“Unthought” keyword

Katherine Hayles

James B. Duke Professor of Literature, Duke University

Keyword: Unthought

Two currents flow into Unthought.  The first comes from Foucault’s comment at the end of The Order of Things (1974) that “the whole of modern thought is imbued with the necessity of thinking the Unthought – of reflecting the contents of the In-itself in the form of the For-itself, of ending man’s alienation by reconciling him with his own essence, of making explicit the horizon that provides experience with its background of immediate and disarmed proof, of lifting the veil of the Unconscious, of becoming absorbed in its silence, or of straining to catch its endless murmur” (356).  Recast in terms of Foucault’s archaeological approach, the Unthought consists of experiences, ideas, concepts, and affects that do not fit into the reigning episteme.  Since, as he has shown, epistemes are historically and culturally specific, this also implies that the Unthought is a moving target; what is Unthought for one age may become accepted doctrine of another.  Foucault’s formulation emphasizes that although epistemic borders may shift, something always remains on the other side, an irreducible otherness that can never be brought into the confines of conscious reflection.

The other current has many manifestations.  My favorite comes from Ursula Le Guin’s science fiction novel The Left Hand of Darkness (1987), in which she imagines the Zen-like cult of the Handdarata.  The Handdarata aspire to ignorance, which they understand not as a simple lack of knowledge but a disciplined undoing of received assumptions, which they call “unlearning.”  That this mode of unthinking—or, as some have called it, thinking without thinking—could have cognitive capacities is a thread running through many literary texts, foregrounded in cognitive literary studies by such scholars as Alan Richardson (2001), Vanessa L. Ryan (2012) and Marcus Iseli (2014).  It has also received popular attention in works such as Malcolm Gladwell’s Blink (2005).

The two currents converge in recent research in neuroscience, cognitive science and neuropsychology (Edelman and Tonino 2000; Dehaene 2016; Damasio 2000; Lewicki, Hill, and Czyzewska 1992), revealing what I call the cognitive nonconscious, a level of neuronal processing inaccessible to conscious introspection but nevertheless performing tasks essential for consciousness to function.  While many literary texts have intuited the existence of the cognitive nonconscious, only within the last couple of decades has it been possible, with more sophisticated methodologies, to determine with precision what those functions are.  They include processing information much faster than consciousness can, discerning patterns too complex for consciousness to perceive, creating a coherent body image, interpreting conflicting or ambiguous information, and drawing inferences from complex data. The cognitive nonconscious differs from background awareness (the mode of perception that broadly monitors environmental information) because it is inaccessible to consciousness; it also differs from the Freudian unconscious because it does not communicate through the coded language of symptoms and dreams.  Rather, the results of its information processing manifest as subtle intuitions that dissipate after about half a second unless they receive top-down reinforcements from consciousness.  Thus the cognitive nonconscious, previously terra incognita, is brought partly within the circle of Thought.  Nevertheless, there remains a gap between understanding these functions intellectually and experiencing them as affective, embodied reality, something that remains forever outside the grasp of consciousness.

The Corporeal Archive

The embodied and enactive brain is the center of our interior archive.  Indeed, it may be considered the ur-archive, since all other archives originate with it.  Just as Foucault’s trajectory led him beyond archaeology to genealogy, so to explore the nature of this interior archive we must go beyond anatomical brain structures to dynamic models of integrated brain processes and their relation of embodied cognition: not the brain as sculpture or plasticized exhibit but as vibrant and transforming interactions: not as noun but verb.

In addition to its dynamism, the interior archive also exceeds the boundaries of conscious reflection.  The cognitive nonconscious reveals, with startling clarity, how extensively our cognitive resources extend beyond consciousness, and so the archive too must go beyond conscious memory into nonconscious processes that underlie memory formation.  Moreover, the archive also reaches beyond the brain into what Lawrence Barsalou (2008)  calls embodied cognition, including muscle memory, affective dispositions, sensations and perceptions in the central and peripheral nervous systems outside the brain, and in such organs as the skin and viscera.  Not just the brain as archive, then, but all the sensing and perceiving sites within the body that receive, process, transmit and store responses to stimuli, so the result may be more properly called a corporeal archive distributed throughout the body and, through the body’s interactions with the environment, into the world (Andy Clark 2008).

