Guides
Part of Why humorous quotes lose their authors
Why 'best humorous quotes' can't be ranked
Why best humorous quotes resist ranking: no fixed text, no single author, and jokes that are types rather than texts. How to cite comedy honestly.
A ranked set of the best humorous quotes is a category error, and it is a category error twice over.
The first problem is no text: a joke exists in the air, differently every night. The print version is one transcription of one performance, often not even that.
The second problem: the credit does not resolve to a person. Scripted comedy is written by one set of people and delivered by another, and both contributions are real, so "who said it" has two correct answers, neither complete.
Neither problem is a reason to give up on comic material. They are reasons to collect it differently.
What to take away
- There is no manuscript for most funny lines, and where a script exists it differs from what was broadcast.
- Scripted credit is assigned by formal process to the written work. It is not a record of who first said the line in a room.
- Most jokes are structures rather than textsthe same joke exists in thousands of wordings with different targets.
- The answerable questions are about wordings, carriers and dates. "Who said it first" usually is not one of them.
No manuscript
A quotation normally rests on an object that many people can open and that has not changed: an edition, a page, a fixed text. Nearly everything that makes quotation possible comes from that stability.
Comic material has almost none of it.
- Live material has no authoritative version.A stand-up set changes nightly, and the recorded special is one take late in a tour, by which point the bit has been rewritten a hundred times in front of audiences.
- Scripts are not what was broadcast.Lines are cut, added and improvised at recording, so a shooting script is evidence of intent rather than of what happened.
- Improvised lines exist only on the recording.If the recording was not kept, the line survives as somebody's memory of it.
- Print humor is the exception, and it is the reason columnists and comic novelists are so much easier to quote than performers. There the text is fixed and the ordinary rules apply, which is what a press made possible in the first place.
The practical consequence: for spoken comedy the recording is the primary source and everything else is derivative, which is a reversal of the usual hierarchy and the reason comic context work starts from the performance rather than the transcript.
Credit does not resolve to one person
In scripted film and television, a formal process decides writing credit: in the United States the Writers Guild of America arbitrates screenplay and teleplay credit under published rules. That process is about the script, not a log of who improvised what on the day, and it is not designed to be.
Where a guild or a production's own agreement governs the credit, consult its rules, not an assumption about how it works.
The arrangement resembles ghostwriting elsewhere in publishing, where the named person and the writing hand routinely differ. The room matters too.
Comedy writing is often collective: a line arrives in a draft, gets rebuilt by four people, and ends up in the mouth of a performer who changes the timing and one word, the change that makes it land. Assigning that sentence to one person means deciding which contribution counts, and reasonable people answer differently.
So there are usually several true statements available, and the useful move is to make the ones you can support:
- The program or film, which is the object the line exists in.
- The credited writer or writing team, if the credit is public.
- The performer, as the person who delivered it.
- The date, which is often the only thing everyone agrees on.
A collection that records those four fields is doing better than one that prints a name and a colon, and the difference is that every field can be checked.
Jokes are types, not texts
This is the deeper problem, and it is not solvable by better record keeping.
Many jokes are structures: a setup shape and a turn, refilled with new nouns. The same joke appears across centuries and languages, with the target swapped, the occupation changed, the object updated to current technology.
Folklore scholarship classifies this material by type, not wording, because wording is the least stable part: the Aarne–Thompson–Uther index of tale types and Stith Thompson's Motif-Index of Folk-Literature catalog stories by structure and motif rather than by sentence. The point generalizes: a joke is a form before a text, and forms outlive their fillings.
A clean example is the light-bulb joke: "How many X does it take to change a light bulb?" The frame never changes, and the fillings rotate through professions, nations, brands and fandoms. No version is the original, no wording is canonical, and the type — not the text — is what can actually be catalogued.
Once you see a joke as a type, several questions dissolve.
Who said it first is usually unanswerable and often meaningless. The structure may be older than any record, and the person credited may have been the four hundredth teller.
What the correct wording is has no answer. There is no canonical form, only a family of forms, and the one you know is the one that reached you.
Whether it was stolen is frequently the wrong frame. Two comedians can arrive at the same structure independently, because the structure was available to both of them.
What remains answerable is when a form is first attested, which is a real question with a real method behind it, described in how to antedate a joke.
Better questions
Why it fails
- Who said this?
- Assumes one author and one wording
- What is the exact quote?
- There is no exact version of an oral text
- Is it really theirs?
- Merges writing, delivering and originating
- When was it first said?
- Unknowable for most jokes
- Is this the original?
- Presumes a fixed original existed
The answerable version
- Who said this?
- Who delivered this wording, on which recording, on what date?
- What is the exact quote?
- Which transcription is this, of which performance?
- Is it really theirs?
- Which of those three are you claiming, and what supports it?
- When was it first said?
- When is this joke type first attested in print?
- Is this the original?
- Is there an earlier form, and what changed between them?
Every question in the right-hand column can be settled or honestly reported as unsettled. Every question in the left-hand column produces a confident answer that will not survive checking. That is how the standard collections, Bartlett's Familiar Quotations among them, came to contain what they contain, as the history of quotation collecting sets out.
Unanswerable vs Answerable Questions
Question people ask
- Who said this?
- Who delivered this wording, when?
- What is the exact quote?
- Which transcription of which performance?
- Is it really theirs?
- Writing, delivering or originating?
- When was it first said?
- When is the type first attested?
- Is this the original?
- Is there an earlier form?
Answerable version
- Who said this?
- What is the exact quote?
- Is it really theirs?
- When was it first said?
- Is this the original?
Writing about a joke without misquoting it
The craft is different from quoting prose, and it is not harder.
- Describe the bit rather than reproducing a line.For most purposes a sentence explaining what the joke does is more useful to the reader than the joke itself, and it does not misrepresent a wording.
- Cite the carrier.Program, release or recording, with a date and a timestamp. That claims exactly what you can support.
- Use type language where it applies."A version of an old joke about" is accurate, informative, and takes the pressure off an attribution that will not hold.
- Name both contributions when the record allows it.Written by one party, delivered by another, is two facts rather than a compromise between them.
- Say when the wording is yours.If you tightened it in the retelling, and everybody does, that is your edit and not theirs.
The pass-or-fail gates that a source-checked collection applies to any entry work here too, with one change: the object you must be able to open is a recording rather than a book.
Common questions
Is anything in comedy cleanly attributable?
Yes: published humor writing. A column, an essay, a comic novel, a printed script that names its author. Fixed text, named writer, a date. Almost all the trouble comes from oral material, and choosing print sources is the easiest way to avoid it.
Two comedians tell the same joke. How do I write about that?
Describe the sequence you can document: who is on the earlier recording, who is on the later one, and whether the structure predates both. Resist the theft framing unless you have evidence of copying rather than of similarity.
Does the audience recording count as a source?
It is evidence that the line was performed that night, which is a real fact with a date attached. Treat it the way you would treat any unofficial document: say what it is, say where it came from, and do not upgrade it to a published record.
