Abstract
This article proposes a structural account of the self as narrator grounded in Living Value Theory (LVT). It argues that cognitive science, philosophy of mind, and literary theory have each been pursuing a version of the same insight — that the self is a narrator — while each has been stopped short of the insight's fullest implications by a shared and unexamined assumption: that the self occupies a fixed narrator position. The article's central claim is that the opposite is true. The healthy self is not a fixed narrator at all. It is an embedded living agent whose primary condition is absorbed engagement with its own life at the level of pre-reflective coping, and whose narrator-position capacity is a selectively deployable resource that expands when the situation requires it and contracts when it does not. This dynamic oscillation is what LVT calls the embedded narrator: not a position but a living movement between embeddedness and graduated narrator-position awareness. The article develops three consequences of this account. First, cognitive science's search for the stable reality of the self has been misled by the same error: it has taken a narrator position that is genuinely available to the self under specific circumstances and treated it as the self's permanent or defining structure. Descartes's cogito is the paradigm case: a genuine and important moment of pure self-recursive narrator awareness that is useful in its proper place but pathological as a permanent condition. Second, the literary theory of narrator positions, developed by Genette, Booth, Cohn, and Phelan, provides cognitive science with the fine-grained typology of narrator-position modes it has lacked, but only when those modes are understood as configurations available to the embedded narrator, not as fixed identities. Third, the history of changing narrator configurations in Western narrative fiction, from Homer through Augustine through Austen through modernism, is not a history of changing selfhood but a history of changing literary conventions for representing the embedded narrator's structure. What changed is what literary genres permitted to be written. The embedded narrator was always there.
I. The Pursuit That Missed Its Own Object
For fifty years, cognitive science and philosophy of mind have been converging on a conclusion that would have seemed paradoxical to an earlier generation: the self is not a substance but a structural achievement. Beginning with David Hume's bundle theory, which denied the existence of any unified self beneath the flux of experience, and accelerating through the second half of the twentieth century, the major traditions of cognitive science have dismantled the Cartesian subject piece by piece. Daniel Dennett described the self as a "center of narrative gravity," a useful theoretical fiction generated by the brain's continuous storytelling rather than any actual homunculus inside the skull. Marya Schechtman developed the Narrative Self-Constitution View, arguing that personal identity is constituted by narrative integration: to be a person is to have woven experiences, actions, and plans into a coherent autobiographical story. Thomas Metzinger went further, arguing that no such things as selves exist in the world at all, only a Phenomenal Self-Model generated by neurocomputational processes — a representation so transparent that the organism mistakes it for unmediated reality. Karl Friston and Andy Clark, working within the predictive processing framework, identified the self as the brain's most robust high-level generative model, a set of predictions about the organism's interoceptive and exteroceptive states that functions as the stable background against which all other predictions are evaluated.
In parallel, literary theory and narratology have been elaborating, with extraordinary technical precision, the structure and varieties of exactly this kind of orientating mechanism as it appears in fiction. Gérard Genette's systematic analysis of narrative discourse distinguished between narration (who speaks?) and focalization (who sees?), producing a typology of narrator positions that has become foundational for literary analysis. Wayne C. Booth introduced the concept of the implied author and the distinction between reliable and unreliable narrators. Dorrit Cohn analysed the modes through which fictional consciousness is represented: psychonarration, free indirect discourse, and interior monologue. James Phelan's rhetorical theory identified progression — the generation, sustaining, and resolution of audience interests and stakes — as the narrator's primary function.
The two fields have arrived at structurally convergent conclusions. Cognitive science finds that the self is a narrative construction, a structural position from which experience is organised. Literary theory finds that the narrator is a structural position from which a storyworld is organised. Both are identifying the same underlying operation: an orientating mechanism that attributes stakes, organises concerns, and maintains coherence across time.
And yet both fields have been stopped short of the fullest implication of what they have found, by a shared and unexamined assumption. Both have tended to treat the narrator position as something the self or the fictional narrator occupies stably — as a fixed mode of being that defines what the self is or does. Cognitive science asks: what kind of narrator is the self? Literary theory asks: what kind of narrator is this text's narrator? Neither has adequately asked: what is the self's relationship to the narrator position, moment by moment, in living experience? And it is in the answer to that question that the most important insight is located. The self is not a narrator. It is a living agent with a narrator-position capacity that it deploys selectively, maintains in dynamic oscillation with its primary embeddedness, and always returns from. This is not a minor refinement. It changes everything.
II. The Error in the Search for the Fixed Self
The error that cognitive science has made is not in concluding that the self is a narrator. That conclusion is correct. The error is in treating the narrator position as the self's primary and stable mode of being — as what the self fundamentally is, rather than what the self does under specific circumstances. This error has led to a systematic misidentification of the self's ordinary condition, and it has generated a set of theoretical puzzles that dissolve once the error is corrected.
