Feedback Loops and Modern Thought

There never was a man who could sit down and say: “Now I am going to be the first man to write.”

—David Diringer

            Poetry precedes philosophy.  This claim is not seriously contested nowadays, with the knowledge that the Homeric epics circulated long before the appearance of Thales of Miletus in the sixth century BCE (not to mention the circulation of non-Western poems such as the Epic of Gilgamesh).  Despite their availability in print today, scholars know that the Homeric epics were transmitted orally for centuries, repeated and adapted by traveling bards.  Their poetic structure proved a useful mnemonic device; after all, it’s easier to remember something when we’re able to phrase it as a rhyme, a rhythm, or some other musical form.  So ancient poetry and epic traditions evolved orally and creatively, unwritten.  They didn’t require alphabetized written language because of their creative form, the poetic techniques.  The talents of the muses were conducive to the retention of content, albeit with new flourishes along the way.  The story remains the same, but the song doesn’t.

            Philosophy is different.  In order for precise and extensive reflection on the contents of thought, it must retain not only its general meaning but its particular appearance—for this is the only way to sustain any successful meditation on the relationship between words and what they signify.  “As soon as one could set words down, examine them, look at them anew the next day, and consider their meaning, one became a philosopher,” writes James Gleick in his impressive book The Information: A History, a Theory, A Flood; “and the philosopher began with a clean slate and a vast project of definition to undertake.  Knowledge could begin to pull itself up by the bootstraps.”  In other words, the advent of writing opens the door to philosophical thought in that it allows the philosopher to begin classifying, categorizing, defining, meditating, reflecting, and theorizing—in short, philosophizing.  Philosophy blooms with the emergence of writing because writing allows us to revisit our thoughts as they were originally conceived: to reconsider and amend them.

            Thought, language, writing—this is one of our species’ most evolutionarily important feedback loops, but it wasn’t until the era we think of as modernity that intellectuals began to conceptualize at length the historical and social impact of writing systems.  Whatever thought was in its infancy, it wasn’t what we tend to think of as modern consciousness—that is, self-reflexive awareness.  In this infant phase of humanity, consciousness emerged not alongside writing, but alongside spoken language; it wasn’t available for self-reflection.  With the development of systematic writing, language could be studied and learned, ideas once thought fleeting in ephemeral speech could be recorded and copied.  And suddenly, as writing took shape, it changed the way humans thought.  They gained the capacity for organized self-reflection, for revisiting and revising their thoughts.  Thinking may not have begun with writing, but writing undeniably changed human thinking forever.  The feedback loop of thought, speech, and writing introduced a new way of being in the world.

            Feedback is a counterintuitive concept despite its deceivingly straightforward explanation and ordinary applications.  In simple terms, feedback describes a process in which a system’s output (its product) is fed back into the system as input.  In common parlance, the term finds multiple uses.  When we attend a concert and experience the displeasure of an electronic screech emanating from the speakers, many of us recognize this phenomenon as feedback.  When working on a project, we might share it with friends or colleagues and ask for their feedback.  Some of us may even employ the more technical phrase “feedback loop” when discussing everyday occurrences: for instance, the tendency of MBTA passengers in Boston to neglect paying their fares, which leads to increased fares, which leads to more people not paying, which leads to higher fares, and on and on.  Many of us likely have some kind of superficial grasp of feedback, or know examples of it; but comprehending it conceptually is another matter, largely because feedback interferes with one of our most basic assumptions about how reality works: causality.  Traditionally, causal reasoning implies an original act, a cause.  A causes B, which causes C, which causes D, and so on.  Causal reasoning is linear.  It moves in one direction, and from one event to the next.  Feedback disrupts linearity, leading some theorists to label feedback (somewhat paradoxically) circular causality: A causes B, which causes C, which causes D, which causes A, which causes B, which causes C, which causes… where does this reaction begin?

            Our perception intuits (usually involuntarily) a starting position, but the physical world is not a canvas for our impressions.  Or rather, it permits a multiplicity of impressions, all of which comprise an actuality beyond any single perspective: “For existence has its own order and that no man's mind can compass, that mind itself being but a fact among others,” as Cormac McCarthy writes in Blood Meridian.  The paradox of circular causality is that there is no longer any starting position, no longer any origin.  The concept of the origin fades into the structure of the loop.  Douglas R. Hofstadter refers to such structures as strange loops, or tangled hierarchies; the only way out of such disorienting structures is to pose an “inviolate level.”  For example, take M.C. Escher’s famous Drawing Hands sketch:

The image constitutes a tangled hierarchy—there’s no telling which hand comes first, so to speak.  Each draws the other into existence.  The sketch’s inviolate level is Escher himself, the artist or author of the image.  Determining an inviolate level doesn’t allow us to untangle the hierarchy of the image (the paradox of the drawing hands remains), but it does allow us to contextualize the image so that its ambiguity is easier to digest.  It imposes linearity and order so that we might better conceptualize the structure of the feedback loop.

