener and Lem speculated. The challenge that Lem knew such intelligence would pose to us has not been made. This failure postpones confrontation with some of the issues Lem raises, but they still await.

As for humanity itself, I do believe Lem underestimated the social structures permitted by the Internet, the spontaneous self-organization of political and social communities united by an interest or an agenda. Whether these organizations will ultimately prove beneficial or harmful remains to be seen, but they do allow for one more potential force against the overwhelming onslaught of unorganized information.

But if social forms and technology can be experimented with sufficiently in an ecological way that lets the most benevolent manifestations flourish, and these experiments can be done quickly enough, perhaps there is yet hope that we won’t drown in the idiocy and violence that evolution has bequeathed to us.

In the years since Lem’s masterwork, Western intellectual thought has been set upon by a vulgar and ill-informed pessimism that shares Lem’s skepticism without possessing any of his vision. In the guise of critiquing culture, critical theorists parrot received ideas and cling to an archaic essentialism about humanity. They look backwards to falsely imagined golden days, rarely bothering to learn the details of the science they are critiquing. Meanwhile, science simply ignores such chatter and proceeds onward unthinkingly.

Lem offers a far more rigorous model for thought. Lem’s sober materialism may seem dehumanizing, but he brings back to the forefront a question that has plagued civilization since the beginning, and whose shifting, always insufficient answers have always signaled revolutions in culture: what is it to be human?


LARB Contributor

David Auerbach is a writer and software engineer. His book Illogical Operator: A Life in Code is forthcoming from Pantheon. He lives in New York.


LARB Staff Recommendations

Did you know LARB is a reader-supported nonprofit?

LARB publishes daily without a paywall as part of our mission to make rigorous, incisive, and engaging writing on every aspect of literature, culture, and the arts freely accessible to the public. Help us continue this work with your tax-deductible donation today!