There was a time in American English when words were like contraptions you could jerry-rig: a bit from here, a bit from there, all bolted together to make a pseudo-classical bit of hickery. Absquatulate and copacetic are two classics – though copacetic is the newer by nearly a century, showing up circa 1910. Absquatulate hit the scene around 1830. So did contraption – a pseudo-classical construction with a trap stuck in it. And so did conniption.

You know what a conniption is, right? It’s a fit: a fainting fit or a hissy fit or some other pique or fright. We often see the redundant (but assonant) phrase conniption fit – in fact, that’s how it shows up in the earliest attested uses.

I think conniption has a good sound; that nip in the middle is fittingly indignant but short; the gathering con could call on confound and condemn and consarn (a fake-swear probably based on concern and usable where one might use goshdarn), and the ption ending brings out not only contraption but corruption, consumption, and conscription – and eruption and exception, among others. And just maybe, the word as a whole has an air of a sneezing fit.

So where did conniption come from? Um, the US… around 1830… and no one’s really sure of anything more than that. There’s speculation, of course, but not even a whole lot of that. It was confected; it fit well; it stuck. If you look at Google ngrams, you’ll see ebbs and flows over the decades.

Mind you, you’ll also see results from the earliest 1800s. Have a look and you’ll find hits like these:

“there are a thousand occasions in which it breaks through its original conniption” —1803

“The economy of injustice is, to furnish resources for the fund of conniption” —1807

“who would be the avengers, not the abettors of conniption” —1811

“the moral conniption of our first parent has been entailed on his whole posterity” —1811

This is rather entertaining. But if you click through and look at the actual photo facsimile, you will find what you may have already guessed: it’s due to bad optical character recognition. This conniption is in every case a corruption… of corruption.

Tsk. It’s enough to give one a conniption.

When intransitives go transitive

This article was originally published on BoldFace, the official blog of the Toronto branch of the Editors’ Association of Canada.

We’ve all learned that there are two kinds of verbs: transitive and intransitive. Transitives take a direct object—“I fry an egg”—and intransitives don’t—“My stomach aches.” But that’s not the whole story. In fact, it’s not actually quite right.

For one thing, there are also impersonal verbs (“It seems to me,” “It rained”), which don’t even have proper subjects, just empty pro forma its.

For another thing, there are different kinds of intransitive verbs. Linguists divide them into unergative, where the subject really is the one doing the thing, and unaccusative, where the subject is treated as being on the receiving end of the action and can be modified by the past participle. We see from the guests are departed and the departed guests that depart is unaccusative; run, on the other hand, is unergative—you can’t say the run horse.

There are also verbs that change from intransitive to transitive or vice versa—several kinds of them. We don’t always think about them. In fact, some details of them are still being argued about by linguists.

I think it’s time for a quick field guide to these changeable verbs, complete with their overstuffed technical names.

Agentive ambitransitives

Some verbs can name the object of the action or not, but they always say who or what is doing the action (i.e., what is the agent). Read is one of these: “What are you doing?” “I’m reading.” “Reading what?” “I’m reading this article on grammar.” These are the nice, simple ones, and we don’t need to worry about them. But worry, now… yes, that verb can worry us a bit more, or we can worry it.


With worry, the object when it’s transitive—“That worries me”—is the subject when it’s intransitive—“I worry about that.” Another one of these is break: “I broke the window,” but “The window broke” and “The window is broken.” And if “I fry an egg,” then “The egg is frying.” Do those look like the unaccusatives I just mentioned? Some say that’s what these are. But some linguists argue that these aren’t true unaccusatives, precisely because they have transitive variants. True unaccusatives, like come and arrive, can’t be used this way. So what do we call these ones? Ergatives (from a Greek root for work). Well, some of us call them that, anyway.

Some people call some of these middle voice. Take for example shave: “The barber shaved me” or “I shaved myself”; “I shaved” means “I shaved myself” and “The barber shaved” means “The barber shaved himself.” Why middle voice? Because it’s not exactly active and it’s not exactly passive—or, we could say, it’s both at the same time.

Preterite causatives

Our real favourites, though (if by “favourites” we mean “favourites to get exercised about”) are a set of verbs that express transitive causation by using the past tense of the intransitive form. We don’t make new preterite causatives anymore, but we have some lying around… not laying around.

