All notes

    Craft

    Dramatica's four throughlines, explained without the jargon

    Dramatica is the most complete and least approachable story theory ever written. Its central claim — that a finished story is a model of a single mind working through a problem — is worth understanding even if you never touch the software. Here are the four throughlines, and the constraint that makes them do real work.

    Josep16 min read

    Dramatica has a reputation, and it is mostly deserved. It is the most ambitious attempt anyone has made to describe what a complete story is, and it is written in a private vocabulary so dense that most writers who encounter it bounce off within an hour. The software costs more than a year of most writing tools. The forums read like a mathematics seminar.

    It is also, underneath, built on one idea that is genuinely useful and can be stated in a sentence — and on a structural constraint that solves a problem most story theories do not even notice. This post is about those two things. You do not need the software to use either.

    The premise: a story is a mind

    Dramatica's founding claim, from Melanie Anne Phillips and Chris Huntley's Dramatica: A New Theory of Story, is that a complete story is an analogy to a single human mind working through a problem. Not a description of events. Not a character's journey. A model of one mind deliberating.

    They call this the Story Mind. Every character, every subplot, every argument in a finished story represents some facet of how that one mind considers the problem — the impulse to act, the impulse to wait, the conviction that the fault lies outside, the suspicion that it lies within.

    The practical consequence is what matters. If a story is one mind thinking, then the parts have to be different from each other, and they have to be different in a structured way — because a mind that considers the same angle four times has not deliberated at all. That is where the throughlines come from. They are not four plots. They are four positions on one question.

    The four throughlines

    1. The Overall Story

    The objective view: what is happening, to everyone, seen from outside. The war, the investigation, the succession crisis. This is the throughline most writers think of as "the plot", and the one most drafts do competently.

    Its characters are functions rather than people — Protagonist, Antagonist, Reason, Emotion, Sidekick, Skeptic, Guardian, Contagonist. Note that Protagonist is a job, not a viewpoint: it means the character pursuing the story's goal. Dramatica separates that from who we experience the story through, which is the next throughline, and the split is one of the theory's most useful exports.

    2. The Main Character

    The subjective view: the person whose shoes the audience is in. Their personal problem, which is not the same as the story's problem.

    In Dramatica's own published analysis of To Kill a Mockingbird, Atticus is the Protagonist — he pursues the goal, he drives the objective story — while Scout is the Main Character. We are inside Scout. The objective story is a trial; Scout's story is about learning to see people as people. Those are different problems, told at once, and the novel is the standard teaching case for the distinction because collapsing the two roles would flatten it.

    Often they are the same character — Luke Skywalker is both, in Dramatica's analysis of Star Wars. But knowing that they can come apart is the thing worth taking away, because "my protagonist feels like a camera" is usually this distinction going unexamined.

    3. The Influence Character

    The one who forces the Main Character to confront their way of seeing. Not an antagonist — an alternative. They embody the opposite answer to the story's central question, and their presence makes the Main Character's position untenable.

    Obi-Wan is Luke's Influence Character: he does not oppose Luke, he presses a worldview on him. The pressure is philosophical rather than physical. This is the throughline most commonly missing from a draft, and its absence has a recognisable symptom — a protagonist who changes because the plot requires it rather than because anything specific made the old position impossible to hold.

    4. The Relationship Story

    The throughline between the Main and Influence Characters — not either person, but the thing that grows or breaks between them. Dramatica treats this as a full throughline with its own arc, which most theories do not.

    It is the reason a story can resolve its plot and still feel unfinished. If the relationship has not moved, a mind has not finished thinking.

    The constraint that makes this work

    Here is the part that gets lost when the four throughlines are summarised, and it is the part that does real diagnostic work.

    Dramatica sorts every throughline into one of four domains: Universe (an external situation — a war, a sentence, a drought), Physics (an external activity — pursuing, building, investigating), Mind (an internal fixed attitude — a prejudice, a grief, a conviction), and Psychology (an internal way of thinking — a manner of reasoning that must change).

    The rule: each throughline occupies a different domain. All four are used, and no two share.

    That single constraint is the theory's best practical export, because it converts a vague complaint into a locatable fault. If your Overall Story is a situation the characters are trapped in and your Main Character's problem is also fundamentally that situation, you have put two throughlines in Universe. Nothing will feel wrong sentence by sentence. The book will simply feel like it is saying one thing repeatedly, and no amount of rewriting scenes will fix it, because the redundancy is structural.

    Read as a diagnostic, it asks: is my external plot pressuring my character in a way that is genuinely different from how their own head is pressuring them? If not, the story is a mind agreeing with itself.

    Leaving a domain out has a name

    The constraint also runs the other way, and this is the part of Dramatica that most repays sitting with. All four domains have to be used. If your story never occupies one of them — never puts anything in Mind, say, or skips Psychology entirely — then a whole class of objection to your argument has gone unexamined. The mind modelled by the story never considered it.

