Character Development | Creative Writing Handbook Vol.006

A character profile can contain fifty facts and still fail to produce one believable decision.

EDK-SETC-CWH-0060 · Return to The Creative Writing Handbook

Pillar support: Character Practice in Creative Writing | Train Desire, Contradiction, Pressure and Choice

Favourite colour. Childhood pet. Birthday. Music. Breakfast. Clothing. Star sign. Greatest fear. Secret dream. These details can be useful. They can also create the illusion that a character has been built when the writer has only built a dossier.

Readers meet characters in motion. Someone wants something, misunderstands something, owes somebody, avoids somebody, protects a value, tells a lie, keeps a promise, breaks one, notices one detail and misses another. The character becomes legible through choices made under conditions that matter.

DESIRE → PRESSURE → CHOICE → COST → CONSEQUENCE → NEW STATE

For a broader explanation of character craft, read How Authors Create Characters. This handbook volume focuses on practice: how to practise character until abstract description becomes believable behaviour.

1. Character is not a trait list

Traits describe tendencies. Stories test them.

“Honest” is not yet evidence. Put the character in a situation where the truth will cost them something. “Generous” is not yet evidence. Make the resource scarce. “Brave” is not yet evidence. Make fear reasonable. “Loyal” is not yet evidence. Give loyalty a competing value.

The pressure test asks a simple question: what does this trait make the person do when doing it becomes expensive?

2. Begin with desire

A character who wants nothing is difficult to move. Desire creates direction. It can be grand—escape a country, expose corruption, win an election—or tiny—finish dinner without discussing a particular subject.

Scene-scale desire is especially useful for training because it is visible. What does this person want in the next five minutes? To get a key back. To avoid answering. To persuade a friend to stay. To leave without appearing rude. To hide a mistake until they understand it.

Maren used to write characters with large life goals and no local intention. Her scenes contained biography but little movement. She now writes one sentence above a scene: “Nara wants Joel to leave the kitchen without seeing the envelope.” The sentence gives every action something to relate to.

3. Want and need are different tools

Writers often learn a model in which characters want one thing and need another. It can be useful and should not become compulsory psychology.

A character may want status and need belonging. They may want revenge and need grief to be acknowledged. Or the “need” imagined by another character may be wrong. Some stories concern people refusing the lesson others assign them. Some characters do not change. Some change in the opposite direction.

Use want-versus-need as a diagnostic question: Is the thing the character pursues actually capable of solving the deeper problem driving them? If not, that mismatch may create an arc.

4. Motive answers why this desire has force

Two people can want the same promotion for different reasons. One wants money. One wants recognition from a parent. One wants access to a project. One wants to prevent a colleague from obtaining power. The external goal is identical; the story is not.

Motive matters because it predicts which costs the character will tolerate. Someone driven by money may abandon the goal when the financial benefit disappears. Someone driven by identity may continue even when the job becomes objectively worse.

5. Ask “why now?”

A desire can exist for years without producing a story. Something changes its urgency.

The parent has always wanted to reconcile with a child; now the house is being sold. The researcher has always doubted a result; now the paper is about to be published. The employee has always wanted to leave; now a new colleague depends on them. The urgency converts background preference into present action.

6. Give the desire an obstacle that can respond

An obstacle is stronger when it changes in response to the character. A locked door is an obstacle. A person guarding the door can interpret, negotiate, suspect and adapt. An institutional rule can be applied rigidly or flexibly by a human operator.

Dynamic resistance produces tactics. The character tries one route, meets a response, changes approach and reveals more of themselves.

7. Pressure should reveal hierarchy among values

A person may value honesty, loyalty, safety, autonomy, family, achievement and kindness simultaneously. The interesting moment comes when two values cannot both be preserved.

If honesty costs loyalty, what wins? If safety costs autonomy, what wins? If achievement costs fairness, what wins? The answer does not need to prove a moral lesson. It reveals the character’s hierarchy at this moment.

8. Cost makes choice meaningful

A choice between a good option and an obviously bad option reveals little. Strong pressure creates trade-offs.

Tell the truth and lose a scholarship. Lie and let a friend accept blame. Leave and protect your career. Stay and protect someone else’s safety. The writer should understand why both options remain available long enough to require a decision.

