Writing Tips Archives

How Can I Write a Joke with AI?

Posted on August 20, 2024 by Joe Toplyn

You need jokes.

Maybe you’re giving a speech
and want a few jokes to help make your points stick.

Maybe you’re writing for social media,
where jokes can grab an audience’s attention.

Maybe you’re a standup comic
looking for some fresh material.

So how do you get jokes?

You could write them yourself. But that’s hard.

You could study how to write jokes, then try again. But that would take a lot of time.

You could pay somebody to write jokes for you. But that could be expensive.

Instead…

Why not generate jokes with AI?

You’ve probably read about all the amazing things a large language model (LLM) like ChatGPT can do. Why couldn’t you get ChatGPT to write jokes for you, for free?

You could. But they wouldn’t be good jokes. If you’ve ever tried using ChatGPT as a comedy writer, you’ve undoubtedly found that it isn’t very funny. As this article in Popular Science says, “Despite their immense capabilities, AI can’t tell a joke.”

Why can’t AI write jokes?

ChatGPT and other AI language models can’t write good jokes because they can only mimic the surface patterns of jokes that they’ve been trained on, jokes that people have already written. That means AI language models can’t write jokes that are actually surprising.

And because a joke has to be surprising to be funny, jokes written by AI alone aren’t funny. As that Popular Science article says about ChatGPT, “…its comedic repertoire was likely learned and memorized during its training phases, and not generated on the fly.”

A green circuit board with two yellow smiley faces as computer chips

But there must be a way to use AI to write jokes.

By themselves, AI language models can’t write good, original jokes.

But an AI system can if it’s explicitly told how to structure a joke.

Enter Witscript.

This app can write a joke.

A stack of books entitled "Comedy Writing for Late-Night TV"

Witscript is an AI-powered web app created by me, Joe Toplyn, four-time Emmy winner and author of “Comedy Writing for Late-Night TV.”

Witscript is a hybrid of an LLM and the professional joke-writing techniques in my book. I’ve taught those techniques to thousands of people through my book and classes, and now I’ve taught them to an AI system.

Witscript is the product of sending an LLM like ChatGPT to joke school.

So to answer the original question, here’s how you can write a joke with AI.

Write original jokes fast with Witscript.

You can use the Witscript app to write the jokes you need,
or as a tool to kickstart your comedy creativity.

For more information about Witscript, click here.

And to see lots of jokes Witscript wrote, check out the sidebar nearby, or click here.

Make Witscript your AI comedy co-writer today!

If You Wish to Be a Writer, Write. Right? Wrong!

Posted on November 2, 2016 by Joe Toplyn

Bust of Epictetus

Aspiring writers often hear the poor advice of the Greek philosopher Epictetus, who is quoted as saying, “If you wish to be a writer, write.”

For example, this blogger states, “If you want to be a writer, then all you have to do is write. That’s all there is to it, just start writing.”

I think that’s poor advice because that’s not “all there is” to writing. That’s like telling somebody who has never driven a car before, “If you want to be a driver, just start driving. That’s all there is to it.”

Writer Stephen King gets much closer to making sense when he recommends this: “If you want to be a writer, you must do two things above all others: read a lot and write a lot.”

But even Stephen King’s advice is incomplete. Read a lot of what?

This is my advice to aspiring writers:

If you wish to be a writer, start by reading books about how to write.

Here’s why my advice makes sense in the context of writing scripts for TV and film:

Writing well, like driving well, means following a lot of rules. It just does. Writing great scripts requires knowing how to structure a story, create characters, build compelling scenes, and so much more.

Figuring out those writing rules by yourself—by reading hundreds of scripts, say—would take a huge amount of time. And most people value their time too much to spend it on reinventing the wheel.

Taking classes to learn those writing rules would cost a lot more money than buying a few books, even if such classes were available.

Writing without knowing those rules will result in your writing badly. You will get demoralized by your lack of success. And you will annoy your friends when you give them your bad drafts to read for their feedback.

