You are currently browsing the category archive for the ‘languagelog’ category.

Suppose you were reading and came to the following line:

“She kept her head and kicked her shoes off, as everybody ought to do who falls into deep water in their clothes.”

Would you …
(a) continue reading, because that’s a perfectly acceptable sentence, or
(b) throw a tantrum and insist that the author is an imbecile speeding the wholesale destruction of the English language?

If you’re a regular reader of this blog, you’re probably answering (a). If you’re answering (b), I regret to inform you that you hate the writing of C. S. Lewis.

And if you’re the sort to answer (b), the sort of person who rages at the alleged grammatical buffoonery of your fellows, I’m sure it’s because you think you’re doing us all a favor, and that your condescending tone is justified because: first, you’re being helpful regardless of the tone you’re using; second, people only learn through negative conditioning, and so it is your duty, however unpleasant, to rub their noses in it to keep them from going on doing it; third, only a truly illiterate mouth-breather would be so moronic as to make such a mistake, and such imbeciles are below contempt and probably don’t even realize that you’re condescending to them anyway; and fourth, given the Heruclean effort you’ve put into learning the English language as impeccably as you did, it’s really only fair that you get to be a little self-satisfied and perhaps even gloat a smidge.

The only problem with this view is that all you’ve managed to learn about English is how to get your brain to release some satisfying endorphins every time you blindly regurgitate some authority figure’s unjustified assertion. You’re not helping; you’re just getting someone to pretend to agree with you long enough to shut you up. Or worse, you’re scaring people into submission to a point where they feel compelled to preface their speech with apologies for any unknown violence their words are committing against the presumed propriety of the language. Never forget, though, that language is the people’s. Your witless superstition will, by-and-large, be ignored by the speakers of the language, and the alleged impropriety will almost certainly win out in the end. Don’t mistake yourself for a brave defender of our language against the barbarians at the gates when, in truth, you’re nothing but a millennialist shouting about the end-times of the English language. Meanwhile, the world spins on, and the language flourishes, hale and hearty.

One great example of this situation is the shouting down of those who use singular they.  I’ve wanted for some time to have one place to send everyone who complains about singular they, a single page that can debunk whatever junk they’re peddling against it. There’s been lots of great stuff written about why singular they is acceptable, but every time I want to smash the arguments against it, I have to waste time jumping through old Language Log posts and books and whatnot, so I figured I’d finally go about summarizing it all. Without further ado, here’s the evidence for singular they, and why you ought to stop “correcting” it.

Historical usage: Geoffrey Chaucer is widely credited as the father of English literature. He was one of the first well-known authors to write in Middle English instead of the prevailing literary tongue, Latin, bringing legitimacy to the language. And, what’s this? Why, it’s a line from The Canterbury Tales, ca. 1400:

And whoso fyndeth hym out of swich blame,
They wol come up […]”

It’s a little hard to tell in the Middle English, but whoso is a quantified expression, like whoever, that is syntactically singular, but then is paired to the syntactically plural they. So, since at least the beginnings of literary Middle English, 600 years ago, it’s been all right to use singular they. It’s been consistently attested since then; Henry Churchyard reports examples from the Oxford English Dictionary in 1434, 1535, 1643, 1749, 1848, and a wide variety of years in between. There has literally been no point since 1400 when singular they went unattested in contemporary English.

Usage by good writers: Lest one counter the historical point by claiming that it was a mistake or an illiterate usage, it should be noted that singular they has been employed by revered writers throughout its history. A list of examples from some such authors (including Chaucer’s and C. S. Lewis’s quotes above) is available on Churchyard’s site. Among the luminaries: Lewis Carroll, Walt Whitman, George Eliot, Shakespeare, William Thackeray, Jane Austen, and Oscar Wilde.   The Merriam-Webster Dictionary of English Usage has still more examples for those who prefer their empirical data to be overwhelming.  And, if you subscribe to Mark Liberman’s one-liner “God said it, I believe it, that settles it,” you’ll be interested to see that the King James Version, along with the Tyndale, Bishop’s, and Geneva Bibles, along a range of other versions of the Christian Bible all employ singular theys.  (I’m not sure of the stance of non-Christian religious texts. I imagine no religion has a commandment disavowing singular they, but I have not studied comparative religion.)

Acceptance by authorities: So it’s historically attested, with usage by great writers. “But great writers are fallible!”, cries the grammaticaster*, ignoring the implication in this that the grammaticaster is substantially more aware of the rules of our language than its best writers. “Grammatical authorities agree that singular they is a barbarism!”