Consider Dennett's center of narrative gravity. The metaphor is powerful and the insight it encodes is real. The self has no fixed neural location, no homunculus that occupies a particular address in the brain. It is a functional abstraction, like a center of gravity, that organises the organism's processing without itself being any particular physical thing. But the metaphor carries a hidden assumption: the center of gravity is a stable structural property of the object. It is always there, always at the same location relative to the object's mass distribution, always available for calculation. If the self is like a center of narrative gravity, it is always operating as a narrative center. The metaphor implies permanence.
This implication is false to the phenomenology of actual lived experience. The self that Dennett describes — continuously generating narratives, continuously functioning as the focal point around which the brain's stories coalesce — is not the ordinary condition of human experience. The surgeon absorbed in a complex procedure is not, at that moment, operating as a narrative center. The farmer reading the soil with their hands is not generating a narrative. The parent attuned to their infant in the first hours of life is not constructing an autobiographical story. These are human beings at their most intelligent, most responsive, most fully alive to their situation — and they are not narrating. The center of narrative gravity is available to them; it is not permanently operating.
The same correction applies to Schechtman's Narrative Self-Constitution View. Schechtman argues that to be a person is to have woven experiences into a coherent autobiographical narrative — that the self is the story, not the entity that has the story. This is a profound and correct observation about one dimension of personhood. But it risks implying that the self is continuously engaged in autobiographical weaving, that the narrative mode of self-constitution is the self's permanent condition rather than a resource it draws on when continuity and coherence need to be actively constructed. Most of the time, persons are not weaving autobiographical narratives. They are living — absorbed, responding, engaging — and the autobiographical capacity is available in the background, ready to be activated when the situation requires it.
Metzinger's account comes closest to avoiding this error. His insistence that the self-model is transparent — that the organism looks through it to the world rather than at it as a model — implies that the self-model is not always foregrounded as a narrator position. The organism in absorbed engagement with its world is not, moment by moment, experiencing its self-model as a self-model. The self-model is doing its work invisibly, in the background, as the ground of orientation rather than as a foreground structure. This is exactly right, and it is what the present article means by the embedded narrator: the organism whose self-model is transparent, working in the background while the organism lives in the foreground of its own experience.
But even Metzinger, having identified the transparency of the self-model as its defining feature, analyses it primarily in the mode in which it becomes visible — in the breakdowns of transparency that constitute depersonalisation and related phenomena. The theoretical framework is built around the moments when the self-model becomes opaque, when the organism begins to experience the model as a model. This is valuable for understanding pathology. It is not adequate for understanding the healthy organism's ordinary condition, in which the self-model is invisible precisely because it is working well.
Friston and Clark's predictive processing account, while the most computationally sophisticated of the cognitive science frameworks, inherits the same limitation. The self as generative model is described in terms of its computational architecture — the hierarchical prediction structure, the minimisation of free energy, the balance between prior and likelihood. This architecture is always running. It does not turn off. The computational framework has no room for the oscillation between embedded absorption and narrator-position engagement that characterises actual living experience. The model is always modelling.
What is missing from all of these accounts is a concept that captures what the self primarily is when the narrator position is not foregrounded: the condition of absorbed, pre-reflective engagement with living that is the self's ordinary mode of being, the condition from which narrator-position engagement departs when required and to which it returns when the situation permits. Living Value Theory names this the embedded narrator, and it is the concept that gives the cognitive science of the self the grounding it has so far lacked.
III. Descartes in His Proper Place
The Cartesian cogito — cogito ergo sum, I think therefore I am — is the founding gesture of modern Western philosophy of mind. René Descartes, having subjected all his beliefs to systematic doubt, discovers that the one thing he cannot doubt is the existence of the doubting itself. Whatever else may be an illusion, the thinking, doubting, experiencing "I" must exist, because the doubt itself is evidence of a doubter. The cogito installs the self-aware thinking subject as the bedrock of philosophical certainty, and in doing so establishes what Charles Taylor identified as the inward turn that has structured Western modernity's understanding of the self: the self is found by going inside, by attending to the contents and operations of one's own mind, rather than by looking outward at the cosmos or the social order.
What LVT can now show is that the cogito is not wrong. It is a genuine and important observation about a specific narrator position that is genuinely available to the self. When Descartes sits in his study and performs the systematic doubt, he achieves something real: a moment of pure self-recursive narrator awareness, in which the self takes itself as the explicit object of its own examination, in which the narrator position is foregrounded and the content of experience is temporarily suspended in favour of attending to the structure of experiencing itself. This is a genuine cognitive achievement. It is what LVT describes as L4 and L5 intelligence applied to the self-recursive domain: the capacity to make the self's own operations the explicit object of analytical attention.
But the cogito is a moment, not a permanent state. Descartes in his study, performing the systematic doubt, is in an exceptional cognitive condition that he has deliberately cultivated for a specific philosophical purpose. He is not in the condition of ordinary human living. And the error — the error that has structured three centuries of subsequent philosophy of mind — is to take this exceptional moment as the definition of the self's primary and fundamental nature. The Cartesian subject, the transparent self-aware thinking thing, is what the self is when it is in one specific and relatively rare narrator-position configuration. It is not what the self always is, or primarily is, or most fully is.
The pathological version of the Cartesian position is exactly what the clinical literature on depersonalisation describes. The person who is permanently in the condition of observing their own thinking, who cannot escape the meta-awareness of their own mental processes, who is always the witness of their own experience rather than the liver of it: this is not a person who has achieved philosophical enlightenment. This is a person in clinical distress. Descartes's cogito, extended from a deliberate philosophical moment into a permanent experiential condition, is depersonalisation. The philosophical tradition's valorisation of the detached thinking subject as the true self has, in effect, been theorising a pathological narrator-position configuration as the norm.
This is not an attack on Descartes or on the valuable philosophical and scientific work that has proceeded from his insight. It is a precise placement of that insight within the full account of the self's actual living structure. The cogito is available. It is useful. It is the condition that makes systematic philosophy and rigorous self-examination possible. But it is useful precisely because it is not the self's permanent condition: it is a capacity that can be activated when philosophical or practical self-examination requires it and from which the self can return to absorbed living when the examination is complete. The mouse that has outwitted the fox can eat its nut. Descartes, having completed his Meditations, could presumably eat his dinner without thinking therefore being for the duration of the meal.
Charles Taylor's monumental account in Sources of the Self traces the genealogy of modern identity through Augustine's inward turn, the Cartesian cogito, Locke's account of personal identity, Romantic expressivism, and the modern valorisation of ordinary life. Taylor's project is to recover the moral sources of the modern self by tracing the philosophical and cultural developments through which the self's current form was constituted. This is a profound and indispensable contribution. But Taylor shares the assumption that has limited the cognitive science accounts: he treats the self-as-narrator-from-within as a modern achievement, a historical form of selfhood that emerged through the specific developments he traces, and that is genuinely different from the selfhood available to pre-modern persons.
LVT challenges this assumption directly. What Taylor traces is not a history of the self's changing structure but a history of the intellectual and literary conventions through which the self's structure has been permitted to be made explicit. The embedded narrator — the self that is primarily absorbed in its living, with narrator-position awareness available on demand, capable of reflective self-examination and of returning to embeddedness — is not a modern achievement. It is the universal form of healthy human selfhood. What changed historically is not the self but the repertoire of positions from which the self could be represented and reflected upon in writing.
IV. What the Self Actually Is: The Embedded Narrator
The embedded narrator is the human "I" as it actually is in healthy functioning across all historical periods and all cultures. It is not a narrator position. It is a dynamic: the ongoing oscillation between absorbed embeddedness and graduated narrator-position awareness, always anchored in the pre-reflective engagement with living that LVT identifies as L1, always capable of ascending toward explicit self-examination when the situation requires it, and always capable of returning to absorbed coordination when the situation permits.
Four features of this dynamic require precise statement.
The first is primacy of embeddedness. The embedded narrator's baseline condition is not self-reflection but absorbed engagement. At L1, the organism is in seamless pre-reflective coordination with its situation: the skilled practitioner whose competence has become second nature, the experienced navigator reading the sea with their whole body, the farmer who knows from the feel of the soil what the season requires, the parent attuned to their infant through embodied cues of timing and tone. This absorbed engagement is not a deficient or primitive state to be transcended by achieving narrator-position awareness. It is the condition of human intelligence at its most efficient and most fully at home in its world. The goal of any form of skill development is not to remain at L4 explicit analysis but to descend from deliberate practice back into absorbed, effortless coordination. The embedded narrator's primary condition is this absorbed engagement. Everything else departs from it and returns to it.
The second is selective narrator-position availability. The embedded narrator has the capacity — distinctively human in its full development — to take up a narrator-position relationship to its own experience. When a situation requires reflection, it can reflect. When a pattern in its own behaviour needs to be named and examined, it can name and examine it. When the interrecursive dynamics of a relationship require modelling from a degree of distance, it can achieve that distance. The narrator-position capacity is available across the full range of LVT's recursivity levels: from L3 articulation of what is happening, through L4 analysis of patterns and significance, to L5 meta-recursive repositioning that can step outside any framework and ask whether a different framing would serve better. What distinguishes healthy functioning from pathology is not the possession of these capacities but their selective deployment — expanding when the situation demands and contracting when it does not.
The third is the constitutive non-abandonment of embeddedness. The embedded narrator does not, in healthy functioning, fully vacate its own lived position to occupy a detached narrator position. Even at its most reflective, even in its most explicit moments of self-examination, the embedded narrator retains the felt reality of its own experience as the ground from which the reflection proceeds. The L5 meta-reflection is taken up from within the L1 ground, not as a replacement for it. When Descartes performs the systematic doubt, he does so from within a body sitting in a specific room in a specific historical moment with specific physical needs and social relationships. The cogito reaches toward the position of pure self-transparent thought, but it never fully arrives: Descartes remains an embedded narrator even in the act of attempting to achieve the position of a pure thinking thing.
The fourth is recursive fluidity as the defining capacity. The embedded narrator's most important cognitive achievement is not the ability to occupy any particular narrator position but the ability to move between positions as the situation demands. LVT calls this recursive fluidity: the capacity to escalate from L1 absorbed coordination through L2 felt misalignment to L3 articulation, L4 analysis, and L5 repositioning when disruption requires it, and to descend back through the levels to L1 when coordination can proceed without friction. Both movements are equally important. A self that can escalate but cannot descend has the underground-man pathology: a narrator trapped in permanent meta-reflection that cannot return to living. A self that can descend but cannot escalate has a different pathology: it is trapped in embeddedness without the capacity for reflective distance when situations require it. Health is the fluid movement in both directions, with L1 embeddedness as the home from which the self departs and to which it returns.
V. The Narrator Horizon
The account of the embedded narrator requires one further concept to be complete: the narrator horizon. This is the degree of narrator-position awareness available to the embedded "I" at any given moment without disrupting the embeddedness that is its primary condition. The narrator horizon is not a fixed feature of the self but a dynamic variable that changes with the situation, the developmental stage, the cultural formation, and the specific challenges the self is currently navigating.
In routine absorbed coordination — the skilled practitioner engaged in familiar work, the person in easy conversation with a close friend, the body moving through a known and safe environment — the narrator horizon is close. Narrator-position awareness is available but not activated; the self lives unreflectively in its embeddedness, and the narrator horizon is not in view. This is not cognitive poverty. It is L1 fluency in a domain where the relevant stakes are being navigated without disruption.
When L2 misalignment occurs — when something goes wrong, when the situation becomes difficult, when familiar coordination is disrupted — the narrator horizon expands. The self moves toward greater narrator-position awareness. It begins to articulate what is happening (L3), analyses the situation (L4), and if necessary repositions its entire framework (L5). The narrator horizon has expanded to bring more of the situation into explicit view. The self is now functioning in the narrator mode that cognitive science has been theorising as its permanent condition, but that is in fact a specific response to a specific need.
When the disruption resolves — when the L2 misalignment has been addressed, when the difficulty has been navigated, when coordination returns to its workable condition — the narrator horizon contracts again. The self returns to absorbed embeddedness. This contraction is as important as the expansion. It is not the loss of a capacity but its appropriate withdrawal. The mouse that has outwitted the fox does not maintain vigilance indefinitely. It eats its nut. The narrator horizon has contracted to its baseline condition because the disruption that expanded it has been resolved.
The narrator horizon concept makes the pathological cases precisely locatable. Depersonalisation is the narrator horizon fixed in the expanded position: the self is permanently in narrator-position awareness of itself, permanently observing rather than living, unable to contract the horizon back into absorbed embeddedness. The person with depersonalisation does not experience enhanced self-knowledge; they experience the breakdown of the structural condition that makes selfhood feel like first-person reality rather than third-person observation. The underground-man pathology is the narrator horizon expanded to L5 and unable to descend: the self is permanently repositioning its own frameworks rather than acting from any of them, paralysed by a meta-recursive capacity that has outrun its capacity for coordination.
What both pathologies share is the inability to return. The defining feature of the embedded narrator's health is not any particular narrator-position configuration but the fluidity of movement between configurations and the capacity to return to embeddedness when the situation permits. Therapy, for both pathologies, is the restoration of this fluidity: not the achievement of a different permanent narrator position but the recovery of the movement between positions that pathology has blocked.
VI. What Literary Theory Provides: A Typology of Available Positions
The foregoing account establishes what the self is: an embedded living agent with a narrator-position capacity that it deploys selectively. What it does not yet specify is the range of narrator-position configurations that the embedded narrator's capacity makes available. Here, literary theory's contribution becomes essential — and here the precise nature of what literary theory contributes needs to be carefully stated.
Literary theory does not describe the self's fixed narrator position. It provides a typology of configurations that the embedded narrator's capacity can assume under different circumstances and in response to different kinds of situations. Genette's focalization types, Booth's reliability spectrum, Cohn's modes of consciousness representation: these are not descriptions of different kinds of selves. They are descriptions of different configurations of the narrator-position capacity that any healthy embedded narrator has available.
Genette's three focalization types describe different configurations of the narrator's access to the interrecursive field it is rendering. Zero focalization — the classical omniscient narrator — is the configuration in which the narrator claims access to all participants' inner states simultaneously. Internal focalization restricts the narrator to one participant's access, placing it inside the interrecursive exchange as a participant rather than as an observer. External focalization observes from outside without accessing any participant's inner state. These are not three kinds of self. They are three configurations of the embedded narrator's access to interrecursive situations, each appropriate to different circumstances.
The healthy embedded narrator deploys all three configurations across the range of its experience. When engaged in absorbed practical activity with others — working together on a shared task, coordinating in familiar roles — the embedded narrator operates in something like internal focalization: it is inside the interrecursive exchange, responding to it from within its own stake position. When reflecting on a past social interaction, it may achieve something like zero focalization: the retrospective narrator who can now access both parties' stake structures and see how they interacted in ways that were not visible from inside the exchange. When managing a difficult relationship with a deliberate degree of analytical distance, it may approach external focalization: observing the other's behaviour without fully inhabiting the exchange.
Booth's distinction between reliable and unreliable narrators maps onto the embedded narrator's self-recursive domain with great precision. An unreliable narrator is one whose values, perceptions, or reports diverge from what external evidence or the implied author's perspective establishes. In the self-recursive domain, every embedded narrator is to some degree unreliable: the self's accounts of its own motivations are reconstructive rather than transparent, its memories are shaped by current concerns rather than faithfully preserved, its evaluations of its own behaviour are influenced by self-serving biases that it typically cannot detect from within. Booth distinguishes three modes of unreliability: factual (the narrator is wrong about what happened), evaluative (the narrator misjudges the significance of events), and motivational (the narrator does not know why it does what it does). These three modes correspond to three distinct and commonly observed phenomena in self-narration: false or reconstructed memories, cognitive distortions in the evaluation of experience, and the gap between the self's conscious account of its motivations and the pre-reflective processes that actually drive its behaviour.
Cohn's three modes of consciousness representation — psychonarration, free indirect discourse, and interior monologue — describe different degrees of managed distance between the narrating position and the experienced content it renders. Psychonarration is the narrator's L4 conceptual apparatus applied to describe L1 and L2 states: the highest degree of symbolic mediation, the greatest distance from the pre-symbolic material. Interior monologue is the attempt to render L1 and L2 consciousness in the character's own L3 idiom, minimising the narrator's own overlay. Free indirect discourse occupies the middle position: the narrator's vocabulary is blended with the character's idiom, creating the double-voicedness that Bakhtin identified as the novel's defining feature.
The embedded narrator moves between all three modes in its own self-narration. When a patient in therapy reports their emotional experience in terms they have learned from previous therapy, using conceptual schemas absorbed from psychological discourse, they are in psychonarration mode: their account is mediated by L4 frameworks that create distance from the L2 felt experience. When the same patient, in a moment of genuine contact with their experience, finds words that fit what they actually feel rather than what the schema predicts they should feel, they are approaching free indirect discourse: the narrating position is closer to the experiencing position, the distance has reduced. When they are simply in the experience itself, not narrating at all, they are at L1 — and when they begin to find the first words for what that experience was, they have produced the opening of interior monologue. The therapeutic movement from psychonarration toward free indirect discourse toward moments of genuine contact with felt experience is the movement from high narrator-position mediation back toward embeddedness, conducted within the narrator-position mode rather than through its abandonment.
VII. Homer, Caesar, and the Convention Error
The most consequential implication of the embedded narrator account concerns the history of the self. The standard view, shared by philosophers from Hegel to Charles Taylor and by literary historians from Auerbach to Ian Watt, is that something genuinely changed in the structure of selfhood between the ancient world and modernity: that pre-modern persons had a different, less inward, less psychologically complex form of self-experience, and that the rich interiority that we associate with modern selfhood — the sense of being a complex inner life, a centre of consciousness with depth, history, and narrative — is a historical achievement rather than a biological given.
This view draws its evidence primarily from literary sources. Homer's characters are externally narrated: their inner states are rendered through their speeches and actions rather than through the reported texture of their inner experience. The divine machinery of the Iliad — the gods who intervene, advise, and direct — is read by Julian Jaynes as evidence that Homeric persons literally heard their inner impulses as external divine voices, lacking the unified inner narrator that modern persons possess. Bruno Snell argued that Homeric characters lacked a unified concept of the self, that the fragmented vocabulary for mental states in Homer reflects the psychological fragmentation of Homeric persons. Erich Auerbach distinguished the Homeric style, which brings everything to the surface and leaves nothing in psychological shadow, from the biblical and modern styles that allow depth, concealment, and interiority.
LVT identifies all of these arguments as committing the same category error: they treat the literary convention as transparent access to the phenomenology of selfhood. The absence of interior narration in a literary tradition is not evidence of the absence of interior experience. It is evidence of the absence of interior narration as a literary convention available to that tradition. These are entirely different things, and conflating them is the source of the error.
Julius Caesar's Gallic War provides the cleanest available demonstration. Caesar refers to himself throughout the work in the third person: "Caesar decided," "Caesar advanced," "Caesar considered the terrain." If literary conventions were transparent guides to phenomenology, this would be evidence that Caesar did not experience himself from the inside — that he was in a permanent condition of external focalization of his own experience, observing himself as a character in someone else's story. This is absurd. Caesar was a Roman general making life-and-death decisions under extreme pressure, managing complex interrecursive dynamics across thousands of soldiers, political allies, and enemies. We can be almost entirely confident that he experienced these decisions from inside — feeling their stakes, weighing his options, managing his fears and ambitions — as any embedded narrator in his situation would. The third-person self-reference is a rhetorical convention of Roman military and historical writing, associated with authority, objectivity, and a specific kind of political gravitas. It tells us about Roman literary conventions, not about Caesar's phenomenology.
The same argument applies to Homer, though the literary distance makes it harder to see. The warriors of the Iliad are externally narrated because the oral epic tradition required external narration: a tradition of composed-in-performance narrative, designed for public recitation before audiences, had very different representational conventions from the private, introspective, written fiction that developed over a millennium later. The convention of external narration does not imply the absence of interior experience in the persons the convention represents. Achilles, at the ships, experiencing the extremity of his anger and grief, was an embedded narrator feeling those states from inside them, with narrator-position awareness available in the way that any human agent in extreme emotional circumstances has it available. Homer's literary convention does not render this interior experience; it does not follow that the interior experience was not there.
The LVT hypothesis, advanced here as a hypothesis requiring further comparative and genealogical investigation, is that the embedded narrator is not a modern achievement. It is the natural and universal form of healthy human selfhood. The evidence for this hypothesis is not primarily literary but structural and evolutionary. The complex social life that hominids have been living for hundreds of thousands of years — the interrecursive coordination of large social groups, the management of alliances and conflicts, the care of dependent offspring, the transmission of cultural knowledge — requires exactly the capacities that the embedded narrator possesses: L1 absorbed coordination for routine life, L2 felt misalignment for detecting when something is wrong, L3 articulation for communicating what is wrong, L4 pattern recognition for understanding why, and L5 meta-reflection for repositioning when the existing framework has failed. These are not capacities that emerged with the novel or with the Enlightenment. They are the cognitive requirements of human social life as such.
What has changed historically is not the embedded narrator but the literary and discursive conventions through which the embedded narrator's structure is permitted to be represented. Augustine's Confessions did not create interior experience; it created a literary form for representing interior experience in an extended, retrospective, self-examining mode. Montaigne's Essays did not create the mobile, self-revising self; they created a literary form for performing the embedded narrator's self-examining capacity in writing. Austen's free indirect discourse did not create the dual embedded-and-narrating configuration of healthy self-reflection; it created a narrative technique capable of representing that configuration with unprecedented precision. Virginia Woolf's stream of consciousness did not create L1 pre-reflective consciousness; it created a literary mode for approximating, in symbolic form, the texture of absorbed experience.
What each of these literary innovations achieved was an expansion of what could be written, and therefore of what could be reflected upon and cultivated. Literary forms that represent the embedded narrator's structure make that structure more available as a resource for the selves that encounter them. The reader who has inhabited Austen's doubly focalised narrator position — simultaneously inside Emma's perspective and aware of its limits — has practised a specific configuration of the embedded narrator's capacity that that configuration was less readily available before Austen made it a literary form. The literary history is not the history of the self's changing structure. It is the history of the expansion of the self's reflective repertoire: the progressive development of literary forms that make more of the embedded narrator's structure available for explicit cultivation.
VIII. The Homology Precisely Stated
With the embedded narrator account established, the homology between the narrator position in fiction and the self position in cognition can now be stated with a precision that neither cognitive science nor literary theory has achieved on its own.
The homology is between the embedded narrator's narrator-position capacity and the fictional narrator's structural operation. Both are instances of the same underlying cognitive operation: the generation of an orientating mechanism from which a field of living stakes is organised, evaluated, and rendered navigable. This operation is what cognitive science has been identifying under various names — narrative gravity, self-model, generative model, autobiographical author — and what literary theory has been analysing under various names — narrator, focalizer, implied author, consciousness representor.
But the homology is between a capacity and a structure, not between two identical things. The fictional narrator is constitutively a narrator: it has no embedded life of its own, no stakes in the storyworld that are not part of the narrating function. It exists only as a narrator. The human "I" is constitutively an embedded agent: the narrator-position capacity is something it has and deploys, not what it is. The "I" exists primarily as a living being, and the narrator capacity is what it uses when living requires it.
This distinction clarifies the homology without undermining it. When we bring literary theory's insights to cognitive science, what we bring is not the claim that the self is identical to a fictional narrator. We bring the claim that the narrator-position configurations that literary theory has identified and analysed with great precision are the configurations available to the embedded narrator when it deploys its narrator-position capacity. The fictional narrator is, in a sense, the embedded narrator's narrator-position capacity operating without its embedded ground — the capacity extracted from the living agent and examined in its own right. Literary analysis examines what the narrator-position capacity can do and how it can be configured. Cognitive science, with the correction that LVT provides, can understand these configurations as selectively available modes of the embedded narrator rather than as fixed descriptions of what the self fundamentally is.
The result is a genuinely bidirectional exchange. Cognitive science provides literary theory with the evolutionary and developmental grounding for why the narrator-position capacity exists at all: it evolved to meet the demands of complex interrecursive social life, it develops through the sequence from joint attention to primary intersubjectivity to autobiographical reasoning documented in the developmental literature, and it is exercised in narrative fiction as a form of practice for the interrecursive intelligence that social life requires. Literary theory provides cognitive science with the fine-grained typology of narrator-position configurations that cognitive science has lacked: the focalization types that describe different modes of interrecursive access, the reliability spectrum that describes different modes of self-recursive distortion, the modes of consciousness representation that describe different degrees of narrator-position mediation over the embedded agent's felt experience.
What neither field has provided, and what LVT now provides for both, is the concept that gives the exchange its proper theoretical grounding: the embedded narrator as the living agent whose narrator-position capacity makes all of these configurations possible, whose primary condition is absorbed embeddedness rather than any narrator position, and whose health consists in the recursive fluidity that allows it to move between configurations as the situation demands and to return to embeddedness when the situation permits.
IX. The Self That Cognitive Science Has Been Looking For
The cognitive science of the self has been looking for the right thing in the wrong place. It has been looking for the stable reality of the self — the fixed structure, the permanent configuration, the enduring neural correlate — that could be identified as what the self is. It has found, instead, a series of functional abstractions: a center of gravity that is useful as a theoretical fiction but corresponds to no physical thing, a self-model that is transparent and therefore invisible to itself, a generative model that is always running but never fixed. Each of these findings is genuine. None of them identifies the stable reality that the search was seeking, because the search was premised on a false assumption: that the self has a stable reality in the sense of a fixed configuration.
The embedded narrator concept dissolves this problem by dissolving the assumption that generated it. The self does not have a fixed configuration because it is not a fixed configuration. It is a dynamic — a living movement between absorbed embeddedness and graduated narrator-position awareness, anchored in the pre-reflective engagement with living that is the organism's primary condition. The search for a fixed self is the search for a resting point in a dynamic that has no permanent resting point. The self is always moving: ascending toward narrator-position awareness when situations require reflection, descending toward absorbed embeddedness when situations permit coordination. What is stable is not the position but the capacity for movement and the anchor that makes movement possible — the L1 embedded life that is both the origin and destination of every narrator-position departure.
This account also resolves what has been called the narrative self's paradox: Galen Strawson's argument "against narrativity," in which he distinguishes between "diachronic" individuals who naturally view their lives as continuous stories and "episodic" individuals who experience life as disconnected moments. Strawson argues that the narrative model of the self, which he takes to imply that persons naturally and healthily live in a mode of continuous autobiographical self-narration, is neither a universal psychological reality nor a normative requirement. He is right that continuous self-narration is not a universal or a norm. But the embedded narrator account does not require it. The embedded narrator's narrative capacity is selectively deployable, not continuously operating. Episodic and diachronic individuals differ not in whether they have the narrative capacity but in how readily and extensively they deploy it. Both are embedded narrators. The diachronic individual deploys the narrative capacity more often and over longer temporal spans. The episodic individual deploys it less frequently. Neither is wrong; neither is the better self. What matters is whether the capacity for movement is maintained — whether the narrative capacity can be activated when the situation genuinely requires temporal coherence, and whether the return to absorbed episodic living is available when it does not.
The search that cognitive science has been conducting is not misguided. The self is a real and important object of inquiry. The narrator-position capacity is genuinely present and genuinely significant. The various frameworks that have been developed to analyse it — the narrative self, the phenomenal self-model, the predictive self-model — are genuine contributions to understanding. What LVT adds is the frame within which all of these contributions can be properly located: the frame of the embedded narrator, the living agent whose narrator-position capacity these frameworks have been analysing, and whose primary condition — absorbed embeddedness in a living mesocosm — those frameworks have systematically overlooked.
The self that cognitive science has been looking for is not a fixed position. It is the being that can occupy many positions selectively, return from all of them, and live primarily in the absorbed engagement with its own life that no position captures but that every position departs from and returns to. This is the embedded narrator. It is what we have always been, in moments. The cognitive science of the self has been describing its narrator-position capacity with increasing sophistication. What it has not yet described is the living ground from which that capacity operates and to which it returns. That ground is the beginning and the end of the inquiry.
References
Bakhtin, Mikhail. The Dialogic Imagination: Four Essays. Translated by Caryl Emerson and Michael Holquist. Austin: University of Texas Press, 1981.
Blanke, Olaf and Thomas Metzinger. "Full-body illusions and minimal phenomenal selfhood." Trends in Cognitive Sciences 13, no. 1 (2009): 7-13.
Booth, Wayne C. The Rhetoric of Fiction. Chicago: University of Chicago Press, 1961.
Botvinick, Matthew and Jonathan Cohen. "Rubber hands 'feel' touch that eyes see." Nature 391 (1998): 756.
Bruner, Jerome. Acts of Meaning. Cambridge, MA: Harvard University Press, 1990.
Caesar, Julius. The Gallic War. Translated by Carolyn Hammond. Oxford: Oxford University Press, 2008 [c. 50 BCE].
Clark, Andy. "Whatever next? Predictive brains, situated agents, and the future of cognitive science." Behavioral and Brain Sciences 36, no. 3 (2013): 181-204.
Clark, Andy. Surfing Uncertainty: Prediction, Action, and the Embodied Mind. Oxford: Oxford University Press, 2016.
