If you want to find out what a word means, you go to the dictionary. But (as we’ve already established) there is no “the dictionary.” Any idiot can publish a dictionary, or, since only an idiot would publish a dictionary nowadays, any idiot can make a dictionary website. Scan in a public domain Webster’s and add a few modern words such as snark and chunder, and if you do a better job of concealing your tracks than say Urban Dictionary does, you are on equal footing with Funk and Wagnalls.
Upon this trembling foundation all our laws and contracts, not to say all our literature and conversation, rest. Language is an anarchic system; no one is in charge of it.
Of course, 60% of statements one hears or reads can be boiled down to “I’m in charge,” so it’s no surprise that many different people think they are in charge of language. There are many reasons why people might call dibs on telling you how to speak: They might want to make their lives easier or they might feel the cruel but omnipresent need to exercise control over you. If they think that language has a “right” way of being spoken, based on traditional rules of grammar, we tend to call them prescriptivists.
This distinction is most often applied to linguistics, although of course it can be extended to many fields of study: a descriptivist looks at language as it is and attempts to analyze and understand it; a prescriptivist looks towards language as it should be, and attempts to reform our wayward tongues.
Some people think a dictionary should be descriptive, and reflect the way people talk (and include, for example, the Chris Traeger definition of literally); some people think dictionaries should be prescriptive, and reflect a standard of English “at its best.”
(I should stress that I’m speaking of English throughout. The French, to take one example, have laws that state that the government is in charge of their language, and the government has weapons and stuff; but things are different in other countries, including most English-speaking ones.)
I’d like to look more closely at this division, because I think things are much more complicated than we usually acknowledge (far more complicated than we have room for here, but I’ll do what I can). One of the many paradoxes inherent in this dictionary model is that people who want the dictionary to be prescriptive are behaving descriptively: this is an accurate description of what dictionaries have traditionally been for, for hundreds of years now; while those who want the dictionary to be descriptive are behaving prescriptively: they are prescribing what a book should be in contradistinction to what it usually has been.
Academic linguists are both the most passionate supporters of description over perscription, and the most ardent in correcting anyone who calls a polyglot a “linguist.”
If you have ever blanched at the Traeger literally or if you would say that prescription was misspelled in the previous paragraph, you are behaving prescriptively.
If a prescriptivist has ever corrected you for saying “ten items or less” and you rolled your eyes and replied, “Everyone says it that way, no one says ‘ten items or fewer,'” you are behaving descriptively. It’s probable that most of us have hedged our bets and have some money on each horse.
One problem is that a prescriptivist is prescribing towards a standard that is by necessity arbitrary (almost no one outside of the tinfoil-hatted neoneoplatonists believe language is not 99% arbitrary), and is therefore contentious. No two prescriptivists agree on which of the arbitrary rules must be followed. Most of our usage panel agree that splitting an infinitive is fine, but generations of Miss Grundys have objected.
I have a friend who scrupulously avoids ending sentences in prepositions, no matter how tangled the required syntax, but consistently confounds its and it’s. I don’t want to make the perfect the enemy of the good, but by my lights this is a flagrant misapplication of resources.
Another prescriptive problem is that language, unlike Augustine’s Law, is custom, and custom changes. If you go by early twentieth-century etiquette books, the most vulgar word imaginable, the one that instantly makes you sound like an uncultured moron, is drapes (the modern equivalent would be irregardless). Fowler, in his Modern English Usage (1926), calls the words couch and close (verb) “genteelisms,” and insists that non-pretentious people would say sofa and shut. I respect Fowler, and I am unhealthily obsessed with Emily Post, but none of this resonates with me in the least, and I would be hard pressed to find anyone who agreed that drapes and couch had class connotations.
Since thou speakest not thus, like Stan Lee’s Thor, clearly language changes, and no prescriptivist demands our speech be the speech of Beowulf, let alone proto-Indo-European; but where to draw the line on acceptable change is always going to be controversial. Probably not one in ten native English speakers can spot the error in Winston’s ad slogan, but in 1954 it was egregious enough to cause a sensation, and Walter Cronkite refused to say it on the air: not because it was a cigarette (Fred Flintstone hawked Winston to kids in those days) but because he categorically refused to use the preposition like as a subordinating conjunction. (Ogden Nash quipped: “Like goes Madison Avenue, like so goes the nation” to commemorate people’s growing switch to a conjunctive like.) It may be as old-fashioned as our snobbery about drapes and our antisnobbery about couches, but the like/as distinction is still a bugbear of formal writing.
Rules are of course, easier to create than to obey, which is why we all spend our time trying to think of reasons why we’re exempt (watch John Scalzi do it here). The most notorious example is the passive forbiddance of passive voice, which is alarmingly common, so I just selected one example at random, from Real World Document Drafting: A Dispute-Avoidance Approach by Marvin Garfinkle:
So much trouble, you may be muttering. Why not be descriptive and make an end of it?