Back to the table

41 Beginner Su Supporting Cast Character

Supporting Cast

The friends, family, and rivals around the lead.

What it is

Your supporting cast is every named, living character in your book apart from the protagonist and the antagonist. The sister, the desk sergeant, the landlady, the rival at work and the friend who tells the lead the hard truth all belong to it. In most books these people make up the bulk of the character list.

They also tend to share one flaw. They come out cardboard. A writer under pressure lets each of them shrink to a function: the one who brings the news, the one who drives the car. Readers notice within a page, even when they couldn't tell you what's wrong.

One habit fixes most of it. Give each of them a want of their own. Keep it tiny, keep it separate from the plot, and show it through what they do. A character with a want of their own reads as a person. A character who wants only what the scene needs reads as a prop. For a walk-on, a single behavior is enough. Here's a quick test: if the protagonist had never turned up, would this character still have something to get on with?

Supporting characters don't all stand at the same height. Some appear once, and some carry a big share of the book. Picture a ladder with four rungs instead of one big group.

Climb past the top rung and you have a second lead. Building one is a different problem, closer to building your protagonist.

Most of the craft lies in putting each character on the right rung, because the rung tells you how much to spend on them. Writers get it wrong in both directions. One gives the ticket seller a heavy backstory. Another gives the best friend of fifteen years the same two lines the ticket seller deserved.

The rule for this element is short. Each supporting character needs a goal of their own, however small, so they feel like people and not props.

Start with the prop. A prop is defined by what the protagonist needs. It hands over a key or mentions a memo, and when the scene ends it stops existing. Fill a book with props and the world around the lead goes thin. The scenes start to feel like a stage set of painted flats with nothing behind them. Readers sense it without knowing why.

A person is defined by a motive of their own. The landlady rushes through handing over the key because her program starts in four minutes. The colleague brings up the memo to be the first one in the know. The plot hasn't changed, and the scene is no longer than before. What you've added is a second want in the room. Two wants in one room make the smallest working unit of conflict.

The word "small" in that rule covers three points.

First, the goal doesn't need any tie to the plot. A night clerk racing to finish his crossword before his shift ends can carry three-quarters of a page. He doesn't need a dead wife or a secret. He's just slow to look up.

Second, you show the goal through behavior instead of backstory. He turns the newspaper face-down when the detective walks in and picks it up again when she leaves. That costs you one clause.

Third, a goal that grates against the protagonist's goal beats a neutral one. A clerk who is eager to help gives you texture. A clerk who wants to get out of filling in an incident form gives you friction, and friction drives scenes. Think of it as scene goal logic, applied to the characters who aren't driving the scene.

Characters on the top rungs need something bigger: a lasting aim they chase all through the book while the lead's project blocks it. An aim like that is the usual source of a subplot. Writers who develop their cast this way tend to find that the middle of the book fills up on its own.

Supporting characters also do work for the book, so name each one's job. Seven jobs come up again and again.

An ally helps the lead, at some cost to themselves.

An obstacle blocks the lead without being the antagonist. A bureaucrat can do it, or a parent who disapproves, or a friend who won't lend the car.

A mirror shows the reader something about the lead just by standing close. When contrast is the main point, the mirror becomes a foil.

Comic relief has a real job, and the job is pacing. A laugh lets the reader breathe out between pressures, and that makes the next bad thing land harder.

A truth-teller can say what nobody else dares, because closeness or low rank gives them license. The model is the Fool in Shakespeare's King Lear, first printed in 1608. Fooling is his trade, so he can tell the king he was wrong to give away his power.

A character who needs protecting puts a human face on the lead's stakes, which would otherwise stay abstract.

A character who knows things is the main way exposition, the background the reader needs, gets into a novel without a lecture.

A job is the service a character does for the plot machinery. A type is a pattern readers already carry around, like the old and wise teacher, the loyal best friend, or the trickster. A type of that kind is an archetype, which has its own element. Two roles grew big enough to become elements as well: the mentor and the love interest. Don't pick a type off the shelf. Work out what the book needs first, then ask who in this particular story would most likely meet that need.

One character can hold two jobs, and the best ones do. An ally who helps the lead with one thing and blocks her on another is worth three flat characters. Comic relief who doubles as the truth-teller is an old and sturdy pairing.

