1. Why We Tell Stories

It is a truth universally acknowledged that people love stories. From the Mahabharata and the Iliad to The Rise and Fall of the Roman Empire and Paul Revere's Ride, from One Hundred Years of Solitude to The Handmaiden's Tale – across times, cultures, and media; in art high and low; in myth, fiction, and history; on college applications, at family reunions and office parties – we are drawn to tell and consume stories. Indeed, the anthropologist John Niles (Reference Niles1999) proposes ‘Homo narrans’ as a more fitting label to capture our essential differentiating characteristic than Homo sapiens.

Why are we such inveterate makers and consumers of story? For many reasons, including the sheer imaginative pleasure of filling our minds with something other than the exigencies of the everyday. But beyond this, it's natural to ask how stories work as ‘equipment for living’, in literary critic Kenneth Burke (Reference Burke1938)'s phrase. In this section, I tease out two functional roles for stories, which relate to time and telling in different ways.

The first role is expressed by the psychologist Jerome Bruner's descriptions of stories as part of ‘our armamentarium for dealing with surprise’ (Reference Bruner2002, p. 29). More specifically, Bruner (Reference Bruner2002, p. 31) says that

Translating this line of thought into Burke's equipment metaphor, we might say that stories are wayfinding instructions for navigating life. They specify recurring types of events, situations, and characters, and offer diagnostics for recognizing them. They tell us what to expect next, and offer strategies for response: reasons to select one path over another, and techniques for getting back on track.

On this model, stories’ functional role is essentially prospective and practical. However, this doesn't explain why we retell stories, or care so much about the details of different stories that instantiate the same basic template. We don't just tell a story until we have internalized its moral. Rather, we retrace past events repeatedly, in loving detail, even with the end clearly in mind and even when we don't expect a similar type of situation to arise again. The philosopher Louis Mink (Reference Mink1970, p. 554) proposes that we engage in such retrospective rumination

On Mink's model, we make and consume stories because they give us a cognitive grip on a sequence of events by knitting them together into a coherent explanatory pattern. As he puts it:

Thus, if Bruner treats stories as wayfinding instructions, Mink construes them as satellite maps. Just as having a map of a spatial region scaffolds understanding, recall, and reasoning in a way that even an exhaustive set of turn-by-turn navigational instructions does not (Camp, Reference Camp2007a), so too does a story's ‘configurational comprehension’ scaffold explanation and prediction by situating events in relation to one another. Moreover, in contrast to Bruner's culturally conventional templates, Mink's ‘configurational comprehensions’ can be highly specific, connecting fine-grained details of the particular events they recount in complex, nuanced ways. This is especially relevant for Mink as a philosopher of history who aims to justify stories as providing a species of understanding distinct from scientific categorization and causal generalization. I think this insight about particularist explanation is profound, not just for history but for the humanities more generally (Ismael, Reference Ismael and Grimm2017), though the nature of explanation is not my focus here.

2. What Is a Story?

In §1, we identified two roles for stories as ‘equipment for living’: as comprehensive configurations and as templates for action. In §3, I'll consider stories as tools specifically for constructing selves. But first, we need to get clearer about how stories function, such that they can perform these two roles. Both Bruner and Mink take stories to be useful because they offer us a structure for understanding a collection of otherwise disparate events. What is this structure? Fully addressing this question would require an extensive journey through narratology; for current purposes we need just three basic points, all following from Aristotle's famous claim that a story has a beginning, middle, and end.

While this observation might appear trivial, it entails a substantive relationship to time, with significant implications for what stories can do. A story is temporally bounded: it selects out a portion of the universe's vast continuum of space-time possibilities. It is also essentially about events: the sorts of things that start and stop at times and involve change. While stories also describe states, these occur as the background conditions and consequences of those events. Further, a story connects multiple events into a temporal sequence. Thus, just as maps represent by placing entities in spatial relation (Camp, Reference Camp2007a) and family trees represent by locating entities in ancestral relation (Camp, Reference Camp and Lurz2009a), so stories represent by placing events in temporal relation.

So, stories describe temporally bounded sequences of events. But they don't just list events, as annals do, or even events related to a single subject, as chronicles do (Carroll, Reference Carroll2001; Fraser, Reference Fraser2021). A beginning is more than a starting point, and an end is more than a stopping point: together, they delimit a span of time as an integrated unit. More importantly, as Mink emphasizes, a story connects the described events into a configuration that renders both the events themselves and their dynamics comprehensible, by placing them in explanatory relation.

Aristotle maintained that the principle of ‘narrative connection’ is causal: later events are necessary or probable consequences of earlier ones. While there is something intuitively right about this, causation cannot be necessary for narrative connection, given that many satisfying stories are driven by remarkable coincidences (Carroll, Reference Carroll2001). Nor is causation sufficient for narrative connection on its own: a recounting of the paths of billiard balls across a table in terms of the transfer of kinetic force constitutes a narrative only in the weakest sense. Rather, recall that Mink speaks of a story's end being rendered ‘necessary’ by the ‘promise’ of the beginning and the beginning by the promise of the end, through a kind of interactive ‘backward reference’.

Retrospective necessity, then, is the second crucial feature of narrative structure: a story's end provides an interpretive principle for selecting, describing, and explaining events in its beginning and middle. In the simplest case, retrospective necessity is achieved by describing one agent pursuing one goal, beginning with the goal's formation and specifying crucial causal steps to its fulfillment. This is the canonical case of teleological explanation, sometimes described as backward causation. Obviously, most stories don't hew to this simple, linear, straightforwardly causal format. However, we just need the weaker point that stories interpret a sequence of events as contributing somehow to the story's ending, where these contributions are related to agents trying to make things happen, and the ending somehow resolves those attempts.

As fictions like Lolita or Ender's Game and films like Rashomon and Sixth Sense relish demonstrating, substantively different narratives of the same temporal interval will be generated by attributing different goals to the same agent, by shifting the explanatory focus to different agents, or by raising different explanatory questions about their actions. Different narrative frames select different events in a sequence as relevant for inclusion in a story. They warrant different descriptions of a given event, by subsuming it under different nexuses of causal ‘joints and sockets’ (Dennett, Reference Dennett1991). And they prioritize and connect events in different ways – as initiating, forestalling, foreshadowing, or otherwise motivating or impeding each other – which may not be causal or even explanatory in a more narrowly metaphysical sense. Moreover, selecting, describing, or connecting particular events in a certain way affects the appropriate selection, description, and connections among other events in a global holistic way, so that the significance of the whole both influences and is influenced by the significance of its constituent events. As a result, shifting a story's narrative focus can dramatically transform, not just which events deserve to be mentioned, but what those events fundamentally mean (Camp, Reference Camp2017).

The third feature of narrative structure is that stories don't merely explain the dynamics of a bounded temporal sequence of events involving agents making and undergoing change. Those agents and events are presented as mattering, in a way that engages emotional and evaluative response. More specifically, stories embed a dynamic sequence of events within an emotional or epistemic cadence – what David Velleman (Reference Velleman2003), following Frank Kermode (Reference Kermode1967), calls the narrative ‘tick tock’. Bruner's ‘sting of fortuity’ displays this cadence in its most basic form: we have hopes and dreams, and feel anger or grief when they are stymied. Again, obviously most narratives are not so simple: we experience emotions on behalf of a story's protagonist that they don't themselves feel, or gloat at a villain's downfall. Stories often also engage epistemic emotions, like curiosity about why the Roman Empire fell (Carroll, Reference Carroll2007; Fraser, Reference Fraser2021). The crucial point is that narrative closure – the ‘sense of an ending’, as Kermode puts it – involves an engaged, dynamic investment in the story's guiding concerns.

Here, then, is where we've arrived. We use stories to achieve a comprehensive understanding of a temporally bounded sequence of events, sometimes at least partly in order to guide action. Stories accomplish this by selecting, describing, and connecting those events into a coherent, holistic configuration that explains them in terms of their contribution to the story's ending, which resolves the story's operative concerns in a viscerally satisfying way. Various genres – comedy, tragedy, etc. – offer more specific templates for types of events and characters, guiding concerns, and resolving cadences. And particular stories implement and tweak those templates by recounting particular characters encountering particular suites of obstacles and opportunities in ways that elicit particular constellations of epistemic, emotional, and evaluative cadences.

3. Stories for Selves

So far, I've been discussing stories as tools for managing our engagement with the world in general. Some theorists treat stories as ‘equipment for living’ in a more specific sense: as tools for building selves. A minimal version of this view follows naturally from Bruner's point that we use stories to navigate unexpected events: if we regularly use stories to guide action, and our selves are the products of our actions, then stories often play a role in creating our selves. And indeed, Bruner claims that ‘self-making is a narrative art’; ‘[i]t is through narrative that we create and re-create selfhood’ (Reference Bruner2002, pp. 65, 85).

