CHAPTER ONE

Neuroscience and Narrative Theory

THE ABILITY TO tell and follow a story requires cognitive capacities that are basic to the neurobiology of mental functioning. Neuroscience cannot of course reveal everything we might want to know about stories, but it is also true that our species would probably not produce narratives so prolifically if they weren’t somehow good for our brains and our embodied interactions with the world. What kind of brains do we have that enable us to tell each other stories? And how do stories configure our brains? Neuroscience and narrative theory have much to say to each other about our species’ facility and fascination with stories. How plots order events in time, how stories imitate actions, and how narratives relate us to other lives, whether in pity or in fear—these central concerns of narratologists from Aristotle to Paul Ricoeur are perhaps surprisingly aligned with hot topics in contemporary neuroscience that I explore: temporal synchrony and the binding problem, the action-perception circuit in cognition, and the mirroring processes of embodied intersubjectivity. The ways in which stories coordinate time, represent embodied action, and promote social collaboration are fundamental to the brain-body interactions through which our species has evolved and has constructed the cultures we inhabit. Crucial questions about how the brain works—how it assembles neuronal syntheses without a central controller—have to do with issues concerning time, action, and self-other relations that are central to narrative. The purpose of this book is to explore and explain these convergences.

Stories help the brain negotiate the never-ending conflict between its need for pattern, synthesis, and constancy and its need for flexibility, adaptability, and openness to change. The brain’s remarkable, paradoxical ability to play in a to-and-fro manner between these competing imperatives is a consequence of its decentered organization as a network of reciprocal top-down, bottom-up connections among its interacting parts. Narrative theorist Seymour Chatman (1978, 47, 45) attributes plot-formation to “the disposition of our minds to hook things together”; as he notes, “our minds inveterately seek structure.” This is, indeed, a basic axiom of contemporary neuroscience. Against the cognitive need for consistency, however, the psychologist William James (1950 [1890], 1:139) describes the brain as “an organ whose natural state is one of unstable equilibrium,” constantly fluctuating in ways that enable its “possessor to adapt his conduct to the minutest alterations in the environing circumstances.” The brain knows the world by forming and dissolving assemblies of neurons, establishing the patterns that through repeated firing become our habitual ways of interacting with the environment, even as ongoing fluctuations in these syntheses combat their tendency to rigidify and promote the possibility of new cortical connections. The brain’s ceaseless balancing act between the formation and dissolution of patterns makes possible the exploratory play between past equilibria and the indeterminacies of the future that is essential for successful mental functioning and the survival of our species.

Stories contribute to this balancing act by playing with consonance and dissonance. Borrowing Frank Kermode’s (1967) well-known terms, Paul Ricoeur (1984a, 65–66) describes emplotment as “concordant discordance”—“a synthesis of the heterogeneous” that configures parts into a whole by transforming the “diversity of events or incidents” into a coherent story. According to Ricoeur, the act of “composing plots” converts “the existential burden of discordance” (33, 31) into narrative syntheses that give meaning to life’s disjunctions by constructing patterns of action. Even in the simplest narratives that approach what Gérard Genette (1980, 35–36) calls the hypothetical “zero degree” of difference between the order of events in the telling and their order in the told, the conjunctions that join together the elements of the plot are invariably disrupted by twists and turns on the way to resolution. What Genette calls temporal “anachronies” (flash-forwards and flashbacks, for example, that disrupt the correspondence between the telling and the told) further play with the competing impulses toward consonance and dissonance that are basic to narrative. The imbalances between pattern formation and dissolution in the brain make possible this narrative play between concord and discord, even as the construction and disruption of patterns in the stories we tell each other play with the tensions in the brain between the competing imperatives of order and flexibility. The neuroscience of these interactions is part of the explanation of how stories give shape to our lives even as our lives give rise to stories.

Stories can draw on experience, transform it into plots, and then reshape the lives of listeners and readers because different processes of figuration traverse the circuit of interactions and exchanges that constitute narrative activity. As I explain in chapter 2, the neural underpinnings of narration start with the peculiarly decentered temporality of cognitive processes across the brain and the body—disjunctions in the timing of intracortical and brain-body interactions that not only make possible but also actually require the kind of retrospective and prospective pattern formation entailed in the narrative ordering of beginnings, middles, and ends. Next, as chapter 3 shows, the strangely pervasive involvement of processes of motor cognition not only in the understanding of action and gesture but also in other modalities of perception suggests why the work of creating plots that simulate structures of action can have such a profound effect on our patterns of configuring the world. Finally, as chapter 4 explains, if stories can promote empathy and otherwise facilitate the cointentionality required for the collaborative activity unique to our species, the power and limits of their capacity to transform social life ultimately depend on embodied processes of doubling self and other through mirroring, simulation, and identification, processes whose limitations are reflected in the strengths and weaknesses of narratives as ethical and political instruments.

Each of the following chapters explores one of these convergences in detail. In all of these areas, narratives configure lived experience by invoking brain-based processes of pattern formation that are fundamental to the neurobiology of mental functioning. To set the stage for these analyses, however, we first need to consider some preliminary questions about the relation between neuroscience and narrative theory. Why should students of narrative pay attention to neuroscientific discoveries about cognition and language, and how should the research program of narratology take account of the science?

Language, the Brain, and Cognitive Narratology

Triangulating our phenomenological experience as tellers and followers of stories with neuroscientific findings about embodied cognition and with narrative theories about plots, fiction, and reading is an attempt to understand the relation between language, cognition, and narrative—a goal that many thoughtful investigators across a variety of disciplines have pursued. One of the reasons why philosophers, literary theorists, and everyday readers have wondered about why and how we tell stories is that narrative seems to hold the key to how language and the mind work. Narratology is now at a turning point in its understanding of the relation between language, cognition, and narrative, poised between (on the one hand) the formalist models of schemes, scripts, and preference rules inherited from structuralism and (on the other hand) pragmatically oriented theories of narrative as embodied, intersubjective interaction. Whether these models can be reconciled and if so how are important, unsettled questions. Understanding the neurobiological bases of narrative may help answer them by showing how the ability to tell and follow stories aligns with how the brain processes language. Not all of the conflicts among students of narrative can be settled by the discoveries of science, but some claims turn out to be stronger than others, and others are just plain wrong. The aims and methods of narratology need to be adjusted in light of what we know about the neuroscience of language and the brain.

The goal of classical narratology was to construct the ideal taxonomy—the classificatory scheme that would identify the fundamental elements of narrative and their rules of combination, based on the model of how grammar and syntax determine meaning by establishing the structural relations between the constituent parts of a logical, ordered system.1 Whether inspired by Saussure’s prioritization of langue over parole (the presumably stable, orderly structures of language as opposed to the contingencies of speech) or Chomsky’s claims about universal grammar (the inborn cognitive structures that constitute what Pinker [1994] memorably calls the “language instinct”), the assumption was that the structures of mind, language, and narrative are homologous, innate, and universal. Classical narratologist Ann Banfield (1982, 234) expressed a view shared by many narrative theorists, for example, when she asserted that “the ingredients for represented speech and thought” in a story “are . . . given in universal grammar” and that the ease with which we create and understand stories is explained by fundamental homologies between the structures of narrative and “the speaker’s internalized grammar.”

Some versions of cognitive narratology still operate within the structuralist paradigm, either tacitly or explicitly. For example, the editors of the recent anthology Stories and Minds: Cognitive Approaches to Literary Narrative assert that “rather than turning away from structuralist narratology, cognitive narratologists . . . build on the insights of structuralism and combine them with cognitive studies” (Bernaerts et al. 2013a, 13). As James Phelan (2006, 286) explains, “cognitive narratology . . . shares with [structural narratology] the same goal of developing a comprehensive formal account of the nature of narrative” and “conceives of its formal system as the components of the mental models that narratives depend on in their production and consumption.” These “mental models” are the frames, scripts, and preference rules that Manfred Jahn defines and explains in his authoritative account of cognitive narratology in the Routledge Encyclopedia of Narrative Theory (2005, 67–71). Articulating the aims of “post-classical narratology,” Jan Alber and Monika Fludernik (2010b, 11) endorse this project: “Cognitive narratologists . . . show that the recipient uses his or her world knowledge to project fictional worlds, and this knowledge is stored in cognitive schemata called frames and scripts.”

Whether these mental constructs can do justice to the cognitive processes they purport to describe is highly questionable, however. The formalist goal of identifying orderly, universal structures of mind, language, and narrative does not match up well with the unstable equilibrium of the temporally decentered brain or the probabilistic processes through which cognitive connections develop and dissolve. There is a growing scientific consensus that the formalist model of language as an innate, orderly, rule-governed structure should be cast aside because it does not fit with what we know about how the brain works. As the science of cognition and language has shifted, so too must narratology revise its methods and aims.

New versions of cognitive narratology have arisen to challenge the structuralist paradigm. Advocates of the 4e (embodied, enactive, embedded, and extended) view of cognition argue that rather than “conceiv[ing] of the mind” as a structure of “abstract, propositional representations” like frames and scripts, narrative theory should understand “the human mind as shaped by our evolutionary history, bodily make-up, and sensorimotor possibilities, and as arising out of close dialogue with other minds, in intersubjective interactions and cultural practices” (Kukkonen and Caracciolo 2014, 261–62). Whereas first-generation cognitive science was “firmly grounded in a computational view of the mind,” with “frames, scripts, and schemata” functioning as “mental representations that enable us to make sense of the world by serving as models of specific situations or activities,” second-generation cognitive science shares with phenomenology and the pragmatism of Dewey and James an emphasis on the interactions between embodied consciousness and the world in “feedback loops” through which “experience shapes cultural practices” even as “cultural practices help the mind make sense of bodily experience” (Caracciolo 2014, 45; Kukkonen and Caracciolo 2014, 267). Rather than prioritizing the construction of taxonomies, schemata, and systems of rules to explain how the mind works and to account for narrative by disclosing its underlying cognitive structures, second-generation narratology “insist[s] on the situated, embodied quality of readers’ engagement with stories and on how meaning emerges from the experiential interaction between texts and readers” (Caracciolo 2014, 4). A quest for structures and rules has been displaced by an emphasis on the interactions between embodied minds, stories, and the world.

Instead of viewing this change as a paradigm shift, some prominent narrative theorists with roots in the first generation have sought ways of reconciling embodied, enactivist narratology with schema theory and formalist, grammatically based models. For example, rejecting the idea that second-generation cognitive science has replaced earlier theories, Fludernik (2014, 406) proposes that they should be seen as informing one another: “A history of cognitive studies might perhaps better start out from an inherent duality in cognitive work—research that is static and abstract flanking research that looks at the body and human experience.” Reminiscent of how structural linguistics juxtaposed synchronic and diachronic approaches to language, this proposal views frames and scripts as “static, abstract” structures that are actuated in experience, much as the structuralists thought the rules of langue are manifested in the speech acts of parole. David Herman seems to have cast aside his earlier project of constructing a “story logic” that reflects transcendental, universal “mental models” (2002, 1–24) in favor of what he calls “discursive psychology” (2010, 156). On this view, meaning is not a product of “mental processes ‘behind’ what people say and do”; “the mind does not preexist discourse, but is ongoingly accomplished in and through its production and interpretation” (156). Still, hoping like Fludernik to rescue formalism and schema theory, he nevertheless asks “how we might work toward a rapprochement between (1) discourse-oriented approaches to the mind as a situated interactional achievement and (2) the work in cognitive grammar and cognitive semantics that likewise promises to throw light on the mind relevance of narrative structures but that focuses on discourse productions by individual speakers” (175). Again echoing structuralism’s opposition between langue and parole, Herman proposes that we think of language as having social and individual sides that could be separately but compatibly studied—but with the switch that pragmatic, interactive theories rather than formal structures would explain the social side, while grammatically based schema theory would provide models for individual mental structures.2

The problem with both of these proposals, however, is that the epistemological assumptions of first- and second-generation cognitive science are irreconcilably opposed: the first views meaning as a manifestation of underlying frames, scripts, and rules, while the second regards it as a product of mutually formative, historically evolving interactions between brain, body, and world. The narratological programs based on these opposing epistemologies are also fundamentally at odds. The second focuses on the figurative, interactive processes through which stories are constructed and experienced, while the first prioritizes the schemes, structures, and rules presumably underlying them.3 The problem is not whether to emphasize what happens in interactions between the body and the world or, alternatively, what goes on in the head. What is at issue is how to correlate neuronally based, embodied cognitive processes with our experience of the social world and with our capacities to tell and follow stories. Two questions are at stake here. How are we to understand the pattern-forming capacities of our cognitive equipment that first-generation cognitive narratologies would formalize into frames, scripts, and preference rules? And how should we understand the regularities of language that formalists would systematize into orderly classificatory schemes and rule-governed structures? Cognitive narratology needs a neuroscientifically sound understanding of language that explains how neuronal and cortical processes interact with our lived experience of the social world.

