The fun of the second act should come from the elements set up by the first act. Conceptual specificity (or Santa Claus vs The Zombies)

Conceptual specificity: Second acts ought to create entertainment using every aspect of the setup in the first act.

I know that’s a long and boring mouthful, so to drive this home I created an exercise called “SANTA CLAUS VS THE ZOMBIES.”

Here’s a weak logline: A man discovers that a rare bug in the rainforest cures cancer. He ends up on the run from the hitmen of a powerful drug cartel.

I hate loglines like this, because it’s all about the first act and it tells me nothing about the second act. Scripts like this tend to spend forty pages setting up some wonky scientific MacGuffin, and then deliver an anemic, setpiece-light second act of generic action/thriller moments. It may be competently written, but it won’t be conceptually specific.

Let’s say there’s an open writing assignment called “SANTA CLAUS VS THE ZOMBIES.” It’s stupid, but it pays well, and to get it, you have to pitch a scene from the second act. The easiest way to do this is to use both parts of the setup – create scenes that borrow from the established lore of both Santa Claus and zombies.

Conceptually specific examples:

  • Zombies attack the elf workshop. Santa and the elves must hold them off with tools and weapons improvised out of toys.
  • Mrs. Claus bakes cookies in the kitchen. A zombie attacks and she has to burn its head off in her oven.
  • Rudolph gets bitten by a zombie. Santa must put him down before he can bite and infect the other reindeer.
  • Santa flies his sleigh to escape zombies. A zombie clings to the runners, trying to climb up to bite Santa.

Here are some bad pitches:

  • Santa Claus talks to his friend Bill. (who the hell is Bill? If you’re setting up Santa Claus, use established parts of his mythology. ‘Santa’s Buddy’ is a movie unto itself).
  • Santa loads the sleigh with toys (this is a first act scene. It happens in the ordinary world, when you’re setting up Santa. The second act happens after the inciting incident, when Santa becomes aware of the zombies).
  • A zombie walks through the snow and bites a scientist who’s researching the poles (this scene could appear in any old zombie movie. A conceptually specific scene would have the zombies vs Santa). *Santa drops the bomb that kills the last of the zombies (leaving aside whether Santa has access to a bomb or not, this is a climactic scene that would probably occur in the second act).


Santa vs. the Zombies is a cheesy, lowbrow idea, but it’s a very clear one. It’s easy for an audience to imagine conceptually specific scenes for it, so the premise feels very fraught. Most beginning writers write their early scripts around very soft ideas that don’t lend themselves well to conceptually specific second act scenes. As such, the second acts are rife with filler, so the work done in the first act feels wasted.

Hopefully you’ll have a better idea than SANTA VS. THE ZOMBIES. Hopefully your idea will allow a reader to envision a second act just as clearly.

Premise Test Examples

Reader question: You talk about your premise test a lot. How would you apply this to a more character driven piece like Five Easy Pieces or Dog Day Afternoon? Or even Reservoir Dogs. How about the Shining or Taxi Driver. What theme do the protagonists of those stories learn?

My premise test:  An <ADJECTIVE> <PROTAGONIST TYPE> must <GOAL> or else <STAKES>. They do this by <DOING> and learns <THEME>.

I have no idea what process the authors of these used; it’s not like they used my premise test. But I can work backwards:

FIVE EASY PIECES: A self destructive piano genius must resolve his issues re his estranged father or else continue his downward spiral. He does this by (events of the movie) and learns that his efforts to normalize or resolve his life are doomed to fail. Hence he continues his drifter lifestyle.

DOG DAY AFTERNOON: A desperate man robs a bank to pay for his lover’s sex change operation, but a hostage situation ensues. He must escape it or else go to jail. He does this by (events of the movie). He fails, and learns that his failings (carelessness and desperation) have doomed him in crime just as they’ve doomed him in all other areas of his life.

RESERVOIR DOGS: A group of criminals (and one undercover cop) pull off a robbery, but have to hide from the heat. They quickly turn on each other and do (events of the movie) in the process leaning that their line of work precludes trust.

THE SHINING: A writer on the edge must resist the seduction of an evil hotel, or be transformed into a family-killing madman. He attempts this by (events of movie) and fails, learning that the human animal is frail, corruptible, and more susceptible to evil than we’d like to admit.