9. Personal stakes beat abstract scale

A world-ending threat can feel emotionally empty if the character has no specific relationship to it. A broken notebook can feel enormous if it contains the last handwriting of someone loved.

Stakes come from value. Ask: what does this outcome mean to this person? What identity, relationship, possibility or memory is attached to it?

10. Contradiction creates behavioural range

Real people are not perfectly consistent trait machines. A generous person can be jealous. A disciplined person can avoid one task. A brave person can fear social humiliation more than physical risk.

Contradiction is convincing when both sides have causes. “Kind but cruel sometimes” is not enough. “Kind in situations where help preserves dignity, harsh when they believe pity creates dependence” gives a coherent internal model that can still surprise.

11. The contradiction matrix

Write four headings: public strength, private weakness, public weakness, private strength. Fill each with behaviour rather than adjectives.

Public strength: can chair a difficult meeting calmly.
Private weakness: avoids opening personal mail.
Public weakness: appears indecisive when asked to improvise.
Private strength: spends hours learning a procedure nobody expects them to understand.

Now create a scene where two quadrants collide.

12. Backstory should alter present behaviour

A detailed childhood is not automatically useful. Backstory earns space when it changes what the character notices, fears, trusts, expects or does now.

A history of scarcity may explain why someone saves damaged objects. A past betrayal may make ambiguous silence feel threatening. Professional training may cause a character to notice a technical inconsistency before an emotional one.

Pressure-test backstory: if you removed this past fact, would any present decision change? If not, it may be decorative.

13. Memory is character evidence and character distortion

Characters remember selectively. They may preserve a humiliating detail and forget the context. Two sincere people can remember the same event differently. A character may reinterpret an old memory after learning new information.

Use memory not only to explain but to reveal how the character constructs selfhood.

14. Belief matters as much as fact

Characters act on their model of the world, not on the author’s complete truth.

A person believes a friend resents them, believes a door is locked, believes the job is secure, believes a parent never read the letter. If the belief is wrong but reasonable from available evidence, the resulting action remains believable.

Iona tracks three columns: fact, character belief, reader belief. This makes misunderstanding causal rather than accidental.

15. Knowledge boundaries create character

What a person does not know can matter as much as what they know. Limited knowledge shapes confidence, fear and strategy.

An expert outside their domain may overgeneralise. A child may understand a relationship but not the institution surrounding it. A manager may know policy and not the practical workaround staff use. Knowledge has topology.

16. Competence changes tension

Incompetence is not the only source of difficulty. Competent characters create rich pressure because they can foresee consequences.

A skilled engineer recognises a failure but lacks authority to stop operations. A trained mediator understands both sides and is personally implicated. A teacher knows exactly why a student is failing but cannot provide the intervention without violating another obligation.

Competence creates responsibility.

17. Limitation should be specific

“Bad at people” is broad. “Interprets indirect requests literally and responds well when expectations are explicit” is more useful. “Clumsy” is broad. “Excellent fine-motor control at a workbench, poor balance on unstable surfaces” is specific.

Specific limitations produce scenes. General labels produce stereotypes.

18. Relationship is a pressure multiplier

The same request changes depending on who asks. “Can you stay?” from a manager, sibling, former partner, child or stranger carries different history and power.

Build characters relationally. Who can disappoint them? Whose approval matters? Who knows an earlier version of them? Who has the right to ask difficult questions? Who misreads them consistently?

19. The relationship ledger

For each important pair, write:

what A wants from B;
what B wants from A;
what A believes B wants;
what B believes A wants;
what neither will say directly;
what past event still changes present behaviour;
what one action could alter the relationship.

Misalignment creates subtext before a line of dialogue is written.

20. Power is not one number

A junior employee has less formal authority and may have more technical knowledge. A child has less legal power and can control whether a parent receives affection. A famous person has social power and may depend completely on an unknown technician in a crisis.

Map power by domain: money, information, status, access, emotional dependence, physical capability, institutional authority, reputation. Scenes become more dynamic when power can shift.

21. Give every major character a local strategy

Desire alone is static. Strategy makes it visible.

To obtain cooperation, one character jokes, another bargains, another provides facts, another withdraws, another appeals to loyalty. When one strategy fails, what do they try next? Tactical change reveals personality under pressure.

22. Tactics should reflect character history

People repeat strategies that worked before. A person raised in an unpredictable environment may overprepare. Someone rewarded for competence may respond to emotional conflict by solving practical problems. Someone whose direct requests were ignored may become indirect.