A character with no job at all, a pleasant name who walks on and walks off, should be merged with someone or cut.

How much inner life can a supporting character carry before they start competing with the lead? That question is what separates a cast from a pile of rival leads, and it has a practical answer.

Interiority is how much of a character's inner life the reader gets to see. You can count what goes into it: thoughts given directly, private history, change the reader watches happen, and time alone on stage.

The biggest factor is whether the narration ever goes inside their head. You make that call through point of view more than through the character.

A good working limit is to give a supporting character as much inner life as you can show from the outside. Use behavior, dialogue, objects, and one choice that surprises the people who know them. The reader rebuilds the private world from those clues. The character can then be as deep as you like and still pose no threat to the lead, since the reader is still standing where the lead stands.

Go inside their head as often as you go inside the protagonist's, and the whole book tilts, whether you meant it to or not. Telling a story through several characters' eyes, called multi-POV, commits you to a structure. You can't treat it as a convenience. A viewpoint chapter promotes a character, and from then on the reader expects that person's arc to be closed by the end.

The standard account of how much weight to give a character is Orson Scott Card's Characters & Viewpoint, from Writer's Digest Books in 1988. The book is still in print in the Elements of Fiction Writing series. Card ranks characters as walk-ons, minor characters and major characters. His guiding idea is that a character's prominence is an allowance you spend on purpose, and he gets specific about which techniques add weight and which take it away. When you suspect you're overspending on someone, read the chapter that covers it. His book is listed below, along with Linda Seger's Creating Unforgettable Characters, whose sixth chapter deals with adding secondary and minor figures on purpose, and Creating Characters from Writer's Digest Books, a collection by working novelists that gives its whole sixth part to secondary figures.

Three signs tell you you're overspending. The character gets more inner thought than the lead over the same stretch of pages. You privately find their side want livelier than the lead's. Your beta readers mention them first when they sum up the book. The last sign isn't always a fault, as the Mercutio example below shows, but it always tells you something.

For your top two or three supporting characters, a safe allowance looks like this. Give each one a single private want, spoken or hinted at once. Give each one a scene where they refuse the protagonist something. Let each one pay one price the story never asked them to pay. That gives them enough inner life for a reader to grieve if they die, and too little to take the book over.

Why it matters

Readers fall for the cast as much as the lead. Ask readers why they love a long series and they'll name the village, the crew, the squad or the family before they get to the lead. When people recommend a book they say "you'd love her sister", not "the detective is interesting". A cast of props gives readers nothing to love. Affection for the people around the lead is what turns a finished book into the sale of the next one.

A cast with wants fills the middle. Picture six named characters, each chasing something of their own. The middle of the book fills up with people acting on private motives, and private motives are what a saggy middle lacks. If you find yourself pushing uphill somewhere past the forty percent mark, look at your cast. Often you'll find a cast of props.

The cast tests what your book is saying. If only the protagonist and the antagonist carry the theme, the book is a debate between two voices. Spread different answers across ordinary people and the book starts to weigh them. Give the lead a neighbor who made the compromise she refuses, and a friend who paid the price she dodges.

The cast decides how long your world lasts. In a series, the recurring ensemble is one of the biggest reasons readers come back, and often the biggest. The lead stays steady by design, so change has to come from the people around them. Treat the cast as an asset with a balance sheet. You can promote people, marry them off, kill them, hand them their own book, or retire them. Plan that on purpose, in your character bible and your series arc, and a series can renew itself at book six. Leave it to luck and you get repetition.

Every name costs the reader attention. Each named character is something the reader has to remember. Readers pay that cost happily for characters who are doing something, and resent it for characters who sit idle. The size of your cast is a budget decision, never a matter of taste.

Real-world examples

The Fellowship of the Ring and the 2001 film: how to trim a crowded cast

No number of characters is the right number. There are reliable signs that you have too many, though. Readers flip back to check who someone is, or they mix two characters up. You catch yourself reintroducing someone in chapter nineteen, which is really a note to yourself dressed up as prose. Two characters have never been in a scene together and never disagree. The deciding test is whether you can say what a character's job is but can't say what they want.

A big cast is fine in itself. Epic fantasy and the sprawling Victorian novel are full of them. Each member is distinct, and each is chasing something. Muddle comes from characters who are too alike and too idle. Headcount has nothing to do with it.

