How Do You Kill Off a Character?
Quick answer: How do you kill off a character? Make the death change the story, earn it with build-up, and write the aftermath. Craft rules, examples and mistakes to avoid.
You kill off a character well by making the death do work: it must change the survivors, redirect the plot, or pay off a meaning the story has been building, ideally all three. The death itself can be quiet or violent, on the page or off it; what cannot be skipped is the earning before it and the aftermath after it. A death with no build-up feels cheap, and a death with no grief feels like a video game respawn.
Getting this right matters because death is the loudest note fiction can play, and readers keep a strict ledger. Spend the note well and they will remember your story for years. Spend it for shock alone and they will distrust every page that follows, because you have taught them that nothing in your world is safe or, worse, that nothing in it matters.
The longer answer: three tests before anyone dies
First, the consequence test: list what changes because this character dies. If the plot and the other characters would proceed roughly the same, the death is decoration, and you should either cut it or promote it. Second, the alternative test: would a maiming, a betrayal, an exile or a failure accomplish the same change? Death is the strongest tool, so use it only where weaker tools fail; a story that kills when it could wound is shouting.
Third, the timing test. Early deaths establish stakes: George R. R. Martin executes Ned Stark near the end of the first book of A Song of Ice and Fire precisely to announce that the story's world does not protect its heroes. Middle deaths reroute plots and harden survivors. Late deaths pay debts and price victories. Know which job your death is doing, because each position asks for different build-up.
One more distinction: on-page or off. An on-page death buys intensity; an off-page death, reported afterwards, buys dread and realism, since life delivers most of its losses by phone call. Louisa May Alcott lets Beth March in Little Women fade across chapters rather than in one scene, and the slowness is the sorrow.
Deaths that work, and why
Beth March again: her death lands because the book has spent hundreds of pages making her the family's quiet centre, so the loss is structural, not sentimental. Charlotte in E. B. White's Charlotte's Web dies off the page, alone at the fairground, and the devastation comes from the aftermath, Wilbur guarding her egg sac, grief converted into duty.
John Steinbeck's Of Mice and Men ends with a death that is also a choice, and the entire short novel exists to make that choice legible: every scene has taught us what the world will do to Lennie if George does not act. That is the earning at its most rigorous. Contrast the deaths readers resent, the beloved figure dropped in a chapter for shock, resurrected two books later, and you can see the ledger at work: consequence honoured versus consequence refunded.
How to write the death scene itself
Underwrite it. The scene is already carrying maximum emotional charge, so plain sentences beat lyrical ones; readers cry at restraint and flinch at violin music. Anchor the moment in one concrete detail, the unfinished cup of tea, the shoe that comes off, rather than in abstract sorrow. Keep the point of view disciplined: grief seen through one set of eyes hits harder than an aerial view of general mourning.
Then write the aftermath, and give it real space. Grief is where the death becomes meaning: the argument at the funeral, the survivor who cannot say the name, the practical horror of the belongings. Skipping from deathbed to next plot point is the single most common way a well-planned death gets wasted. In a short story, where a contest word limit is breathing on your neck, the aftermath can be one paragraph, but it cannot be nothing; judges scoring emotion, as Writory's monthly Short Story Contest rubric does, can tell a felt loss from a plot event in a single read.
Common mistakes
Fridging: killing a character, often a woman, purely to motivate another character, with no life or arc of her own on the page. The fix is not never kill anyone the hero loves; it is make sure the dead had their own story first. Shock-value deaths that trade long-term trust for a single gasp. Fake-out deaths that teach readers your story refunds its losses.
Also beware the sanitised death, where the passing is so gentle and meaningful that it lies about loss, and the over-choreographed one, where a dying character delivers a two-page speech with collapsing lungs. Real deaths interrupt. The most honest thing a death scene can do is leave something unsaid, permanently.
What poets can take from this
- List what the death changes; if the answer is nothing, cut it or promote it.
- Choose death only when a wound, loss or betrayal cannot do the same work.
- Underwrite the scene: one concrete detail beats a page of elegy.
- Give the aftermath real space; grief is where death becomes meaning.
- Never refund a death you have made the reader pay for.
Try it yourself
Write the aftermath of a death without writing the death: begin the scene an hour after, in the point of view of the person who must now tell someone else. Let one household object carry the grief.
Questions
Should a main character die at the end of a story?
Only if the ending's meaning requires it, and the story has priced the death in from early on. A protagonist's death should feel inevitable in hindsight, never like a trick. If the theme resolves equally well with survival, survival usually reads as the braver choice.
How do you make readers cry when a character dies?
Do the work before the scene, not during it: build the relationship, then write the death plainly and anchor it in one specific physical detail. Tears come from accumulated attachment plus restraint, not from the scene's own vocabulary of sorrow.
By the Writory editorial team, reviewed by working poets. Updated July 2026.
Enter a Writory poetry contest
Enter the August contest and have your own poem read and scored.