History becomes alive through tactic, not exposition alone.

23. Pressure forces tactic changes

A good scene often contains several attempts. Ask politely. Joke. Bargain. Threaten. Leave. Return. The order matters because escalation tells us what the character prefers before desperation changes them.

If the character begins with their most extreme tactic, ask what justifies the speed.

24. Voice emerges from repeated social choices

Voice is not merely vocabulary. Does the character answer questions directly? Use humour to evade? Overexplain technical details? Understate emotion? Repeat other people’s words? Correct small inaccuracies?

A character voice becomes believable when language reflects attention, relationship and strategy.

25. Dialogue can reveal what profile pages cannot

Put two characters with incompatible goals in a room. Let neither state the conflict directly. What they avoid, repeat, mishear and redirect can reveal more than biography.

For the wider dialogue mechanism, route to How Authors Write Dialogue. Here the training question is narrower: does this person speak as someone trying to do something to another person?

26. Silence is a character action

Not answering can protect, punish, invite, resist, conceal, buy time or reveal powerlessness. Silence means little without context.

Write the same unanswered question in three relationships. The silence changes because history changes.

27. Body language should be character-specific

Trembling hands and racing hearts cannot carry every emotion for every person. Ask what this character does when attention narrows.

One cleans. One becomes unusually formal. One checks exits. One makes tea. One speaks faster. One becomes very still. Behaviour should arise from person, situation and learned coping, not a generic emotion thesaurus.

28. Objects can reveal character through use

What does a person repair, preserve, discard, lend, label, hide or refuse to replace? Objects become character evidence when behaviour around them has consequence.

Maren gives a character an old calculator. The useful question is not “Why is the calculator symbolic?” It is “What does the character do when someone offers a new one?” Choice earns meaning.

29. Money creates concrete values

Characters often float economically. Give them budgets, debts, obligations and opportunity costs when relevant. The same £50 or $50 means different things to different people.

Money can reveal priority without requiring a speech. What someone spends, saves, hides or refuses can be character action.

30. Time is a character resource

Who arrives early? Who assumes others will wait? Who has five minutes and who has the whole afternoon? Time pressure changes social behaviour.

A character with care responsibilities, multiple jobs or chronic uncertainty may experience time differently from someone with discretionary control. Research life conditions rather than using punctuality as a moral shorthand.

31. Space is a character resource

Who has a room of their own? Who shares? Who controls the door? Who is allowed behind the counter? Who knows the shortcuts? Physical access reveals social structure.

Character and setting meet through affordance: what the environment permits this person to do.

32. Status changes interpretation

The same blunt sentence can read as confidence from a senior person and insubordination from a junior one because readers understand social context. Characters know this too and may adapt speech strategically.

Pressure-test status by moving the character between rooms where their standing changes.

33. A character should have something they are wrong about

Perfect self-knowledge makes development difficult. A character may misjudge another person, misunderstand their own motive, hold an outdated belief, or interpret a strength as a weakness.

Do not make everyone deluded. The error should arise from available evidence and history.

34. A character should have something they are right about

Flawed does not mean wrong about everything. A character can be socially difficult and technically correct. They can misread their family and accurately understand an institution. Competence keeps characters from becoming lesson-delivery machines.

35. Secrets matter because of relationships

A secret is not interesting merely because information is hidden. Ask who has a legitimate claim to know, what disclosure would change and why concealment feels necessary.

A medical secret, romantic secret, financial secret and harmless surprise carry different ethical structures.

36. Fear should change strategy

Fear is character-active when it alters behaviour. A person afraid of public failure rehearses excessively. A person afraid of dependence refuses help. A person afraid of conflict agrees too quickly and resents later.

Do not confuse fear with cowardice. Fear is information about perceived cost.

37. Shame and guilt produce different pressures

Guilt concerns action: “I did something wrong.” Shame concerns self: “Something about me is unacceptable.” They can overlap but often produce different strategies.

A guilty character may confess or repair. A ashamed character may hide, attack the standard or seek proof of worth. Treat emotional distinctions as behaviour generators, not vocabulary decoration.

38. Love is not one motive

Love can protect, control, release, idealise, forgive, demand, sacrifice or avoid. Two characters can both sincerely love someone and choose opposite actions.