Not everything, to be sure, in first-generation cognitive narratology need be abandoned. Jahn (2005, 67) describes “ ‘seeing X as Y’ as a foundational axiom” of cognitive narratology, and this idea is indeed scientifically sound. Configurative processes of categorization and pattern formation—what existential phenomenologist Martin Heidegger (1962 [1927]) similarly calls the “as-structure” (“Als-Struktur”) of understanding—are crucial to embodied cognition and narrative, but they need to be understood in nonschematized, interactive form. One reason why gestalt theory has been a resource from which neuroscientists like Semir Zeki (2004), cognitive psychologists like James J. Gibson (1979), and phenomenologists like Maurice Merleau-Ponty (2012 [1945]) have all repeatedly drawn is its appreciation of the role that figuration or seeing-as plays in cognition. This is, for example, the epistemological moral of the famously ambiguous rabbit-duck gestalt (with the beak of the duck shifting if we see the shape as a rabbit, a new part-whole configuration transforming it into a pair of ears). This gestalt is a model of cognition because the circular, recursive work of configurative pattern building (seeing-as) animates not only vision but cognitive processes of all kinds. Making a case for what he calls “carnal hermeneutics,” Richard Kearney (2015, 20) similarly observes that the “ ‘as-structure’ is already operative in our most basic sensations.” This is because, as Merleau-Ponty (2012 [1945], 162) points out, “the slightest sensory given cannot be presented except as integrated into a configuration and as already ‘articulated.’ ” It is consequently a basic principle of contemporary neuroscience that “categorization (or conceptualization) is a fundamental process in the human brain. . . . There are ongoing debates about how categorization works, but the fact that it works is not in question” (Lindquist et al. 2012, 124).

The as-structure of categorization—how seeing always entails seeing-as—is also evident in the circularity of literary interpretation (see Armstrong 2013, 54–90). Literary theorists have long recognized that interpretation is inherently circular because understanding a text or any state of affairs requires grasping in advance the configurative relation between part and whole. Any act of interpretation sets in motion a reciprocal interaction between part and whole because a detail makes sense only if it can be seen as somehow relating to the entire text, even as the whole can only be understood by working through its parts. This epistemological theory about the need for pattern—the idea that part and whole are reciprocally constructed and together construed as a configurative relation of some kind—is common ground between the humanities and the cognitive sciences.

It is a mistake, however, to reify these configurative processes into mental modules that bear no relation to the anatomy of the embodied brain or to posit linear logical models of cognitive decision making that do not correspond to the reciprocal, to-and-fro movements of figuration in experience, in the cortex, or in the interactions between brain, body, and world. These are some of the problems with the terminology of frames, scripts, and preference rules employed by cognitive narratology. As Jahn (2005, 69; see 1997) acknowledges, these notions were developed by artificial intelligence theorists “to replace the concept of context by more explicit and detailed constructs” that “aim at reproducing a human cogniser’s knowledge and expectations about standard events and situations”—with frames referring to “situations such as seeing a room or making a promise” and scripts encompassing “standard action sequences such as . . . going to a birthday party, or eating in a restaurant.” The brain is not a computer, however. As Hubert Dreyfus (1992) points out, computers lack context, background, and prior experience that we as embodied conscious beings typically employ in testing hypotheses about how to configure a situation we encounter, whether in a text or the world, and replacing this deficiency by positing preset mental constructs that do the work only displaces the problem that needs to be solved. Rather than explaining the processes whereby the embodied brain configures experiential contexts, these constructs instead call attention to what computers can’t do.

Seeing-as sets in motion interactions between brain, body, and world that are fluid, reciprocal, and open ended, and preset schemata like frames and scripts are too rigid and linear to do justice to these sorts of dynamic, recursive processes. This is why psychologist Richard Gerrig (2010, 22), whose work on reading is widely (and rightly) respected among cognitive narratologists, has recently parted company from what Jahn describes as the mainstream view, in the process rejecting the term “schema” as too rigid and formulaic. Gerrig prefers instead to speak of “memory-based processing,” a concept that recognizes that “readers’ use of general knowledge” is “more fluid and more idiosyncratic” than the terminology of frames and scripts can capture.

The linear, overly tidy notion that cognition is governed by preference rules also needs to be abandoned. According to Jahn, “A preference rule is usually cast in the form Prefer to see A as B given a set of conditions C” (69). In its favor, the notion of preference is not absolute and leaves a little wiggle room for probabilistic variation, but the problem with structuring preferences into “rules” is that they posit a linear chain of decision making, following the form of a logical proposition: if C, then A implies B. This linear, mechanical, logical structure is not an adequate representation of how cognitive decision making happens either in neurobiology or experience. Neurobiologically, it bears little relation to the interactive, top-down, bottom-up processes of the dynamical systems of synchronization and desynchronization in the brain. Neuronal assemblies form and dissolve according to patterns of habituation that result from the reciprocal reinforcement of connections that can be displaced by other syntheses, and these interactions are not like linear, mechanical algorithms (see Edelman 1987). Experientially, the unidirectional logic of preference rules is unable to capture the to-and-fro circularity of seeing-as in the phenomenological process of configuring part-whole relations in a text or in life. Reading is not linear logical processing, and embodied cognition cannot be adequately modeled either by ordered hierarchies of modules or mechanical, linear algorithms.

The work of seeing-as is not localizable in any particular region of the cortex but extends across the brain, the body, and the world. It is not governed by rules but develops habitual patterns through repeated experiences and is consequently always open to disruption, variation, and change. The formalist goal of identifying orderly, universal structures of mind, language, and narrative doesn’t match up well with the messiness of the brain or with how cognitive patterns emerge from our embodied experiences of the world. The consensus among neuroscientists is that the brain is a bushy ensemble of anatomical features whose functions are only partly fixed by genetic inheritance and are to a considerable extent plastic and variable depending on how they connect in networks with other, often far-flung cortical areas. These connections develop and change through experience according to Hebb’s law (2002 [1949]), a fundamental axiom of neuroscience: “Neurons that fire together, wire together.” As Stephen E. Nadeau (2012, 1) points out, “Brain order is chaotic rather than deterministic; rules are not defined but instead emerge from network behavior, constrained by network topography” and connectivity (not all parts of the cortex can do everything, and they cannot interact if they are not linked by the axons through which neurons exchange electrochemical charges). Whatever order can be found in language and cognition results, he explains, from patterns of reciprocal relationship “acquired through experience,” and these patterns are attributable less to innate, genetically determined anatomical structures than to “statistical regularities of experience.”4

The brain, in short, is not an orderly structure consisting of rule-governed relations between fixed elements like a computer with hard-wired connections between components that operate according to logical algorithms. Much messier, more fluid, and more open to unpredictable (if not unlimited) developments than the linear, mechanical model assumes, the brain is an ever-changing ensemble of reciprocally interacting parts whose functions may vary according to how they combine with other elements. Once popular when artificial intelligence concepts dominated first-generation cognitive science, modular models of the brain (see Fodor 1983) have fallen out of favor because cortical regions are not autonomous and orderly. As Shaun Gallagher (2012, 36) observes, the brain is “a dynamical system [that] cannot be explained on the basis of the behavior of its separate components or in terms of an analysis that focuses on the synchronic, or static, or purely mechanical interactions of its parts”; “the parts of a dynamical system do not interact in a linear fashion” but rather “in a non-linear way, reciprocally determining each other’s behavior” (see Kelso 1995, 2000). Patterns of relationship can become established over time as particular interactions recur and reinforce existing connections or propagate and strengthen new ones, but how repeated experiences lead to the formation of habits through Hebbian “firing and wiring” is a better model for understanding these patterns than the genetically fixed, orderly structures assumed by the epistemological formalists. Preprogrammed modules and linear algorithms are not a good model for understanding the workings of the brain.

The structures of neural anatomy are limiting but not ultimately defining. Different cortical locations have particular functions that can be disabled if they are damaged, but no region works alone, and the role of each region can vary according to how it reciprocally interacts with other areas. Function and connectivity can change with experience. The visual cortex of a blind person, for example, can adapt and become responsive to touch when he or she is reading braille (see Changeux 2012, 208), and some sight-deprived people as well as animals have been shown to have superior sound localization because the unused parts of their visual cortex are recruited for auditory functions (see Rauscheker 2003). These instances of plasticity may seem exceptional, but they are examples of the general rule that, as a study by Kristen A. Lindquist and colleagues (2012, 123) explains, the “function of individual brain regions is determined, in part, by the network of brain regions it is firing with.” According to Lindquist’s group, this is why there is “little evidence that discrete emotion categories can be consistently and specifically localized to distinct brain regions” (121). Their review of the experimental evidence shows, for example, that the amygdala is not uniquely and exclusively associated with fear but is also active “in orienting responses to motivationally relevant stimuli” that are “novel,” “uncertain,” and “unusual” (130). Various studies have similarly shown, they point out, that the anterior cingulate cortex, typically connected with disgust, “is observed in a number of tasks that involve awareness of body states,” including “body movement,” “gastric distention,” and even orgasm (133–34).

This research calls into question Patrick Colm Hogan’s claim (2010, 255; see 2011a, 2011b) that “emotion is . . . the response of dedicated neurobiological systems to concrete experiences, not a function of the evaluation of changing situations relative to goals.” Lindquist is a member of Lisa Feldman Barrett’s laboratory that has led the challenge to the theory of basic emotions promulgated by Paul Ekman and Silvan Tomkins (also see Colombetti 2014, 26–52). As Barrett (2017, 22, 23, 33) explains, a large and growing body of neuroscientific and psychological research has challenged the view that emotions are universal classes with objective biological markers:

Overall, we found that no brain region contained the fingerprint for any single emotion. . . . Emotions arise from firing neurons, but no neurons are exclusively dedicated to emotion. . . . [A]n emotion word, like “anger,” does not refer to a specific response with a unique physical fingerprint but to a group of highly variable instances that are tied to specific situations. . . . [T]he emotions you experience and perceive are not an inevitable consequence of your genes. . . . Your familiar emotion concepts are built-in only because you grew up in a particular social context where those emotion concepts are meaningful and useful, and your brain applies them outside your awareness to construct your experiences.

Emotions are mixed products of biology and culture that are better thought of as variable, internally heterogenous populations than logical categories or universal classes with fixed neurobiological foundations.

