How to Write Antagonists Who Believe They're the Hero

“Every villain is the hero of their own story” is the most repeated piece of antagonist advice in circulation, and it has produced a decade of villains with tragic childhoods who are still, structurally, obstacles.

The problem is that writers hear it as an instruction to add sympathy. It isn’t. It’s an instruction to build an argument.

A story is a proposition. The protagonist embodies it; the antagonist is the strongest available case against it. If that case is weak, the protagonist wins by default and the ending proves nothing. A sad backstory doesn’t fix that - it decorates it.

This article covers how to build an antagonist whose position actually holds, and what changes in your draft when it does. It sits under our character architecture guide, which covers the interior structure this applies to.

What makes an antagonist work?

An antagonist works when their goal is incompatible with the protagonist’s, their reasoning is internally coherent, and they are correct about at least one thing the protagonist is wrong about. The first condition creates conflict. The second creates credibility. The third is what turns a conflict into an argument.

Most drafts satisfy the first, roughly half satisfy the second, and the third is where memorable antagonists separate from functional ones.

Antagonist vs. villain: not the same job

These get used interchangeably and they describe different things.

An antagonist opposes the protagonist. A villain is morally bad. They overlap often enough that the distinction feels academic, until you need one without the other.

Javert is an antagonist and arguably not a villain - he is upholding the law with total integrity, and the tragedy is that his integrity is the problem. Miranda Priestly is an antagonist who is mostly just excellent at a demanding job. Meanwhile a story can contain a genuinely evil character who never obstructs the protagonist at all, and that character is a villain doing no antagonist work.

Why this matters practically: writers who conflate the two default to making the opposition evil, because evil feels like the safe way to guarantee opposition. It isn’t. Opposition comes from incompatible goals, and incompatible goals are available between two decent people.

The self-justification principle

Nobody experiences themselves as the antagonist. That’s not a writing tip - it’s a fact about consciousness that your character has to obey to feel real.

Which means the antagonist needs a version of events in which their behaviour is not merely excusable but correct. Not “I know this is wrong but I have to.” That’s a character who agrees with the protagonist and does it anyway, which is weaker.

The test: write the antagonist’s account of the story’s events in a paragraph, from inside their head, with no irony. If it reads as obviously monstrous, you haven’t found the justification yet. If it reads as uncomfortably reasonable, you have.

The uncomfortable part is the goal. An antagonist whose logic you can dismiss is an antagonist who costs the protagonist nothing to refuse.

Four kinds of self-justification

Most working antagonists run on one of four engines. Knowing which one you’re using tells you what the protagonist has to defeat.

The greater good

The antagonist has done the arithmetic and accepts the cost. Harm now, benefit later, to more people. Thanos is the blunt version; institutional antagonists run on this constantly.

Strength: genuinely hard to argue against on its own terms. Failure mode: the maths is usually wrong in a way the story never examines, so the position collapses under scrutiny the film never applies. If you use this engine, make the protagonist actually engage the arithmetic.

The legitimate grievance

The antagonist was wronged, the wrong was real, and the response is disproportionate. Killmonger is the reference case: the injustice he names is accurate, the history is accurate, and the method is the only thing the story disputes.

Strength: the most destabilising engine, because it makes the protagonist’s victory partial by design. Failure mode: the story concedes the grievance in dialogue and then never lets it change anything.

The pragmatist

Someone has to do the ugly necessary thing, and the antagonist has decided it’s them. Often the most quietly effective engine, because it’s the one most readers recognise from work.

Strength: requires no melodrama and scales down to domestic-sized stories. Failure mode: drifts into competence porn if the story never shows the cost of that pragmatism to the antagonist themselves.

The true believer

The antagonist holds a principle above people, and applies it without exception including to themselves. Javert. The zealot, the purist, the incorruptible.

Strength: unbribable and therefore genuinely frightening. Failure mode: brittle. A true believer needs one moment where the principle is tested against something they love, or they read as a machine.

Build the antagonist with the same architecture

The most reliable fix for a thin antagonist is to give them the treatment the protagonist got.

The wound and the rule. Same construction, same discipline: a formative event, and the false conclusion drawn from it. The difference is only that nobody in the story is helping them revise it.

The want that collides. The antagonist’s external goal must be incompatible with the protagonist’s, not merely inconvenient. Two characters who want the same singular thing, or want mutually exclusive states of the world, produce structural conflict. Two characters who dislike each other produce friction.

The Truth they’re partly right about. This is the piece most drafts omit, and it’s the one that matters.

Find the thing the antagonist understands correctly that the protagonist has not yet accepted. The protagonist should have to absorb it, not defeat it. That’s the difference between an ending where the hero wins and an ending where the story concludes something.

If the antagonist is wrong about everything, the protagonist learns nothing from opposing them, and the arc is decorative.