Pressure reveals the model of love each person carries.

39. Ambition needs a specific object

“Ambitious” is vague until you know what counts as success. Title? Money? Craft mastery? Public impact? Escape from a background? Approval? Freedom?

The same promotion can satisfy one ambition and violate another.

40. Kindness needs a boundary

A kind character becomes interesting when kindness conflicts with exhaustion, justice, self-respect or another person’s autonomy. Endless kindness without cost feels abstract.

Write a scene where helping would make the helper’s own obligation fail. What do they do?

41. Secondary characters need independent vectors

Supporting characters should not exist only to deliver advice, admiration or obstacles. Give them local desires that would continue even if the protagonist left the room.

A friend helping Maren’s protagonist move house also wants the move finished before an appointment. That small independent goal changes how help is offered and creates friction without villainy.

42. Antagonism is a function, not a personality type

The antagonist is whatever meaningfully opposes the protagonist’s goal. A person can be loving and antagonistic in one scene. An institution can obstruct without malicious intent. Weather can block a journey. The protagonist’s own competing desire can create internal opposition.

When the antagonist is human, give them a coherent positive goal of their own. “Stop the protagonist” is rarely enough.

43. Villains require motive without moral excuse

Understanding why someone acts does not require approving the action. Avoid the false choice between cardboard evil and exoneration.

A harmful character can have recognisable motives, self-justification, competence and tenderness in some domains while remaining responsible for harmful choices.

44. Archetype can be starting geometry

Mentor, rival, trickster, outsider, caregiver and rebel are useful relational functions. They become stereotypes when function replaces personhood.

Ask what the mentor wants besides mentoring, what the rival admires, what the caregiver refuses to provide, what the rebel preserves. Complicate function with independent life.

45. Stereotype audit

Before drafting a character from a group you know poorly, list what information came from direct knowledge, research, media convention and assumption. Then expand sources.

Ask who gets interiority, competence, error, humour, desire and change. If one group exists only as background or moral lesson, revise the distribution of complexity.

46. Research professions through decisions

Do not research only vocabulary and clothing. Ask practitioners what decisions repeat, what novices misunderstand, what errors are expensive, what work is invisible and what trade-offs feel routine.

Profession shapes character when it shapes attention and behaviour.

47. Research culture through internal diversity

No culture produces one personality. Learn differences within communities—generation, region, class, profession, language, religion, migration history, family, urban or rural experience—without treating any axis as destiny.

Specificity is more respectful and more interesting than generic “cultural traits”.

48. Disability is not a single plot function

Disabled characters need desires beyond teaching others a lesson, overcoming themselves or symbolising tragedy. Research lived experience, accessibility, social barriers, adaptations, competence and ordinary life.

The character pressure test should ask what this person wants in this story, not how their body can serve another character’s growth.

49. Age changes resources and social permission

Children, teenagers, adults and older people do not simply have different vocabularies. They have different authority, mobility, dependence, experience, obligations and ways others interpret them.

Research the actual age context. Do not write “a teenager” as one universal type.

50. Character change must be earned through evidence

If a character becomes trusting, what happened that made trust newly rational or newly necessary? If they become selfish, what pressures and rewards reinforced it? If they forgive, what changed in their model of the offence?

An arc is not a label attached to the ending. It is a sequence of decisions that changes what future decisions become possible.

51. Change can be positive, negative, flat or ambiguous

Some characters learn. Some deteriorate. Some remain stable while changing others. Some appear to improve and later reveal the change was strategic. Do not force every story into self-improvement.

The useful test is whether the final state has been prepared by the pressures experienced.

52. The choice ladder

List five important choices from smallest to largest. Does each one reveal or alter something?

chooses not to answer a message;
chooses to attend anyway;
chooses to conceal why;
chooses to protect someone who misunderstands them;
chooses to accept a permanent cost.

The ladder helps build arcs from behaviour rather than summary.

53. Reverse-outline the character

After drafting, ignore the profile. List what the character actually did. What did they choose, avoid, risk, give, take, say and hide?

If the evidence contradicts the planned character, decide whether the draft discovered something better or whether behaviour needs revision. Character sheets are hypotheses. Scenes are data.

54. Track state, not just arc labels

At major points, record five variables: wants, knows, believes, fears, can do. A meaningful event changes at least one.