Anatomical location and cortical structure alone cannot explain embodied cognition. Brain-body-world interactions can affect not only internal connectivity but also the functions of particular cortical regions. To understand a complex cognitive phenomenon like vision, emotion, or language, it is not enough to identify structure and modularity (as the formalist models assume); it is necessary, rather, to trace the configurative, nonlinear, to-and-fro processes through which various components of our dynamic cognitive systems interact and reciprocally constitute each other.

A good example of the brain’s combination of anatomical specialization and openness to change through experience is the manner in which the visual cortex adapts inherited functionalities in order to support the unnatural, culturally acquired capacity to read written texts (see Armstrong 2013, 26–53). As Stanislas Dehaene (2009, 4) points out, we learn to read Shakespeare by adapting cortical capacities that our species acquired on the African savannah. This is an instance of what M. L. Anderson (2010) calls “neural reuse”—the capacity of cortical regions to acquire functions for which they did not first evolve. New, unpredictable experiences with the world may set in motion variable interactions between different areas of the brain and the body as well as with other members of our species that can produce fundamental changes in cortical structure and functionality. As the visual and auditory cortices interact during the often arduous processes through which beginning readers learn to associate word shapes with phonetic sounds (also activating parts of the motor cortex associated with the mouth and the lips that fire not only in the articulation but also during the recognition of speech), connections get established and reinforced between different regions of the brain that have the effect of converting a specific area of the visual cortex to a culturally specific use (the recognition of visual word forms) for which it was not innately, genetically predetermined. The acquisition of the ability to read may be an extraordinary cultural and neurobiological accomplishment, but as an example of neural reuse, it is simply an illustration of what Anderson calls “a fundamental organizational principle of the brain” (245).

Our species’ development of the capacity to read illustrates the dual historicity of cognitive functions (see Armstrong 2015). Some of our epistemological equipment is based on long-term, evolutionarily stable capacities like the responsiveness of the visual system to edges, orientation, lines, and shapes, but these capacities are open to change depending on learning and experience—they can be recruited, in this case, to identify alphabetic signs—because the function of a cortical region depends on how it interacts with other components of the dynamical system in which it is engaged. The brain can be molded by cultural institutions (like literacy) that adapt particular areas and capacities for their purposes, but as with reading, these capabilities need to be relearned with each generation until or unless the neural reuse through which they are repurposed becomes evolutionarily adapted into the biological makeup of the species. The structures and functions of the brain are historical, not universal, because they are the products of evolution, but some capacities are more enduring than others and are shared by members of our species across time and around the globe, even as they get reshaped and repurposed through particular, historical, culturally situated experiences of learning.

Language is what neuroscientists call “a biocultural hybrid” (Evans and Levinson 2009, 446) that develops through the interaction of inherited functions and anatomical structures in the brain with culturally variable experiences of communication and education. Although some parts of the brain are known to be linked to language (lesions in Broca’s and Wernicke’s areas, for example, can disrupt syntactical or semantic processes), Nadeau points out that “linguistic function taps the entire cerebrum” (83), and recent fMRI-based research has confirmed that language entails far-flung syntheses of cortical areas and connections between the brain and the body (see Huth et al. 2016). There is no single module that governs language and no discrete, anatomically identifiable set of regions that would constitute the grammar unit predicted by structural linguistics. As Nadeau explains, “The grammar anyone of us uses is not intrinsically universal. . . . Instead it is based on the statistical regularities of our own linguistic experience (instantiated in neural connectivity), which have been determined by the modest community of people we have conversed with or read” (164–65). Cases of aphasia in different languages reveal not an anatomically based, universal grammar system that gets knocked out with the loss of language function but rather what Nadeau calls “graceful degradation” (17). Everything doesn’t simply collapse and disappear, but some functions are more or less strongly preserved in different patterns of vulnerability that depend on cross-cerebral connections and redundancies and that vary between linguistic communities. This evidence is better accounted for by the stochastic, probabilistic regularities established through Hebbian connections and developed through experience than by a logically ordered, innate grammar.

Such a probabilistic model also helps to explain the duality of language as a set of regularities open to innovation, variation, and change. As Jean-Pierre Changeux argues, the Hebbian explanation of stochastic regularities offers a better account of the creative capacities of language than prefixed formal systems can provide (see 2012, 206–7, 316–17). On the one hand, language is a set of shared codes, evident in its recurring patterns, that support intersubjective communication and well-formed sentences. On the other hand, the irregularities of language are also vitally important because they make possible unpredictable if constrained linguistic innovation through rule-governed or rule-breaking creativity. In accord with a probabilistic model, structures do not completely decide in advance all the ways they can be used (innovation within the rules is possible), and sometimes new configurations can emerge as previous connections are replaced by new ones (transgressing existing rules is not always wrong, as with a novel metaphor that at first may seem like a mistake but then becomes accepted and gets adopted into the lexicon [see Ricoeur 1977]).

If language and narrative are biocultural hybrids, any transcultural, transhistorical regularities in their functions and forms are a product of variable but constrained interactions between brain, body, and world and not universals that are homologous to logical structures of the mind. The sources of these regularities are typically both biology and culture; it’s not simply that nature is fixed and culture variable. Similarly, any recurring patterns in the stories we typically tell each other are the mixed products of interactions between our species’ neurobiological equipment and repeated experiences we are likely to undergo. If stories across the world have recurrent forms, this is not a result of narrative structures that reflect universal cognitive schemata. Rather, as biocultural hybrids, the patterns identified by various narrative theories have probably developed because evolved cognitive proclivities shared by members of our species have interacted with recurrent, typical experiences to produce configurative relations between brain, body, and world that demonstrate statistical regularities. These patterns are not logical structures but habitual configurations that are variable but constrained within limits that are attributable to the regularities of both biology and experience.

Consider, for example, Hogan’s claim (2003, 230–38) that certain narrative universals characterize the mind and its stories—story structures that he identifies as the romantic, the heroic, and the sacrificial. How to understand cross-cultural “universals” is a notoriously difficult matter. For example, arguing that the claims of relativism overstate the differences between cultures, Donald E. Brown (1991, 9–38) carefully distinguishes between different kinds and degrees of universality—universals of “essence,” attributable to the biological characteristics of our species, as opposed to universals of “accident,” produced by widely shared experiences, some of which may be “near universals,” probably all encompassing but at least broadly evident, and “statistical universals” that may not be omnipresent but are more common than would be predicted by chance. Hogan (2010, 48–49) admits different kinds and degrees of universality, but he thinks and talks like a structuralist: “Hierarchies of universals are defined not only by the schematization of techniques and by a receding series of explanatory abstractions but by a series of conditional relations. . . . Much as unconditional universals may be subsumed into hierarchies of abstraction, implicational universals may be organized into typologies.” This kind of logical formalism is not a good way of thinking about the messy, probabilistic development of regularities that characterize biocultural hybrids like language and narrative.

If narrative patterns like those identified by Hogan recur across cultures, that is not because they reflect universal cognitive structures. They are better understood as biocultural hybrids—recurrent configurations that develop because certain repeated characteristics of our species’ shared experiences of birth and death, collaboration and competition, propagation and violence interact with biologically based cognitive proclivities to produce statistically discoverable regularities in cultural institutions, including the stories we circulate in our communities. Given the commonalities in the basic experiences members of our species typically undergo in their journeys from birth to death, it would be surprising if the cognitive configurations established through Hebbian connectivity between our brains, bodies, and worlds did not demonstrate various regularities that would show up in our narratives. Members of our species fall in love and have sexual relations, engage in conflicts that produce winners and losers, and form communities that join some members and exclude others, and the configurative powers of pattern formation based on the connective capacities of our embodied brains build narratives about these experiences that may evince various regularities (Hogan’s stories of romance, heroism, and sacrifice).

It is misleading to call these “narrative universals” or to attribute them to a structural logic of “the mind and its stories,” because these terms are too static, orderly, and ahistorical to do justice to the messy, dynamic processes through which biocultural hybrids get produced in the interactions of brain, body, and world. These interactions may produce patterns that demonstrate regularities because habitual connections are established through Hebbian processing and neural reuse that are then passed on by cultural sharing of the kind through which, for example, literacy is developed and handed down. But formalist terminology and structuralist models are not good tools for describing these processes because such concepts misrepresent the way habitual patterns of connection and configuration get made and transformed in experience and in the brain. Formal taxonomies are not sufficient to explain these interactions.

Figuration in Narrative and Cognition

Figuration, or seeing-as, is the key concept joining cognitive science and narrative theory, and it is central to the neurophenomenological model of narrative that I construct in the following chapters. The configurative powers of embodied cognition are a primary concern of phenomenological theories of how we tell and follow stories, from Roman Ingarden’s early analysis of the polyphonic concretization of the multiple strata of a literary work to the theory of reading as a process of consistency building and gap filling developed by Konstanz School theorists Wolfgang Iser and Hans Robert Jauss.5 Paul Ricoeur’s monumental work on time and narrative similarly foregrounds the role of configurative processes in cognition and aesthetic experience.6 Explaining the figurative processes at work in a phenomenological model of narrative is one of the aims of this section.

A second, related aim is to show what a focus on cognitive figuration can contribute to narrative theory. Sometimes narrative theory can seem like an exercise in the generation of classificatory schemes for their own sake, without regard for their practical purpose, unfortunately resulting in what James Phelan (2006, 283) describes as “the large Terminological Beastie looming over the field.” Even if we need to reject the structural, formalist epistemology implicit in the taxonomical project, that does not mean we have to discard all narrative terms as useless. Challenging this epistemology should, rather, foreground the question of what their uses are, and triangulating narrative theory, our experience with stories, and the neuroscience of embodied cognition is a way of posing that question. Many of the ambiguities in the lexicon of narrative theory are a result of the ways in which figurative activity crosses back and forth between aspects of narrative that narratological distinctions attempt to discriminate but can’t always keep tidily in place. Attending to the passage of figurative activity across different aspects of narrative with the help of a phenomenological model can clarify some of these ambiguities.

Ricoeur describes emplotment as a response to the enigmas of time that famously troubled Augustine (1961 [397–400], 264): “What, then, is time? I know well enough what it is, provided that nobody asks me; but if I am asked what it is and try to explain, I am baffled.” Time is a scandal for speculative thought because various aporias and paradoxes immediately arise as soon as we try to provide a rigorous theoretical account of the lived experience of time passing. “How can time exist,” Ricoeur asks, “if the past is no longer, if the future is not yet, and if the present is not always?” (1984a, 7). What William James calls “the specious present” (1950 [1890], 1:609)—our everyday sense of the moment as “a duration, with a bow and a stern, as it were—a rearward- and a forward-looking end”—is more bewildering than we ordinarily realize. As Merleau-Ponty (2012 [1945], 441) points out, “My present transcends itself toward an imminent future and a recent past, and touches them there where they are, in the past and in the future themselves.” Hence Husserl’s (1964 [1928], 48–63) description of the passing moment as bounded by “retentional and protential horizons,” a metaphor meant to suggest the paradoxical connectedness of the present to past and future experiences that we sense across its borders even if we cannot reach them. But the horizon metaphor describes what still needs to be explained: how can the present “actually” touch the past and the future? Or, as Shaun Gallagher and Dan Zahavi (2008, 75) ask, “How can we be conscious of that which is no longer or not yet?” We respond to these enigmas, according to Ricoeur, by telling and following stories. Finding in Aristotle an answer to Augustine’s question, Ricoeur argues that “speculation on time is an inconclusive rumination to which narrative activity alone can respond,” not by resolving these aporias theoretically but by making them productive—“put[ing them] to work—poetically—by producing” plots that transform the paradoxical disjunctions and connections of lived time into coherent figures of “discordance and concordance” (6, 22).