After you’ve climbed way up the learning curve by reading a few books about how to write, then read something else: actual produced scripts in your favorite genre. Study those real-world examples of how professionals followed the rules that you’ve learned.

Many TV and movie scripts are available for free online, like at Simply Scripts and The Daily Script.

Car crashed into telephone pole After you’ve read a lot of how-to books and scripts, then, and only then, write something yourself.

With a firm grasp of the writing rules of the road, drive your script safely down entertaining avenues to a satisfying destination instead of into a bad neighborhood or a telephone pole.

Here are a few of my favorite writing how-to books, to get you started:

For TV writers:

The TV Writer’s Workbook, by Ellen Sandler

Writing Television Sitcoms, by Evan Smith

For movie writers:

The Anatomy of Story, by John Truby

Writing Screenplays That Sell, by Michael Hauge

The Writer’s Journey, by Christopher Vogler

Plus my friend Greg DePaul just wrote the book Bring the Funny: The Essential Companion for the Comedy Screenwriter. I’m eager to read it because he’s a hilarious guy with solid screenwriting credentials.

And, of course, if you want to write short-form comedy—the kind you see on late-night TV—get my book, Comedy Writing for Late-Night TV.

Should You Include the End in a Movie Synopsis?

Posted on October 10, 2016 by Joe Toplyn

An early step in selling a movie script you’ve written is often to send a brief synopsis of the script to a reader. If the reader likes your synopsis, he or she may ask you to send the entire script.

It’s a big mistake to give away the ending of your movie in your synopsis.

Plenty of screenwriting advisers disagree with me. For example, in this InkTip article entitled “Your Synopsis Should Not Be These 3 Things,” Michael Kim recommends, “Spoil the ending in your synopsis.”

And Allen B. Ury writes in an article for Greenlight My Movie, “Should you reveal your ending? Absolutely.”

Here’s why I strongly disagree with both advisers.

A big pile of movie scripts

The purpose of a script synopsis is not to make a reader’s life easier.

The purpose of a script synopsis is not to give away all the key elements of your script, for free, to a total stranger who is hungry for new movie ideas.

Instead, as Kim himself points out, the purpose of a script synopsis is “to compel the reader to read the script.”

Spoiling the ending in the synopsis will not help to compel a reader to read the script.

Consider this. If your synopsis gives the reader the main characters and the entire story, including the end, the only reason for the reader to read your script is to see how you handle dialogue.

But dialogue is a minor part of any movie script. Premise, characters, and story are almost everything.

So sending the reader a synopsis that includes the end is basically giving away your entire script to the reader. Why then would an overworked reader bother reading your actual script?

movie_trailer

Here’s a show business analogy. The purpose of any movie trailer is to compel the public to spend time and money to see the movie. That makes a trailer the movie marketing equivalent of a script synopsis.

Does a movie trailer ever include the ending of the movie? Never!

A trailer that spoiled the end would give people a reason not to spend time and money to see the movie. And a studio marketing executive who insisted that a movie trailer reveal the ending would be laughed at, and possibly fired.

So don’t include the end of your screenplay in your synopsis.

Instead, make your synopsis as compelling as possible without the ending.

Prove to the reader with your few well-crafted paragraphs that your script has an intriguing premise, engaging characters, textbook story structure in the first two acts, rising conflict, stakes that are raised, and all that good screenwriting stuff.

But omit the ending. Make the reader of your synopsis think:

“This writer seems like a total professional and this script sounds amazing. I really want to know how it ends. I guess I’ll just have to read the whole script to find out.“

Any reader who thinks like that will ask you to send your script. And your synopsis will have done its job.

As successful screenwriter William Martell says about writing a synopsis: “It’s designed to make the reader want to grab my script and start reading, NOT to give them all of the story beats so they can ‘pass’ without reading it.”