            Feedback loops aren’t merely a part of our daily lives; they constitute the structure of modern consciousness, and their inviolate levels aren’t always so easy to find.  One of the central conceits of Jacques Derrida’s theory of deconstruction (hardly a theory of consciousness) was something called arche-writing: the notion that writing, in some anachronistic way, precedes thought.  Derrida’s isn’t a viable scientific theory, but it does strike at a key truth about modern self-awareness.  “Logic might be imagined to exist independent of writing—” Gleick writes, “syllogisms can be spoken as well as written—but it did not.  Speech is too fleeting to allow for analysis.”  Gleick’s The Information offers perhaps one of the most thrilling speculations of what it must have been like for ancient humans during the shift from orality to literacy, outlining the unseen difficulties that awaited them: “It is a twisting journey from things to words, from words to categories, from categories to metaphor and logic.  Unnatural as it seemed to define tree, it was even trickier to define word, and helpful ancillary words like define were not at first available, the need never having existed.”  What possibility was there for fledgling civilizations working to systematize the inconsistent sounds they made into something like an alphabet?

            The possibility was feedback, of which ancient civilizations could know virtually nothing.  It wasn’t until the nineteenth century, when people like Charles Babbage and Ada Lovelace began contemplating the possibility of calculating machines, that feedback took on a more concrete, material form.  In their infancy, Babbage’s inventions may seem quaint to us today, but they embodied a paradigm shift in the making.  Put simply, Babbage realized that a calculating machine—what he eventually called his Difference Engine—could perform multiple operations at the same time.  In Reckoning with Matter: Calculating Machines, Innovation, and Thinking about Thinking from Pascal to Babbage, Matthew L. Jones writes that “[w]hile the machine was, in its basic design, limited to constant differences [i.e. differences, or subtractions, that were tabularly fixed], Babbage experimented with procedures for computing with variable differences in a process he called ‘eating its own tail” (46).  The reference to eating its own tail invokes the figure of the ouroboros, an emblem of circular causality.  Babbage was effectively imagining a calculating machine whose calculations (its output) would serve as input for future calculations.  In other words, the machine would consume its product in order to produce more; it would eat its own tail.

            Babbage didn’t describe this process as feedback, but he was already anticipating it in the nineteenth century.  He understood the concept even if he didn’t know it by the name we use today.  Our contemporary terminology comes from Norbert Wiener, who placed feedback at the center of his theory of cybernetics—the study of communication and control in animals and machines.  For Wiener, feedback was the key to machine learning.  It wasn’t a matter of starting with some baseline set of knowledge and adding elements along the way.  Wiener wanted machines to be able to circle back, to reflect on their behaviors and limits, and to adapt:

[F]eedback is a method of controlling a system by reinserting into it the results of its past performance.  If these results are merely used as numerical data for the criticism of the system and its regulation, we have the simply feedback of the control engineers.  If, however, the information which proceeds backward from the performance is able to change the general method and pattern of performance, we have a process which may well be called learning.

As engineers and scientists began to experiment with methods and processes of learning in artificial machines, they began to intuit parallels between mechanical behavior and organic behavior.  This led to the profound, if somewhat overexcited, interdisciplinary conversations of the Macy Conferences on Cybernetics, in which engineers and mathematicians debated with biologists, anthropologists, and sociologists about the potential applications of cybernetic theory beyond machine intelligence.  The intellectual community was discovering the underlying presence of feedback loops throughout social order, dictating phenomena as varied as traffic patterns, food chains, alcoholism and climate change.

             The primary meaning of “modern thought” in this post’s title is that of post-Enlightenment knowledge.  With the rise and passing of the twentieth century, feedback loops have become a crucial feature of theoretical models across multiple disciplines; but we still struggle with the logic of circular causality.  It’s no surprise that we find some of the most profound and lucid attempts to work through the puzzle of feedback in the machinations of literary narrative.  From various science fiction conceits to the literary experiments of Ralph Ellison’s Invisible Man, William Burroughs’s Naked Lunch, Samuel Beckett’s Happy Days, and Vladimir Nabokov’s Pale Fire, to the feminist explorations of Kathy Acker and Shelley Jackson, feedback became a driving force in literature.  A quintessential example is the work of Thomas Pynchon, whose Gravity’s Rainbow (1973) is a feedback loop in its very form, its ending leading back to its beginning.  Throughout the narrative, Pynchon explores the dynamics of feedback, often in scientific scenarios.  Pavlovian researchers examine causality in their experiments, a theme that returns throughout the book.  As the novel’s plot grows denser, all vestiges of linear narrative begin to dissipate: “there’s a feeling about that cause-and-effect may have been taken as far as it will go,” one character says; “That for science to carry on at all, it must look for a less narrow, a less… sterile set of assumptions.  The next great breakthrough may come when we have the courage to junk cause-and-effect entirely, and strike off at some other angle.”  In a strong sense, Gravity’s Rainbow is about the emergence of a new mode of scientific thinking: the acknowledgement of circular causality and its importance for modern thought.

            It’s easy to perceive an impasse in the challenge of circular causality—a stumbling block of intellectual thought that permits a concession to unknowability.  In fact, literary studies has occasionally been accused of entertaining precisely this position.  The point of nonlinear models like feedback isn’t to subvert scientific rigor, attention, and experiment, but to introduce the possibility of alternative perspectives.  Human scientists may not be able to internalize a feedback worldview (we are causally perceptive beings, after all), but our machines and instruments of measurement are already running feedback circles around us.  It’s likely that in the decades to come, we’ll continue to encounter feedback models in the simulation results of climate change, social complexity, and cognition.  The more we study and understand the emergence of feedback in the history of science, the better prepared we’ll be for its continued impact on our daily lives.  It might be that poetry precedes philosophy, but that doesn’t mean poetry needs to be without philosophy.  We must find ways to incorporate rational thinking and experimentation into the speculative processes of fiction.  We need to acknowledge the feedback loop between creativity and critique.