Cohn, Dorrit. Transparent Minds: Narrative Modes for Presenting Consciousness in Fiction. Princeton: Princeton University Press, 1978.
Dennett, Daniel C. "The Self as a Center of Narrative Gravity." In Self and Consciousness: Multiple Perspectives, edited by F. Kessel, P. Cole, and D. Johnson. Hillsdale, NJ: Erlbaum, 1992.
Dennett, Daniel C. Consciousness Explained. Boston: Little, Brown, 1991.
Descartes, René. Meditations on First Philosophy. Translated by John Cottingham. Cambridge: Cambridge University Press, 1986 [1641].
Dreyfus, Hubert L. Being-in-the-World: A Commentary on Heidegger's Being and Time, Division I. Cambridge, MA: MIT Press, 1991.
Dreyfus, Hubert L. and Stuart E. Dreyfus. Mind Over Machine: The Power of Human Intuition and Expertise in the Era of the Computer. New York: Free Press, 1986.
Fludernik, Monika. Towards a "Natural" Narratology. London: Routledge, 1996.
Friston, Karl. "The free-energy principle: a unified brain theory?" Nature Reviews Neuroscience 11, no. 2 (2010): 127-138.
Gallagher, Shaun. "Philosophical conceptions of the self: implications for cognitive science." Trends in Cognitive Sciences 4, no. 1 (2000): 14-21.
Gallagher, Shaun. How the Body Shapes the Mind. Oxford: Clarendon Press, 2005.
Genette, Gerard. Narrative Discourse: An Essay in Method. Translated by Jane E. Lewin. Ithaca: Cornell University Press, 1980 [1972].
Genette, Gerard. Narrative Discourse Revisited. Translated by Jane E. Lewin. Ithaca: Cornell University Press, 1988.
Habermas, Tilmann and Susan Bluck. "Getting a life: The emergence of the life story in adolescence." Psychological Bulletin 126, no. 5 (2000): 748-769.
Herman, David. Story Logic: Problems and Possibilities of Narrative. Lincoln: University of Nebraska Press, 2002.
Hume, David. A Treatise of Human Nature. Oxford: Oxford University Press, 2000 [1739].
Ingold, Tim. The Perception of the Environment: Essays on Livelihood, Dwelling and Skill. London: Routledge, 2000.
Jaynes, Julian. The Origin of Consciousness in the Breakdown of the Bicameral Mind. Boston: Houghton Mifflin, 1976.
McAdams, Dan P. "The Psychology of Life Stories." Review of General Psychology 5, no. 2 (2001): 100-122.
McAdams, Dan P. The Stories We Live By: Personal Myths and the Making of the Self. New York: Guilford Press, 1993.
Merleau-Ponty, Maurice. Phenomenology of Perception. Translated by Colin Smith. London: Routledge, 1962 [1945].
Metzinger, Thomas. Being No One: The Self-Model Theory of Subjectivity. Cambridge, MA: MIT Press, 2003.
Metzinger, Thomas. The Ego Tunnel: The Science of the Mind and the Myth of the Self. New York: Basic Books, 2009.
Phelan, James. Reading People, Reading Plots: Character, Progression, and the Interpretation of Narrative. Chicago: University of Chicago Press, 1989.
Phelan, James. Experiencing Fiction: Judgments, Progressions, and the Rhetorical Theory of Narrative. Columbus: Ohio State University Press, 2007.
Ricoeur, Paul. Oneself as Another. Translated by Kathleen Blamey. Chicago: University of Chicago Press, 1992.
Ricoeur, Paul. Time and Narrative, 3 vols. Translated by Kathleen McLaughlin and David Pellauer. Chicago: University of Chicago Press, 1984-1988.
Schechtman, Marya. The Constitution of Selves. Ithaca: Cornell University Press, 1996.
Seth, Anil K. Being You: A New Science of Consciousness. New York: Dutton, 2021.
Snell, Bruno. The Discovery of the Mind: The Greek Origins of European Thought. Translated by T. G. Rosenmeyer. Oxford: Blackwell, 1953.
Strawson, Galen. "Against Narrativity." Ratio 17, no. 4 (2004): 428-452.
Taylor, Charles. Sources of the Self: The Making of the Modern Identity. Cambridge, MA: Harvard University Press, 1989.
Tomasello, Michael. The Cultural Origins of Human Cognition. Cambridge, MA: Harvard University Press, 1999.
Tomasello, Michael. Becoming Human: A Theory of Ontogeny. Cambridge, MA: Harvard University Press, 2019.
Trevarthen, Colwyn. "Communication and cooperation in early infancy: A description of primary intersubjectivity." In Before Speech: The Beginning of Interpersonal Communication, edited by M. Bullowa. Cambridge: Cambridge University Press, 1979.
Zahavi, Dan. Subjectivity and Selfhood: Investigating the First-Person Perspective. Cambridge, MA: MIT Press, 2005.
Zunshine, Lisa. Why We Read Fiction: Theory of Mind and the Novel. Columbus: Ohio State University Press, 2006.