    Dramatica's term for the complete version is a Grand Argument Story: one that examines the problem from every angle and therefore earns its conclusion. A story that leaves an angle out has not argued, it has asserted. And the theory is blunt about what an assertion dressed as a story amounts to — it is propaganda. Not as an insult, as a category: a one-sided case that forecloses the objection instead of answering it.

    That word is doing real work, so it is worth being precise about the trade it names. Propaganda is not weak. It is frequently more memorable than a balanced argument, and more moving in the moment, exactly because it is not spending any of its energy on the strongest case against itself. Plenty of admired, effective, deeply loved fiction is one-sided in exactly this way, and plenty of it was written that way on purpose.

    The diagnostic is not "you left a domain out, fix it." It is: if you leave one out, notice that you have written something that will land hard and persuade fast, and that a reader who already disagrees with you will feel handled rather than convinced. That is sometimes precisely what you want. Dramatica's contribution is that it makes the choice visible instead of accidental.

    What lies beyond this, and why most people stop here

    The four throughlines are the top of a much larger structure. Beneath the domains sit Types, then Variations, then Elements — four levels of increasing specificity. A complete storyform is the set of choices across all of them, plus appreciations like Story Outcome (Success/Failure), Story Judgment (Good/Bad), Main Character Resolve (Change/Steadfast), Growth (Start/Stop), Approach (Do-er/Be-er), and Story Limit (Timelock/Optionlock).

    These are not independent. Choosing some constrains others, which is what the software exists to compute — the claim being that only certain combinations describe a story that actually holds together, and Dramatica calls a story that does a Grand Argument Story.

    This is where most writers leave, and the reason is not laziness. The vocabulary is unusually unforgiving: terms like "Contagonist" and "Optionlock" carry precise meanings that cannot be guessed, and the payoff arrives only after you have learned enough of them to see the system whole. It is a genuinely steep curve for a genuinely real idea.

    One note for anyone reading older Dramatica material and finding something that reads oddly: the appreciation now called Problem-Solving Style, with values Linear and Holistic, was originally published as "Mental Sex" with values Male and Female. The terminology was changed; the underlying distinction — whether a character resolves problems by isolating a cause or by rebalancing a whole relationship — is what the current terms describe, and it is not a claim about anyone's gender. If you meet the old phrasing in an archived thread, that is what it is referring to.

    Where we think it is genuinely strong, and where it is not

    Strong: as a completeness check on a finished draft. The four-domain constraint catches a specific, common, hard-to-see failure — redundancy between plot and interior — that no amount of line editing addresses. The Main Character / Protagonist split resolves a whole class of confusion about viewpoint. And the insistence that the relationship is its own throughline explains a lot of endings that technically work and feel hollow.

    Weak: as a generator. Dramatica's own framing is that a storyform describes a complete argument, not that filling one in produces a good novel — but the software's structure invites the opposite reading, and writers do get stuck trying to specify their way to a story. It also has close to nothing to say about prose, voice, or scene-level craft, which is most of the actual work.

    And it is contested. Plenty of serious people regard the taxonomy as elaborate beyond what the evidence supports — a very precise map of a territory whose borders were chosen rather than discovered. That criticism is fair, and it does not much affect the two exports above, which stand on their own whether or not the full model does.

    Using the throughlines on a draft tonight

    1. Name the Overall Story problem in one sentence, from outside.
    2. Name the Main Character's personal problem in one sentence. If it is the same sentence, that is your finding.
    3. Identify who forces the Main Character to question their position — not who opposes them, who argues with them by existing. If nobody does, your Influence Character throughline is missing.
    4. Ask what changed between those two people by the end. If nothing, the Relationship throughline never ran.
    5. Sort each of the four into Universe, Physics, Mind or Psychology. Any collision is a redundancy you can now name.

    That is an evening's work on a finished manuscript, and it does not require buying anything.

    If this way of thinking is useful to you, the same distinction runs through Truby's seven steps from a different direction — his opponent-wants-the-same-thing rule is doing similar work to the domain constraint — and the Chubbuck technique covers the scene-level layer Dramatica deliberately ignores.

    We built a Dramatica analyst, and we are not sure who wants it

    It takes a story and works out its storyform — throughlines, domains, the appreciations that follow — and explains its reasoning instead of handing you a verdict. It runs today. It is not open, because we do not yet know whether the handful of people who think in these terms actually want a machine to do this, and building further before finding out is how you end up with a finished product nobody asked for.

    If you would use it, say so. That is the whole signal we are working from, and it will be a paid tool rather than a free one.

    One email when there is something to try, and nothing else. If enough people join, we build it out; if not, we say so and stop. Privacy.