The standard fix is to merge two characters who each do half a job into one who does a full job. Say a sister supplies family history and a colleague supplies pressure at the office. Make them one cousin who works in the same building. You lose nothing you were using, and you gain three things. She gets twice the page time, so the reader remembers her. Two demands pull on her, which hands you inner tension for free. The manuscript gets shorter without losing a scene.

You can see a merge done in public in a well-known adaptation. In J.R.R. Tolkien's The Fellowship of the Ring, the elf-lord Glorfindel meets Frodo on the road, puts him on his own horse and sends him across the Ford of Bruinen. Peter Jackson's 2001 film has no Glorfindel. Arwen takes over the job and rides with Frodo to Rivendell.

One character took on another's duty. The film gained someone the audience had reason to care about later, and it avoided adding a new name to a crowded story. The plot didn't change.

One warning. Never merge two characters whose clash the book depends on. If your lead needs one person pushing and another holding back, that takes two people. Fuse them and the pressure disappears.

Telling a cast apart on the page

Making characters easy to tell apart and making them feel real are two different skills. At the level of a single line, the test is to cover the dialogue tags and see whether you can still tell who's speaking. That test lives under character voice. At the level of the whole cast, the question is bigger. Can a reader keep five or ten people separate across two hundred pages and long absences?

Four tools help, from strongest to weakest.

The strongest is what they want. An agenda separates characters better than a trait does. Two characters who want opposite things stay distinct even if they look alike on the surface.

Next comes how they talk: formal or casual, long sentences or short, the words they use and the subjects they avoid. A cast needs range across its voices. Accents don't matter much. The details are in character voice.

Third is their place in the group. Readers hold a cast as a shape: who owes whom, who fears whom, who defers to whom. Sketch that web for yourself. The relationship web covers it in depth.

Weakest of all are physical or habitual markers, like a limp, the cigarettes, the endless cups of tea, the yellow coat. Beginners reach for these first. Used alone, they give you a cast of tics. Laid on top of the first three tools, they help a reader find someone again after eighty pages offstage.

Two small tricks help too. When you name characters, avoid a shared first letter, a shared rhythm, or a rhyme. Readers find their way by the outline of a name more than its spelling, so Bram, Bran and Brandt in one cast melt into a blur. And give every recurring character one strong entrance that shows them going after something. A reader who sees that once will keep hold of them for the rest of the book.

Little Women (1868 and 1869): a household where everyone wants something

Louisa May Alcott's Little Women came out from Roberts Brothers of Boston in two parts, the first on September 30, 1868 and the second in April 1869. Her publisher, Thomas Niles, had asked for a book aimed at girls. Alcott wasn't keen. She said she knew little about girls apart from her own sisters. Even so, the book became the best-known demonstration in English of the goal rule.

In chapter thirteen of part one, "Castles in the Air", the young people lie on a hillside and each one names an ambition. Meg wants a lovely house full of pleasant things. Jo wants horses, a room full of books, and to write something that will last after she's gone. Amy wants to go to Rome and be the best artist in the world. Beth wants everyone to stay well and together, and wants nothing else for herself. Laurie, the neighbor, wants to go to Germany and be a musician, and his grandfather means to stop him. The free text is in the resources below.

Three lessons come out of that one scene.

The ambitions differ in size on purpose. Beth's is the smallest, and later it hurts the most, because a person who wants nothing for herself can still be thwarted. The uneven sizes make the group feel like a family instead of four rivals.

Every ambition pulls against another one. Amy and Jo both hunger for attention, and there isn't enough to feed both. When Amy burns Jo's manuscript, in the book's best-known episode, her spite has a cause: two ambitions have collided. When wants overlap, plot happens without the author arranging it.

Laurie's want runs into an obstacle, his grandfather, and that gives him a story of his own beyond being the boy next door. Later in the same chapter he decides to give up his castle and stay with the old man who needs him. A supporting character making a choice that costs him personally is the last item in the allowance described above. That choice goes a long way toward explaining why Laurie lasts so well in a novel about someone else.

