I’ve built a tool that checks for misquotes. It’s called Quote Checker: www.quote-checker.com

When I was translating David Deutsch’s book The Beginning of Infinity, our copy editor recommended providing missing sources for some of the quotes in the book. While researching those sources, I realized that the book contained several misquotes. We were too close to publication at that point to change the translation beyond small alterations, but I remember wishing I had had a tool to help me check for misquotes.

As a programmer, my eye is already trained for small changes. It may sound insignificant, but a single missing semicolon can make the difference between a completely working and a completely broken program. That’s why so-called diffing software is among a programmer’s most important tools. But that software isn’t made for quoting specifically, it’s made for highlighting changes in code, and even a trained eye can easily miss smaller changes. Just as in programming, a single missing or improperly added semicolon can make or break a quote.

What I needed was a smart diffing tool that could not only show me differences between original texts and corresponding quotes but also explain which differences were done properly vs improperly: if a quote differs from the original, the changes need to be indicated so that the reader isn’t misled.

The result is Quote Checker. It’s free, at least for now. It compares a given quote to the original text and highlights any issues.

Quote Checker has two main customers: readers and writers.

Readers can use the tool to check whether some quote they read is accurate. If it isn’t, they can hold the quoter accountable by publishing a diff of the misquote.

Writers – authors, publishers, journalists, students, etc. – can use the tool to quote properly in the first place. They should paste their quotes into Quote Checker before publication. Writers who are themselves misquoted can publish a diff to explain and make light of the misquote.

In short, Quote Checker aims to increase intellectual conscientiousness among readers and writers.

To learn more, visit www.quote-checker.com. I’ve already documented a bunch of misquotes in the past few days since its release – you won’t believe how common misquotes are! I also paste an article from the Quote Checker site explaining the functionality and rationale below for posterity. (Future changes may not be reflected.)

What’s the big deal?

I get that one shouldn’t alter a quote too much. But if someone just made a small mistake, what’s the big deal?!

Consider this passage from Marta Stanton’s The Right to Misquote¹ (bold emphasis added, footnote removed):

Webster’s New College Dictionary defines quotation marks as punctuation marks “used chiefly to indicate the beginning and the end of a quotation in which the exact phraseology of another or of a text is directly cited.” When quotation marks are used, the reader automatically attributes the exact words within the quotation marks to the person cited. The reader believes that the quotation marks represent that the author is accurately conveying language used by another without paraphrasing or fabricating. The reader is on guard when quotation marks are lacking that the author may have paraphrased the exact wording or inserted his own rendition of the statement. However, the reader lowers his guard at the sight of quotation marks and believes that the exact language must have been used or the author could not have placed quotation marks around the words.

I should give a fair warning that I’m about to disagree with Stanton in two regards below, but I wholeheartedly agree with the quote above. In other words, quotation marks (and blockquotes) not only indicate another’s speech but signal intellectual conscientiousness on the part of the quoter. When this conscientiousness is signaled but not delivered, that’s a type of fraud.

If the meaning of the quote is retained, what’s the harm?

The harm is difficult to gauge. Conveying the intended meaning is difficult enough when a quote remains unchanged; alterations typically increase this difficulty, especially if they are not indicated.

Even the author of the original text may not always be in a position to determine whether an altered quote accurately conveys the meaning he intended. This is because no text is ever obvious and every reader has to recreate the meaning while reading. (This is an insight from Popperian epistemology explained by physicist David Deutsch in his book The Beginning of Infinity, chapter 10.)

Should the author consent to the publishing of inaccurate quotes of his original text, or should he retroactively deem them no big deal or choose to be associated with them, this doesn’t necessarily help readers trying to understand his text, nor does it solve the fraudulent nature of misquotes. (I disagree here with Stanton, p. 448., that an author’s chosen association with a misquote makes the misquote acceptable.)

But all I did was omit a comma! (Or something like that.)

When people hear the term ‘misquote’, they typically think of putting words in someone’s mouth or grossly misrepresenting someone’s speech. While there are more and less egregious cases of misquotes, even minor changes to punctuation can, when done improperly, turn a quote into a misquote. (I use a stricter standard than Stanton (see p. 448 under "B. Acceptable Alterations of Quotations"), who, on the given page, contradicts the quote above.)

Such mistakes may seem small, but they can easily compound when a misquote is itself misquoted and so on.

Additionally, moving, adding, or omitting even small things can greatly change the meaning of some quotes. The size of some change in text does not always correspond to the size of that change in meaning.

Also observe that things we humans consider small can be important to machines. Here’s an example, taken from a real quote. The content doesn’t matter, focus only on the changes at the beginning of each line:

1-.-+)+ Creativity is necessary and sufficient for consciousness/sentience to arise.
2-.-+)+ Animals are not creative.
3-.-+)+ Therefore, animals are not sentient.

All the quoter did was change periods to closing parentheses. That may not be a big deal to humans, but it will break the formatting rendered by certain programs processing this text (such as markdown formatting, in this instance). We cannot possibly know or anticipate everything that depends on the accurate rendition of a quote, so it’s best not to change it improperly in the first place.

Fair enough. But if I do need to make changes to a quote, how do I make them properly?

There are different style guides out there, such as Chicago and MLA, and the specifics will depend on which style guide you choose. The general rule is that your quote must reproduce the exact wording of the original text letter for letter while retaining the formatting (italics, bold, underline, strikethrough, etc.). When you do need to omit parts of the original, use ellipses (e.g. ‘[...]’); surround additions by square brackets (e.g. ‘This is a [great] quote.’). When you remove or add formatting, indicate the changes in your own text, as I did above when I quoted The Right to Misquote, or add brackets inside the quote indicating the change.

Any other tips?

First of all, use this tool! It gives you a detailed view of the differences between a given original text and the corresponding quote; it also lists issues you need to fix in order to avoid a misquote.

Never quote from memory, always consult the original source. Whenever possible, copy/paste, ensuring that the formatting carried over. Don’t assume that you can judge whether an unindicated change to a quote results in a change in meaning or ‘isn’t a big deal’. Remember the meaning of quotation marks and be true to that meaning.