Central to understanding this larger sense of the corporeal archive is the work of Walter J. Freeman III, a UC Berkeley neuroscientist and polymath who died in 2016 after making important contributions to a variety of fields, including neural biology, cognitive neuroscience, philosophy of brain research, and mathematical modeling of brain dynamics.  He was the son of Walter Freeman II, a brain scientist who advocated for and practiced lobotomies with mentally ill patients; Freeman II was the principal force behind the (supposedly therapeutic) medical practice of destroying parts of the human frontal cortex (for an account of his legacy, see Day 2013).  Ironically, his son dedicated himself to understanding how the cortex uses interactions with the environment to create meaning.

I will foreground three of Freeman III’s contributions that most strongly connect Unthought with the book’s central theme of uncertain archives.  By tracing the experiments and arguments that form the basis for his work, we can get a better sense of how uncertainty is defined in cognitive brain research and how this research connects it with the ability of brains to think new thoughts.  Since, as Einstein observed, God is in the details, this journey will necessarily take us into the minutiae of specific models and theories.  The point is not to elevate these theories as “truth” but rather to demonstrate the emergence of an epistemological understanding of brain processes that emphasized their flexibility, potential for change, and essential openness to the world.

Unlearning

At a time when neuroscience was split between using microelectrodes to study individual neurons and technologies like electroencephalograms to study cortical activity as a whole, Freeman pioneered the study of mesoscopic cortical research, focusing on coalitions of cortical units, called Hebbian cell-assemblies (named after Donald Hebbs [1949]), that work together to create memory traces and other cognitive functions involved in learning. Experimentally, as Kozma and Noack (2016) explain, this meant that he “took the electrode out from the interior of the individual neuron and placed it in the interstitial matrix between cells . . .  [using electrocorticogram (ECoG) recording to] listen to not only one single neuron but to the collective effect of thousands of neurons in a local pool” (3).  Freeman’s distinction contribution (Societies of Brains, 1995) was to argue for the concept of “unlearning,” showing on a neurobiological level how unlearning dissolved existing patterns of rigid and stereotyped behavior to make way for new forms of behavior.  Moreover, he emphasized that these new forms could be constructed communally through such shared activities as group singing or partners learning to dance together, thus making clear the adaptive advantages of unlearning for the evolution of community cohesion and bonding.  It is unlearning, he concluded, that opens cortical pathways and makes them receptive to novel kinds of experience emerging from cooperative activities and shared affects.  In this sense, he modeled the brain as an “uncertain archive” capable of undoing static forms and creating new possibilities, much as Le Guin imagined in the “unlearning” of the Handdarata.

Chaotic Processes in the Brain

A second contribution emerged from his modeling of brain dynamics as chaotic processes (Skarda and Freeman, 1987), work that established a dynamical and mathematical basis for the process of “unlearning.”  Working with Christine Skarda, a philosopher with an appointment at UC Berkeley as “Lab Associate in Charge of Philosophical Analysis of Models for Interpreting Data,” Freeman conducted a series of experiments on the olfactory bulb in rabbits (a frontal portion of the rabbit brain that detects, analyzes, and recognizes odors), using electrodes implanted along one side of the olfactory bulb and testing the rabbits’ reactions to known versus new odors.  Skarda and Freeman obtained results consistent with the connectionist model that now forms the basis for advanced forms of artificial intelligence such as AlphaGo and AlphaGoZero (DeepMind).  They also identified mechanisms not present in (contemporary) connectionist models that “may be necessary to solve problems critical to the efficient functioning and survival of any system that has to behave adaptively in an environment subject to unpredictable and often violent fluctuations,” suggesting that these could provide clues for further refinement and development of connectionist models (Skarda and Freeman, 161).

Essentially, their data suggested that the base state for olfactory neurons was low-level chaotic dynamics.  Chaotic systems differ from noise or randomness because chaotic systems are deterministic rather than stochastic and operate within attractor basins (Hayles 1990).  Their specific trajectories within the basins cannot be predicted (and in this respect resemble random movement), but unlike random processes, the movements of a chaotic system, as mapped with phase state diagrams, remain within the attractor basin unless or until their energy levels change, as with the famous butterfly shape of the Lorenz attractor.  “Chaos,” Skarda and Freeman explain, “is controlled noise with precisely defined properties.  Any system that needs random activity can get it more cheaply and reliably from a chaotic generator than a noise source” (165).