This prevents abrupt turns. A character cannot make a decision based on information they never acquired. They cannot suddenly fear a consequence that has not entered their model unless the story explains the shift.

55. The one-scene character test

Before writing a biography, write the person returning a damaged item to a shop, asking a neighbour for help, receiving unexpected praise or being corrected publicly.

Small social tasks expose voice, pride, strategy, assumptions and relationship to rules. You may learn more in 800 words than in a six-page questionnaire.

56. The no-backstory test

Write 1,000 words without explaining the past. Make the reader infer enough from behaviour, objects and interaction to understand present pressure.

Then restore only the backstory that changes interpretation. The exercise teaches what history must be told versus what can be embodied.

57. The cost escalation test

Give the character an easy chance to demonstrate the claimed trait, then a moderate one, then an expensive one. Honest when nothing is at stake. Honest when embarrassment is at stake. Honest when someone they love will be harmed.

The sequence reveals where the value breaks or transforms.

58. The opposite-choice test

Take a planned choice and force the character to choose the opposite. What would need to be true for that version to become believable?

This exercise exposes hidden assumptions. Sometimes the “opposite” version produces a more interesting character because it requires a deeper motive.

59. The stranger test

Write the character interacting with someone who has no history with them. Which behaviours remain when family expectations disappear? Does the character become more direct, more polite, more performative?

Identity is partly relational. Different rooms reveal different selves.

60. The old-friend test

Put the character with someone who remembers an earlier version of them. What old nickname, habit, debt or interpretation survives?

The scene can test whether change is real or merely context-dependent.

61. The authority test

Give the character authority. Do they become clearer, kinder, controlling, avoidant, procedural? Then remove authority in another scene. What remains?

Power exposes values because choice expands.

62. The dependence test

Make the character need help from someone they dislike, distrust or have harmed. Which value bends first: pride, suspicion, independence?

Dependence creates strong pressure because it changes the normal direction of power.

63. The misunderstood test

Have another character interpret the protagonist’s action incorrectly. Does the protagonist correct them? Use the misunderstanding? Become offended? Question whether the interpretation is partly true?

This reveals relationship to reputation and self-knowledge.

64. The praise test

Unexpected praise can reveal as much as criticism. Does the character deflect, accept, distrust, compete, become embarrassed, raise the standard?

Pressure is not always negative. Positive events can threaten self-models too.

65. The small-loss test

Take away something minor but personally meaningful: a seat, notebook, routine, tool, saved message. Large trauma is not necessary for revelation.

Small losses show what a person had invested without announcing high drama.

66. The success test

Give the character exactly what they said they wanted. What do they do next?

Success can expose mismatched need, fear of maintenance, guilt, changed identity or the fact that the goal was a proxy. Stories do not require failure to create pressure.

67. The boredom test

What does the character do when nothing demands performance? Scroll, clean, repair, call someone, walk, make lists, seek stimulation, sleep? Boredom reveals default regulation.

Use lightly. Every ordinary habit need not become symbolic.

68. The embarrassment test

Embarrassment sits at the intersection of self and audience. Create a small public mistake. Does the character joke, apologise, blame, disappear, overcorrect, challenge the norm?

The response can reveal social strategy quickly.

69. The unfair-rule test

Put the character under a rule they believe is unfair. Do they comply, appeal, exploit loopholes, organise others, leave, or obey while recording evidence?

The choice reveals relation to institutions, risk and collective action.

70. The scarce-resource test

Give two people one available seat, one dose, one ticket, one hour, one opportunity. Scarcity creates selection. But avoid manufactured cruelty if realistic alternatives exist.

Good pressure comes from the world’s actual constraints, not the author secretly removing obvious solutions.

71. Maren’s character problem: biography replaces action

Maren loves building history. Her character documents are magnificent. Her first scenes begin with explanation because she wants the reader to appreciate the work.

Her rule becomes: every important backstory fact must first be connected to a present decision. If she cannot find the decision, the fact remains private planning.

Her characters become less explained and more visible.

72. Iona’s character problem: coherence becomes predictability

Iona wants every action to be perfectly justified. Her characters become logical enough that readers can predict them.

Her practice adds competing values. A coherent person can surprise when two coherent motives collide. The surprise comes from hierarchy being revealed under new pressure, not from random behaviour.

