Word by Word: The Secret Life of Dictionaries

by Kory Stamper

The bloody battle to defend English and champion “good grammar” hasn’t always been in existence; in fact, prior to about the middle of the fifteenth century, there was very, very little thought given to English as a language of discourse, officialdom, and permanence. Prior to that, most official documents were recorded in Latin (the gold standard for Languages of Record) or French.*1 Sure, there had always been anonymous writers (and a few onymous*2 ones, too, like Geoffrey Chaucer) who chose to preserve their wisdom—or fart jokes, in the case of Chaucer—in English, but it wasn’t taken seriously as a literary language until Henry V suddenly began using it in his official correspondence in 1417. Within a few decades, English had become the language of the English bureaucracy, replacing French and Latin almost completely. The problem with this shift was that both French and Latin, having been used as languages of record for a while, were already comparatively standardized, and English was not. Latin and French had written forms that stood independent of their pronunciation; English, on the other hand, was entirely phonetically spelled. That meant that while Medieval Latin had one way to spell the word that we know as “right” (rectus) and Old French as used in English laws and literature had six (drait, dres, drez, drettes, dreyt, and droit), Middle English, the form of English in use when it became an official language of record, had a whopping seventy-seven recorded ways to...

Vocabulary boomed in the sixteenth century, and many of those new words were words borrowed from lovely, literary languages on the Continent—Latin, Italian, and French. The Romance-language borrowings weren’t without controversy—Shakespeare himself made fun of people who piled on the highfalutin foreignisms just to sound smart*5—and by the end of the century the language was growing so quickly, both with borrowed words from other languages and with foreign speakers attempting to get their mouths around this burgeoning language, that a handful of native speakers stepped in to provide order. In 1586, William Bullokar, a man who was interested in regularizing and reforming English, published the first English grammar (appropriately titled Bref Grammar for English); in 1604, Robert Cawdrey published what is held to be the first monolingual English dictionary. The concern was that English was becoming terribly unruly, and it needed some reining in. Some called for a large-scale remedy—an academy of English that would not only prescribe good usage but proscribe bad stuff out of English. By “bad stuff,” they meant not just words that people thought were uncouth but all forms of the language—styles, uses, poetic meters, the whole kit and caboodle—that were deemed inelegant and unlovely.

The biggest problem with this sort of grammar, however, is that it sounds logical but it’s based on a faulty logic. Take the oft-repeated injunction to get “its” and “it’s” straight. Everyone claims it’s remarkably easy to remember that “its” is possessive and “it’s” is a contraction. But logic tells us that in English, ’s attached to a noun signals possession: the dog’s dish, the cat’s toy, the lexicographer’s cry. So if English is logical, and there are simple rules to follow, why doesn’t “it’s” signal possession? We know that ’s also signals a contraction, but we don’t have any problems with differentiating between “the dog’s dish” and “the dog’s sleeping”—why should we suddenly have problems with “it’s dish” and “it’s sleeping”?

So where do these rules come from, if not from actual use? Most of them are the personal peeves, codified into law, of dead white men of yore. Take, for example, the rule that we’re not to end sentences with prepositions. It’s one that is drummed into most young writers at some point in their careers, and failing to heed it will result in some teacherly knuckle smacking (literal or figurative). If you ask a modern adherent to this rule why, exactly, you aren’t supposed to end a sentence with a preposition, they merely goggle at you as if you had just asked why you aren’t supposed to lick electrical sockets. Because it’s objectively better not to, that’s why. The rule itself was first articulated by the seventeenth-century poet and literary critic John Dryden. He had used the terminal preposition in his early works, but as he aged and gave himself wholly over to the glories of Latin, he decided against its use: I cast my eyes but by chance on Catiline; and in the three or four last pages, found enough to conclude that [Ben] Jonson writ not correctly….The preposition in the end of the sentence; a common fault with him, and which I have but lately observed in my own writings. When his works were reprinted later in his life, he took the opportunity to tidy up some of the follies of youth, and the terminal preposition was one such folly. Later editions of his work are carefully scrubbed of terminal prepositions: “the age which I live in” became “the age in which I live” and so ...

You can, of course, choose not to end your sentence with a preposition, but that is a stylistic choice, not a grammatical diktat from on high. The fact is that many of the things that are presented to us as rules are really just the of-the-moment preferences of people who have had the opportunity to get their opinions published and whose opinions end up being reinforced and repeated down the ages as Truth.

Lynne Truss’s book “eloquently speaks to the value of punctuation in preserving the nuances of language,” slobbers one adoring reviewer—one among many—and yet Truss commits oodles of punctuation errors throughout her own usage book on punctuation, including one on the cover: there should be a hyphen between “Zero” and “Tolerance.” Humanity sets up rules to govern English, but English rolls onward, a juggernaut crushing all in its path.*7

Many of the rules that have been codified into “grammar” uphold an ideal, not a reality. The grammarians of the seventeenth century onward weren’t interested so much in preserving the language as it was used as in perpetuating a re-formed idea of what language should be. The first soldiers in the fight to preserve English radically changed English, not according to the best practices of the great writers of the language, but according to their own views of elegance and correctness. What they wanted to preserve and promote didn’t, for the most part, actually exist: it was a convenient fiction that was painted in moral terms, thereby ensuring its own propagation. Let me say that again: Standard English as it is presented by grammarians and pedants is a dialect that is based on a mostly fictional, static, and Platonic ideal of usage.

But that’s not the worst of it. I have also, as Gil pointed out, gendered the definition when I didn’t need to. He struck the “his” from “his running mate.” “It is conceivable,” he said afterward, while going over my definitions with me, “that a woman will run for president at some point, and if she does, this definition will need to be revised. So why not write it in a way that the gender of the candidate doesn’t matter.” I was gobsmacked: here I was, a recent graduate of a women’s college, getting schooled on gendered language by an old guy. And rightly so: I brought to the definition my own assumptions about the gender of people who have in the past run for president. But a good lexicographer weighs the untold future as well as the told past and present: Is it so inconceivable that a woman could run for president?

Finding a suitable quotation for a dictionary entry is near impossible, because quotations used in dictionaries need to meet three main criteria: they need to illustrate the most common usage of the word; they need to use only words that are entered in that particular dictionary; and they need to be as boring as humanly possible. Writers generally want to catch and hold your attention, and so they write things that are full of narrative interest, clever constructions, and tons of proper names. These things are a delight to read, and because of that they make for the worst example sentences possible. The goal of a dictionary is to tell people what words mean and show them how they are used in the most objective, dispassionate, and robotic way possible. People do not come to the dictionary for excitement and romance; that’s what encyclopedias are for. They just want to glance at an entry, get a sense of what the word they’re looking at means, and then get back to finishing their homework, love letter, or all-caps, keyboard-mashing screed.

You must also excise all potential double entendres from the book; they say that the best editors have a sharp, sharp eye and a filthy, filthy mind, and they are right. Editors are, at heart, twelve: if we can construe something as a fart or sex (or a fart and sex) joke, we will. This is a double-edged sword as you write verbal illustrations: the elevation of your adult duty is constantly pulling against the gravity of your native gutter thinking. Duty must prevail because duty ostensibly pays the bills, and so <I think we should do it> gets changed to <I don’t want to do that>; <That’s a big one!> becomes <That’s a big fish!>; <He screwed in the lightbulb> becomes nothing at all. After a while, you see double entendres everywhere: you remove every single verbal illustration at “member” and “organ” for obvious reasons, you scrutinize the verbal illustrations at “wind” extra hard just to make sure there’s no secret fart joke in there, and you wonder if you should science-up the verbal illustrations at “organism” because it’s just a few letters off from “orgasm,” and no doubt someone will read sex into <complex organisms>. (And yet here I must confess that I am aware of two good double entendres that have made it into our dictionaries and have stayed there. The first is in a paperback dictionary at the entry for “tract” and references a boob joke from Monty Python and the Holy Grail: <huge tracts of land>. The second is in our middle-school dictionary, at the entry for “cut....