This appeal to imagined authority wouldn’t be convincing regardless, but it rings especially hollow when you realize it’s patently false. Certainly many prescriptivists assert that singular they is an affront to the language. Some even put it in books. Eric Partridge, for instance, says it’s so in Usage and Abusage, supplying exactly no argumentation for his opinion.

But The New Fowler’s, 3rd Edition, which carries on its front cover the subtitle “The acknowledged authority on English usage”, takes a neutral-to-positive stance on singular they, calling the issue “unresolved” but noting that it “is being left unaltered by copy editors” and that aside from pedants, “such constructions are hardly noticed any more or are not widely felt to lie in a prohibited zone.” [p. 776] (This is an especially interesting stance because it goes against Fowler’s own original position from 1926.) Grammar Girl also comes down unambiguously in favor it, if she’s your cup of tea.

Some old style guides even saw the light a century ago. An English Grammar by Baskervill & Sewell, originally published in 1896, states that while he is preferred to singular they in general, they is “frequently found when the antecedent includes or implies both genders. The masculine does not really represent a feminine antecedent […]” (Italics in original.) Further, as an exercise, they give examples of singular they, and tell the reader, “In the above sentences, unless both genders are implied, change the pronoun to agree with its antecedent.” (Again, italics in original.)

There was even an article in Robert Hartwell Fiske’s fervently prescriptivist Vocabula Review arguing for singular they. The money quote: “We have seen that history is not on the side of those who would ban singular they from written texts; neither is logic; nor is majority rule.” If you needed an authority figure to tell you that singular they was all right, well, I hope you might find it harder to find one against singular they.

Singular/plural syntactic disagreement: Then, of course, there’re the self-styled logicians who say that they can’t be used with an indefinite pronoun like everybody because they have different numbers.  After all, you say they are but everybody is, and so that proves it.  A moment’s reflection shows that this argument is fallacious, especially if in that moment’s reflection you think of a sentence like

(1) My family stops by regularly, and they always bring pizzas.

My family is syntactically singular in American English, as seen in the conjugation of stopsThey is syntactically plural, as seen in the conjugation of bring.  And yet, (1) is a well-formed sentence, and the other option (“My family stops by regularly and it always brings pizzas”) sounds absurd.  The key point here is that it’s not the syntactic number, but rather the semantic number that matters.  And everybody is semantically plural, just like they.  Don’t believe me?  Consider this trio from Geoff Pullum:

(2a) Everybody knows each other.
(2b) They know each other.
(2c) *He knows each other.

Each other is a reciprocal pronoun that requires a plural antecedent, or in non-linguistic terms, whoever each other refers to has to be plural.  So it works in (2b), where it can refer to the semantically plural they, and it doesn’t work in (2c) with the semantically singular he.  Since (2a) is a grammatical sentence, we know that everybody can be semantically plural.  Since everybody can be semantically plural, we know that there’s nothing wrong with using they with it.  And, as we’ll see in the next section, this agreement only matters if you insist that everybody and they have a pronoun-antecedent relationship, which probably isn’t the right way of looking at it.

It’s not really a pronoun relationship anyway: The above argument supposes that they is a pronoun referring to a syntactically plural but syntactically singular quantified expression like everybody.  But what if you’ve got a semantically singular one like anybody? Is it essential that they and the quantified expression agree in number at all?  Steven Pinker argues that it isn’t:

The logical point that everyone but the language mavens intuitively grasps is that everyone and they are not an antecedent and a pronoun referring to the same person in the world, which would force them to agree in number. They are a “quantifier” and a “bound variable,” a different logical relationship. Everyone returned to their seats means “For all X, X returned to X’s seat.” The “X” is simply a placeholder that keeps track of the roles that players play across different relationships: the X that comes back to a seat is the same X that owns the seat that X comes back to. The their there does not, in fact, have plural number, because it refers neither to one thing nor to many things; it does not refer at all.”

