What completion gives a story that continuation never can
Last time, we talked about mystery and mystique — the two contradictory stories the Joker tells about his own scars, a vest, a pair of dice, and the surname Solo spends an entire film quietly unraveling until a myth turns into a receipt. That piece ended on an idea we didn’t sit with long enough: that some things are better left unresolved, that certain silences do more characterizing than any scene could. It closed out six weeks of thinking about character, moving through structure and genre and landing back where it started — at the space around a person that a story deliberately refuses to fill.
But there’s a question this series has circled without ever asking directly, and it’s time to ask it head-on. Why does a story need to end at all?
Put that bluntly, it sounds almost naive — the kind of question a first-year film student asks before learning which questions are worth asking. Of course stories end: films run their length, novels close on a last page, someone eventually stops writing. But that isn’t an answer. It only describes the fact that stories, as objects, have a final page. It doesn’t explain why that page matters, or why a story that never produced one would be lesser, different, or perhaps not a complete story at all.
So here is the claim, stated as plainly as it can be stated, before anything else gets in the way: a story without an ending cannot produce meaning. Not a diminished meaning. None. It can still produce incident, character, feeling, even real pleasure — but not meaning, which turns out to be narrower than we usually treat it, and which depends entirely on a condition only an ending can supply.
That claim runs against how we currently talk about stories. We live in a culture more devoted to narrative than any before it and less willing to let one finish. A film ends, and the question waiting outside the theater is what happens next, not what did that mean. A show gets renewed before its own writers know what its ending would have looked like. A character we watched die comes back because audiences wanted them back, and wanting them back turned out to be reason enough. The working assumption is that a successful story should keep going for as long as people want more of it, and that stopping early is close to a failure of nerve. That sounds generous, sounds like it’s on the audience’s side. But it treats a story the way you’d treat a product whose value rises the longer it sits on the shelf, and that has never been how meaning works inside a narrative. The rest of this article tries to show why.
Start with a distinction the culture has mostly stopped making, because the two things it confuses look, from a distance, identical. A story can stop. Or a story can end. These aren’t the same event, even though both look, from the outside, like the same last page turning, the same last frame holding before it cuts to black.
A story stops when whatever was generating it — attention, momentum, a writer’s patience, a studio’s calculation — runs out. Stopping is imposed on a story from outside itself, the way a conversation stops when someone has to leave the room, whether or not everything that mattered got said. A stopped story can, in principle, always be resumed; that’s the tell. If a story can plausibly pick back up exactly where it left off, it hasn’t ended. It has only paused, and paused stories are, by definition, incomplete — not because anyone failed, but because incompleteness is what a pause structurally is.
An ending is different in kind, not degree. It isn’t imposed from outside; it’s produced by the story itself, out of everything that came before it — the moment a narrative has finally said the thing all of it was building toward, and having said it, has nothing left to withhold. Stopping is an interruption. Ending is a completion. The two can look identical from far enough away, but one is arbitrary and one is necessary — and audiences feel the difference even when they can’t name it, the same way an earlier article in this series noticed that audiences can feel a manufactured ending without ever saying exactly what gave it away.
Aristotle got here first, and said it better than most of what’s been said since. A whole action, he wrote, has a beginning, a middle, and an end — not as three convenient sections of an outline, but because a beginning needs nothing before it to make sense, a middle both follows from what came before and leads into what comes after, and an end follows from everything before it while requiring nothing after it. That last clause holds the whole argument in miniature. An ending isn’t defined by where it sits on the page. It’s defined by the fact that once it arrives, nothing further is dramatically necessary. The action has closed itself.
He adds a second condition, easy to skip past: a whole needs a magnitude the mind can actually hold — not so short it can’t be grasped as a shape, not so long the mind loses the thread between beginning and end. A story that never stops accumulating events eventually loses this quality no matter how good any individual scene is. It stops being a shape the mind can hold, because it was never built to be held that way.
A third condition does the real philosophical work. The events inside a whole action, Aristotle insisted, have to follow one another by necessity or probability — not merely one after another, but one because of another. A king dies, and a queen dies: that’s a sequence, two facts connected by nothing but time. A king dies, and the queen dies of grief at his death: that’s a story, because the second event happens on account of the first. Causality turns “and then” into “and therefore.” An ending is the point at which a story finally proves that everything in it was “and therefore” all along — that the last event was necessary, not merely sequential. A story that refuses to end never has to offer that proof. It can keep generating “and then” indefinitely.