For more of my advice about writing get my book Comedy Writing for Late-Night TV. The book "Comedy Writing for Late-Night TV"

How to Write a Roast

Posted on September 21, 2016 by Joe Toplyn

A roast is a series of insulting jokes about a particular person, the roastee. You write each roast joke using the same techniques that you’d use to write a joke about a topic in the news.

Roast of Rob Lowe The only difference is that for a roast joke, the topic is “I’m paying tribute to [Name of the Roastee].”

So here’s how to roast someone. Start by brainstorming as many associations of the roastee as you can. An association is something that most people in your audience would know about the roastee or would accept as true about the roastee.

Rob Lowe Roast

An association could be anything: something the roastee did or said, a physical characteristic, a family member. Research the roastee if you have to. Associations are the main building blocks of jokes, so the more associations a roastee has, the easier it will be to generate multiple jokes about him or her.

Of course, because you’re writing roast jokes, be sure to come up with plenty of associations that are unflattering, embarrassing, or negative in some way.

Then draw on your list of associations to create roast jokes using the joke-writing techniques that I call Punch Line Makers. I describe those techniques, and many others, in my book Comedy Writing for Late-Night TV. Write your punch lines so that the laugh triggers are negative associations of the roastee.

I’ll show you how the process works by analyzing some jokes from the Comedy Central Roast of Rob Lowe. Here’s a joke about actor Rob Lowe himself:

Rob Lowe as Soda Pop

“Rob was in a movie called The Outsiders, playing a character called Soda Pop…which made sense since he was about 98-percent coke.” — David Spade

That joke was written using Punch Line Maker #1: Link two associations of the topic. One handle of the topic, Rob Lowe, has the association “Soda Pop,” which has the sub-association “Coca-Cola.”

Another, negative, association of Rob Lowe is “did a lot of cocaine,” which has the sub-association “coke.” The punch line, “98-percent coke,” links those two sub-associations in a surprising way.

When assembling your jokes into a roast of somebody, make sure your jokes don’t become repetitive and therefore less surprising and less funny. That means including jokes that rely on as many different associations as possible and eliminating some jokes that rely on the same association.

An eighty-minute show like the Comedy Central Roast of Rob Lowe minimizes joke repetition by also roasting the other celebrity guests on the dais.

Ann Coulter at mike

For example, here’s a joke about conservative writer and commentator Ann Coulter:

“Ann Coulter has written eleven books…twelve if you count Mein Kampf.” — Nikki Glaser

That joke is a product of Punch Line Maker #3: Ask a question about the topic. The topic “Ann Coulter has written eleven books” invites the question “What are their titles?”

The writer answered that question by using an association of Ann Coulter that the writer believed most audience members would accept as true: “Hitler,” who has the sub-association “Mein Kampf.” That answer to the question became the punch line.

When you’re writing a roast joke, you can also start with a punch line and work backward to create a topic that sets up that punch line. For example, take this joke about Rob Lowe:

Rob Lowe in a Lifetime movie “Rob has been in some of the most successful movies of our lifetime…I’m sorry, on Lifetime.” — David Spade

The writer started by associating Rob Lowe with all the movies he’s made for the lesser-prestige Lifetime network. Then the writer decided to create a punch line around “Lifetime” by using Punch Line Maker #4: Find a play on words in the topic.

A second meaning of “lifetime” is the duration of one’s life. The writer relied on that second meaning to create a topic sentence that sounds like something you’d say when paying tribute to somebody. With the addition of that topic, the joke was complete.

After you’ve written the rough draft of a joke, edit it using what I call the Joke Maximizers. This next joke, about distinctive-looking Ann Coulter, demonstrates the use of Joke Maximizer #11: Don’t be too on-the-nose.

Ann Coulter at the roast “Ann Coulter, if you’re here, who’s scaring the crows away from our crops!?” — Pete Davidson

The rough draft of that joke must have referred to Ann Coulter as a scarecrow. But just calling her a scarecrow would be too on-the-nose. The edited punch line says the same thing but in a less direct, and therefore more surprising and funnier, way.

Because roast jokes are, by definition, insulting, think hard about whether they’ll be acceptable to your anticipated audience.