The Long Way to a Small, Angry Planet (2014): a whole book run on private wants

Becky Chambers raised money on Kickstarter in 2012, asking for a few thousand dollars so she could work part-time and finish the novel. She self-published it in 2014. Hodder & Stoughton brought out an edition in 2015, and Harper Voyager published it in the United States. It was shortlisted for the Kitschies Golden Tentacle, the best debut award, the first self-published book on that list. It made the 2016 shortlist for the Arthur C. Clarke Award. In 2019 her Wayfarers series won the Hugo Award for Best Series. If you publish independently, that path should interest you.

The book comes close to a controlled experiment in supporting cast. It follows the crew of a tunneling ship on a long contract. The plot is gentle by design, and there is no villain. The ensemble carries the whole book. Every crew member wants something private, and almost none of it has to do with the contract.

Sissix, the reptilian pilot, longs for the kind of family her species builds, which she can't have on a mostly human ship. Jenks, one of the two engineers, is in love with the ship's artificial intelligence and wants to get her a body, which is against the law. Ohan, the navigator, carries a symbiotic infection that will kill him, and refuses a cure because the infection gives him the way he perceives the world. Corbin, the algaeist, wants to be left alone with his tanks, and the crew resent him for it. Rosemary, the new clerk whose arrival starts the book, took the job to get away from her family name.

Private wants with no link to the plot still make plot. Nobody's agenda is about tunneling, yet every major event grows out of one of them.

The book also shows what weight costs. When many characters have real inner lives, the whole ensemble sits near the top of the ladder. Chambers did that on purpose, and she paid for it with less forward drive. Readers who want plot-driven space opera complain. A thriller writer shouldn't copy that balance. The technique underneath, one private want for every named crew member, works in any genre and costs little.

Spoiler: near the end, another crew member forces Ohan to take the cure over his refusal. The scene is one of the hardest in the book, because the reader has spent three hundred pages learning what he wanted and why. Goals are what give a scene its sting.

Mercutio in Romeo and Juliet (1597): the friend who had to die

Romeo and Juliet, first printed in 1597, holds the best-known scene-stealer in drama. It also prompted the earliest surviving craft comment on the problem.

Mercutio is the hero's friend. He has no part in the feud, no part in the romance, and no job in the plot machinery. He is livelier, quicker and better company than either of the lovers. He dies in act three, scene one, wounded under Romeo's arm, and the tone of the play changes for good. The scene is free to read in the resources below.

In 1672 John Dryden wrote the "Defence of the Epilogue", added to his play The Conquest of Granada. In it he reported that Shakespeare had admitted he was forced to kill Mercutio in the third act, or Mercutio would have killed him. Dryden wrote that fifty-six years after Shakespeare's death in 1616, so his report is secondhand, an anecdote that has never counted as a record. Nobody knows whether Shakespeare really said it. The idea has lasted ever since 1672 anyway, because working writers recognize it.

The root of the problem is that supporting characters are more fun to write, because they carry less weight. They're allowed to be extreme, comic, cruel or absurd. The book's central question doesn't hang on them, and they don't have to change. The lead has to make choices and pay for them, which is a heavier and narrower job.

So the fun character grows. Let them keep growing and readers start reading for them instead of the lead. From the reader's side the symptom sounds like this: the book sagged after chapter twenty, the chapter where the best friend left.

You have exactly three choices. You can remove the character, which is what the play does; it's drastic and it works very well. You can contain the character by capping their page time and inner life, so they stay a delight without becoming the center. Or you can promote them, which is the next example.

Jeeves (1915 to 1974): the walk-on who took over

Jeeves first appeared in P. G. Wodehouse's "Extricating Young Gussie", printed in The Saturday Evening Post on September 18, 1915, and then in The Strand Magazine the next January. His part was tiny. He comes in, announces a caller, asks which suit his employer will wear, and leaves. His whole function is to show in Bertie's Aunt Agatha. In his preface to The World of Jeeves in 1967, Wodehouse looked back on that debut. He recalled that Jeeves never pushed himself forward, and that by his own count Jeeves had just two lines that first time.

Then came thirty-five short stories and eleven novels, the last finished one being Aunts Aren't Gentlemen in 1974. Across nearly sixty years, from 1915 to 1974, the valet's name moved onto the covers, and the nominal lead became the man Jeeves manages.

The promotion didn't take more page time. What it took was a want. Wodehouse worked out what Jeeves wants, which is order in a specific and slightly sinister sense: the right clothes, suitable engagements, no unsuitable fiancées. Then he made that want the engine of the stories. The walk-on became a person, and the person became what drove the stories.

