A Lexicographer’s Memoir.

Adrienne Raphel reviews Kory Stamper’s Word by Word: The Secret Life of Dictionaries for the New Yorker; I’ll quote the start to give you an idea:

One morning in 2001, Kory Stamper, a lexicographer for Merriam-Webster, arrived at work and was given a single word: “take.” She set to work hunting down examples of where the verb form of the word had been used in the wild, from American Literary History to Us Weekly to Craigslist, and organizing these citations by part of speech and usage. Normally, editors will work on several words in a batch. But smaller, more common words are used so often and in so many different ways that a single one can be an incredible headache to revise. As Stamper explains in her recent book, “Word by Word: the Secret Life of Dictionaries,” such words “don’t just have semantically oozy uses that require careful definition, but semantically drippy uses as well. ‘Let’s do dinner’ and ‘let’s do laundry’ are identical syntactically but feature very different semantic meanings of ‘do.’ ” Lexicographers know that when they’ve been assigned a notorious small word—“do,” “run,” “about,” “take”––they’ve arrived.

This was the most ambitious and slippery project Stamper had taken on, and, at times, as she parsed the differences between “take first things first” and “take a shit,” she felt herself “slowly unspooling into idiocy.” It took two weeks to organize the verb form alone into a hundred and seven different senses and sub-senses; after a month, “take” was finally ready for the eleventh edition of Merriam-Webster’s Collegiate Dictionary. In the world of words, however, spending a month perfecting an entry is nowhere near the extreme. At a conference in 2013, a lexicographer from the Oxford English Dictionary told Stamper that when he revised “run” it took him nine months. Dictionary editors trade word stories the way élite marathoners collect courses. For Emily Brewster, one of Stamper’s colleagues, a career highlight was discovering a previously unrecorded sense for the indefinite article “a”: “used as a function word before a proper noun to distinguish the condition of the referent from a usual, former, or hypothetical condition.” Stamper gives as an example, “With the Angels dispatched in short order, a rested Schilling, a career pitcher 6-1 in the postseason, could start three times if seven games were necessary against the Yankees”: “a rested Schilling” tells us that, in contrast to his current rested state, he is not usually rested, or he had not been rested previously, but now he is. Each lexicographer has stories like this: epiphanies that reflect the evolution of language.

Isn’t that fun? Sounds like a wonderful book. Thanks, Trevor!


  1. I hope they are paid by time, not by source word count…

  2. I enjoyed the book, but I like her blog even more, which is less edited and more uninhibited.

  3. I hope they are paid by time, not by source word count…

    By Zipf’s law.

  4. Squiffy-Marie van 't Blad, Dutchman-at-large says

    I will certainly have to take a look at this take on ‘take’.

  5. Stephen C. Carlson says

    I’ve read the book. I predict that it will be enjoyable to most who read this blog.

  6. “used as a function word before a proper noun to distinguish the condition of the referent from a usual, former, or hypothetical condition.”

    But what about something like “supporters of appeasement faced an eloquent Churchill in the afternoon’s debate”. Churchill’s eloquent, he’s always been eloquent, no one’s suggesting that he wasn’t going to be eloquent; the point is not to distinguish his condition from a usual, former, or hypothetical condition, the point is to draw attention to his eloquence as being particularly relevant, rather than, say, to his height or the colour of his hat.

  7. No, I disagree; if one wanted simply to draw attention to his eloquence one would say “the eloquent Churchill.” To me, the use of “a” implies that he was not always eloquent or that he was particularly eloquent at that time.

  8. “supporters of appeasement faced an eloquent Churchill in the afternoon’s debate”

    This suggests to me that the supporters did not expect to face an opponent as eloquent as Churchill was known to be (and demonstrated on that particular occasion), so the debate was a fiasco for them.

  9. “an unexpectedly smooth landing”

    Does this count?

  10. @Bathrobe:

    No, since there’s no proper noun involved. (It’s distinctly odd to use “a” with a proper noun, but perfectly standard to use it with a normal noun like “landing”. Also, the phrase you offered refers to a distinct, singular event — one particular landing — and not to the altered or unusual state of some pre-existing entity.)

  11. “The episode infuriated an already angry Smith, who was heard on the stump microphone disagreeing with the officials and was caught on cameras shaking his head.”

    Does this one count?

  12. Sure, that one makes sense to me.

  13. Graham Asher says

    I have started reading Kory Stamper’s book and am immediately disappointed. I hope it gets better. The reason for my disappointment is the incorrect pronunciation of Old Norse given on page 6. Stamper reports her prof’s explanation of it at university, where she studied, among other things, mediaeval Icelandic family sagas. What the prof gives is in fact the pronunciation of modern Icelandic (which may well be what is used when Old Norse is taught in American universities; I don’t know). Why should this annoy me? Well, for one, I’m a pedant, but seriously, this sort of thing is very easy to check, and assertions about language in a book on language should be checked.

  14. True, but your indignation should be directed at her professor rather than her — you can hardly expect everyone to double-check everything they were told by professors, who can be presumed to know what they’re talking about. It would be nice, of course, if the MS had been reviewed by someone who could have pointed that out, but we no longer live in such times. And I remind you that she is a lexicographer, not a historical linguist.

  15. Latin as Italian, Ancient Greek as Modern Greek, Old Norse as Icelandic. None of these things are actual evils, and of them all the third is probably the least so, as the sound changes separating them are for the most part smooth, not provoking restructuring that would, for example, distort the meter of verse or the rhythm of prose.

  16. Data point: The late Foster W. Blaisdell of Indiana U taught Old Norse with Icelandic pronunciation.

  17. It is a disappointment, since I am used to everything Stamper says being perfect.

Speak Your Mind