According to Ricoeur, this work of transforming temporal aporias into structures of discordant concordance is evident in each of the three parts of Aristotle’s well-known definition of a plot as 1) an imitation of an action that 2) combines a beginning, middle, and an end into a unified whole, which then, in the case of tragedy, 3) induces pity and fear in the audience. “It is only in the plot that action has a contour” and “a magnitude,” Ricoeur argues, as the immediacy of lived experience is shaped into dramatic form (39). Plots configure action by negotiating the competing claims of change and pattern in the tensions (for example) between the accidental and the necessary in decisions about what is probable (and not merely possible) or in the opposition between the expected and the unexpected in the resolution of a story’s conflicts. Similarly, Ricoeur claims, “it is only in virtue of poetic composition that something counts as a beginning, middle, or end” (38). This structuration of succession into a logic of connection governed not by chance but by “conformity to the requirements of necessity or probability” (39) transforms the one thing after another of lived time into a differentiated order that endows some events with the status of beginnings that meaningfully project us forward toward other moments that relate backward to them as ends. Finally, whatever the much-disputed term “catharsis” may entail, the emotions of pity and fear it arouses also provoke opposed movements of relatedness and disconnection as we empathize with the travails of the characters only to undergo the repulsions of terror, in both cases experiencing the paradoxical doubleness of an identification of ourselves with who we are not. In each of these dimensions of emplotment, the disjunctions and connections between the horizonal moments of time passing make possible the formation of narrative configurations that synthesize the heterogeneous.

Narrative is a productive response to the aporias of time because stories reorganize what Ricoeur (1984b, 338) calls “the ordinary experience of time, borne by daily acting and suffering.” Stories give intelligible form to the lived immediacy of our interactions with the world, embodied experiences that are already meaningful but that we may not fully comprehend. The work of narrative, according to Ricoeur (1980a, 151), is “the unceasing passage . . . from a prefigured world to a transfigured world through the mediation of a configured world”—a circuit of reciprocal, mutually transformative interactions that he calls “mimesis1,” “mimesis2,” and “mimesis3” (see 1984a, 52–87). “Mimesis” is, Ricoeur acknowledges, a potentially misleading term for what he has in mind. “If we continue to translate mimesis by ‘imitation,’ ” he explains, “we have to understand something completely contrary to a copy of some preexisting reality” or “some redoubling of presence” through representation because “artisans who work with words produce not things but quasi-things; they invent the as-if” (1984a, 45), “a kind of simulacre, . . . in the sense precisely that fiction is fingere, to feign and figure, or better configure” (1980a, 139; also see Iser 1993). The language of figuration, configuration, and refiguration is preferable to the terminology of imitation, representation, and copying not only because the latter terms are heavily freighted with referential associations but also because the term “figure” suggests the activity of constructing gestalts or patterns that is basic to embodied cognition at all levels—from neuronal assemblies to the interactions of brain, body, and world—and that is also integral to the process of telling and following stories.

Different processes of figuration cross and interact as stories transform experience into narrative patterns. Mimesis2 is the configurative work of creating stories by fashioning incidents and events from life into plots, but it in turn draws on prefigurative patterns of experience (mimesis1), the intentional activity of our nonreflective being-in-the world that includes the “implicit categorization of the practical field” by the “cultural stock” of inherited norms and prior narratives that the poet utilizes (1984a, 47). Because action itself is characterized by preinscribed “temporal structures that evoke narration,” Ricoeur describes “life as an activity and a desire in search of a narrative” (1987, 434). But life is always also culturally and symbolically mediated, he notes, and “without myths that have been passed on, there would be nothing to transform poetically” (1984a, 47). This configuration of life into plots and stories is, however, not an end in itself but only a stage along the way: “Structuration is an oriented activity that is only completed in the spectator or the reader” (48) through the potentially transfigurative experience of comprehending the narrative (mimesis3). The reception of the narrative can in turn bring about a reshaping of the audience’s emotional, embodied, and culturally mediated sense of its world.

And so the circuit is completed, only to stand ready to begin again, as culturally shared and shaped patterns of being-in-the-world (mimesis1) are taken up and refashioned by poets, writers, and storytellers of all kinds (mimesis2) who offer refigured narratives to ever-new audiences that may play with and transform the configurations through which they experience the world (mimesis3). It may be, as Ricoeur observes, that “stories are told and not lived” whereas “life is lived and not told” (1987, 425), but there is a circuit between living and telling that is mutually formative and potentially transformative, and that is because the work of figuration crosses and joins the three modalities of mimesis. Stories have the power to pattern and repattern our brains because the configurative activity of narration across the circuit of the three types of mimesis sets in motion and can shape and reshape fundamental processes of embodied cognition.

An analysis of the configurative powers of narrative and the embodied brain can clarify a number of conundrums in narrative theory. Focusing on the ability of stories to forge patterns of concordant discordance out of our heterogeneous temporal experience can dissipate at least some of the terminological fog by showing what matters experientially and cognitively in some notorious disputes. Consider, for example, E. M. Forster’s (1927) well-known but highly problematic distinction between a story and a plot. These simple terms can seem hopelessly confusing until one understands their relation to the production of concordant discordance.7 According to Forster’s oft-quoted definition, “The king died and then the queen died” is a story because it recounts events in time, whereas “the king died and then the queen died of grief” is a plot because it supplies motivation and causality. Whether the report of the king’s and queen’s successive deaths is by itself a story is unclear, however. Anyone hearing of these events would want to know “why?” Is there a connection? Or are these deaths merely accidents, contingencies, random happenings? What is missing from the initial, question-begging account is the as-structure of figuration. The one thing after the other of temporal succession only becomes a story when the sequence of events is made meaningful and understandable by answering these questions, thereby configuring them into a pattern that makes the discordant concordant.

As Ricoeur (1980b, 171; original emphasis) compactly if somewhat cryptically explains: “A story is made out of events to the extent that plot makes events into a story.” In other words, a story needs a plot in order to become a story because emplotment synthesizes the heterogeneous—building figurative consistency between elements that otherwise sit inertly side by side, awaiting connection and explanation. When plot transforms a mere sequence of events into a story, a narrative pattern makes concordant the juxtaposition of events (like these two deaths) that might otherwise seem random, accidental, and discordant—and this configurative connection is what the addition “of grief” supplies. Other explanations would produce different stories by arranging the events into different patterns—for example: “as the assassin turned his gun from the monarch to his bride” or “as the poison they had taken gradually did its lethal work.” Each of these potential plots configures events into a different story by projecting a different mode of seeing-as.

Another related ambiguity that often troubles narrative theorists is whether the plot is an element of the telling or the told. The arrangement of events in the telling is typically regarded as belonging to the discourse, borrowing Seymour Chatman’s well-known terminology, whereas the story consists of the events that are told. Similarly, narratologists sometimes employ the terms “fabula” and “sjužet,” first introduced by the Russian formalists, to differentiate between “the order of events referred to by the narrative” and “the order of events presented in the narrative discourse” (Brooks 1984, 12). The problem with these distinctions, of course, is that we only know the story through the discourse, and the telling shapes the told. Is the explanation of the queen’s grief part of the telling or the told, the sjužet or the fabula? Is it an interpretation added by the discourse or a key element of the story? A narrator’s insertion of a psychological inference as he or she recounts what happened or a crucial connecting element between otherwise inexplicable events in what is told? As Peter Brooks observes, “the apparent priority of fabula to sjužet is . . . a mimetic illusion, in that the fabula—‘what really happened’—is in fact a mental construction that the reader derives from the sjužet, which is all that he ever directly knows” (13).

It is easy to get lost in the terminological fog here. Once again, however, the key issue to keep in mind is how narratives perform the work of figuration. Because the arrangement of events in the discourse is itself a way of creating syntheses of the heterogeneous, the boundary between story and discourse is inherently blurry and unstable. Both entail configurative processes (modes of seeing-as) that can mix, mingle, and play off of each other. Discordant concordance is a characteristic not only of emplotment (how events in the story are configured into meaningful patterns) but also of the act of narration (how they are organized in the recounting, with different patterns of presentation distinguishing how various storytellers see the same events). The role of seeing-as in narration also explains why Gérard Genette’s much-discussed distinction between mood (“who sees”) and voice (“who speaks”) is notoriously slippery. The narrator’s voice is itself necessarily a way of seeing—a perspective on the events he or she narrates by shaping them into one configuration or another that sees them as a particular gestalt, a reciprocal arrangement of parts in a whole.8

The interplay of configurative processes between the telling and the told—between the shaping, perspectival activity of narration and the patterns of action organized by the plot—is open to infinite permutations that defy the taxonomist’s best efforts to sort them into separate categories. The interactions between these patterns of configuration refuse to stay neatly in one narratological box or the other. Hardly unfortunate, however, this merging and mingling is productive because the interplay of configurative processes between telling and told makes possible unlimited narrative innovation. From Homer’s Odyssey to Joyce’s Ulysses, for example, the same configuration of events can be repatterned in ever-changing acts of narration. The boundary crossing that confuses narratological classification is evidence of how semantic creativity occurs by reshaping configurative patterns.

The configurative work of building cognitive patterns characterizes not only the telling and the told but also the relations between them, and that is another reason why the distinction between story and discourse is so hard to pin down. To follow a story, readers not only need to understand the events and appreciate the storyteller’s artful ways of presenting them; they must also align the sometimes congruent but sometimes disjunctive, even occasionally bewildering relations between the story and the discourse. Think, for example, of the difference between the Odyssey and Ulysses, where Joyce’s experiments with style provide an ironic commentary on the journeys of his heroes, experiments that sometimes, in their antic playfulness, seem to take on a life of their own—hence one Joyce critic’s well-known description of the novel as an “odyssey of style” (Lawrence 1981). The fun or the frustration (depending on your point of view) of reading Ulysses is that these levels do not line up as straightforwardly as T. S. Eliot claims in his assertion that Joyce’s “mythic method” holds in parallel Homer’s story and the events of Bloomsday in Dublin. The challenge of making sense of these ironic, playful allusions that typically refuse to settle down belongs neither to the story nor to the discourse but to how they interact and align (or don’t quite fit together). If reading is an activity of consistency building, as Wolfgang Iser (1978) argues, the work of constructing configurative patterns in our experience of stories occurs not only within but also between these interconnected domains that narrative theory only artificially separates.

Configurations of discordant concordance characterize story, discourse, and their interactions in our experiences of narratives. Complex patterns of figuration may occur even in Genette’s so-called zero-degree narratives (1980, 35–36)—stories where the telling seems to be a transparent reflection of the told and the events related are interesting primarily because of the connections between them and not because of the artfulness of the telling. Meir Sternberg’s (1987) description of narrative as a construct based on curiosity, suspense, and surprise holds even for such simple tales because these responses can be evoked by the story itself—the conjunctions and disjunctions in the patterning of the told. Hence Aristotle’s (1990 [355 BCE], 12–13) well-known concern with reversals and recognition in the construction of plots—changes of fortune like the fall of the tragic hero, for example, or the revelation of unexpected facts or relationships—and his claim that “the finest recognition is the one which occurs at the same time as the reversal, like the one in Oedipus [Rex]” where the hero’s demise is precipitated by his discovery of his parentage. Moments of reversal and recognition are crucially important in the construction of plots because of their role in disrupting and re-establishing configurative patterns. Zero-degree narratives are still narratives to the extent that they are structures of concordant discordance and play with the competing cognitive imperatives of pattern and change that stories set in motion and manipulate.