And that’s the weird thing.  Here’re these pedants crying about how English has to adhere rigidly to logic, and they don’t notice the one time the language actually behaves like a system of formal logic.  The point is that singular they can behave non-referentially; it’s an entirely different word from the standard referential pronouns he or plural they in these cases.  In fact, Pinker notes that some other languages have different words for the two meanings.  Since this they doesn’t pick out any specific entity or entities, it functions like the variable x in the mathematical expression 2(x + 7).  Can he be used in the same way as they, as a bound variable?  Sure, but that leads to the next point.

he isn’t gender-neutral: Some claim that singular they is unnecessary because he is gender-neutral, and that this whole kerfuffle about singular they being in any way good or useful only came about when “arrogant ideologues began recasting English into heavy artillery to defend the borders of the New Feminist state.” That’s from an article in The Weekly Standard by David Gelernter, a computer science professor at Yale. See,

“Ideologues can lie themselves blue in the face without changing the fact that, to those who know modern English as it existed until the cultural revolution and still does exist in many quarters, the neutral he ‘has lost all suggestion of maleness.'”

Yep, back before the evil, scary cultural revolution of the 1970s, no one ever saw anything odd about gender-neutral he.  And we see this by the fact that back in 1896, when women couldn’t vote in the U.S., Baskervill and Sewell thought that he sounded weird with mixed company. And we see evidence in the fact that singular they has been used since Chaucer’s time.  No, wait, that’s the opposite of his claim!  Nuts!

If you really think that he is gender-neutral, you ought to find nothing wrong with the following sentences:

(3a) At the funeral, everyone was dressed to the nines, each wearing his swankest tie or nicest dress.
(3b) Is it your brother or your sister who can hold his breath for four minutes?

Geoff Pullum came up with (3b), and I think it’s the clincher. I just can’t picture any competent speaker of English saying it and thinking it correct.  Sometimes it might be the case that he is approximately gender-neutral, but it’s not so in the general case.  There are many such examples where he sounds bad compared to a truly gender-neutral pronoun.

Equal ambiguity: Some others, often members of the “Don’t start a sentence with since!” set, complain that another problem with using they with a quantified or generic expression is that it introduces ambiguity. For instance, who does they refer to in

(4) Everyone meeting the royal family said that they were gracious?

Yes, that’s ambiguous as to whether the visitors or the royal family were gracious. Yes, replacing they with he removes the ambiguity. But what about

(5) Everyone meeting the new principal said that he was gracious?

What’s this? He has led to an ambiguity?  Inconceivable!  Note that (5) wouldn’t be ambiguous with a singular they.  Like the Oxford comma, sometimes singular they introduces an ambiguity, but just as often it avoids an ambiguity. Ambiguity is par for the course with pronouns with multiple referents, anyway:

(6a) Bob asked Jim if he was fat.
(6b) The Romans befriended the Gauls, but they slew them.

These sorts of ambiguities are common, even in edited writing, because the surrounding sentences give context to the ambiguous sentence.  Pilgrim’s Progress, for instance, one of the most prominent books in English literature, has almost 40 examples of “they * them” (e.g., they overtook them, they seek to stifle them).  That’s a lot more examples than one would expect if this sort of ambiguity were so crippling.  So ambiguity in singular they isn’t a deal-breaker either.

Summary: You don’t have to use singular they yourself.  You can go ahead and re-work your sentences to avoid it. You can employ he or she, or s/he, or a made-up gender-neutral pronoun of your own devising like xe.  You can even just stubbornly plow on, using he as a gender-neutral pronoun until you grow tired of people pointing out that it isn’t really.  I don’t care, and you’re not grammatically wrong.  But you’re just making a fool of yourself when you go around telling users of singular they that they’re wrong, because they’re not.

*: Grammaticaster, by the way, is one of my new favorite words, learned from the book Dimboxes, Epopts, and Other Quidams. It refers to a “petty, self-styled export on grammar, usually a niggling, precise type who can stab a bony finger at a dangling participle or split infinitive but lacks a true appreciation of writing in all its riches and varied styles. The rule-conscious pedant who sees writing not as good or bad but as right or wrong.” Or as the OED more briefly puts it, “A petty or inferior grammarian. (Used in contempt.)”

**:The information above was compiled from a number of sources, most of which are mentioned in the post, but here’s a few others that I found useful and may or may not have linked to above:
Grammar Girl: Generic Singular Pronouns
Language Log: Shakespeare used they with singular antecedents so there
Language Log: Singular they with known sex
Language Log: “Singular they”: God said it, I believe it, that settles it.
Language Log: Lying feminist ideologues wreck English, says Yale prof
The Lousy Linguist: Singular ‘they’ is old, logical, and grammatical
Wikipedia: Donkey pronoun

I am not the sort of person who receives an inordinate number of invitations, likely due in no small part to my propensity to swing conversational topics away from things like popular movies or good books and over to the specifics of the language by which one talks about such things. As such, it is not in the cards for me to be picky about the tenor of an invitation. I never understood the people who refuse to go to a party because they were invited at the last minute. My response is always, “I’ll be ready in three minutes, thanks thanks thanks.” This may be because I was — and this may surprise some of you — not one of the popular kids in high school.