By itself, the core idea that we use stories to create selves might just amount to the claim that stories create selves by guiding action at critical junctures of our lives. But this invites the question of what unites the self we thereby create into a coherent whole. And given what we've said so far, a natural answer is that this unity is provided by an overarching narrative of the self's life. Thus, Marya Schechtman claims that ‘a person creates his identity by forming an autobiographical narrative’ (Reference Schechtman1996, p. 93); ‘constituting an identity requires that an individual conceive of his life as having the form and the logic of a story […] where “story” is understood as a conventional, linear narrative’ (Reference Schechtman1996, p. 96).

By itself, the core idea might also just amount to the claim that stories are one tool within a richer ‘armamentarium’ for self-construction. But again, given what we've said so far, it's natural to take a further step. In §2, we saw that narrative structure is essentially teleological, in the broad sense of selecting, describing, explaining, and evaluating constituent events in terms of an overarching structure defined by ‘retrospective necessity’. Stories in general need not be teleological in the narrow sense of recounting an agent's pursuit of a goal. But it is plausible that stories focused on selves will focus on the goals and concerns of those selves. Moreover, one might think that any description of actions as actions will be teleological in the stricter sense of explaining them in terms of an agent's intention to achieve a goal. Together, these points might seem to suggest that, as Alisdair MacIntyre says, ‘narrative history of a certain kind turns out to be the basic and essential genre for the characterization of human action’ (Reference MacIntyre1981, p. 208).

If we amplify the core idea in these two ways, we end up with a view on which articulating an overarching autobiographical narrative is the essential mechanism for constructing selves as agents. It follows from this amplified view that it is essential to having a self that one actualize such a life narrative. It also follows, perhaps less directly but still fairly naturally, that selves have value insofar as they do implement that identity-constituting narrative. As MacIntyre (Reference MacIntyre1981, p. 203) puts it:

I want to note three things about this amplified view of stories as tools for self-construction. First, it is remarkably strong: it holds that the essential principle of a self is the explicit articulation of a narrative of a lifelong quest in pursuit of an overarching goal. While the overarching goal that drives an agent's lifelong quest may be complex and the path to its achievement indirect, failing in that quest means failing to be a self in the full sense of the term. For many of us, this is starkly counter-intuitive. We ordinarily take many selves that fail to satisfy this criterion to be successful, including ‘Episodics’ who live for the moment (Strawson, Reference Strawson2004); ‘characters’ who simply instantiate a stable identity across a long period; and those with cognitive deficits that prevent them from articulating an overarching quest either prospectively or retrospectively (Camp, Reference Camp2011).

Second, we arrived at this amplified view through a series of plausible, if hardly incontrovertible steps: starting from the core observation that story-telling plays a central role in our ordinary practices of self-interpretation and action-guidance; noting that narrative structure is teleological in the broad sense of being defined by retrospective necessity; and adding a pair of motivations for interpreting that broadly teleological structure in a more robust way, as a lifelong quest. We also moved from a claim about interpreting selves to a claim about constituting them, justified by the implicit assumption that well-grounded interpretations reflect reality, and from there to a claim about evaluating selves.

Third, strong as it is, the conclusion is not just a philosopher's abstract concoction: it articulates an intuition that runs deep and wide. The idea of a Hero's Journey is especially prevalent in Western Romantic culture, as reflected in the long-standing admiration for figures like Alexander the Great, Napoleon, and Wordsworth. But practitioners of ‘mythopoetics’ including Carl Jung, Joseph Campbell, and Robert Bly have traced out variations on the basic genre across cultures. More recently, it has thrived in the neo-Jefferson American myth of the self-made man, evinced by figures like Abraham Lincoln, Booker T. Washington, Henry Ford, and Elon Musk. And it animates various self-help therapies and movements with broad contemporary popularity, like The Purpose-Driven Life.