Yes, lay is one of these. “I lie down today,” “I lay down yesterday”; “Now I lay me down to sleep” (reflexive), and “I lay down the law of grammar” (transitive). We wanted something to express “cause another thing to lie down,” and we just used our past tense of the intransitive for the present of the transitive (and then made a new double past from that: lay gets a d to be laid). I’m sure many of you wish we hadn’t.

Another one like this is fell. This isn’t an ergative—if it were, you could have “I am felling the tree; the tree is felling.” Nope. “The tree falls,” “The tree fell”; “I fell the tree today,” and “I felled the tree yesterday.”

Cognate object constructions

There’s one more especially fun case: verbs that are intransitive—and in some cases always and everywhere intransitive and never taking an object—except when the object is a nominalization of the verb. You die, and you don’t die something, but you can die a death. You can die the death of a hero; you can die a happy death or a sad death. Likewise, you can smile, and you can’t smile me and I can’t smile you and neither of us can smile our faces (not in standard English, anyway), but we can smile a smile. I can smile an aimless smile that hovers in the air and vanishes along the level of the roofs (to steal from T.S. Eliot). And then perhaps you can smile that same smile.

What do we call these? What we probably should call them is a term Iva Cheung made up for them: self-transitives. But in case you haven’t noticed, linguists sometimes like ugly terms a bit too much, and so it turns out that the technical term for this sort of thing is cognate object construction, because the object has to be cognate (coming from the same source) with the verb. I wouldn’t blame you for preferring Iva’s term, though.

cymotrichous, leiotrichous, ulotrichous

When I was a little kid, certain adults would tell me to eat the crusts on my bread because they would make my hair curly.

This did not make me want to eat the crusts on my bread.

Seriously, what was so much better about curly hair? I was perfectly happy with my hair, which was fine and straight. (It still is, though I have since discovered that if I grow it to 24 inches it develops a whorl at the bottom.)

Nonetheless, one time at age 4 or 5 when I was at the home of a friend of mine, her mother was curling her hair and asked if I would like a curl, and when I said yes she put a curl right in the middle of the top of my head. I think it lasted a few days. I probably looked like a soft-serve ice cream cone.

Some words are like that curl: unnecessary ornaments used just because someone thinks they will look good: “A longer, hairier word would go better here.” I have nothing against ornamental words, of course – I have a massive collection of them – but I also don’t think they are intrinsically better. There is no prima facie reason to think that a polysyllabic Latin-Greek confection is a truer, more accurate name for a thing than two syllables of Anglo-Saxon. But words are known by the company they keep, and some words just look like they belong to the best clubs.

Today’s triplet of words are a veritable Huey, Dewey, and Louie – or maybe Shadrach, Meshach, and Abednego – or, hmm, Larry, Curly, and Moe – of ostentatious sesquipedalian pseudo-classicism. They are words that you will probably encounter in three places: newspaper articles written by the same sort of people who feel compelled to call a cucumber an indehiscent pepo and to call a pumpkin a gourd four times for every once they call it a pumpkin; “did you know” articles and lists passed around by the kind of people you’d really like to unfriend on Facebook but don’t want to cause bad feelings; and spelling bees.

All three of these words are made from Greek parts and have been around in English for about a century and a half. They were made up by a French naturalist who wanted to classify humans into types by hair, because if you’re going to classify things you have to classify them in Latin or Greek or it’s not science!

Because of the time when they came into English, their pronunciation follows the old-style English-oriented pronunciation of classical words. For one thing, the stress is on the antepenult – the third-last syllable – in all three. For another, cymo is said like “sigh mo,” leio is said like “lie oh,” and ulo is said like “you low.” (That makes my hair stand on end. People! These are Greek roots!)

Any guesses as to what they mean, what they classify? Try the trichous half; it’s the less tricky part. Have you seen the word trichotillomania? It’s a compulsion to pull one’s hair out. (Endemic to the editorial profession, if figuratively.) The source is the Greek τριχ trikh root, which refers to hair.

So. I am leiotrichous. This may sound like the self-introduction of some ancient monster or warrior, but it just means I have straight hair – Greek λεῖος leios ‘smooth’. Certain adults of my childhood thought it was better to be ulotrichous: to have curly hair – Greek οὖλος oulos ‘crisp, curly’. Many people favour being cymotrichous: having wavy hair – Greek κῦμα kuma ‘wave’. I like all sorts of hair, and all lengths from ankle to none. I’m fine with what I have.