If you’re roasting a non-celebrity, keep your roast jokes fairly gentle. Joke about associations of the roastee that neither the roastee nor your audience is likely to be sensitive about. And make sure the roastee has agreed to be roasted and will probably have a good sense of humor about it. That way the audience won’t feel too guilty to laugh.

If you’re roasting a celebrity, your jokes can be a lot harsher because in that case you’re “punching up.” The audience for a celebrity roast feels that because the celebrity is richer and more famous than they are, it’s okay to laugh at brutal jokes. It’s doubly okay because the celebrity has obviously agreed to be roasted.

Want help writing roast jokes? Make AI write jokes for you with my AI-powered joke writer app, Witscript. Explore Witscript here.

Why Do People Not Like Puns?

Posted on July 15, 2016 by Joe Toplyn

A lot of people hate puns. They think puns aren’t funny. And often they’re right.

That’s because many puns lack one or more of the characteristics of a good joke.

For example, here’s the pun that, as I write this, has received the most votes on OneLineFun.com: “I’m glad I know sign language, it’s pretty handy.”

“I'm glad I know sign language, it's pretty handy.”

The reason that joke isn’t funny is because its topic—“I’m glad I know sign language”–isn’t true.

A good joke topic is true.

If you start a joke with a topic that your audience doesn’t believe, they’ll be focusing on that lie instead of paying attention to your joke.

Silvio Berlusconi signing

Your audience won’t be engaged by a topic that’s not true. Instead they’ll be thinking, “That guy doesn’t know sign language. He’s only saying he does to set up some tricky punch line to show us how clever he is.” Thus annoyed and distracted, your audience won’t laugh.

Or take this captioned-photo joke, #20 in an article on CollegeHumor.com entitled “20 Puns Even People Who Hate Puns Will Admit Are Good”:

A wall of china plates

The punch line of that joke–the photo of the plates–is visual. A visual punch line can theoretically get a laugh. But that visual punch line won’t because it’s not surprising.

A good punch line is surprising.

In a well-constructed joke, the topic and the angle (also known as the setup) lead the audience to expect something. The punch line is unexpected, so the audience is puzzled at first. Then they are surprised when they realize that the punch line actually makes sense, and they laugh.

But in the joke above, the setup (the words “Great Wall of China”) and the punch line (the photo of the plates) are both visible at exactly the same time. So the audience has no opportunity to form any expectations about what the words mean; their meaning is right there in the photo. The audience isn’t surprised, so they don’t laugh. The captioned photo is just, again, a demonstration of the writer’s cleverness.

But a punning joke that is well-constructed can get a big laugh.

For example, I wrote this joke and posted it on Comedywire.com:

Sheen pun

One reason that joke got a lot of votes is that it follows the rules of good joke-writing. The topic rings true: celebrities actually did pose for those photos. Who would make up an odd news item like that?

And the punch line is surprising. The wordplay–“crabs”–arrives at the very end of the joke, only after the topic and angle have led the audience to expect something about sea creatures.

So don’t hate puns and other wordplay jokes. They’re not all bad, just the ones that are badly written.

3d2

Finding a play on words in the joke topic is what I call Punch Line Maker #4. Learn five other Punch Line Makers, and many more comedy creation techniques, from my book, “Comedy Writing for Late-Night TV.”

Can a Computer Write a Joke?

Posted on May 5, 2016 by Joe Toplyn

Before long, a computer will write jokes as funny as those created by professional comedy writers.

I’m no expert in computational humor, but I am the author of Comedy Writing for Late-Night TV. It’s possibly the most practical guide to writing jokes and other short-form comedy ever written.

I’ve also been reading up on the rapid advances in artificial intelligence (AI).

Take a look at these AI milestones:

1997: IBM’s Deep Blue computer system beats
a human world champion in a chess match.

2011: IBM’s Watson computer system beats
human champions in Jeopardy.