4. Narrative Hostage-Taking

We stand, then, on the edge of a dark forest, having followed a fairly well-trodden path, with a hidden dragon's lair marked out just around the bend. My primary interest here is not to assess the narrative conception of self on metaphysical or normative grounds. As we noted in §3, there is ample room to worry about whether the narrative conception, in the amplified form it does and arguably needs to take to perform the explanatory work assigned to it by theorists like Schechtman and MacIntyre, is committed to an implausibly pervasive error theory about our ordinary practices of describing and valuing selves (Camp, Reference Camp2011). There is also room to worry that narratives are epistemically suspect because they lead us to confabulate an unwarranted grounding basis for the emotional cadences that make them so viscerally compelling (Velleman, Reference Velleman2003), when really they are just epiphenomenal fictions (Dennett, Reference Dennett1988). I want to put these more general concerns to the side, to focus on how the narrative conception interacts with stories’ ability to perform the functional roles we identified in §1.

The fundamental problem is that the narrative conception of selfhood saddles stories with an explanatory burden that compromises their capacity to perform the functional roles of configurational comprehension and guiding action. Because narrative structure involves retrospective dependence, a narrative's overall meaning is only determined, and only accessible, given the story's ending. By defining selves in terms of life narratives, the narrative conception holds the meaning and value of those selves hostage to their lives’ ends, in a way that leaves radically underdetermined not just who we might become, but who we are and what we should do right now.

More explicitly, as we saw in §2, the comprehensive configuration offered by narrative is complex, holistic, and broadly teleological: it selects, individuates, and connects a sequence of disparate events in terms of their contribution to the story's ending. A constituent event can contribute to that ending in many ways: in the simplest case, by causally facilitating the protagonist's focal goal, but also by explaining or motivating the ending in non-causal ways, or by reinforcing or qualifying its broader epistemic or emotional significance. What matters is just that the complex, holistic structure for an entire life narrative depends essentially on its end. The consequence is that, as the literary theorist Peter Brooks (Reference Brooks1985, p. 95) puts it, ‘All narrative may be in essence obituary in that […] the retrospective knowledge that it seeks […] stands on the far side of the end, in human terms on the far side of death.’

Epistemically, this means that as we move through our lives, the narrative meanings we assign to past and present events can be no more than guesses about their ultimate contribution to our life's end. Brooks says that we read novels ‘in anticipation of retrospection’ (Reference Brooks2005, p. 23). This basic point also applies to life narratives. As we might put it, in interpreting our lives just as in reading a novel or history, we are attempting to solve a multivariable equation in which the value of each particular event is constrained by the values of other events in relation to the whole, by way of their relation to the end.

But this basic similarity cloaks a significant difference. An essential condition on the anticipation of retrospection we perform while reading a fictional or historical narrative is our trust that an intentional agent has constructed or imposed a comprehensive, epistemically and emotionally satisfying structure on the narrative's constituent events. Again, we are often deeply unsure how the author will manage to build that satisfying structure; and they can manage it in many, often unexpected ways. We may also question, while reading and at the end, whether our trust is well placed. Still, the act of reading a completed narrative constitutively involves a basic level of trust that the equation does have a satisfying solution. At the most basic level, this trust imbues the story's constituent events with a minimal level of significance: we assume that if an event has been mentioned at all, it must contribute in some substantial way to the story's end.

When we transfer narrative understanding from these sorts of authoritatively authored texts to real life, the anticipation of retrospection that governs our provisional interpretations of unfolding events becomes considerably more fraught. Some people experience a basic authorial trust for real life, typically via belief in a personal God; but many do not. But absent a faith that the events of our life will ultimately be subsumed into some coherent holistic configuration, we lose even the provisional constraints on events’ narrative meanings that we deploy while reading novels and histories. Indeed, unless we assume that some intentional agent has engineered a coherent ending which is already determined albeit inaccessible to us, the unfolding events of our lives can't even have the sort of robust, descriptively thick, and emotionally resonant meaning that narrative promises to provide until our lives have ended. And in turn, if life narratives define selves, then this further entails that stories can only bring selves into being at life's end.

The facts that the narrative meanings of unfolding events are contingent on faith that those events contribute to a satisfying ending, and that their meanings are constrained only by our assumptions about what that ending might be, also makes trouble for narratives’ capacity to guide action. The basic challenge here is, in Kierkegaard's pithy (Reference Kierkegaard1843, p. 306) motto, that ‘life must be understood backwards but lived forwards’. At any given moment, I must act based on my understanding of my current situation. A narrative conception of selfhood promises to assist in this task by individuating and prioritizing a profile of options for action for me, defined in terms of my particular life. But the anticipatory retrospection of narrative comprehension in medias res entails that this profile of options is only defined in terms of its contribution to my life's end. Thus, I can at best guess at what I should do now by guessing at an ending that is determined but as yet unknown. And the more indeterminate that ending is, the more indeterminate what I should do now becomes and the less guidance a life narrative can provide.