Okay, actually, I’m lying. In truth, I am picky about the invitations I accept, just because many of the things that my friends enjoy doing hold no inducement for me. Bars, dancing, sunny day beach trips, all not my cup of tea. Unless there’s cheap food or a thrift store involved, I’m out. But when I reject an invitation, I always have a valid reason: it sounds boring. Some other people do not; instead they complain about the fact that they have not been given an invitation, but rather an invite. This is because those people assume invite is either just a recent truncation of the full and more proper invitation, or the recent co-opting of the verb invite into a noun. In either case, it’s unacceptable. As Eric Partridge writes in Usage and Abusage:

invite for (an) invitation is incorrect and ill-bred and far too common”

A sharp dismissal. Except, wait, what the hell does it mean for a word to be “ill-bred”? The only meaning I can come up with is that the word was formed through improper means. But that’s patently false, as nominal invite comes from verbal invite by the same means as some uncontroversial nouns like command and request, both of which started life as verbs according to the MWDEU. In fact, this method (zero-affixation) of forming nouns from verbs used to be quite commonplace.  Arnold Zwicky has found that nominal request took the place of nominal ask, which first showed up a millennium ago.  Adam Albright found the following words in the OED as nouns:

adorn, disturb, arrive, destroy, relate, pray, recede, announce, ask, think, amaze, depart, reduce, produce, maintain, retain, detain, deploy, retire, acquit, greet, defend, divulge, startle, entertain, vanish

The attestations of these are all in the past; it’s likely few people would consider all (or even many) of these valid nouns nowadays. But I think it gives some evidence that invite isn’t ill-bred; it’s attested back to the 1600s in the OED, and it was formed by what used to be a pretty productive rule. So it’s not incorrect, it’s not ill-bred, and since neither of the first two hold, there’s no reason to complain about its commonness.  Sorry Eric Partridge, but zero-for-three.

Now, there does seem to be some truth to the claim that invite is less formal than invitation; the MWDEU’s historical examples of nominal invite are often from the mouths of lower-class characters or light writing. But being informal is not the same as being bad grammar, no matter how badly the prescriptivists want that to be the case.

Summary: Nominal invite, as in I got an invite, isn’t a recent piece of bad grammar. It’s been attested since the 17th century and it came from a previously common grammatical rule. At worst, it’s informal. I’d use invitation if you don’t feel like a fight, but when you’re in a bad mood, use nominal invite and tear into anyone who dares object.

Ben Zimmer has once again written a cutting post about Global Language Monitor, its absurd claim that the English language is about to get its millionth word, and the news sources that blindly regurgitate GLM’s warmed-over press releases about that.   I know it’s become cliche, upon reading an article that one disagrees with, to ask “So this is what passes for journalism these days?”   But articles like the BBC’s really demand that question. Here’s another, from the Telegraph, touting an obviously false claim: “One millionth English word could be ‘defriend’ or ‘noob’.”

First off, to the reporter’s credit, he manages to answer one question about GLM’s methodology; a word is a word by their count once it has been attested 25,000 times “by media outlets, on social networking websites and in other sources.” This information is not available on GLM’s website — I searched for 25,000, 25000, “twenty-five thousand”, “twenty five thousand”, “twentyfive thousand”, and “25 thousand” on the GLM website and didn’t get a single hit.  So kudos to the reporter for getting this nugget out!

But then the whole enterprise falls apart. The article notes that among the words GLM is “currently monitoring which could take English to the one million threshold” is noob. If that’s the case, then GLM’s monitors are incompetent.  I popped over to MySpace, which surely would be included in any reasonable list of social networking sites, and lo! 145,000 hits. It’s already a word by GLM’s arbitrary standard!  Who is GLM using to monitor the social sites? Clearly they ought to be fired. If noob, which has been in wide use by computer folks since the turn of the millennium, managed to slip under their nose, think of how many other unnoticed words there are! For all we know, English might have already passed this made-up milestone a month ago!  To call this possibility a tragedy is an unacceptable understatement.  And the claim that noob hadn’t been yet used 25,000 times on the Internet — where it was born all those years ago! — didn’t set off any alarms at the Telegraph?