TAXI DRIVER: A border-line psychotic, alienated war vet takes a job driving a taxi in 70’s New York. He must cope with his growing violence and madness, and try to find hope in a world that seems dingy and fallen or else give in to madness and violence/wreak evil. He does this by (events of movie) and eventually learns to channel his madness into something that almost resembles heroism.


I’ve purposely avoided the doing section, because it’s been a while since I’ve seen any of these and I don’t want to dilute my point by getting the details wrong. Note though, the importance of the doing section. Even if you keep all the other options the same, you can radically change any of these stories by substituting in “winning a surfing contest,” “turning into a wereshark,” or “by taking care of six loveable orphans.” You can boil any of these stories down to a basic premise. If you can’t do this on your premise, you might not have enough material yet.

Even your “character driven” movies have plots. Their characters flourish because they have something to do and a world/story to push against. Just because I’m stating premise first doesn’t mean I’m ignoring character. At some point, I’ll continue with the examples and show the character stuff I do before starting.

Some people read the test and think “this is soulless and will yield nothing but cookie cutter ideas.” The point is to know how to bend it to make it work for you. Almost every story has a character (or group of characters), and they’re going to have a goal and do things. This is formulaic only in the sense that sentence are formulaic for having nouns and verbs.

Most loglines are weak because they’re all about the first act setup and give no clue for how the story will be resolved. Example: A time traveler escapes to 2014, but is tracked by a time cop who wants to kill him to prevent a time crisis.

This tells me the first act, but doesn’t hint how the second act will unfold. I want to know how the story is resolved. Do they fight across the time stream, do they end up in a time jail, does it turn into the second act of Looper? Is the story about love, car chases, gun fights, sword fights, or battle by giant robot?. Each of those choices birth a different movie.

So the real question with the doing part of a premise is “Can I see filling a 60 page second act with this idea?”

The Premise Test

An (ADJECTIVE) (ARCHETYPE) must (GOAL) or else (STAKES) by (DOING) and learns (THEME) .

The premise test is simple, useful, and much harder to fill out than you’d think. Its brevity glosses over a lot of complexity and theory. In practice, it’s a powerful and merciless tool that exposes flawed or incomplete thinking.

Adjective: The chief trait of the character. A teen might be sex-crazed (Porkies), an outcast (Ghost World), a type-A overachiever (Election), or a shy nebbish (Perks of Being a Wallflower). (More)

Archetype: How you’d sum up the character. Major Payne is a badass major, as that’s his principal role in his story. Major Dad is a sitcom dad who happens to be a major. (More)

Goal: The rooting interest of the script. This can evolve or change, but it’s nice to have something to follow. Goals can be straightforward, like blowing up a comet (Armageddon) or more abstract, like getting over personal demons (Good Will Hunting).

Stakes: What happens if the character fails? These can be clear-cut, like “all life on Earth dies” (Armageddon) or abstract, “live life alone.” (Good Will Hunting).

Doing: The hard part. A character must make a million or lose his house or the Earth explodes is a setup. You only get a sense of the movie when you establish if he does it by shooting people, shooting a porn, or shooting baskets well enough to make the NBA.

Theme: The world view that’s established by all that doing. If a character grows and succeeds, that’s one kind of theme. If he stagnates and doesn’t, that’s a different theme. Or vice versa. The theme is not necessary what the character learns, more what the audience takes from the story.

Oh, and one more thing, many stories need to preface everything with “In a world where,” like, “In a world where ghosts are traded on the stock market, an <ADJECTIVE> <PROTAGONIST TYPE> must <GOAL> or else <STAKES>. They attempt this by <DOING> and learns <THEME>.”

2014-06-25 18.02.42

Most second acts suck. Here’s a tip on how to fix that.

I read scripts for money. I enjoy it. I like reading, I like teaching, and reading has given me insights that have helped my own craft.

But 90% of the time, I end up writing some variation of this paragraph:

The script starts late – it spends 35 or so pages setting up the whys and wherefores of its complicated setup, and then does nothing with it. The second act only spends two scant setpieces exploring the ostensible main idea, and spends the rest with talky, pro forma scenes that could be swapped into almost any other movie of the genre.