The set of "Jeopardy," with two human champions and IBM's Watson computer

2016: Google’s AlphaGo computer program beats
a top-ranked human in Go.

Each of those milestones was previously considered unattainable by a computer. Each of them was considered to be a true test of a computer’s ability to think like a human. And a computer raced past each one of them.

The ability to write a good joke will be the next barrier for AI that a computer leaps over. That’s because writing a joke involves executing the kind of algorithms that computers can apparently already execute.

Here’s how I believe a computer system like IBM’s Watson will soon be able to write a joke. (What I know about Watson I get from The AI Behind Watson: The Technical Article.)
A humanoid robot serves breakfast to a senior citizen.

Say you’re a senior citizen sitting at home with no one but your personal robot for companionship. You’re reading the news and you say, “Huh. Some airline employee was arrested with over two hundred eighty two thousand dollars in his backpack.”

In a flash, the computer brain in your robot companion takes these steps:

STEP 1: The computer identifies the most unusual keywords and sentence fragments in the potential joke topic you uttered. Those are the linguistic elements–I call them handles–that made the topic interesting enough for you to comment on.

This process of parsing the joke topic into handles seems similar to what IBM’s Watson did every time it “decomposed” a Jeopardy question into “subclues.”

A Delta passenger jet on the runway In our hypothetical joke topic, the two handles are “airline employee” and “two hundred eighty two thousand dollars.”

STEP 2: For each of those two handles, the computer generates a list of what I call associations. An association is a word or phrase that’s somehow related to the handle.

The food court at an airport

One association of “airline employee” is “airport restaurant,” which has the sub-association “costs a lot of money.”

And one association of “two hundred eighty two thousand dollars” is “a lot of money.”

This process of generating associations seems similar to what IBM’s Watson did when it found phrases that were statistically related to keywords by using its “hypothesis generation” algorithms.

STEP 3: The computer makes a connection between the two associations “costs a lot of money” and “a lot of money” and it decides to build a punch line around that strong connection.

This process seems similar to what IBM’s Watson did when it evaluated candidate Jeopardy answers by using its “evidence retrieval” algorithms.

Here’s a diagram of Watson’s question-answering (QA) architecture:

A diagram of the question-answering architecture of IBM's Watson computer system

STEP 4: The computer uses its automated reasoning skills to devise what I call an angle leading from the joke topic to the punch line. The computer might reason this way: The airline employee has a lot of money, so he can dine at an airport restaurant, which costs a lot of money.

STEP 5: The computer uses its natural language generation (NLG) skills to transform the topic-angle-punch line data of the candidate joke into conversational English.

Some NLG systems are currently used to transform numerical data into sentences in English. And other computer programs known as chatbots now use NLG systems to simulate human conversation.

STEP 6: The computer decides that the resulting candidate joke is probably surprising enough to be funny. It does this by searching through its store of human knowledge and finding that a meal at an airport restaurant is extremely unlikely to cost two hundred eighty two thousand dollars.

This process seems similar to what IBM’s Watson did when it performed “confidence estimation” and “ranking” on a candidate answer.

STEP 7: The computer directs your robot companion to tell you the joke.

Sure, those are a lot of steps for a computer to execute. But remember, IBM’s Watson beat human champions in Jeopardy. That strongly suggests that a computer could do all of the above really fast, as fast as a professional comedian.

And it may be even easier for a computer to get a laugh than to win dollars in Jeopardy. That’s because there is only one way to win a Jeopardy question: figure out the one correct answer and buzz in first.

But there are several ways to get a laugh from any given joke topic. A fertile joke topic–one with a lot of associations–may spawn three or more different jokes. And those jokes don’t have to be of professional quality to get a laugh; they just have to be reasonably funny.

A man sits at an airport restaurant
So when you comment to your robot companion, “Look at this. Some airline employee was arrested with over two hundred eighty two thousand dollars in his backpack,” the robot might wisecrack:

“He probably needed the money to buy lunch at the airport.”

And you, the senior citizen, might chuckle and feel a little less lonely.