When a supporting character starts hijacking your book, how much you love them doesn't matter. Ask two questions. Could their want carry a book of its own? Could this book survive their promotion?

If both answers are yes, promote them on a plan. Give them a subplot now, a viewpoint if you can afford the cost, and their own book later. If the answer to the second question is no, contain them in this book and write them into your character bible as a future lead. What fails is promotion by accident, creeping in across the middle third, which is what happens in most drafts.

Julia Quinn's Bridgerton novels (2000 to 2006): a cast laid out like a release calendar

Julia Quinn's series is the textbook case of a cast planned like a production schedule. It began with The Duke and I, Daphne's book, from Avon in 2000. The Bridgertons are eight siblings whose first names run through the letters A to H, and there are eight main novels, one for each. In order of publication: The Duke and I (2000) for Daphne, The Viscount Who Loved Me (2000) for Anthony, An Offer from a Gentleman (2001) for Benedict, Romancing Mister Bridgerton (2002) for Colin, To Sir Phillip, With Love (2003) for Eloise, When He Was Wicked (2004) for Francesca, It's In His Kiss (2005) for Hyacinth, and On the Way to the Wedding (2006) for Gregory. The Bridgertons: Happily Ever After followed in 2013, a collection of second epilogues.

The technique works well beyond Regency romance. Three points carry over.

A set of siblings, or a squad, gives you a series structure. The supporting characters in each book are the leads of other books. One family squabble entertains the reader and advertises the next volume at the same time. A reader who has grown fond of Colin over three books arrives at his book already invested. Read-through, a reader going on to the next book, comes from the characters, with no cliffhanger needed to buy it.

Supporting characters have to be able to survive promotion. A character whose only job is to tease the heroine can't carry three hundred pages. Plant a real want early, even years ahead if you have to. Penelope Featherington is the best case. She hovers offstage through books one to three, and she works as the heroine of the fourth book because of the silences readers noticed in those first three. Spoiler: the identity of whoever writes the town's gossip sheet turns out to be the pivot of book four.

An ensemble needs a shared device that keeps its shape clear. Lady Whistledown's gossip column runs through the whole series. It reminds readers who everyone is, where they stand, and where the family ranks among the others. Any genre has its own version: a running letter, a ship's log, a group chat, a newspaper, a case file. Any of them beats reintroducing eight people in chapter one.

Place can do the job that family does here. Louise Penny's village of Three Pines and Richard Osman's retirement village hold their recurring ensembles together through geography. What all of these share is that returning readers come back for the cast, and not only for the lead.

An everyday scenario: seventeen names, two that mattered

Here's a made-up case. Say you've finished book two of your crime series. The manuscript runs 94,000 words, and the prose is clean. Your three beta readers all say the middle felt crowded and they lost track of who was who. One asks which one Dov was.

You count every named character and get seventeen. Next to each, you write a job and a personal want. Every job gets filled in. The want column has nine blanks.

Three of the blanks are walk-ons: the pathologist, the desk sergeant and the barman. They're fine as they are, since a walk-on needs a behavior and not a want. You give each one a clause. The pathologist is hoping to get to her daughter's concert. The sergeant is hoping his shift ends. The barman is hoping to clear the detective out before the football starts.

That leaves six real problems, and you deal with them three ways.

You merge two. Dov supplies the neighborhood history. Lorne drives the detective around and knows everyone. They have never shared a scene or clashed. You fold them into one Dov, a minicab driver with thirty years on the estate. His new want is to keep his brother's name off the file. Now the detective has a source who lies to her, and two scenes exist that weren't there before.

You cut two: the second cousin of a character called Ines, and a colleague named Sunniva. Both were there only to hand over facts in chapter eleven. You give those facts to someone already in the scene, who now has a reason to hold them back. The chapter gets better.

You fund two. Both sit high on the ladder, and both have been coasting. Pilar, the detective's partner, has a promotion board in March, and this case is her file. Now she and the lead want different things from the same evidence. Hedda, the detective's mother, craves being kept informed, and when she isn't, she goes digging on her own.

The plot hasn't changed. The cast drops from seventeen to fourteen, and the middle now has three people pushing in different directions. In the second round, your beta readers ask no questions about who anyone is.