73. Leonie’s character problem: plot pushes too hard

Leonie is an efficient plotter. Sometimes an event happens because the outline needs it, and the character follows obediently.

Her repair is the refusal test: at each major beat, ask what happens if the character refuses the planned action. If refusal is obviously more believable, the plot must change or the preceding pressure must be rebuilt.

74. A character workshop in 60 minutes

10 minutes: write one desire, one motive, one competing value.
10 minutes: place the character in a low-cost decision.
10 minutes: rewrite with the cost increased.
10 minutes: add another person with an incompatible goal.
10 minutes: remove direct explanation and let behaviour carry the model.
10 minutes: write a new scene in which the same character faces a different pressure.

The final scene tests transfer. The writer should not merely become good at the original setup.

75. A seven-day character practice

Day 1: desire and motive.
Day 2: contradiction under low pressure.
Day 3: relationship ledger.
Day 4: power shift.
Day 5: costly choice.
Day 6: consequence scene.
Day 7: reverse-outline actual behaviour.

76. A thirty-day character build

Week 1 establishes present desire, motive, knowledge and contradictions. Week 2 develops relationships, power and social tactics. Week 3 tests the person under increasing cost, misunderstanding, authority, dependence and success. Week 4 drafts connected scenes and revises the planned character against actual behaviour.

Do not spend thirty days filling profile fields. Every week should produce scenes.

77. Forty character pressure prompts

Tell the truth when a lie would protect a friend.
Accept praise for work partly done by someone else.
Return an object the owner does not want back.
Ask help from a rival.
Refuse help from someone trustworthy.
Choose between punctuality and care.
Receive an apology that arrives too late.
Discover a rule was applied unfairly in your favour.
Be corrected by a younger person.
Be corrected by someone you trained.
Win something you no longer want.
Lose something you pretended not to value.
See evidence that challenges a family story.
Keep a promise that now causes harm.
Break a promise for a defensible reason.
Choose whether to expose a colleague’s mistake.
Discover your enemy was right about one thing.
Discover your mentor was wrong.
Meet someone who remembers you differently.
Be mistaken for someone more important.
Be mistaken for someone less important.
Have authority in a room where you expected none.
Lose authority in a room you normally control.
Need a resource owned by someone you harmed.
Be offered exactly the compromise you demanded.
Learn that a sacrifice was unnecessary.
Learn that a small kindness created a large obligation.
Choose whether to correct a flattering misunderstanding.
Be asked to keep a secret you think should be disclosed.
Find evidence that does not prove what you hoped.
Be given one hour alone in a place you wanted to enter.
Meet a stranger who shares your private habit.
Receive a gift you cannot afford to reciprocate.
Choose who gets the only available seat.
Be asked for advice on the mistake you are currently making.
Have a practical skill become unexpectedly valuable.
Have your strongest skill create a new problem.
Watch someone else make the choice you avoided.
Realise your stated goal has been achieved.
Decide what to do next when nobody is watching.

78. Character and plot are reciprocal

Plot pressures character. Character choices generate plot. If only one direction exists, something is missing.

A flood forces relocation. That is external plot. Which person refuses, who leaves early, who rescues records, who protects pets, who uses the disruption for another purpose—those choices create the human story. The choices then change what happens next.

79. Character and scene are reciprocal

A character model is tested one scene at a time. A scene becomes meaningful because somebody inside it has a specific relationship to its pressure.

The next volume turns this relationship into a scene laboratory: goal, friction, turn, consequence, space, time and exit.

80. The final character operating model

BUILD LESS PROFILE → CREATE MORE PRESSURE → WATCH BEHAVIOUR → TRACK CONSEQUENCE → REVISE FROM EVIDENCE.

Maren stops asking whether she knows enough about the character’s past and asks whether the past changes the next choice. Iona stops demanding perfect predictability and learns that contradiction can remain coherent. Leonie stops forcing characters through plot gates and lets refusal expose weak architecture.

The aim is not to make every character dramatic. It is to make their behaviour intelligible enough to carry surprise. Readers do not need a full psychological file. They need evidence arranged across pressure and time.

Continue the handbook

If you want to know who a character is, do not ask only what they believe about themselves. Give them two things they care about, make it impossible to keep both unchanged, and watch what they protect.

The Creative Writing Handbook Vol No.006 | The Character Pressure Test: Train Desire, Contradiction, Choice and Consequence