Another reason why the boundary between story and discourse is blurry is that the arrangement of events in the story is itself artful, even in the case of the simplest tales. Even in stories where the telling seeks to disappear and the discourse does not call attention to itself, Sternberg’s curiosity, suspense, and surprise may be enhanced or spoiled by different modes of emplotment. This is why, after all, Aristotle thinks he can give useful advice about what makes for a better or worse plot, more or less artfully patterned. There is an art to making good plots, and there is an art to telling good stories, and it is sometimes difficult to tell the difference between emplotment and narration because both entail acts of configuration.

Despite these ambiguities and complications, the distinction between story and discourse is still useful. The line separating story and discourse can be hard to draw because the configurative work of constructing patterns of discordant concordance crosses back and forth across it, but it is sometimes useful to ask what side of the line we’re on. Sometimes our interest in a narrative has mostly to do with one side or the other—that is, with the relations between narrated events themselves or with how they are assembled in the process of narration. The elements in what is told may seem intrinsically interesting because their connections and disjunctions seem curious, anomalous, or surprising, or the artfulness of the telling may transform an otherwise unremarkable set of events into a narration that holds us in suspense or keeps our curiosity alive by the way information is offered, implied, or withheld. The same story can be recounted more or less well by different tellers who are trying not to embellish or digress.

The opposition between story and discourse can help to disclose different modes of configuration that can interact in a variety of ways. Even if the terms themselves refuse to settle down, the distinction they suggest can make visible various typical, recurring patterns of interaction. Once again, what matters is not the construction of a logically impregnable classificatory scheme but understanding how narratives undertake the cognitive work of figuration. The interactions between these aspects of narrative matter more than precisely delimiting and controlling their boundaries.

Consider, for example, what can happen when the discourse provides patterns of explanation that complement, assist, and support the configurative work of the story—as when Balzac’s omniscient narrator in Père Goriot declares that “all is true!” and asks us to trust him as he explains “the laws of Paris” that Rastignac must master in order to achieve the social success he strives for; or as when George Eliot’s sympathetic if nevertheless critical narrative voice in Middlemarch guides the reader’s judgments about Dorothea Brooke’s noble if lamentably misplaced commitments. When the configurative patterns in the telling and the told complement each other in these ways, the result can be (as with Balzac) to encourage the reader’s immersion in a lifelike world by building coherent, consistent illusions or (as with Eliot) to reinforce the pedagogical lessons of the narrative as the teller’s judgments draw out the moral implications of the told. Consonant configurative patterns between the telling and the told can enhance the illusion of being transported to another world or encourage us to learn certain practices of seeing-as.

The dissonances between the various modes of figuration set in motion by a narrative can also be productive even if (or sometimes especially when) it is difficult to determine whether to attribute them to the story or the discourse. For example, the interaction between different story lines in a multiple-plot novel like Anna Karenina can be artfully constructed so that its effects go beyond what can be traced back to either of the stories alone or to the discursive pronouncements of the narrator. The famous opening line of this novel (“All happy families are alike”) is an authoritative declaration by what Isaiah Berlin memorably calls the voice of the “hedgehog” in Tolstoy, who knows simple, singular truths, but the interplay of the Anna plot and the Levin plot resists these reductions and produces a counterpoint of sympathies pointing in different, not always mutually compatible, directions that Tolstoy the “fox” seems to recognize cannot be easily reconciled. This novel would be much less interesting if the two plots were disentangled and published separately—the decline and fall of the adulteress in one volume and the moral triumph of the happily married man in the other. As the two plots interact and play off each other, the flaws and virtues of these two characters, the pressures on their lives, and the opportunities open to them can be differently configured by readers in ways that the novel’s narrative structure sets in motion but cannot predict or completely control. Each story offers a perspective through which to view the other, and the interaction of seeing-as in this interplay of perspectives is not attributable to either story alone. Nor is it simply a product of the narrator’s discourse. The interaction between the plots occurs in the figurative space of reading, where story and discourse meet and the reader attempts to align the patterns they offer.

The consonances or disjunctions between discourse and story can have a variety of effects by playing with the figurative workings of cognition. If congruent processes of figuration in the telling and the told can facilitate immersion or instruction, disjunctions between discourse and story can defamiliarize and foreground cognitive processes that may ordinarily go unnoticed in reading or in life because we are absorbed by our habitual ways of knowing. In Lord Jim, for example, Joseph Conrad juxtaposes Marlow’s telling and the titular hero’s tale in ways that call attention to the role of configurative patterns in cognition by blocking the narrator’s (and the reader’s) attempts to formulate them. The inconsistencies between the different perspectives Marlow receives on the enigmatic title character resist synthesis into a coherent point of view and consequently leave him frustrated and bewildered: “The views he let me have of himself were like those glimpses through the shifting rents in a thick fog—bits of vivid and vanishing detail, giving no connected idea of the general aspect of the country” (1996 [1900], 49).

Conrad’s narrator presents Jim to the reader through fragmentary and disconnected perspectives—conflicting interpretations offered by multiple informants that refuse to coalesce, pieces of evidence with opposing implications that cannot be reconciled—and these disjunctions in the discourse prevent a single, coherent story from emerging. Is Jim a romantic hero or an egotistical coward? Is Marlow’s sympathy for him a noble example of fellow feeling or an unconsciously motivated wish for the wrongdoer to escape punishment?9 The refusal of these perspectives to synthesize foregrounds the drive to build consistency among elements in a pattern that is necessary for lucid comprehension. The incongruities between Marlow’s fragmentary, disjointed discourse and Jim’s elusive story produce an ambiguous text that resembles the classic rabbit-duck figure, oscillating between possible configurations, interrupting the building of an illusion in order to call attention to the processes of pattern formation it sets in motion but prevents from settling down. Once again, the value of the distinction between story and discourse has less to do with its ability to support a classificatory scheme than with its usefulness as a tool for analyzing the configurative workings of narrative.

The examples I have been discussing illustrate how this distinction relates to narrative configurations of action, but it is also relevant to temporality and intersubjectivity, the two other dimensions of narrative that are the topics of the following chapters. For example, the relation between discourse and story is often complicated by the anachronies (to use Genette’s terminology) that disrupt the temporal correspondence between the telling and the told. If simple, zero-degree tales unfold seamlessly from beginning to end, artful telling can manipulate chronology to provoke curiosity, suspense, or surprise through temporal disjunctions that play further with the tension between concord and discord. Conrad’s sometime collaborator Ford Madox Ford (1924, 136) complained that “what was the matter with the Novel, and with the British novel in particular, was that it went straight forward, whereas in your gradual making acquaintanceship with your fellows you never do go straight forward.” By keeping a rough parallel between the chronology of presentation and the sequential order of events, however, the novels Ford criticizes assist the reader’s efforts to discover and build patterns and thereby actually encourage the immersion in an illusion on which realism depends. Ford’s point, however, is that this continuity disguises the processes it manipulates. In getting to know any state of affairs, we “never do go straight forward” inasmuch as we are always going back and forth between expectations about what lies beyond our horizons and corrections of previous guesses in light of evidence that has since come into view.

Temporal anachronies can serve many purposes, one of which is to foreground the temporal processes of consistency building that otherwise go unnoticed in the configurative work of storytelling or everyday cognition. In Ford’s novel The Good Soldier, for example, the narrator John Dowell criss-crosses back and forth over his past as he tries to make sense of events and relationships that, as he is surprised to find, he had completely misconstrued, and the result is a bewildering narration that constantly jumps around in time, challenging the reader to piece together in an orderly way the shifting, disconnected, incomplete perspectives it offers: “I have, I am aware, told this story in a very rambling way so that it may be difficult for anyone to find their path through what may be a sort of maze. I cannot help it. . . . I console myself with thinking that this is a real story and that, after all, real stories are probably told best in the way a person telling a story would tell them. They will then seem most real” (1990 [1915], 213). As with many other things, Dowell is probably not right about this. The notorious time-shifts in The Good Soldier may actually thwart the consistency building necessary to create illusions that can “seem real” because their coherence encourages our suspension of disbelief—but the gain from these disruptions is the opportunity they afford to reflect about temporal processes of cognition that are invisible when they function smoothly. By making the bewildered reader work harder and more reflectively than with continuous narration to build coherent story patterns out of the scattered bits and pieces in Dowell’s discourse, Ford transforms anticipation and retrospection from implicit cognitive processes into explicit issues in the experience of reading. Where consonance between the temporalities of story and discourse can facilitate mimetic immersion, disruptions to the temporal work of pattern formation can expose how configuration works.

Different degrees of consonance and dissonance between discourse and story can similarly be deployed to play games with the intersubjectivity of stories. According to Henry James (1970b [1883], 237–38), reading a narrative “makes it appear to us for the time that we have lived another life—that we have had a miraculous enlargement of experience.” Narratives whose consonances encourage immersion in an illusion can promote such identifications of self and other. The experience of following a story does not collapse the differences between selves, however, but entails a doubling of my consciousness with the intentionality held ready by the narrative, thereby enacting the paradoxical combination of community and solipsism that characterizes intersubjectivity. Such an identification is a paradoxical experience in which I configure another world as if it were my own—which, of course, it is not, so that a sense of the foreignness of the cognitive patterns I am temporarily inhabiting invariably shadows my immersion in them, even when I feel them in my own body in the fears and tears and other impassioned responses the story may set in motion. The pity or fear I may feel for the tragic hero is an identification with a quasi person who is not identical to me, after all, and is (strangely) not even a real human being, although I may intensely care about what happens to her (and may shed actual tears about her misfortunes).

Playing off the odd if everyday sensation that other worlds are both complementary and inaccessible to our own, various kinds of congruences and dissonances in the ways narratives stage these configurations may seek either to disguise or to foreground the doubleness of my world with another that occurs when I identify with lives that are not mine. James’s own famous experiments with point of view are marked by a peculiar doubleness that exploits this ability of narratives to transport us outside of ourselves even as he calls for reflection about the epistemological limits it oversteps. By projecting the reader into the world of the character whose perspective he re-creates (Strether’s bewildered fascination with the Parisian Babylon, for example, as he begins his ambassadorial mission), James gives us a rare view of another life from the inside, experienced by another for him- or herself. But in projecting ourselves into this other life, we experience as well the gap between its perspective and other points of view that remain obscure and mysterious to varying degrees (we never know but can imagine what Mrs. Newsome thinks back in Woollett, Massachusetts, about the no doubt alarming letters Strether sends home). This double movement of transcending and reencountering the gap between selves dramatizes in the reader’s own experience the combination of relatedness and opacity that makes the alter ego paradoxical.

Narratives double self and other by configuring and reconfiguring the relations between worlds in to-and-fro interactions between storytellers and their recipients. This can be done variously by first- and third-person narrations and by different kinds of focalization (the narratological term for point of view) or narrative voice (narrators with different attitudes or degrees of authority). The person of the narrator or the particular technical device that sets this doubling in motion matters little in itself. The same linguistic or narratological structure can be deployed for a variety of purposes. What counts is how the exchanges between storyteller, story, and recipient put into play various interactions between worlds, and these are dynamic, back-and-forth processes of configuration and refiguration that cannot be adequately accounted for by constructing a static classificatory scheme or by mapping narrative techniques onto a taxonomic grid. Naming a technique or a structure is only the first step in analyzing the configurative processes of pattern formation and disruption it sets in motion in any particular narrative and in describing the cognitive purposes these interactions serve.