The contradiction. Antagonists need this as much as protagonists - the tenderness inside the ruthlessness, or the cruelty inside the principled. Without it, self-justification reads as programming.

For documenting all of this across a long project, the character bible template has fields for each element; antagonists drift more than protagonists because they’re on the page less.

Do antagonists arc?

Some do, most don’t, and the choice should be deliberate.

An antagonist with a negative arc entrenches as the story pressures them - this is the most common and the easiest to execute. An antagonist with a flat arc holds their position throughout and forces change on everyone else, which is what makes ideological antagonists so effective. An antagonist with a positive arc is unusual and is generally a redemption structure, which has requirements of its own.

The taxonomy is covered fully in character arcs explained, and Redemption Arcs handles the villain-to-hero case specifically.

The mirror problem

There’s a strong convention that the antagonist should reflect the protagonist - same wound, different resolution. It’s a genuinely powerful structure and it’s also the single most overused antagonist design of the last fifteen years.

Use it when the story is about a choice. If your protagonist could plausibly have become the antagonist, the mirror makes the near-miss visible and the victory costly.

Avoid it when the story is about a system. A mirrored antagonist personalises everything, which is exactly wrong for stories where the real opposition is institutional.

The technique itself - contrast between characters as a structural device - extends well past antagonists, and Character Foils covers it as its own discipline.

When the antagonist isn’t a person

Not every story has a human opponent, and the self-justification principle still applies in modified form.

Institutions. The most common non-human antagonist and the hardest to dramatise, because institutions have no interiority. The solve is to give the institution a face - a functionary who personally believes the institution is right. The self-justification lives in them.

Nature or circumstance. No justification available, so the antagonist’s work has to be done elsewhere - usually by a human character who represents a wrong response to the circumstance.

The protagonist themselves. Internal antagonism, where the false belief is the opponent. This works, but it needs external pressure to make the internal conflict visible, or the story becomes a character sitting still.

Six ways antagonists fail

Evil as motivation. “Wants power” or “wants chaos” is a placeholder, not a want. Ask what the power is for.

The monologue. An antagonist who explains their philosophy at length is an antagonist whose philosophy hasn’t been demonstrated by their behaviour.

Competence collapse. The antagonist is formidable for two acts and makes an uncharacteristic error so the protagonist can win. Readers forgive this less than writers assume.

Sympathy substituted for logic. A sad backstory attached to unchanged behaviour. Sympathy is not the same as a position.

No cost to their own side. An antagonist who sacrifices nothing for their goal doesn’t believe in it.

Too much presence. Antagonists usually work better with less page time than instinct suggests. Their pressure should be felt in scenes they aren’t in.

Antagonists across media

Brief, because the constraints differ more than the principles.

Screen externalises everything - the justification has to arrive through behaviour and subtext, since there’s no interiority available.

Serial fiction and television face the sustain problem: an antagonist must remain credible across dozens of installments without either winning or being defeated. The usual solution is escalation of stakes rather than escalation of villainy.

Games invert it entirely. A player who spends forty hours opposing an antagonist forms a relationship with them, and antagonists designed to be argued with - rather than merely beaten - tend to land hardest.

Keeping the antagonist’s case visible

Antagonists degrade quietly. They’re on the page less than the protagonist, so their logic gets less maintenance, and by the third act the coherent position from chapter four has become whatever the climax needed.

The practical defence is keeping their case somewhere you can see it while drafting - the justification paragraph, the thing they’re right about, the cost they’ve accepted.

Skrib keeps that adjacency by design: the Board is an infinite canvas where the antagonist’s argument can sit as a card beside the protagonist’s, with the collision points mapped between them, while the Draft stays open next to it. Everything exports out to .docx, .md, .fdx or .fountain when you’re done.

Frequently asked questions

What is the difference between an antagonist and a villain? An antagonist is defined structurally - they oppose the protagonist. A villain is defined morally - they’re bad. A character can be one without the other: a decent person with an incompatible goal is an antagonist and not a villain, while an evil character who never obstructs the protagonist is a villain doing no antagonist work.

How do you write a believable villain motivation? Give them a position rather than a desire. “Wants power” is a placeholder; the motivation is what the power is for, and why they believe they’re the right person to hold it. The test is whether you can write their account of the story with no irony and have it read as reasonable.

Should the antagonist be sympathetic? Not necessarily. They need to be coherent - their reasoning should hold from inside their own head. Sympathy is optional and is a separate craft problem; internal logic is not optional.

How much page time should an antagonist get? Usually less than instinct suggests. Antagonists tend to be strongest when their pressure is felt in scenes they aren’t in, and over-exposure reduces the sense that they’re operating on their own initiative rather than waiting for the protagonist.

Can the antagonist be right? They should be, about at least one thing. An antagonist correct about something the protagonist has not yet accepted forces the protagonist to absorb it rather than simply defeat it - which is what makes an ending feel like a conclusion rather than a scoreline.