When a rabbit encounters a new odor, their model indicates that the animal’s neural system is propelled from its base chaotic state into a “high-level chaotic state,” one with more degrees of freedom.  The emergence of this new basin of attraction then “enables the system to avoid all of its previously learned activity patterns and to produce a new one” (p. 171).  Thus they conclude that “without chaotic behavior the neural system cannot add a new odor to its repertoire of learned odors.  Chaos provides the system with a deterministic ‘I don’t know’ state within which new activity patterns can be generated, as shown by what happens when the system encounters a previously unknown odor” (171).

The philosophical implications include a shift from the then-current “pattern completion model,” in which a connectionist system is given partial information and then learns how to complete it, to a model of destabilization that leads to emergent novelty, in which it is not a matter of completing a pattern (for how would the animal know in advance that the data would fit a pattern?).  Rather, they envision that “in an alert, motivated animal, input destabilizes the system, leading to further destabilization and a bifurcation to a new form of patterned activity” (p. 172).  Uncertainty and destabilization thus become the allies, rather than the enemies, of corporeal archives, enabling them to adapt to new situations and learn from them.  In this sense, Freeman’s research has something in common with Gregory Chaitin’s discovery of Omega numbers (2001, 2006), which as Luciana Parisi (2015) has noted, reveals an irreducible presence of randomness in mathematics.  Like Chaitin, Freeman was interested in the positive aspects of uncertainty and destabilization as escape routes from a prison of order so excessive it became rigid and stereotypical.

Around 1988, I heard Feeman give a presentation on this research, when I was in the midst of writing Chaos Bound (1990) and thus very much into chaos and complexity theory.  He concluded with a gesture, unique in my experience, indicating his commitment for going beyond stratified protocols into new forms of behavior and learning.  He thanked the rabbits.

Corporeal Archives in Environments

          A third contribution of Freeman’s work was to develop a model of how an animal’s perceptual and sensory systems related to environments that transformed the animal’s role from passively receiving information to actively seeking it.  As Kozma and Noack (2017) note, Freeman linked the creation of knowledge and meaning to “sequences of oscillatory patterns of activity in sensory network distributed across time and space” (5), a topic developed in Freeman’s book on neurodynamics (2000), which integrated results from EEG (electroencephalogram) and ECoG (electrocorticogram).  As he writes in that book’s preface, “the patterns [of electrical fields detected by EEG and ECoG] are enigmatic, ephemeral, easily dismissed as noise, and by most accounts epiphenomenal . . . Yet, some of the patterns are neural correlates of intentional actions, specifically the perception and discrimination of sensory stimuli by alert, aroused human and animal subjects” (vii).  (The idea that cortical electrical fields could have implications for the construction of intention and meaning has subsequently also been proposed by  Nicolelis and Cicurel, 2015).

Working from this premise, Freeman developed a model of action-perception interactions that moves from searching through the environment to receptors, progressing through sub-cycles that includes loops for proprioception, control, and reafference.  Essentially, as Kozma and Noack summarize, this model moved from “posterior sensory cortices toward the frontal motor cortices, then out in the environment and back into the sensory cortices” (p. 6).  It begins with an animal’s brain reaching out to an environment by searching for a specific sensory stimulus (an “efference signal”), as when a dog or rabbit sniffs the air to detect prey or predator.  This information initiates the cycle, which re-orients the animal’s bodily disposition toward appropriate action.  “Reafference” occurs when the animal has developed certain anticipations about stimuli (based on previous iterations of the cycle) and then experiences something very different.  It represents moments within the cycle when the vast repository of the Unthought breaks through and creates a breach in the boundary of Thought that opens the organism to new experiences and new kinds of cognitions.  In epistemological terms, the model gives priority to curiosity as an intimation of the great unknown beyond, while also foregrounding internal processes that interpret the results of curiosity’s search to create meaning for the organism, which incrementally changes the corporeal archive through the formation of new memories and the creation of novel predispositions.