Push this one level deeper than words like completion or shape, because those words can start doing more work than they’ve earned. Meaning is not contained in events. Events simply happen, one after another, in life exactly as much as in fiction — a person wakes, goes to work, has a conversation, goes home. None of that means anything by itself; it’s just occurrence. Meaning lives in the relationship between events — in the fact that this one caused that one, that this one matters only because of what it leads to. And a relationship between events can’t be fully known until the sequence it belongs to is finished. You can suspect a relationship while a story is still unfolding. You cannot confirm one. Until the last event arrives, every relationship inside the story stays provisional, still open to revision by whatever comes next. An ending is the moment those relationships stop being provisional and become fixed. That is why a story without an ending cannot produce meaning. It isn’t that it produces less. It produces events. The relationship among them — where meaning actually lives — never gets to close.
Which is what makes meaning, in a story, necessarily retrospective. You cannot fully understand what a scene means while you’re still inside it; you can only understand it once you know what it was leading toward. A man carefully explaining chemistry to a room of strangers doesn’t mean much on its own. It becomes a different scene entirely once you know he’ll eventually die in the very lab that first lecture was preparing him to build, having gained an empire and lost everything else in the process. Breaking Bad’s finale doesn’t work because it resolves plot mechanics. It works because it’s the moment every previous hour becomes legible at once, retroactively, in a way none of those hours could be legible while they were only happening. The pilot means something different once you’ve seen the finale than it meant the first time through — not a trick of hindsight, but the ending reaching backward and fixing, for the first time, the relationships among everything that came before it.
None of this requires a happy ending, or even a comforting one. Taxi Driver ends with Travis Bickle back behind the wheel of his cab, having changed nothing essential about himself. No Country for Old Men ends with a sheriff describing a dream instead of catching the man he spent the whole film chasing. Neither ending is happy. Both are complete, which was always the only thing an ending owed anyone.
Reflection depends on this kind of completion, because reflection requires distance, and distance requires that the thing being reflected on has stopped moving. Consider the long grief that follows the climax of The Lord of the Rings, which doesn’t arrive the moment the Ring goes into the fire. Tolkien understood that the Ring’s destruction is the story’s climax, not its ending — a climax and an ending are no more the same event than stopping and ending are. What follows is nearly a hundred more pages: the long road home, the discovery that the Shire has been quietly ruined in the heroes’ absence, and finally Frodo’s departure for the Grey Havens, too wounded by what he carried to ever fully live again in the world he helped save. The length was never a flaw sitting on top of the argument. It is the argument. A wound that size can’t be metabolized in the same scene that inflicts it. You cannot know the book was really about the cost of victory, rather than the defeat of a dark lord, until the war is over and there’s nothing left to distract from what happened to Frodo.
Pause on the fact that this is a three-volume work, because at first glance it might look like a counterexample rather than an illustration — proof that length and multiplicity are compatible with an ending after all. They are, but only under a specific condition, and the condition is the whole point. Start with what those three volumes actually are, since “trilogy” gets the history backward. Tolkien wrote one continuous narrative, structured internally into six books rather than three; it was his publisher, Allen & Unwin, who grouped those into three volumes to manage length and cost, and who published them separately, the last more than a year after the first — a decision Tolkien resisted, insisting the book was not a trilogy at all, that the division was, in his words, a “fudge,” and that no volume of it was ever meant to stand alone. Yet Tolkien knew, before he had written the last page, that there would be a last page, and every volume before it was built as a middle in the Aristotelian sense: following from what came before, leading into what had to come after, contributing to a single action aimed at completion since its beginning. A single story cut into volumes after the fact, like this one, meets that condition the same way a genuine series would, provided the whole of it was aimed at a conclusion from the start.
This is also where the real line between a single, complete work and a franchise comes into focus, and it isn’t length. A multi-volume story can produce meaning as completely as a single film, provided the whole of it was built toward an ending from the start — provided every volume functions as an Aristotelian middle. A franchise engineered to keep expanding for as long as it stays profitable is a different kind of object, whatever individual satisfactions it produces along the way. The difference was never short versus long. It’s planned completion versus indefinite extension — and naming that difference isn’t nostalgia, isn’t a claim that stories used to end and somehow don’t anymore. It’s a claim about what happens to a story’s capacity to mean something once the expectation of an ending disappears from how it’s built in the first place.
Indefinite extension is, increasingly, the default. Sequels exist, in most cases, not because the first story structurally required a second one, but because the first one generated enough attention to justify producing more. Extended universes are built explicitly around the premise that no single story inside them needs to close anything, because the universe itself is the product being sold, and any individual film is a chapter that keeps the larger structure open for the next one. A story that knows, while it’s being written, that it won’t be permitted to complete itself is a fundamentally different object than one built toward completion from its first page. It can’t organize itself around an ending it will never reach, so it organizes itself around momentum instead: keep the audience curious, keep the door open, resolve enough of the immediate hour to satisfy without resolving so much that nothing is left to extend.