Let’s unpack this:

* First acts should be efficient. We want to sketch out the rules of the world, the main goal, and the main opposition, then we’re off to the races. If you’re still explaining the difference between quantum cyborgs and nexus cyborgs by page 41, your premise is already dead in the water.
* Second acts are usually underpopulated. When I was starting out, I was always proud of my first acts, which felt fun and writerly, but I would sort of bullshit my way through the second act on my way to the big finish. This seems to be a general trend with writers.
* Premise is your friend – if a premise is working, the movie is working. If you do a story about a werewolf cop, the story is clicking whenever his werewolfing is complicated by his policing. If you put in something that’s not related to the core concept, you’ll have to work twice as hard.

Let me be perfectly clear: the second act basically is the movie. If you don’t have 4-8 dynamite sequences that relate to your concept in your second act, you’ve basically written an overstuffed short, something that could be written in 10-20 pages. Shorts are a noble art form and everyone should try writing one, but they won’t make you rich enough to change your life, so let’s move on.

Redundant metaphor: if a script is a sandwich, act one is bread, act three is bread, act two is the meat that the bread contains. You don’t make an anemic sandwich better by adding a third slice of bread.

Before anyone says “yeah, but I write art movies so spare me your commercial hackery,” the idea of premise also applies to art movies. If you spend 25 pages explaining why the Maori woman must go to the sacred rock to atone for her miscarriage, I want the second act to explore Maori stuff and human drama, I don’t want to see a bunch of talky bullshit that could come from any other movie.

Ultimately, a movie is its second act. That’s the money part, that’s where the premise is explored. When someone pitches a comedy with a premise like “Zombie OKCupid,” they’re making an implicit promise that they can find enough funny moments in the second act to justify whatever inane setup that movie would require. If the zombie Okcupid stuff is funny, the comedy is succeeding, if all the jokes come from two human characters, the premise is a wash.

So second acts are important, and they’re mostly made of setpieces. Big moments, movie moments. If it’s an action movie, it’s gotta be action. If it’s a talky drama set in Regency England, the costumes have got to be gorgeous, the talk eloquent, and the drama dramatic. If it’s a zombie OKCupid movie… you get the idea.

A good premise yields 4-8 obvious moments. A good premise is one where even your non-writing mother gets excited and pitches you an idea that could work in your story. A bad premise is one that only yields one or two ideas.

People talk of premises being light, loose, or “soft for development.” What this means it that the premise doesn’t yield enough ideas to populate a second act and that it’s hard to imagine more. So if you want to write a movie about zombies attacking a farm, make sure you can think of enough specific, fun moments to make your story worth telling. If you can’t, you’re better served by picking a different premise, one that’s more fraught with evident possibility.

TLDR: Before you write an idea, make a list of 4-8 sequences that logically flow from your main idea. If you can’t, your premise might be too soft for development.

This can be fixed with the help of a diagnostic logline.

The premise of a movie is like a machine that generates entertaining scenes, setpieces and ideas. These are largely explored in the second act.

Your movie concept combined with the genre of movie creates the means by which entertainment is made. For instance, a time travel comedy would probably have a lot of moments where the existence of time travel led to funny set pieces. An avalanche action movie would probably have a lot of gunplay that somehow involved avalanches.

Generally, the concept of a movie is the implicit promise to the audience. If you went to see the new Godzilla movie, you’d be justifiably disappointed if Godzilla didn’t appear until the last twenty minutes. If you’re selling the promise of a giant monster wreaking havoc, it’s fair for an audience to expect a movie ticket price’s worth of giant monsters wreaking havoc.

A bad movie about, say, a vampire attacking Antarctica, would spend over half the script setting up the base, and then bring the vampire in after midpoint. This is kind of a cheat, and when I see it I feel like the writer is self-conscious about having enough ideas re: the core concept. The fear is understandable, but you shouldn’t have a problem coming up with 4-8 fun ideas off a high concept, if you have that much trouble, the topic might be too soft for development.

Scripts tend to work better when they explore one idea to the hilt, rather than explore two or more ideas in a more shallow way. So when you have a concept you’re proud of, set up the first act quickly, and then milk as much entertainment value out of the premise in the second act, because you won’t have much time for play in the third act, which is largely pro forma.