Connecting Unthought to Thought

         In larger terms, Freeman’s work shows with paradigmatic clarity the reflexive dilemma that humans face as they contemplate their existence:  the embodied brain working to understand how the embodied brain can understand.

In Unthought (2017) I focus on cognition as a key term through which we can re-orient humanistic understanding so that it encompasses nonhumans as well as humans, technical as well as biological cognitions.  Searching for a definition that would have a low entry threshold for something to count as cognitive but also be able to scale upward in complexity, I arrived at the following:  “cognition is a process that interprets information in contexts that connect it with meaning” (22). The formulation emphasizes cognitive processes as interfaces where the great unknown is encountered and transformed into meanings vital to an organism’s survival, in a continuing process of altering and extending the corporeal archive.  This definition implies that all lifeforms have some cognitive capacities, including unicellular organisms; it also positions plants, which comprise 95% of the world’s biomass, as cognitive entities.  In addition, it acknowledges that technical devices, specifically computational media, are capable of nonconscious cognition and thus opens possibilities for critical analyses of complex human-technical systems as cognitive assemblages.

Within this framework, Unthought has multiple significances.  It names the cognitive nonconscious within biological lifeforms as a process inaccessible to consciousness, and only recently admitted into scientific theory; it affirms the existence of extended cognitions throughout the body; it provides a way to think about cognitive technical media in terms that link them to human cognition and agency; and it locates humans within a planetary ecology less anthropocentric and more balanced about the cognitive capacities of nonhumans.  In the most encompassing terms, Unthought names the great unknown itself, considered as an inexhaustible reservoir vast orders of magnitude more complex than any cognitive system, and ultimately the source of all cognitions.

This version of Unthought connects with many other articulations:  virtuality in the work of Deleuze (Deleuze and Guattari 1987; Massumi 2002); transindividuation in Simondon (Simondon 1989; Combes 2012); process reality in Whitehead (Whitehead 1978; Hansen 2015); trace and differance in Derrida (Derrida [1976] 2016); autopoeises in Maturana and Varela (1980); reflexive differentiation in Luhmann (1996) and Wolfe (2009); the “great outdoors” in Meillassoux (2010).  While each of these frameworks has its specific emphases and terms of elaboration, they share a family resemblance in aspiring to articulate and understand how that which exceeds all attempts at articulation and understanding animates and gives vitality to our efforts.

Distinctive to Unthought as I have sketched it here is an emphasis on cognitive interfaces, including but also extending beyond the brain, and on acts of interpretation and meaning as central to the formation of corporeal archives in biological lifeforms and cognitive technical systems.  As I have shown through the example of Walter Freeman III’s work, critical to the creative potential of biological corporeal archives are uncertainty, chaos, and undecidability, traces and intimations of Unthought within Thought.

 

Works Cited

Barsalou, Lawrence W. 2008. “Grounded Cognition.” Annual Review of Psychology 59: 617– 45.

 

Chaitin, Gregory. 2001. Exploring Randomness. Heidelberg: Springer.

———. 2006. MetaMath! The Quest for Omega. New York: Vintage.

Clark, Andy. 2008. Supersizing the Mind: Embodiment, Action, and Cognitive Extension. London: Oxford University Press.

Damasio, Antonio. 2000. The Feeling of What Happens: Body and Emotion in the Making of Consciousness. New York: Mariner Books.

 

Day, Elizabeth. 2013.   “He was bad, so they put an ice pick in his brain . . . “, The Guardian, Jan. 13.               https://www.theguardian.com/science/2008/jan/13/neuroscience.medicalscience.  Accessed Jan. 12, 2018.

Dehaene, Stanislas. 2014. Consciousness and the Brain: Deciphering How the Brain Codes Our Thoughts. New York: Penguin.

 Deleuze, Gilles, and Félix Guattari. 1987. A Thousand Plateaus: Capitalism and Schizophrenia. Translated by Brian Massumi. Minneapolis: University of Minnesota Press.

Foucault, Michel.  1974.  The Order of Things: An Archaeology of the Human Sciences.  New York: Vintage.

Gladwell, Malcolm. 2005. Blink: The Power of Thinking without Thinking. NewYork: Little, Brown.

Combes, Muriel.  2012.  Gilbert Simondon and the Philosophy of theTransindividual, translated Thomas LaMarre.  Cambridge: MIT Press.

DeepMind.  2017.  “AlphaGo: Learning from Scratch.”  Available athttps://deepmind.com/blog/alphago-zero-learning-scratch/. Accessed December 26, 2017.

Derrida, Jacques.  2016.  Of Grammatology, translated Giyatri C. Spivak.  Baltimore: Johns Hopkins University Press.

Edelman, Gerald M., and Giulio Tononi. 2000. A Universe of Consciousness: How Matter Becomes Imagination. New York: Basic Books.

Freeman, Walter J.  2000.  Neurodynamics: An Exploration in Mesoscopic Brain Dynamics.  New York: Springer.

———.  1995.  Societies of Brains: A Study in the Neuroscience of Love and Hate.  Milton Park, UK; Psychology Press.

Hansen, Mark B. N. 2015. Feed- Forward: On the Future of Twenty- First Century Media. Chicago: University of Chicago Press

Hayles, N. Katherine.  2000.  Chaos Bound: Orderly Disorder in ContemporaryLiterature and Science.  Chicago: Chicago University Press.

———-.  2017.  Unthought: The Power of the Cognitive Nonconscious.  Chicago: Chicago University Press.

Hebbs, Donald O. 1949.  The Organization of Behavior: A Neuropsychological Theory. New York: John Wiley & Sons.

Iseli, Marcus.  2014. “Thomas De Quincey’s Subconscious: Nineteenth Century Intimations of the Cognitive Unconscious,” Romanticism 20.3: 294-305.

Kozma, Robert and Raymond Noack.  2017.  “Freeman’s Intentional Neurodynamics. Available at https://binds.cs.umass.edu/papers/2017_Kozma_Noack_Freemans_Intentional Neurodynamics.pdf

Le Guin, Ursula K.  1987.  The Left Hand of Darkness.  New York: Ace Books.

Lewicki, Pawel, Thomas Hill, and Maria Czyzewska. 1992. “Nonconscious Acquisition of Information.” American Psychology 47 (6): 796– 801.

Luhmann, Niklas.  1996.  Social Systems, translator John Bednarz, Jr.  Stanford: Stanford University Press.

Massumi, Brian.  2002.  Parables for the Virtual: Movement, Affect, Sensation.  Durham NC: Duke University Press.

Maturana, Humberto R., and Francisco J. Varela. 1980. Autopoiesis and Cognition:The Realization of the Living. Dordrecht: D. Reidel Publishing.

Meillassoux, Quentin. 2010. After Finitude: An Essay on the Necessity of Contingency.London: Bloomsbury Academic.

Nicolelis, Miguel A. and Ronald M. Cicurel.  2015.  The Relativistic Brain: How It Works and Why It Cannot be Simulated by a Turing Machine.  N.p.  CreateSpace Independent Publishing Platform.

Parisi, Luciana.  2015. “Critical Computation: Digital Philosophy and GAI.” Presentation at the “Thinking with Algorithms” Conference, University of Durham, UK,  February 27. Forthcoming in Theory, Culture and Society.

Richardson, Alan.  2001.  British Romanticism and the Science of Mind.  Cambridge: Cambridge University Press.

Ryan, Vanessa L.  2012.  Thinking without Thinking in the Victorian Novel.  Baltimore Johns Hopkins University Press.

Simondon, Gilbert. 1989. L’individuation psychique et collective: À la lumière des notions de forme, information, potentiel et métastabilité. Paris: Editions Aubier.

Skarda, Christine A. and Walter J. Freeman.  1987.   “How Brains Make Chaos in Order to Make Sense of the World,” Behavioral and Brain Sciences, 102:161-173.

Whitehead, Alfred North. 1978. Process and Reality: An Essay on Cosmology.  Edited by David R. Griffin and Donald W. Sherburne. New York: Free Press/Macmillan.

Wolfe, Cary.  2009.  What Is Posthumanism?  Minneapolis: University of Minnesota Press.