Notice what’s happened, in the process, to the cliffhanger. It used to be a device applied to a middle — a way of ending an episode without ending the story, because the story itself was still headed somewhere. What’s common now is a story that ends, formally, on a question rather than an answer, and treats the unanswered question itself as the payoff. That inverts Aristotle’s definition almost exactly. An ending, in his account, is the thing that requires nothing further. A story engineered to end on a question is built, deliberately, to require something further. It hasn’t failed to complete itself. It was never trying to.
What a completed story asks of an audience that an endless one cannot is to look back — to hold the whole shape of what was just experienced in the mind at once, the way a shape can only be held once it has stopped moving. What continuous narrative offers instead is consumption: something taken in without the pause that would let it settle into understanding. That isn’t a claim about which is more enjoyable. It’s a claim about which produces meaning, and meaning requires a story to stop somewhere long enough to be turned around and looked at.
There’s a subtler cost too, harder to name than plot or character. Call it a kind of trust between a story and the person experiencing it. A story that ends has made a claim, whether it announces it or not: this is what it was, here is the whole shape, finally, of the thing you spent your attention on. A story that never ends never has to make that claim. It can always defer the question of what it was about to some future installment that may or may not answer it — and in practice frequently doesn’t, because answering it would mean the continuing had to stop.
Most people who consume a great deal of ongoing narrative have felt this without being able to place where it comes from. It isn’t disappointment, exactly, and it isn’t boredom. It’s closer to the sensation of realizing, partway through a long conversation, that it was never actually going anywhere — that you’ve been listening for a while, enjoying the listening, without ever getting the moment where all of it would click into a single shape. That recognition is a kind of loss, worth sitting with as a loss rather than translating immediately into a grievance against anyone in particular. Nobody did this to the audience. It’s simply what happens, structurally, whenever completion is designed out of a narrative from the start.
All of this puts a very specific responsibility back on the person writing the thing, and it is not the responsibility the culture currently assigns to writers, which is usually understood as extending a story’s life for as long as an audience will tolerate it, or a market will support it. The actual responsibility runs almost in the opposite direction: it is the responsibility of recognizing the moment a story has already become whole. Not deciding that it should become whole. Not manufacturing that wholeness by bolting an artificial climax onto material that never earned one. Recognizing it — the way a person in conversation sometimes keeps talking past the point where everything that needed to be said has already been said, purely out of discomfort with the silence that would otherwise follow. A story, unlike a conversation, doesn’t come with an obvious social cue for when it’s finished. The writer has to notice it alone, usually while still convinced there’s more to do.
This is where a writer can genuinely be fooled, and the mistake is worth naming precisely, because it rarely announces itself. Everything written after the real ending has already occurred still looks like story. It has scenes, dialogue, incident, forward motion. Nothing about its surface tells you it’s superfluous. The only way to catch it is to ask, of each new scene, whether it answers a question the story is still actually asking, or whether the question has already been answered and the scene only delays the answer’s arrival. Activity added after the shape has already closed can feel like momentum. Structurally, it’s closer to noise.
Almost anything can continue. That is the trap hiding inside the cultural assumption this article opened by naming. The fact that a story could keep going is never, on its own, evidence that it should. A story can always generate one more complication, one more antagonist, one more season, one more return of a character an audience missed. Possibility is not the same thing as necessity. An ending isn’t the moment a story runs out of possible next events. It’s the moment none of the possible next events are dramatically necessary any longer, because the thing the story existed to say has already been said, completely, with nothing held back for later.
Return, then, to Aristotle, whose idea turns out to hold the entire argument once you’ve walked all the way around it. A whole is a thing with a beginning, a middle, and an end, built to a magnitude the mind can hold, its events connected by necessity rather than sequence — and an end is the part that requires nothing further. That was never a rule about running time. It’s a description of what completion actually is: the point at which a story has organized everything before it into relationships that no longer need anything added to be whole.
This is why the endings that stay with an audience longest are rarely the ones that resolve everything comfortably. They’re the ones that feel, once you arrive at them, both surprising and completely inevitable — surprising because no one could have predicted the exact shape the completion would take, inevitable because once it’s visible, no other ending could have been true to everything that came before it.
There is a version of this argument that could be mistaken for grief, as though the point were to mourn something the culture has lost and isn’t getting back. That isn’t quite right. The point isn’t that stories used to end and now they don’t. The point is narrower, and more useful: an ending was never a formality attached to a story’s conclusion. It was the mechanism by which the story became legible as a whole in the first place. Understanding that doesn’t require indignation. It only requires noticing what completion was doing, all along, underneath the plot mechanics.
Stories lose nothing when they end. They lose almost everything when they don’t, because a story that never ends is a story that never fully becomes what it was reaching to be. It remains permanently in the condition of becoming, generating event after event that never quite organizes into significance. The ending was never the death of the story. It was always the only thing that could make the story fully alive — the single event, out of everything the story ever did, that finally let it mean something instead of merely continuing to happen.