High-quality computational humor is the next Holy Grail for AI. That’s because a computer with a sense of humor is a more human computer. And a more human computer is worth billions of dollars.

UPDATE #1: A Computer Can Write a Joke

Soon after I wrote the above post I decided to dust off my engineering degree and teach a computer to be funny myself.

I learned how to program in Python and how to use many readily-available tools of natural language processing (NLP), natural language generation (NLG), and machine learning (ML).

Then I used those tools to transform some of the techniques in my book, “Comedy Writing for Late-Night TV,” into computer code.

Now I think it’s fair to update the second paragraph of my original post to say that I am an expert in computational humor.

I also think it’s fair to answer the question posed by the title of my original post this way: Yes, a computer can write a joke.

I call my joke-writing system Witscript (U.S. Patent Nos. 10,642,939; 10,878,817; and 11,080,485). Right now the jokes it makes aren’t very sophisticated.

But as research in artificial general intelligence (AGI) advances, Witscript can learn to create smarter jokes. Witscript is a road map to a computer with a sense of humor, like a witty human companion.

Witscript also represents a new approach to computational creativity.

On November 22, 2019, I presented a poster about Witscript at the Natural Language, Dialog and Speech (NDS) Symposium in New York City.

The poster is here.

And below is the poster abstract. For a demo of Witscript, contact me through my company website.

Witscript: Computational Methods for Generating Contextually Integrated Jokes
Joseph E. Toplyn, SB, Twenty Lane Media, LLC

Research shows that people prefer to interact with an artificial intelligence (AI) that exhibits a sense of humor. But most conversational AIs have only a limited ability to make jokes. For example, they only output jokes written by humans, they mostly generate nonsense that requires human editing, they only tell jokes on request, or they tell jokes unrelated to a context. In contrast, we present Witscript, a novel approach to computational humor that independently generates original, contextually integrated jokes. A user inputs text, which may be an utterance in dialog, into the Witscript system. Witscript uses natural language processing tools and word embeddings to identify which pair of keywords in the input text is most likely to be useful in creating a joke response. Word embeddings are then used to list words associated with each of the two keywords. The two word associations that have the best wordplay are linked to create a punch line. Finally Witscript uses a language model fine-tuned on a dataset of jokes to fill the gap between the input text and the punch line with bridge text in a natural-sounding way. This bridge text and the punch line comprise a joke response to the original input text. The system outputs the joke response to the user only if it exceeds a preset score. With Witscript software to simulate a humanlike sense of humor in conversation, chatbots can be more engaging and personal robots can be friendlier companions.

UPDATE #2: This App Can Write a Joke

I’ve launched my AI comedy co-writer, Witscript, as a web app. Now anyone can get AI to write a joke quickly and easily.

Want help writing jokes?
Make AI write jokes for you with
my AI-powered joke writer app.
Explore Witscript here.

How to Fill the Blank Page

Posted on April 22, 2016 by Joe Toplyn

I’ve heard that some writers get writer’s block when they face a blank page. They don’t know how to start writing. They have no ideas. They get paralyzed.

I’ve never had writer’s block.

That’s because I understand that the page is never blank. Okay, I’m exaggerating a little. The page is blank, but only for a few seconds. Here’s what I mean.

To find something to write about, I start by reading. I surf the Internet—news websites, blogs, Facebook—paddling around in an endless sea of stimuli, until an item provokes an emotional response in me.

The response could be any of these:

“That’s interesting.”

“That’s dumb.”

“That’s surprising.”

“That’s annoying.”

“That’s weird.”

“That’s hypocritical.”

“That’s disgusting.”

Then I copy the item, open a blank page in Word, and paste the item onto the page. Or I’ll type one or two sentences summarizing the item. Voila—no more blank page.

Because an item grabs your interest, it will probably grab other people’s interest, too. That means it’s probably worth writing about. It’ll probably be easy to write about, too, because you care about it.

Now that the page isn’t blank, all you have to do is develop the item into something more entertaining: ask “what if” about it, put a twist on it, build a story around it. Sure, you’ll have to do some hard work, but at least you got over your writer’s block.

If you have trouble fleshing the item out, go back to reading until a new item jumps out at you. Then copy that item onto your page. So, again, the page is never blank.

Here’s how that process worked recently when I wanted to post a joke on Twitter. I cruised around news websites until this item jumped off the screen:

Did the news item provoke an emotional response? Sure it did. A solid gold toilet? Wait, what? That’s weird, surprising, and kinda disgusting.

So I copied the item into a blank Word document, used what I call Punch Line Maker #4 (“Find a play on words in the topic”), and came up with this:

Take that, blank page.

To learn more about what makes a good topic for comedy, get my book Comedy Writing for Late-Night TV.

How to Write a Funny Caption

Posted on November 11, 2014 by Joe Toplyn

Say you want to write a funny caption for this cartoon, which was the subject of Caption Contest #447 in The New Yorker.

You write funny captions for uncaptioned cartoons the same way you’d write jokes about a Found Photo, which is an unaltered photo that wasn’t originally intended to be funny. Examples of Found Photos are the animal photos on the website I Can Has Cheezburger?…

…and the athletes’ headshots in the comedy piece “Superlatives” on The Tonight Show Starring Jimmy Fallon.

Follow these steps to write a funny caption for a cartoon, photo, or other picture:

A) Briefly describe the scene in the picture using its handles. The handles are the one or two most distinctive people or things in the picture and the action that those people or things are performing. This descriptive sentence will bridge the visual picture and a verbal punch line, i.e., the funny caption.

In the case of the New Yorker contest cartoon, the descriptive sentence would be something like, “A guy’s pets have psychotherapy.”

B) Use that descriptive sentence and the Punch Line Makers to create a funny caption. The Punch Line Makers are the six proven techniques for creating punch lines that I cover in my book Comedy Writing for Late-Night TV.

In the case of the New Yorker contest cartoon, the winning writer, Alonso Cisneros, apparently used Punch Line Maker #1: Link two associations of the topic. Here’s how he went about it.

The topic—the descriptive sentence—is “A guy’s pets have psychotherapy.”

Brainstorming on the handle “pets” produces associations like “hair on the furniture,” “shelter animal,” “adoption,” “cat food,” “veterinarian,” “do tricks,” and “leash laws.”

Brainstorming on the handle “psychotherapy” yields associations like “depression,” “fifty-minute hour,” “my parents hate me,” “childhood trauma,” “I was adopted,” “couch,” and “Freud.”

At least one association, the one involving adoption, appears in similar form on both lists. Linking those two associations results in the winning caption:

“My pets found out they were adopted.”

C) Edit the caption using the Joke Maximizers. The Joke Maximizers are twelve tools for editing jokes to make them as funny as possible; I list them in my book. The writer of the winning caption used these Punch Line Makers:

Punch Line Maker #1: Shorten as much as possible. The caption doesn’t have any unnecessary words or syllables.

Punch Line Maker #2: End on the laugh trigger. The most surprising word in the caption—“adopted”—is at the very end.

Punch Line Maker #4: Make everything clear. Every word in the caption is simple. Not only that, the words “my pets” make it clear that the man on the couch is talking. That’s important because it’s not immediately clear from the drawing itself which human or animal is talking, if any.

Want help writing funny captions?
Make AI write funny captions for you with
my AI-powered joke writer app, Witscript.
Explore Witscript here.

How to Write a Desk Piece

Posted on October 11, 2014 by Joe Toplyn

“Children grow up so fast these days. That’s the reasoning behind this latest addition to a beloved series of children’s books. It’s [HOLDS UP FAKE BOOK] Curious George and the hole in the wall of the girls’ locker room.”

That’s an example of a joke in a Desk Piece, a type of short-form comedy that’s popular on many comedy/talk shows. A Desk Piece is a segment of fully-scripted comedy that the host performs by himself while sitting at his desk.

Here’s how I wrote that joke. The process was very similar to that of writing a topical monologue joke, a process I cover in detail in my book Comedy Writing for Late-Night TV.

I started with the topic of the Desk Piece. The topic is typically a collection of things in a particular category, in this case “New Books.” Next I brainstormed a long list of angles off that topic, that is, types of real books. Here’s a partial list of those angles:

cookbooks * dictionaries * puzzle books * coffee table books * The Lord of the Rings * thesauruses * novels * atlases * children’s books * Mark Twain * celebrity memoirs * paperbacks * etiquette books * Curious George * biographies * encyclopedias * guidebooks * pop-up books * how-to books * photography books * manuals

To create the punch line I chose one angle–Curious George–and used my Punch Line Maker #3 on it, asking the question “What might a curious person do?” I answered that question using a surprisingly unwholesome association of “curious,” which is “spy on people.”

Finally I used my Joke Maximizer #9 (“Get specific”), devising a very specific Peeping George scenario, and arrived at the punch line: “Curious George and the hole in the wall of the girls’ locker room.”

“New Books” is what I call a Graphic/Prop Piece, one of the seven types of Desk Pieces I analyze in my own new book, Comedy Writing for Late-Night TV.

A footnote: I wrote that Curious George joke for Late Show with David Letterman in May 1998 but you’d never know it. It’s an example of evergreen comedy, comedy that has a long shelf life, usually because it’s not based on something topical but on a more lasting pop culture phenomenon.

If you’re preparing a writing sample to submit to a comedy/talk show, include a generous portion of evergreen comedy. That way your submission won’t seem too dated if it winds up sitting on the credenza of some head writer for months before it’s read.

How to Pick a Good News Item for Writing Jokes

Posted on August 17, 2014 by Joe Toplyn

Some news items are easier to write multiple jokes about than others. Monologue writers like those news items because cranking out several jokes about the same news item takes less time than hunting down other promising news items and then writing jokes about those.

So what makes a news item a good candidate for multiple jokes, or, as I say, fertile?

The answer lies in the handles of the news item. The handles are the most attention-grabbing words or phrases in the news item, those details that stand out the most.

Take this headline that I saw recently on the ABC News website:

“Spain: Woman Arrested With Coke in Breast Implants”

One reason that headline lit up on my joke topic radar was because it has two handles and each handle has a lot of associations, which are words or phrases that come to mind when you think about something. Two handles with a lot of associations make a news item fertile because linking handles using their associations is one of the main ways to create punch lines. So the more associations, the more possible punch lines.

Here’s how that principle works in practice. In the above news item the two handles are “coke” and ”breast implants.” The associations of “coke” include “Colombia,” “snort,” “Coca Cola,” and lots of others. The associations of “breast implants” include “D cup,” “silicone,” “Pamela Anderson,” and many more.

Linking associations like those using Punch Line Maker #1, described in my book, I wrote this joke and posted it on Facebook:

“In Madrid, police arrested a woman with 4 lbs. of cocaine hidden in her breast implants. Her bra was from Victoria’s Secret Stash.”

Then I added a pop culture association to the mix, used my Punch Line Maker #2, and posted another joke:

“In Madrid, police arrested a woman with 4 lbs. of cocaine hidden in her breast implants. The good news: she’s now engaged to Charlie Sheen.”

My comedy writer friend Gabe Abelson supplied a third punch line, this one using my Punch Line Maker #4:

“The good news about keeping four pounds of cocaine in your breast implants is they’ll stay up for about 50 years.”

And another comedy writer friend, Wayne Kline, also turned to Punch Line Maker #4 for this one:

“The clever DA lowered the thermostat in the trial room so the evidence will stand up in court.”

The point is that jokes flowed freely from that news item because it has two handles with a lot of associations. So focus first on news items like that when you’re hunting for possible joke topics.

Learn more about my six Punch Line Makers in my book Comedy Writing for Late-Night TV.