Try it yourself

Call this a cast audit. Plan on an hour or two for a finished draft, or about twenty minutes for an outline. You need a spreadsheet, or columns ruled by hand.

  1. List every named character. Include the ones who show up only twice. The list is usually longer than you expect. Note the chapter where each one first appears. Give each character one row, with these six columns:

Name | Rung (walk-on / functionary / minor / major) | Job in four words | What they want for themselves | Scene where that want makes them unhelpful | Reusable in later books?

  1. Give each one a job. Use four words at most. The labels from earlier work well: ally, obstacle, mirror, comic relief, truth-teller, needs protecting, knows things. Your own wording is fine too. Flag anyone with no job to merge or cut, and save the argument for later.
  2. Write down what each one wants. Keep it to a short phrase. Judge it by what they'd be doing if the protagonist had never shown up, not by what they want from her. If you don't know, leave it blank. The blanks are your result.
  3. Fill the blanks by rung. Walk-ons get one behavior each, like the newspaper turned face-down, the coat already on, or the glance at a phone. Minor and major characters get a real want, plus at least one scene where that want makes them unhelpful.
  4. Merge the halves. Look for pairs who never share a scene, never clash, and each hold half a job. Merge them. Then check that you haven't fused a pair the book needs for their clash.
  5. Say the names out loud. Flag any pair that shares a first letter, a rhythm or a rhyme, and rename one of each. Do this once, after the merging.
  6. Read your most crowded group scene with the names covered. Anyone you can't recognize is missing a want, a register or a clear place in the group. Register is a matter of character voice. Place is a matter of the relationship web.
  7. Check the inner life. Take your three biggest supporting characters and count the passages of direct thought each one gets. Compare that with the protagonist over the same chapters. If anyone comes close, make it a deliberate point of view decision: contain them, or promote them and plan it.
  8. Find your scene-stealer. Name the character you most enjoy writing. Ask whether readers would rather be with them than with the lead, and whether their want is strong enough to carry a book. Then choose: remove, contain or promote.
  9. Write the series note. If you're writing a series, list which supporting characters can come back, which are used up and which could lead a book. File the list in your character bible. Don't leave it to memory.

Common mistakes and misunderstandings

Building characters from their function. The Helpful Librarian, the Gruff Sergeant, the Best Friend Who Listens. The prose can be perfectly competent, but these people were built from the plot outward. Fix each one with a single question: what is this person going after in the scene that has nothing to do with the protagonist?

Mistaking a quirk for a want. A man who collects owls, says "right you are" and hates rain has a label, nothing more. Tics help a reader find someone again, but they can't stand in for an agenda. Here's the check: does the quirk ever push him against what the scene needs?

Giving the lead a fan club. When everyone in the cast is an ally who agrees with the lead, you have a fan club. At least one ally should be badly wrong about something that matters. At least one should want an outcome the lead can't give them.

Blaming a crowded book on too much plot. Writers start cutting scenes. Count the names first. Often the extra weight is four idle people, not four scenes you don't need. Merging characters is cheaper than restructuring the book.

Keeping the wrong character after a merge. Keep the one with the stronger want. Don't choose by the nicer name or the longer file of notes.

Letting a character promote themselves. The pattern runs like this. A monologue creeps in, then a viewpoint chapter, then a subplot that drifts loose from the main line. In the end the reader can't tell whose book it is, and the ending has to settle two stories with room for one. Promotion is fine when you plan it.

Giving interiority out of habit. A thought or two from a supporting character early on, then nothing, is worse than none at all. You've set up an expectation and then taken it away. Decide who gets a viewpoint and stick to it. Both point of view and multi-POV go into that choice.

Spending a character too early in a series. Kill off the best friend in book one for an emotional beat, and you've lost that asset for books two to eight. Make sure the beat is worth it.

Using the cast as an information desk. A character who shows up only when the reader needs a briefing reads as a device, however well written. Give them a reason not to want to tell. The same scene turns dramatic.

Thinking an ensemble is easier than a lead. Spreading attention across six equal characters multiplies the work, since each one needs a want, a voice and an ending. An ensemble is a perfectly good form, and the expensive one.

Forgetting the cast is why readers come back. In series work above all, the cast is a big part of what you're selling. Give it the care you give the protagonist and the fourth book gets easier to write.

Related elements