Lived Worlds and Narrative Worldmaking

The term “world” is often employed in narrative theory to describe these interactions, but it is a slippery notion that deserves some analysis. For theorists in the traditions of phenomenology and pragmatism, the term refers to the configurations of meaning-making activity that characterize experience—what Merleau-Ponty calls the unreflective, “operative intentionality” that we find already at work when we reflect on our lives and discover various patterns of relationship that give shape to our typical, habitual interactions with people, places, and things (see 2012 [1945], lxxxii). As Shaun Gallagher explains, this “intentionality” (a key term in the lexicon of phenomenology) has an as-structure: “All consciousness is consciousness of something as something” (2012, 67; original emphasis). According to Gallagher, “Actions have intentionality because they are directed at some goal or project, and this is something that we can see in the actions of others”—hence our tendency “to attribute intentionality to geometrical figures moving in certain patterns on a computer screen . . . as well as to non-human animals and human infants whom we do not regard as following specific social norms” (77, 75).

Recognizing a world entails understanding and interpreting patterns of intentionality—as-structures that configure states of affairs in a meaningful way in our own experience or in the experiences we construe others as having, analogous to our own. As Ricoeur (1984a, 61) explains in tracing the notion of world back to Heidegger’s Being and Time, “The structure of being-in-the-world . . . is more fundamental than any relation of a subject to an object.” Against Cartesian dualism, subject and object are not separate but are always already joined in patterns of configuration that make up our lived worlds. The hyphens joining “being-in-the-world” are the patterns of intentionality set in motion as we project ourselves into futures of possibility based on the circumstances in which we find ourselves situated (hence Heidegger’s description of existence as a “thrown projection,” or “geworfener Entwurf” [see 1962 (1927), 185]).

Narratives bring worlds into relationship as patterns of configurative activity cross back and forth in the circuit joining lived experience, the construction of stories, and their reception by listeners and readers, Ricoeur’s three-fold mimesis. As Ricoeur (1987, 430–31) explains, “The intersection of the world of [the] text and the world of the reader . . . opens up a horizon of possible experience, a world in which it would be possible to dwell.” The process of telling and following stories sets in motion interactions between the patterns of configuration that characterize these different worlds.

Although the term “world” may seem slippery as it passes from one domain to another, this boundary crossing is a necessary and by no means lamentable consequence of the openness of stories to experience. As Ricoeur (1984b, 349) points out, “A text, actually, is not a self-enclosed entity. It has not only a formal structure; it points beyond itself to a possible world, a world I could inhabit, where I could actualize my own possibilities in so far as I am in the world.” The recurrence of the term “world” in these different contexts is an indication of the openness of stories to the experiences they reconfigure and to the existential possibilities of their recipients. Pragmatically oriented psychologist Jerome Bruner (1986, 66) similarly invokes this term to characterize stories not as logical, formal structures but as aspects of our lived experience—projections of “possible worlds in which action, thought, and self-definition are possible (or desirable).” What matters for theorists like Ricoeur and Bruner in the phenomenological and pragmatic traditions is how the patternmaking powers of stories contribute to what Nelson Goodman (1978) memorably calls our “ways of worldmaking,” a concept also adopted by phenomenological reading theorist Wolfgang Iser in his “literary anthropology” (1993, 152–70).

It is a mistake to reify worldmaking, however, by reducing it to formal, schematic models of the sort sometimes proposed by cognitive narratologists to map the structures of “story worlds” or to classify the logical markers that differentiate “actual” and “possible worlds.” The interactions between worlds as we tell and follow stories cannot be reduced to grids and schemes. For example, the good thing about David Herman’s (2013, x–xi) somewhat awkward, confusing description of “narrative worldmaking” as “worlding the story” and “storying the world” is its recognition that the configurative work of narrative is a dynamic process. But the taxonomic drive to construct classificatory schemes to account for these processes is problematic. Herman’s much-discussed book Storytelling and the Sciences of the Mind (2013) is filled with one classificatory scheme, diagram, and taxonomy after the other and proliferates terms, categories, and distinctions in an almost manic attempt to reduce to an orderly system the dynamism of worldmaking. This elaborate edifice of maps, grids, and definitions misrepresents the configurative activity through which worlds are projected and interact in narratives and lived experience and fails to capture the processes it attempts to reduce to static schemes.

The fundamental problem with this approach is evident in two of Herman’s central claims (2013, 56) about narrative worldmaking—namely, that “interpreters map textual patterns onto WHO, WHAT, WHERE, WHEN, HOW, and WHY dimensions of storyworlds” and that “the patterns in question emanate from reasons for (text-producing) actions” that can be systematically categorized. A static map of positions on a grid charting the answers to these questions cannot do justice to the configurative processes of meaning making through which worlds are experienced and exchanged. Herman’s interrogatory map is reminiscent of the stock questions that newspaper reporters are instructed to ask as they gather material for their stories, but any journalist knows that the answers they jot down in their notebooks are not the story but only the bits and pieces out of which it must still be put together when they return to the newsroom. What is missing from Herman’s map are the configurative processes of pattern formation that connect the dots and fill in the blanks. As I read The Ambassadors, for example, I can chart on a grid the answers to all of these questions about Strether’s mission to retrieve Chad from Paris, but I will still miss the patterns of longing, regret, and desire that animate his impassioned plea to Little Bilham in Gloriani’s garden to “live all you can! It’s a mistake not to.” The patterns of intentionality that animate a world resist reduction to a classificatory scheme and cannot be explained by positions on a map.

A similar problem afflicts Mark Turner’s (1996) diagrams of what he calls “conceptual blending,” none of which can do justice to the interaction between a word and a context through which metaphorical innovation occurs. This interaction is characterized by what Nietzsche (2015 [1873]) describes as “das Gleichsetzen des Nicht-gleichen,” the “setting equal” of what is “not the same.” As Ricoeur (1977, 252, 197; also see 21–22, 170–71) explains, a novel metaphor is “a calculated error,” “comparable to what Gilbert Ryle (2009 [1949], lx) calls [a] ‘category mistake,’ which consists in ‘the presentation of facts belonging to one category in the idioms appropriate to another.’ ” According to Ricoeur, a novel metaphor is surprising and perhaps initially confusing because it “speak[s] of one thing in terms of another” (197) that is both like and not like it, in an act of comparison that transgresses the patterns of configuration to which we are accustomed. This incongruity—“a planned category mistake” (197)—is not simply dismissed as an error, however, but instead produces new meaning because of the interpreter’s adjustments through which its initial incoherences are made consistent. After they are reconfigured into new patterns of similarity and difference, these incoherences may come to seem “right” in unanticipated ways. A novel metaphor can then become dead when its incongruities are so assimilated and conventionalized that they are no longer noticed. Only then is it a “blend”—but at this point its meanings are literal, not metaphorical, because the interactions set off by the incongruities between like and not like have ended and its suggestive powers have faded and died.

Turner proposes elaborate schemes to classify these interactions and develops complex spatial diagrams to map their components (also see Fauconnier and Turner 2002). These structures necessarily miss the interaction—the category mistake and the readjustments it provokes—that constitutes the metaphorical innovation. Instead of capturing the configurative activity that creates new meaning, his taxonomies reduce the processes of disruption and reorganization to a series of charts and distinctions. The mistake here, analogous to the problem with Herman’s maps, is that classifying kinds of “blends” according to their structures misses the interactions through which the surprising combination of like and not like provokes new modes of consistency building. By the same token, Turner’s much-used term “blending” does not convey just how necessary incongruity and discordance are to the production of new congruence. The maps, grids, and schemes that Herman and other structurally oriented narratologists use to characterize the concordant discordance of narrative are similarly destined to miss the processes of figuration, configuration, and refiguration that they reduce to static taxonomies.10

Herman tellingly employs a static, structural metaphor to describe worldmaking: “The process of building storyworlds . . . scaffolds a variety of sense-making activities” (x; emphasis added). This is a concept sometimes invoked by 4e theorists (for example, see Clark 2011, 44–60), and it is a term to which Herman repeatedly returns. To speak of the process of building storyworlds as “scaffolding” is awkward and confusing, however. A scaffold is a structure that assists in the activity of constructing the building. It does not capture the essential work of making it, and it is then discarded when the building is finished and made available to be used and inhabited, opening the way for further unpredictable, open-ended interactions with other members of our species who pass in and through its structures and repurpose them for goals, relationships, and projects that the builders may not have had in mind. The metaphor of a scaffold misrepresents the work through which narrative worlds are built and the uses to which they can be put by reducing a dynamic process to a schematic structure. This confusing locution prioritizes structures over the events of meaning that the process of employing them makes possible.

The metaphor of a scaffold may appeal to a structurally minded imagination, but its awkwardness and blindnesses are evidence of its reification of the processes through which worlds are made, remade, and interact as we live our lives and tell and follow stories. The configurative activities involved in the construction and interaction of narrative worlds are not reducible to a map of locations, a logic of reasons, or a scaffold of structures. Metaphors like these are traces of the ghost of structuralism that haunts formalist cognitive narratology. To avoid reifying the intentional activity that constitutes lived and narrated worlds, it is necessary to keep focused on the processes of pattern formation that the metaphor of world entails, figurative processes foregrounded in the phenomenological and pragmatic traditions.11

This schematic approach is a legacy of the first-generation prioritization of frames and structures. By contrast, various forms of second-generation narratology focus on the actions and interactions through which stories shape the patterns of our worlds. For example, Terence Cave (2016, 4, 5) describes “thinking with literature” as a collaborative process of inquiry, improvisation, and conversation that “conscript[s] our capacity for cognitive inference” and may “alter the cognitive environment of the reader in ways that are powerful, potentially disturbing, and not at all self-evident.” Cave’s model of reading, based on relevance theory (see Sperber and Wilson 1995 [1986]), emphasizes the “bold and highly precise modes of underspecification” (27) of literature—what a phenomenological theorist like Iser or Ingarden calls the blanks or indeterminacies that require improvisatory responses beyond what can be captured by mapping dimensions of a storyworld onto a spatial grid. For Cave, “thinking with literature” is a cocreative response to a literary work “not as [a] neutral text but as an animated affordance” (9) that encourages and makes possible but does not fully determine our interpretations. According to Cave, “What happens when we redescribe literary conventions as affordances,” borrowing Gibson’s (1979) well-known term, is that “what was static and merely constraining” turns out to open up “all kinds of unexpected possibilities, ways of breaking out into new territory” (55–56).

Affordance is a better metaphor than scaffolding to describe the actions and interactions in the environments that various tools and resources may be used to build, a process Andy Clark (2011, 61–68) describes as “niche construction.” Building a niche creates an environment that in turn affords a variety of uses. Similarly, the storyworld that is configured by employing various techniques, conventions, and other narrative resources (the scaffold) then affords an open-ended, unpredictable history of responses by recipients who engage in activities of meaning making that it makes possible (also see the analysis of narrative affordances in chapter 3). Mapping a world as points on a grid misses this dynamism—what Cave (77, 33) calls “the human ability to think beyond the immediate demands of [our] environment” that a text invokes through its “implicatures,” the various “intended meanings that can be derived inferentially from a given utterance.”

