Industry
Part of Inspirational quotes: the directory, with corrections
A working brief on inspirational quotes misquotes
Polish grows with retelling and credit flows uphill. Those two patterns triage a suspect inspirational quote before you spend an hour in the archive.
Two rough regularities do most of the work in this field. Held together, they tell you which suspect lines to check first, because you are not going to check everything.
Our account of what a quotation is actually asserting sets out why the wording and the name have to be tested apart. The first regularity: the better a line is made, the less likely the credited person made it. The second: credit only ever flows toward greater fame, never away from it.
Neither is a law. Together they are a prior, and a prior is what you need when a toast is in three days.
What to take away
- Polish is a transmission artifact, not evidence of authorship. A line nobody could improve has usually been improved by many people.
- Credit moves one direction only. A line credited to an obscure person gets recredited upward; a line credited to a famous one almost never moves down.
- The counterintuitive rowan awkward quotation attached to a famous name is more likely genuine than an elegant one, because transmission does not preserve clumsiness.
- When you publish a correction, reprint the circulating version beside the sourced one. The folk sentence will outlive your correction either way.
- No source means no use in a book epigraph or a leadership deck. A toast can carry a folk version if you label it as folk.
Real sentences are worse than famous ones
People do not speak in epigrams. Even good writers produce uneven prose, with clauses that run long, a word repeated by accident, and a rhythm that does not resolve.
Speech is worse. Hesitations, restarts, sentences that change direction halfway through.
A line carried by oral tradition for a few hundred retellings has none of that, because each retelling is an edit. The edits are not random. The version with the extra clause gets forgotten; the version with the strong word at the end gets repeated.
So when a quotation is perfectly balanced, exactly as long as it needs to be, and impossible to improve, you are usually looking at the output of that process, not the input to it.
If your reaction to a line is "you could not write that better," raise your suspicion rather than your confidence.
What transmission optimizes for
These are the features to look for, because they are the ones repetition selects: short clauses, a strong word at the end, parallel structure, concrete images, direct address, and the removal of anything that ties a sentence to one particular occasion.
What retelling selects for
- Brevityshorter every version
- Balancematching halves, added symmetry
- Reversalsecond half inverts first
- Concretenessimages replace abstractions
- Hard endingshort concrete final noun
- Universalitynames, places, dates sanded off
That last one does the most damage and gets noticed least. A sentence about one particular situation becomes a sentence about life, and nobody in particular does the generalizing. It happens over decades.
Credit flows uphill
The second regularity is about names rather than words. When a line detaches from its author, which happens constantly because most lines are heard secondhand, somebody eventually supplies a name. They draw from the small stock of names available to them, and that stock is famous people.
The flow is one-directional. A line credited to a famous person almost never gets recredited to an obscure one, because nothing propels that move. The obscure attribution is less interesting, less memorable, less likely to be repeated. A line credited to an obscure person is recredited upward all the time.
This is the ordinary accumulation of credit toward the already prominent. Run it over a few generations and the result is arithmetically inevitable.
A handful of names end up carrying more material than any person could produce, which is what our material on comic attribution describes for jokes. That concentration is not evidence they were unusually quotable. It is evidence they were unusually well known.
Three lines you have probably met
The regularities are easier to trust with cases attached. Each of these travels with a name and without a source, and each has a paper trail that says something else.
- "Be the change you wish to see in the world" travels as Gandhi's, and it appears in none of his collected writing. What he argued in his newspaper Indian Opinion is conditional: change yourself, and the world's attitude toward you changes. A slogan replaced an argument.
- "The reports of my death have been greatly exaggerated" travels as Twain's. The note he sent to a reporter who came to check on him reads "The report of my death was an exaggeration" — singular report, no "greatly." The popular version is a small upgrade.
- "The only thing necessary for the triumph of evil is for good men to do nothing" travels as Burke's. The sentence behind it is John Stuart Mill's, from his inaugural address at the University of St Andrews: bad men need nothing more to compass their ends, than that good men should look on and do nothing.
- "Everybody is a genius. But if you judge a fish by its ability to climb a tree, it will live its whole life believing that it is stupid" travels as Einstein's. No source in his writing has ever been produced for it, and that absence is the whole case against it.