This indeterminacy does not undermine Bruner's less ambitious point that stories are templates for action. For Bruner, stories are conventionalized tropes that guide action by assimilating my particular situation to a more generic one, which is itself defined by its contribution to a generic ending. Thus, if I assume my life is a comedy, then it makes sense that this must be the ‘meet-cute’ with my future spouse with whom I will live happily ever after; while if I assume that it is a tragedy, then this must be the fateful error that ultimately unravels my career and leaves me slobbering drunk in a dank hotel. Because Bruner's ‘cultural coins’ don't traffic in the narrative conception's highly individualized selves, the options for action they deliver are also commensurately generic.

So, the retrospective dependence of narrative structure entails that stories can offer specific guidance for action in medias res only insofar as they treat life's ending as itself determinate and specific. Beyond its epistemic and metaphysical implications, this point has an important practical consequence. Precisely in order to individuate and prioritize what we should do now, narratives must foreclose other options for action, defined by other potential endings. In this sense, the very feature that makes stories powerful equipment for living also makes them constricting shackles: they highlight a particular path forward only by concealing alternative paths. Given an assumed life story, it can palpably feel like this is what will happen next, and so that this is what I must do now. If events don't then transpire as anticipated, this raises the stakes for future action: it may seem that this is the only remaining path for me to achieve the ‘right’ ending. And if my life's guiding telos becomes unattainable, I may feel that I am an utter failure – echoing Marlon Brando's Terry Malloy in On the Waterfront: ‘I could have been a contender. I could have been somebody instead of a bum, which is what I am, let's face it.’

The risk is that by blinding us to alternative paths, an assumed life narrative becomes a self-fulfilling prophecy. This dynamic is particularly insidious for those who undergo trauma, like rape or war. Trauma is by definition a profound disruption of the anticipated course of events, in which a person's agency is undermined at least temporarily. This is inherently disorienting. But this acute local disruption often generates a broader narrative disruption, which compounds the initial insult by threatening to undermine the person's entire self-conception. Moreover, the traumatic event's acute local significance also makes it an especially tempting locus for a new narrative, one that imposes the additional penalty of holding the rest of the person's life and self to its intuitive narrative schema, often in the form of self-blame (Brison, Reference Brison2002).

5. Antidotes to Narrative Foreclosure

In §1, I identified two functional roles for stories: configurational comprehension and action guidance. In §2, I identified the basic contours of narrative structure: a complex, holistic configuration of a temporal sequence of events, presented as mattering in virtue of their contribution to an emotionally and epistemically satisfying ending. In §3, I introduced an apparently natural extension of the core idea that stories are ‘equipment for living’: the view that selves are constituted by narratives of their lives. In §4, I argued that given the retrospective dependence of narrative structure, this view ends up holding selves interpretively and practically hostage to their life's endings.

For many people, the retrospective dependence of a life story may be benign or beneficial. Some have an abiding faith in Fate or a personal Author. Some stumble into a critical event that they take to have been highly contingent but that sets them off on a quest they fully embrace. However, we also saw in §3 that many people's lives fail to conform to a familiar narrative template, or to ever come together into any overarching structure defined in terms of a satisfying ending. For them, defining their selves in terms of their lives’ stories can be disorienting or defeating.

The problem is that we do need ‘equipment for living’. We need to construe an overwhelmingly complex world in a way that renders it intuitively coherent and personally meaningful; and we need to act fluently in real time in service of our personal goals and values (Camp, Reference Camp and Grimm2019). If narrative is our only tool for doing this, then the price of assuming a hidden ending engineered by a remote author may not be too steep to pay. Even if it dooms many of us to being erstwhile contenders, at least our lives have some meaning, insofar as there is somebody we might have been and something we should have done.

But must we pay that price? In this section, I suggest that the narrative conception is driven by two plausible but pernicious assumptions: that narrative is our only tool for making meaning; and that selves are defined by their lives. We can achieve a healthy flexibility for constructing selves, and rehabilitate stories as equipment for living, if we abandon these assumptions.