Other cognitively oriented second-generation theorists similarly focus on these kinds of interaction between modes of figuration. For example, drawing on phenomenological and pragmatic theories of text-reader interaction, Marco Caracciolo’s “enactivist approach” (2014, 4) emphasizes “how meaning emerges from the experiential interaction between texts and readers”—“stories offer themselves as imaginative experiences because of the way they draw on and restructure readers’ familiarity with experience itself”—a to-and-fro, temporally unfolding, dynamic relationship between the world of the reader and the world of the text in a mutually formative experiential transaction. Describing narrative as “a purposeful communicative exchange between authors and readers,” Phelan (2015, 121) offers a “rhetorical theory” that similarly foregrounds processes of interaction. Phelan’s “rhetorical poetics of narrative” is rightly regarded as a kind of cognitive narratology (hence his inclusion in Zunshine’s 2015 anthology) because of his attention to the ways audiences take up and respond to the communicative acts of the author. Carrying on and transforming the tradition of his teachers at the University of Chicago from before the days of cognitive literary criticism, Phelan develops a theory of rhetoric as a purposeful communicative exchange that deserves “second-generation” status because it foregrounds the configurative, mutually formative processes of the text-reader interaction. His definition (2017, ix) of narrative as “somebody telling somebody else on some occasion and for some purposes that something happened” traverses Ricoeur’s circuit of figuration between experience, story, and recipient.12

The error of reifying configurative processes also undermines the formalist project of reducing narrative actions to a logic of “reasons” from which they “emanate.” According to Herman (2010, 169–70), “readers are able to understand the characters’ behaviors as actions in part because of the models of emotions on which they rely to interpret the text”—what he calls an “emotionology,” defined as “the collective emotional standards of a culture as opposed to the experience of emotion itself”: “An emotionology specifies that when an event X inducing an emotion Y occurs, an agent is likely to engage in Z sorts of actions.” Once again adopting a model based on scripts, frames, and preference rules, Herman contends that “the characters’ activities can be construed as more than just a series of individual, unrelated doings because of the assumption, licensed by a model of emotions, that those behaviors constitute a coherent class” (170; original emphasis).

There are several things wrong with this taxonomic, rule-based approach to character, action, and emotion. To begin with, as I have explained, the best contemporary neuroscience of emotions (see Barrett 2017) suggests that anger or embarrassment should be viewed not as a coherent, homologous “class” but rather as a “population” of related, overlapping, but diverse subjective states. Further, the research on the relation between real and imagined action suggests that our responses to the action staged in a text are less like the linear, logical application of a rule from a class than the sort of bodily based resonances that Guillemette Bolens (2012) describes in her kinematic theory of narrative. The intuitive, embodied resonances that emerge and unfold over our engagement with a text cannot be adequately described by a logic of models and rules. Bodily responses to the intentionality of the text are as-relations that have the power to reconfigure our sense of the world because they are not simply applications of schemata we already know. They are, rather, dynamic and unpredictable enactments of the paradox of the alter ego, the doubling of the “real me” of my kinematic sensations that I experience while reading and the “alien me” of the world I set in motion as I empathize and identify with the actions of the text.

A scheme outlining an underlying logic of actions cannot do justice to the processes through which these configurations of embodied intentionality emerge, develop, and change across the horizons joining past, present, and future in to-and-fro, reciprocal interactions. To recall James’s novel The Ambassadors, Strether’s adventure in Europe surprises him not just because his new friends are not completely honest with him but, even more, because the values and possibilities revealed to him in this new world bring out feelings he had not anticipated and is uncertain how to act on. His past sets the stage for these reactions—his regret for his “unlived life,” as it is called in the James criticism—but it does not determine how he will respond to them. Furthermore, as readers of The Ambassadors find their worlds interacting with Strether’s, no map or chart can predict whether the consonances between them will make them identify and sympathize with this poor sensitive gentleman (as I do!) or whether instead the dissonances between his reactions and theirs will lead them to fault his hesitancies, refusals, and acts of resignation (and perhaps then also blame the author for these, as many hostile critics have done).

What matters is not only where actions come from but where they are headed—not just their sources but their goals and directions—and it is the variable, often unpredictable interaction between these that make actions dynamic. For example, as Elaine Auyoung (2013, 60) observes, narrated actions can seem lifelike because their gaps and indeterminacies draw on “our readiness to contend with partial representational cues in everyday, nonliterary experience.” With the actions represented in stories as in those we encounter in everyday life, she notes, we fill out what lies beyond our limited perspective by our expectations about their future course and direction. Similarly, as Karin Kukkonen (2016) points out, the various kinds of action set in motion by a text—not only the characters’ behaviors (the action of the plot) but also our responses to stylistic cues (the action of the narration)—are less like the unilinear application of preference rules than affordances that make possible but do not fully prescribe our responses, guiding our actions but leaving open room for improvisation, innovation, and surprise. Drawing on a Bayesian, predictive-processing model of embodied cognition, Kukkonen explains that the narrative environment makes some actions probable and others less so, thereby motivating not only the development of the plot but also our expectations about the course of the narration in a dynamic, interactive process of feedback loops that cascade into each other, reinforcing or disrupting the patternmaking work set in motion by different kinds of narrative figuration. The eventfulness of worldmaking and the unpredictability of the interaction of worlds in the experience of narrative that these second-generation theorists are attempting to describe are essential to how storyworlds work, and these are necessarily lost in spatial maps and classificatory schemes.

A reifying, reductionist formalism sometimes similarly afflicts narrative theories of actual versus possible worlds. For example, in Marie-Laure Ryan’s analysis of fictional worlds and virtual reality there is an interesting and revealing contradiction between her attempt to do justice to the dynamic processes of “immersion and interactivity” and her resort to logical frameworks to categorize the ontological differences between worlds. On the one hand, she recognizes that the effects of immersion in the simulations of virtual reality depend on their “dynamic character”—how, for example, “it takes . . . a movable point of view to acquire a full sense of the depth of an image” (2001, 53)—and she consequently invokes Merleau-Ponty’s description of how “embodied consciousness” entails “both mobility and virtuality” (71) to explain how the as-if configurations of fictional worlds can acquire the illusion of plenitude. “When my actual body cannot walk around an object or grab and lift it,” she explains, “it is the knowledge that my virtual body could do so that gives me a sense of its shape, volume, and materiality. Whether actual or virtual, objects are thus present to me because my actual or virtual body can interact with them” (71). This is sound phenomenological analysis of the dynamic processes and interactions through which configurative patterns emerge.

On the other hand, however, Ryan also attempts to develop “an ontological model” that schematizes “in semantic and logical terms” (99) the differences between actual and possible worlds. This model, she explains, is based on “the set-theoretical idea that reality—the sum total of the imaginable—is a universe composed of a plurality of distinct elements, or worlds, and that it is hierarchically structured by the opposition of one well-designated element, which functions as the center of the system, to all the other members of the set” (99). Here the phenomenologist is displaced by the structuralist, and an analysis of figurative interactions gives way to a quest for a well-ordered taxonomy. And so her important book Narrative as Virtual Reality alternates between compelling descriptions of how immersion in virtual worlds draws on configurative patterns of embodied cognition familiar from everyday experience and overly simplistic charts and schemes that reduce these processes to formal structures based on various ontological distinctions. Rather than sharpening and clarifying fictional worldmaking, as she hopes, the formal and schematic categories of the logicians obscure the phenomenological experience of immersion and interactivity by reducing it to an abstract set of hierarchically ordered structures that necessarily miss the interactions and the configurative processes of pattern formation through which worlds are constructed.

Narratology needs to break with its structuralist legacy and embrace the paradigm shift proposed by the various pragmatically oriented, phenomenological theories of narrative that have contested the formalist program. If we want to understand stories, logical structures and taxonomies won’t do the job. What we need to know, rather, as the second-generation theorists recognize, is how elements combine into patterns through their interactions in lived experience and embodied cognition. How narratives participate in the formation and dissolution of patterns in the embodied brain’s interactions with the world is the right question to ask if what we have is not a logically ordered, formally structured mind but a bushy brain that is an ensemble of relationships that get fixed over time but are open to a future of variation. Those interactions are the means by which stories help the brain negotiate the tension between pattern and flexibility thanks to the play of their concordant discordances. The terms in the lexicon of narrative theory are useful insofar as they help to elucidate these interactions, but concepts and locutions that obscure or misrepresent them should be discarded. The to-and-fro processes of figuration and refiguration through which we tell and follow stories is the dynamic, ever-shifting ground on which neuroscience, narrative, and narratology meet.

Natural and Unnatural Narratives

A neuroscientifically grounded account of the relation between stories, language, and the brain can clarify what’s at issue in the debate about “natural” versus “unnatural” narratives. This dispute arose when advocates of cognitive narratology called attention to the so-called natural aspects of the production and comprehension of stories. The question, of course, is what counts as natural, and the science on these matters is not irrelevant. Proposing a theory of natural narratology, Monika Fludernik (1996, 12–52) argues that narrative is based on its ability to render what she calls “experientiality” through modes of exchange that reflect the “natural schemata” of embodied cognition and the “natural parameters” of language in conversational storytelling. In her model, “the term ‘natural’ is not applied to texts or textual techniques,” she explains, “but exclusively to the cognitive frames by means of which texts are interpreted” (12; original emphasis). Frames, scripts, or schemas are natural, she argues, because they are “aspects of language which appear to be regulated or motivated by cognitive parameters based on man’s experience of embodiedness in a real-world context” (17), and what she calls “narrativity” is a natural process that “emerge[s] from spontaneous conversational story-telling” and “naturalizes” unfamiliar experience “by the sheer act of imposing narrativity on it” (15, 34).

This is partly wrong and partly right. As I have explained, frames and scripts are not natural features of embodied cognition. They are, rather, artifacts of an artificial intelligence model that do not match up well with the anatomy and functions of the brain except insofar as they are metaphors for the activity of seeing-as. The problem with these terms, as I have argued, is that they reify figurative activity into substantive modules and unilinear mechanisms that misrepresent the to-and-fro processes through which patterns are formed, broken, and remade in lived experience and in the neurobiology of mental functioning. The other part of Fludernik’s theory has merit, however. Her emphasis on the transactional exchanges of conversational storytelling is a useful call to structurally minded narratologists to focus on the interaction between worlds that occurs in the circuit of exchanges of experience as we tell and follow stories. Although the coinages “experientiality” and “narrativity” are somewhat awkward and opaque (the terminological beast rises again), their pragmatic value is that they foreground the circuit between the configurative processes of lived experience and their refiguration in the concordant discordances of narrative that Ricoeur describes.13

These processes are both natural and unnatural, however, in different ways and varying degrees. The stories we tell each other are biocultural hybrids that mix together, sometimes indistinguishably, what is natural about embodied cognition and what is artificial, cultural, and learned. For example, a recent anthropological study of residents of a geographically isolated area of Turkey who had little experience with film or television found that they were baffled by some but not all film-editing techniques. Neuroscientist Arthur P. Shimamura (2013b, 19) concludes that “these findings demonstrated that our understanding of the ‘syntax’ of movies is to some extent a learned phenomenon.” But they also suggest the opposite conclusion, namely that narrative understanding is to some extent natural.

Stephan Schwan and Sermin Ildirar (2010, 974–75), the authors of this study, reported that subjects with little or no experience watching films were able to come up with the “standard interpretation” of a film clip exhibiting a series of typical film-editing techniques only 36–41% of the time, whereas 90% of experienced viewers were able to do so. The details behind these percentages are interesting and important: “The inexperienced viewers did describe the individual shots appropriately, but they had difficulties in relating the shots to each other,” and they had particular difficulty with “clips portraying a situation or an event from the visual perspective of a certain actor. . . . Thus, the notion that so-called point-of-view shots resemble natural perception to the highest degree and therefore should be easiest to understand . . . was not confirmed by our data” (975). One of their key findings, however, was that “films containing a familiar activity led to a significantly higher percentage of standard interpretations”: “The presence of a familiar line of action is essential for the interpretation processes of inexperienced viewers; a line of action helps them overcome unfamiliar perceptual discontinuities” (975). Summarizing, they conclude that “it is not the similarity to the conditions of natural perception,” such as point-of-view editing techniques, “but the presence of a familiar line of action that determines the comprehensibility of films for inexperienced viewers” (970).

These findings are consistent with the well-established phenomenon that modernist recreations of a character’s consciousness or point of view, as in the novels of James Joyce, William Faulkner, or Virginia Woolf, although perhaps closer to “natural perception” in their methods of representation, are typically harder for most readers to understand than conventional, plot-driven dramatizations that may lack epistemological verisimilitude but that facilitate the reader’s ability to follow the line of action. Mrs. Dalloway and To the Lighthouse may be more natural than David Copperfield or Jane Eyre because Woolf’s stylistic experiments seek to recreate the to-and-fro of embodied cognitive processes, but readers of Dickens and Brontë typically find that the plots of these novels promote immersion in a lifelike world (and seem “natural”) because they support the construction of patterns of concordant discordance. Perhaps paradoxically, techniques not only in film but also in fiction that mirror the cognitive workings of perception can seem by contrast unnatural and bewildering.

The Turkish villagers, like many readers of Faulkner and Woolf, may find it difficult to understand a narrative depicted through the point of view of a perceiver because they have not learned this convention, and it consequently may seem unnatural even though it reenacts “natural perception.” This is not surprising inasmuch as techniques in films and novels that foreground how experience is perceived may interfere with emplotment by calling attention to how a pattern is made rather than facilitating the work of fitting things together. As Schwan and Ildirar’s study suggests, however, such viewers and readers are nevertheless able to make some sense of a confusing state of affairs if they can follow “a familiar activity” and track “a familiar line of action.” This, apparently, is what is natural—the ability to follow sequences of events when they are arranged in patterns of action that are recognizable from the perceiver’s past experience. Emplotment, the ability to configure actions in a narrative sequence, is a natural cognitive capability that can be put to use even when the conventions of editing and narrating are unfamiliar, which explains why the Turkish villagers were still able to make sense of bewildering films more than a third of the time. Similarly, readers unacquainted with modernist techniques can still often follow the events of Clarissa Dalloway’s day as she walks through London, buys flowers, mends her dress, and otherwise prepares for her party.

Familiarity also matters. Even when the mode of rendering seems odd or strange, finding ways of grafting the unfamiliar onto the familiar can make it comprehensible. It is a well-established neuroscientific fact that Hebbian firing and wiring can habitualize and automatize perception. Patterns may seem natural, then, either because they can be emplotted in easily recognizable action sequences or because their recurrence in everyday experience has made them familiar, engraining their configurations through repetition in the cognitive habits of our brains and bodies so that they become invisible. Both of these cognitive phenomena—how action sequences facilitate emplotment and how familiarity promotes comprehension—help to explain how, in Fludernik’s terms, narrativity has the power to naturalize a state of affairs.14

Conventions that promote the cognitive work of constructing patterns in these ways can seem natural, then, even when they are artificial. This conclusion is also supported by a recent eye-tracking study that Tim Smith and colleagues (2012, 107–8) conducted on “the Hollywood style of moviemaking,” an artificial set of techniques characterized by “focused continuity” between scenes and “consistency in cues.” Comparing the movement of saccades (rapid, jerky jumps in the eye’s point of focus) in natural perception and in viewing films, this study showed that Hollywood editors have evolved editing techniques that work—that is, promote understanding and the formation of illusions—because their “formal conventions . . . are compatible with the natural dynamics of attention and humans’ assumptions about continuity of space, time, and action” (107). Observing striking parallels between saccadic eye movements in ordinary seeing and in watching Hollywood films, Smith’s group concluded that, “by piggybacking on natural vision cognition, Hollywood style presents a highly artificial sequence of viewpoints in a way that is easy to comprehend, does not require specific cognitive skills, and may even be understood by viewers who have never watched film before” (108). Hollywood movies are unnatural in the sense that they are based on a stylized, artificial set of conventions, but artifices that can be naturalized as they become familiar and habitual, and this process of naturalization is all the easier when, as here, these conventions take advantage of natural ways in which the eyes move and the visual cortex constructs a scene.

The call for a “natural narratology” almost predictably produced arguments in favor of “unnatural narratives.” These sorts of stories are only comprehensible, however, to the extent that their unnatural experiments have a natural foundation in biologically based cognitive processes and familiar cultural conventions. According to the editors of a recent collection of essays on the poetics of unnatural narrative (Alber, Nielsen, and Richardson 2013, 1), “unnatural narrative theorists oppose what one might call ‘mimetic reductionism,’ that is, the claim that the basic aspects of narrative can be explained primarily or exclusively by models based on realist parameters” (also see Richardson 2015, 3–27). As Jan Alber (2009, 79) argues, narratives “do not only mimetically reproduce the world as we know it. Many narratives confront us with bizarre story-worlds which are governed by principles that have very little to do with the real world around us.” There are cognitively based limits to how far this can go, however. Experimental modern and postmodern narratives may test the extent to which dissonance can disturb the configurative activity of emplotment without lapsing into nonsense and noise, but discord without some degree of concordance is incomprehensible. Like “natural” stories, “unnatural” narratives also set into motion and play with the brain’s need for patterns to make sense of the world. Their manipulation of our natural quest for cognitive coherence is how they achieve their unnatural, antimimetic effects.

One of the most unnatural narratives ever written is James Joyce’s Finnegans Wake, but even its bizarre experiments with language and narration are only comprehensible—or enjoyable (and they are both to some readers, but not of course to all)—to the extent that readers can configure them into patterns that relate the text’s many enigmas to familiar contexts and constructs and that shape its events into recognizable structures of action. The text’s repeated portmanteau words, for example, are riddles that challenge the reader to decipher the components they condense by discovering and testing various possible configurations that make sense of their Jabberwocky-like non-sense (even the seemingly innocent word “Finnegan” can then be seen as a Vicoian pun on cycles that “end,” from the French “fin,” only to begin again, while the middle phoneme recalls the “egg” Humpty Dumpty that fell and couldn’t be put back together: Finnegan=fin-again + egg-again).15 The characters (such as they are) of Humphrey Chimpden Earwicker (HCE) and his family may not be lifelike personages, but they relate to one another and to other entities in the text in various configurations of action that call on readers to invoke natural patterns of affection and conflict, rising and falling, and other cycles of transformation that play with our configurative powers. From its title onward, which alludes to a well-known song about a hod carrier who is knocked out only to revive when splattered by whiskey at his own funeral party, this strange text invokes familiar stories of all kinds, from “This is the House that Jack Built” to “The Ant and the Grasshopper,” encouraging readers to search out configurative patterns. Similarly, the famous “Anna Livia Plurabelle” chapter from which Joyce recorded an extended passage is a natural narrative inasmuch as it enacts an extended gossip session between washerwomen on the model of Fludernik’s conversational storytelling, even as its encyclopedic allusions to the rivers of the world are an elaborate, artificial linguistic game.

Natural and unnatural narratives depend on and play off of each other, and this is not only a recent phenomenon that characterizes modernism and postmodernism. It is not accidental, for example, that the rise of the novel in the eighteenth century saw the production of mimetically realistic works like Moll Flanders, Clarissa, and Tom Jones and also the irreverent Tristram Shandy, which pokes fun at the unnaturalness of many of the conventions of realism, such as the pretense of Samuel Richardson’s epistolary style to “write to the moment” and render events as they happened (Tristram lamenting that his very attempt to describe moment to moment the misfortunes and accidents marring his birth takes so long that he has fallen behind even before his narrative gets under way).

In their classic The Nature of Narrative, Scholes and Kellogg (2006 [1966], 4) argue that “for writing to be narrative no more and no less than a teller and a tale are required” (4). This fundamental opposition is basic to narrative because the difference between teller and tale offers all sorts of possibilities for configuring discordance into patterns of concordance not only in the story but also in the discourse and, even more, in the back and forth between them. This opposition makes possible what Scholes and Kellogg call the “disparity among viewpoints” that is fundamental to “narrative irony” (240) in its many different forms. The discordances of irony that call on the reader to synthesize their juxtapositions into patterns of concordance can be as gentle and subtle as the decorous humor of Jane Austen in the famous opening of Pride and Prejudice (“It is a truth universally acknowledged that a single man in possession of a good fortune must be in want of a wife”), or they can be as kaleidoscopic as the multiple styles of Ulysses and or as unruly as the linguistic play of the word games of Finnegans Wake. Narrative irony is an unnatural, artificial state of affairs that can be played with and manipulated in any number of ways. But the natural basis of its many permutations is the ability of narrative to configure concordant discordance in syntheses of the heterogeneous that play with the brain’s balancing act between order and flexibility, pattern and openness to change.

Although the correlations between neuroscience, narrative theory, and our experience of stories can illuminate many matters of interest to both scientists and humanists, it is important to keep in mind what cognitive science cannot tell us about narrative. The explanatory limits of neuroscience have to do with the so-called hard problem (see Chalmers 1995) of how electrochemical activity at the neuronal level produces consciousness and embodied experience. It is undeniable that if someone is having an experience, there must be neuronal activity of some sort correlated to it. What happens across the divide between neurons firing and lived experience is mysterious, however, and how consciousness emerges from brain-body processes at the cellular level is a question no one can as yet answer. Thomas Nagel (1974) has famously argued that empirical measures alone cannot capture “what it is like” to have an experience, and he has recently (2012) speculated that the question of how neuronal activity gives rise to consciousness probably requires a disciplinary paradigm shift (à la Thomas Kuhn) that we cannot fully imagine because we are still on this side of it (how life emerges from chemical interactions is, he claims, a similar quandary awaiting an as yet inconceivable revolution in the explanatory frameworks available to science). Experience and neuronal activity can be correlated, but neither can completely explain the other or the alchemy of their interaction.

There is consequently a disjunction between the levels of analysis at which neuroscience and the humanities approach cognitive issues. This disjunction is what neurophenomenologists call the “explanatory gap” that separates neurobiological and phenomenological accounts of consciousness—the gap between the neural correlates of consciousness that neuroscience has identified and the lived experience of the perceptual world (see Thompson, Lutz, and Cosmelli 2005). This gap may be an obstacle to what E. O. Wilson (1998) calls consilience between cognitive science and the humanities, but it also makes possible conversation between them from positions of distinctive disciplinary difference (see Armstrong 2013, 7–8). We can compare and correlate experience, neuronal processes, and narrative theories across this divide, and this work can be instructive in all kinds of ways and for all sorts of reasons, but these triangulations are exactly that—comparisons that chart convergences and divergences—and not solutions to the hard problem or explanations of the mysteries of emergence.

One obstacle to undertaking such triangulations is the bogeyman of Cartesian dualism that haunts literary studies. To inquire about the cognitive workings of the brain, it is sometimes feared, is to commit the fallacy of assuming that reality is constituted in the mind of an ego that thinks, thereby overlooking the fact that the cogito is always situated in a body and a social, historical setting and that cognition entails interactions across the boundaries joining brain, body, and social world. Advocates of so-called enactive embodied cognition sometimes similarly worry that asking about processes in the brain may wrongly neglect its situation in a body and a world of natural and socially constructed affordances (for example, see Cook 2018). Enactivism risks becoming a distorting dogma, however, if it refuses to investigate cortical and neuronal processes inside the skull on the grounds that cognition is not only a matter of what happens in the head.16 By no means mutually exclusive, these perspectives are interdependent and inextricably linked. Usefully reminding us that “the brain is one element in a complex network involving the brain, the body, and the environment,” Alva Noë (2004, 214, 222) advises that we need both “to look inward, to the neural plumbing” that gives rise to experience and “to look outward, too, to the way that plumbing is hooked up to the world.” Narrative offers neuroscience perspectives that open in both directions. The configurative interactions through which we tell and follow stories deploy fundamental cognitive processes that connect our brains through our bodies with the people, places, and things in our worlds. “Look both ways” is useful advice when you’re at an intersection of any kind, and that includes the conjunction between neuroscience and narrative.