- "Blood, sweat and tears" is how Churchill is remembered. What he said was "blood, toil, tears and sweat," and the remembered version is shorter and easier to say.
Run any of them through the table: a polished sentence, the most famous name available, no source given. That is the modal misattribution, and every one of these lands in that row.
The triage table
| What you are looking at | How suspicious to be |
|---|---|
| Polished line, very famous name, no source given | Highly. This is the modal misattribution |
| Polished line, obscure name, no source given | Moderately. Nobody upgrades an attribution downward, so the odd credit may be the real one |
| Awkward line, famous name, no source given | Low. Awkwardness survives transmission poorly, so it suggests a real text |
| Polished line, famous name, with an edition and page | Low. Open the book and confirm |
| Any line whose punch depends on who said it | High. That structure is exactly what motivates a false credit |
The third row is the useful counterintuitive one. A clumsy quotation with a famous name attached is more likely genuine than an elegant one, because nothing in the transmission process preserves clumsiness. If it survived, it probably came from a text.
How suspicious is this quote?
Polished line, famous name, no source?
Highly suspicious: modal misattribution
Check the next pattern
Triage in ten seconds
Before you spend real time on a line, ask three questions.
- Is the phrasing better than the person's other sentences? Compare it to anything else you have read by them.
- Is the name the most famous available candidate in that category? Famous wit, famous physicist, famous founder, famous poet.
- Does the line arrive with a source, or only with a name?
Three yeses mean check it before you use it. The search procedure takes about fifteen minutes and usually gives a clear answer. The catalog of specific failure mechanisms names the handful you keep meeting: a qualifier dropped in paraphrase, a line pulled toward a more famous name, a translation that restores a punch the original never had, and a remark about one occasion generalized into a rule about life.
Where to check
Quotation dictionaries first, because they have already done the sourcing. Bartlett's Familiar Quotations and the Oxford Dictionary of Quotations key each line to an author and to a page in a named edition, and both print the earliest source their editors could find. The Yale Book of Quotations does the same with a fuller apparatus, and it is the one to reach for when a line has a long history behind it.
Then the full-text archives, because the dictionaries only cover lines somebody already chased. The Internet Archive and HathiTrust both search inside scanned books, which is how you find the sentence the popular version smoothed over. For anything that ran in a newspaper, the Library of Congress maintains Chronicling America, a searchable archive of American papers. When the line was spoken, the record you want is a transcript or a recording held by an archive, not a recollection printed decades later.
Publishing the correction
Suppose you do the work, find the original, and publish it. The circulating version usually keeps circulating. Understand that rather than resent it.
You are proposing to replace a sentence optimized by an enormous number of independent tests for memorability with one that has passed none. The folk version is good at being repeated. That is the only quality it was selected for, and it is very good at it. The original is longer, more qualified, less satisfying.
Two responses work. Quote the real one anyway and let it land less punchy; accuracy is a style in writing, and readers notice.
Or label the folk version as folk: "the line that circulates is X, though what he actually wrote was Y."
That beats either version alone, because it gives the reader the good sentence and the honest one together.
For a book epigraph or a leadership deck, the sourced version is the only defensible choice. For a toast, a labeled folk version reads better than a stiff original.
Common questions
The questions below come up in nearly every review, usually in this order.
Doesn't this unfairly doubt people who really were quotable?
Somewhat, and that is the cost of a prior. Genuinely epigrammatic writers exist and their real lines are excellent. The point is not to disbelieve them but to check them first, because they are also the names that attract everyone else's material. Verification resolves the unfairness in both directions.
If a line has been improved, whose is it?
Nobody's, in the useful sense. It is a collective artifact, like a folk song with a named original composer. The accurate framing names both the source it came from and the fact that the circulating wording differs. Assigning one owner to a sentence thousands of people edited is the mistake, not the answer.
Can polish ever be evidence of authenticity?
For a working aphorist, yes: someone whose entire body of work is tight sentences. That is exactly why the first triage question is comparative. Polish relative to the person's own prose means something. Polish in the abstract does not.
What do I do when the holding archive disagrees with the popular version?
Present the evidence trail and let the archive's record stand. Name the edition, the page and the collection you checked, and say plainly that the popular wording differs. Do not issue a ruling on authenticity; that belongs to the holding institution or the primary document.