How credulous can one be? Here’s the lead paragraph of the Telegraph article:

“The milestone will be passed at 10.22am on June 10 according to the Global Language Monitor, an association of academics that tracks the use of new words.”

And the last paragraph:

“The organisation first predicted that the millionth English word was imminent in 2006, and has repeatedly pushed back the expected date. Other linguist[s] have expressed scepticism about its methods, claiming that there is no agreement about how to classify a word.”

Of course if the first guess was only off by three years, it’s totally reasonable to assume the current guess is off by less than a minute.

Also, “other linguists” implies that Paul Payack is a linguist. He is not. I’m not even convinced he or his merry monitors can be called academics. They are entrepreneurs at best, and they are peddling nothing worth acknowledging.

I haven’t posted anything in a while because it was the end of the quarter and, even without any classwork to speak of, I had to get a few components of my research together before spring break. And now I’m on break, so I’m having trouble putting together the energy to concoct a proper post. However, there are four other posts that I found semi-recently that I was so very fond of that I had to share them with you all.

The first is Mark Liberman’s Language Log post “Teaching Zombie Rules“, which offers a potential answer to the problem I find myself in quite a lot: how should one deal with grammar rules that aren’t really rules?  Sure, it’s an easy question once you’re a professor or even a grad student.  I use the grammar that I believe to be best justified, and if anyone tells me I’m wrong, I present the facts that back up my usage.  If a pedant insists I’m wrong, it doesn’t matter, because they don’t hold any power over me.  But what if you’re a student preparing for a grammar test that includes zombie rules (the SAT, for instance)?  Even worse, what if you’re a tutor preparing someone else for a grammar test?  How do you teach a rule you know to be wrong?  Liberman’s answer is great, in part because it recasts the problem in terms of audience design.

The second is this year’s Grammar Day post from John at Bradshaw of the Future.  John points out that all these grammar points that we all care so much about are just insignificant pieces of the whole.  The core of English (or any other language’s) grammar is essentially the same across all its users.  A few people saying “between you and I” isn’t going to change the fact that English is an Subject-Verb-Object language or that it has singular and plural morphology, but not dual morphology (as in American Sign Language).  This is why you shouldn’t get up in arms about the horrendous English these kids today speak — virtually everything they say is grammatically correct anyway.  (John has a history of good Grammar Day posts; last year’s was a gem as well.)

The last is a two-pack: Arnold Zwicky’s Grammar Day post from last year on Language Log, and this year’s version on his own blog.  There’re a lot of good points in these posts, but I’m just going to mention the minor one that prescriptivists have this infuriating tendency to constantly couch their opinions in light absurdity so that when someone complains that their beliefs are ill-founded, they can point to the absurd part and saying “Can’t you tell I’m joking?!?!?”  It’s like when you’re talking to someone about their spouse and they growl, “Sometimes I just want to wring his/her neck,” and then after they stare into the middle distance for a second, they sort of chuckle.  Sure, they’re probably chuckling about the absurdity of the statement, but then again, you have to wonder if they were if they were really chuckling at the mental image.  So too with prescriptivists; I think they think they’re joking, but having dealt with them and occasionally incurred their wrath, I’m not so sure they are.

I hope you enjoy those links as much as I did.

Post Categories

The Monthly Archives

About The Blog

A lot of people make claims about what "good English" is. Much of what they say is flim-flam, and this blog aims to set the record straight. Its goal is to explain the motivations behind the real grammar of English and to debunk ill-founded claims about what is grammatical and what isn't. Somehow, this was enough to garner a favorable mention in the Wall Street Journal.

About Me

I'm Gabe Doyle, currently a postdoctoral scholar in the Language and Cognition Lab at Stanford University. Before that, I got a doctorate in linguistics from UC San Diego and a bachelor's in math from Princeton.

In my research, I look at how humans manage one of their greatest learning achievements: the acquisition of language. I build computational models of how people can learn language with cognitively-general processes and as few presuppositions as possible. Currently, I'm working on models for acquiring phonology and other constraint-based aspects of cognition.

I also examine how we can use large electronic resources, such as Twitter, to learn about how we speak to each other. Some of my recent work uses Twitter to map dialect regions in the United States.



@MGrammar on twitter

Recent Tweets

If you like email and you like grammar, feel free to subscribe to Motivated Grammar by email. Enter your address below.

Join 975 other followers

Top Rated

%d bloggers like this: