Skip to content
How do we come up with words?
We live in an age in which you can Google, BlackBerry, blog, podcast and spam -- yet none of these words existed (at least in their current senses) just a few years ago.
The addition of vocabulary to the English language is, of course, nothing new. Every word in the dictionary was originally the brainchild of some wordsmith, lost in the mists of history, whose coinage caught on and was passed down the generations.
Words can be coined in several ways. Most new words are simply assembled out of old ones. We can figure out what a defragmenter is thanks to our familiarity with de-, fragment and -er. The last decade has also given us deshopping (buying something to use it once and return it), gripesite (where you post comments about deficient products) and green washing (in which companies cover up polluting practices with eco-friendly PR).
But where do the raw ingredients of words come from? The most obvious source, of course, is onomatopoeia -- when a word resembles what it sounds like, as in oink, tinkle, barf and woofer and tweeter. But onomatopoeia only applies to noisy things, and the resemblance is usually in the ear of the beholder. A more fertile source of new words is the phenomenon called phonesthesia, "the feeling of sound," in which snippets of vowels and consonants vaguely remind people of something because of the way they are pronounced.
Many words beginning with sn-, for example, have something to do with the nose, because you can almost feel your nose wrinkle when you pronounce it. They include words for things associated with the nose (sneeze, sniff, snore and Snuffleupagus) and for looking down your nose at someone (snarky, sneer, snicker, snippy, snooty). Another example: cl- for a cohesive aggregate or a pair of surfaces in contact: clam, clamp, clap, clasp, cleave, clench, cluster, etc.
Why do words that share a teeny snatch of sound also sometimes share a teeny shred of meaning? These clusters grow from a nucleus of similar words that have coalesced for any number of reasons. They may be fossils of a linguistic rule that was active in an earlier period, or in a language from which the words were borrowed, or they might arise by sheer chance. But once similar words find themselves rubbing shoulders, they can attract or spawn new members owing to the associative nature of human memory.
We can infer that phonesthesia was the source of recent words like bling, bungee, glitzy, glom, gonzo, grunge, humongous, scuzzy, skank and wonk. They are not built out of preexisting parts like prefixes, suffixes and roots, and their sounds either remind people of their referents (as in bungee and glom) or vaguely resemble words with related meanings (as in glitter, glamour and ritzy for glitzy, or scum, scuff and fuzzy for skuzzy).
What kinds of things call out to be named? New words, one might guess, should materialize to name a concept that people need to talk about. That's why every hobby and profession quickly develops a jargon. Even casual computer users command an impressive lexicon of new technical terms like modem, reboot and upload. And in an age that professes to treat women and men as equals, what would we have done without Ms.?
But strangely enough, many concepts we long to name remain stubbornly nameless. We still don't have a good word for unmarried romantic partners, or for the current decade, nor a gender-neutral pronoun to replace "he or she." And wouldn't it be handy to have a word for a fact you can learn a hundred times without remembering it, or the early morning insomnia in which your bladder is too full to allow you to fall back to sleep but you're too tired to get up to go to the bathroom?
Every year, the American Dialect Society predicts which new words will catch on. But a follow-up of their picks from the 1990s shows they are about as accurate as tabloid psychics. Some of the words were political barbs that died with the careers of their targets (the verb to gingrich). Others were bets on the wrong name for an innovation, like W3 (that's what they thought we'd call the World Wide Web), information superhighway (waytoo Al Gore) and Infobahn (yuck).
Though most whimsical neologisms go nowhere, others, mysteriously, can take root. Blog, from "Web log," taps the amorphousness of blob and bog and insouciantly cuts a word against its syllabic grain in the style of 1970s slang like shroom (mushroom), strawb (strawberry) and burb (suburb). Earlier decades gave us yuppie (young urban professional, a play on hippie, yippie and preppy), couch potato, palimony, qwerty (technological inertia, from the keys on the top row of a typewriter) and, the strangest of all, spam.
This technical term for bulk e-mail is not a metaphor for something cheap, plentiful and unwanted. It was inspired by a Monty Python skit in which a waitress recites a menu: "There's egg and spam; egg bacon and spam; spam bacon sausage and spam; spam egg spam spam bacon and spam," and so on. The mindless repetition inspired 1980s hackers to use spam as a verb for flooding newsgroups with identical messages. A decade later, it spread to the populace.
Silly coinages are not a new phenomenon. Soap opera is from the 1930s, hot dog from the 1890s (from a campus joke about its alleged ingredient) and gerrymander from an 1812 cartoon showing a political district that had been crafted by Gov. Elbridge Gerry into a tortuous shape resembling a salamander.
With most new words, it's up to the community of English speakers -- not the word's coiner -- to decide whether a word catches on. But there's one kind of naming in which the namer has a big say: naming a baby. Yet even here, chance plays a big role, as we see in the changing fashions in baby names -- the reason you can guess that Ethel is a senior citizen, Linda is a baby boomer, Jennifer a thirtysomething and Chloe a child.
Many people assume these fads are inspired by celebrities (Marilyn Monroe made Marilyn popular) or social trends (biblical names are popular during religious revivals; androgynous names are a legacy of feminism). But sociologist Stanley Lieberson has pored through naming data and disproved every one of these hypotheses. The cause of baby names is other baby names. Parents have an ear for names that are a bit distinctive (as if to follow Sam Goldwyn's advice not to name your son William because every Tom, Dick and Harry is named William) without being too distinctive (only celebrities can get away with naming their children Moon Unit or Banjo). The trends arise when everyone tries to be moderately distinctive and ends up being moderately distinctive in the same way.
Pundits often treat a culture as if it were a superorganism that pursues goals and finds meaning, just like a person. But the fortunes of words, a cultural practice par excellence, don't fit that model. Names change with the times, yet they don't fulfill needs, don't reflect other social trends and aren't driven by role models or Madison Avenue. A "trend" is shorthand for the aggregate effects of millions of people making decisions while anticipating and reacting to the decisions made by others, and these dynamics can be stubbornly chaotic.
This unpredictability holds a lesson for our understanding of culture more generally. Like the words in a language, the practices in a culture -- every fashion, ritual, common belief -- must originate with an innovator, must then appeal to the innovator's acquaintances and then to the acquaintance's acquaintances, until it becomes endemic to a community. The caprice in names suggests we should be skeptical of most explanations for other mores and customs.
Steven Pinker is a professor in the department of psychology at Harvard University and author of "The Stuff of Thought: Language as a Window into Human Nature."