You must avoid any hint of perceived bias anywhere in the verbal illustration. Your illustration for “conservative” that reads <the conservative party blocked the measure> will be read as saying that people who identify as conservative are obstructionist; change that to <he votes a straight conservative ticket>, and you’ll hear from people who think that the dictionary is trying to tell them how to vote. Even something you consider to be as innocuous as possible—<I love pizza a lot>—could end up garnering unwarranted criticism. Never mind that this illustrates “lot”: someone will write in to ask, “How can you love pizza; are you a pervert?”*3 Further, be aware that writing a verbal illustration like <tomorrow is supposed to be sunny> may be problematic, because there are supposedly people out there who will assume that if they look up “suppose” in the dictionary and read <tomorrow is supposed to be sunny>, that means tomorrow will, in fact, be sunny, and they will write in and complain when tomorrow is not sunny. According to one senior editor who worked on the Third, Gove himself prohibited these types of verbal illustrations on the grounds that the reader wouldn’t be able to distinguish between a dictionary and a Magic 8 Ball (my paraphrase) and would take us at our word. What if tomorrow isn’t sunny?

It is unfortunate that the entries that take up most of the lexicographer’s time are often the entries that no one looks at. We used to be able to kid ourselves while tromping through “get” that someone, somewhere, at some point in time, was going to look up the word, read sense 11c (“hear”), and say to themselves, “Yes, finally, now I understand what ‘Did you get that?’ means. Thanks, Merriam-Webster!” Sometimes, in the delirium that sets in at the end of a project when you are proofreading pronunciations in six-point type for eight hours a day, a little corner of your mind wanders off to daydream about how perhaps your careful revision of “get” will somehow end with your winning the lottery, bringing about world peace, and finally becoming the best dancer in the room. But nowadays, thanks to the marvels of the Internet, we know exactly what sorts of words people look up regularly. They generally don’t look up long, hard-to-spell words—no “rhadamanthine” or “vecturist” unless the National Spelling Bee is on TV.*3 They tend to look up words in the middle of the road. Some of the all-time top lookups at Merriam-Webster are “paradigm,” “disposition,” “ubiquitous,” and “esoteric,” words that are used fairly regularly but also in contexts that don’t tell the reader much about what they mean.

"You shall know a word by the company it keeps" (Firth, J. R. 1957:11)Apparently, not always an easy task. Also, makes me wonder whether rare words in language models could benefit from use of their dictionary definitions in a way. Maybe also there's a way to recognize that not all context are equally beneficial and some context could somehow be given more weight than others, in case it's not already done.

the smallest words, like “but” and “as” and “make,” are not looked up either. Most native English speakers know how to navigate the collocative waters of “make” or don’t need to figure out what exactly “as” means in the sentence “You are as dull as a mud turtle.” They recognize that it marks comparison, somehow, and that’s it. But that’s not good enough for lexicography.

Why do we say that someone’s “worth their salt”? Because in the ancient world salt was such a valuable commodity that we used to pay people in it (and this is why you also get a salary).

Peter Sokolowski remembers being in the hallway outside the pronunciation editor’s office one day and hearing from within the office a very measured voice say, as blandly as possible, “Motherfucker. Motherfucker. Motherfucker.” It was one of our old pronunciation editors, trying to get the intonation right for the audio file. He left a few years later to become a priest.

The linguistic process by which “nuclear” became \ˈnü-kyə-lər\ is called “metathesis,” where two phonemes within a word switch positions.*10 This is the process that gave us the standard pronunciations of “iron” (“EYE-urn” instead of “EYE-run”) and “comfortable” (“KUMF-ter-bul” instead of “KUM-fert-uh-bul”) and other nonstandard pronunciations like “PURR-tee” for “pretty.” Some lexicographers and linguists posit that “nuclear” underwent metathesis because there are not any other common English words that end in \-klē-ǝr\ (just “cochlear”),*11 but there are a good number of words like “molecular” and “vascular” that end in \-kyǝ-lǝr\, and the gravitational pull of these words and this more common speech pattern dragged “nuclear” into \-kyǝ-lǝr\ orbit. But the “how” is not as interesting to people as the “why,” and we don’t have a compelling “why.” Why do people who have gone through some of the most rigorous graduate programs in the world use a pronunciation that’s just not right?