And now we have three more words to stick on the knick-knack shelf. They are the kind of word you will always need to define on first use unless you’re talking to a true in-group. They’re like that odd mystical little object that looks rare and special and pricey but that is unidentifiable until you explain to your dinner guests that it is the trigger assembly from a Qin Dynasty crossbow. Then they all nod sagely and are impressed.

But you may also want to tread a little carefully. These words are now used (when used at all) simply as descriptives for kinds of hair, but words that began as means of racial classification can sometimes have a bit of an off odour to them – like burnt hair, maybe.

Hello, LA, this is your future talking

My latest article for The Week, “What Americans will sound like in 2050,” has drawn some attention. In particular, it caught the attention of some folks at KPCC, an NPR radio station in southern California. They did a live interview with me this afternoon (this morning their time). They also recorded it and transcribed some of it. It’s 7:23 long, so it won’t eat too too much of your time…

Predicting the future of American English



I was at a very good party last night. I was barely in the door before I was being acquainted with a gimlet, which turned out to be the drink of the evening. I had heard of it before (probably first in magazine ads in the 1970s) but to my recollection had never had one. Well, I had two last night. A gimlet is made with gin and lime cordial – the bartenders at this party used Bombay Sapphire (Broker’s probably would have been better, or Tanqueray; Bombay is a little delicate) and Rose’s Lime Cordial, plus a little lime juice, a cucumber garnish, and – heretically – a mint leaf. It was nice to have a classic cocktail that was also a relief from the usual drill. It augured well: the party was not boring.

Which was ironic. After all, a gimlet – the thing the drink is named after – is a small hand drill for boring holes. It’s like an auger, but smaller. Once the bit bites in, it keeps digging with each twist, spiraling the wood out as it goes. So it’s sharp and piercing, like the lime juice in the drink and like a look from a squinty eye – a gimlet eye, as they are sometimes called. A gimlet eye is not like being sloe-eyed (which is good, because there is no sloe gin in a gimlet). It’s an eye that may seem to throw down a gauntlet but more likely is just drilling you.

The g on this word, in case you’re not sure, is pronounced “hard” like the one in give, not “soft” like the one in gibe. It comes to us from Old French guimbelet, which is the source of modern French gibelet, which is not to be confused with Old French gibelet, the source of modern French gibelotte and modern English giblet, which has a “soft” g. (This is what you get for drilling down to the giblets.) The source of that Old French guimbelet is also the source of our modern English wimble, which means ‘gimlet’ and is not to be confused with wimple. There is also an unrelated adjective wimble ‘nimble’. Wimbledon is unrelated and it’s not my problem if you find lawn tennis boring.

So anyway, a gimlet – the drink – is for people who want to recast their gin and tonic with lime. Fair enough, since gimlet anagrams to lime GT. It’s maybe more like a lime Tom Collins, though – just replace the lime with lemon and you’re there. (Who was Tom Collins? It’s disputed but most often pointed at an Irish activist of the 1700s. On the other hand, I can tell you that the martini was originally called Martínez.) Now, if you want a different citrus, no need to go off on a tangerine, I mean a tangent; if you’d rather fill holes than make them, just use orange juice in place of the lime cordial – and vodka in place of the gin – and you have a screwdriver.

And, on the other hand, if you decided that the gimlet-eyed person is really sloe-eyed, you can take comfort in that – and complete the assembly – by adding sloe gin and Southern Comfort to your screwdriver and having a drink called a slow comfortable screw. I’m not making this up.

antanaclasis, polyptoton

Imagine lettering these letters on a sheet of letter paper, or articulating them in an article: antanaclasis with its forays of four a’s – see those two articles an an in an article, appearing as is – and polyptoton with its two p’s to tease (and two t’s too), like a pair of polyps until appearing in toto. Such repetition with variation – forms varying as they repeat and repeating as they vary. If you could map them to a map you might imagine an image of Antananarivo, perhaps, or some proximate topology (like the tsingy). But have these word forms landed on the page to inform us about land forms? Is antanaclasis doing its eye-breaking break-dancing to slide in in place of some slide about a landslide? Does polyptoton fall like some fell waterfall, pooling in a pool of manifold loops, so many loopy topoi like so many folds?

In fact, though the results echo by sheer reflex, though the shape reflects that echo and faces you like a sheer rock face, they are not geographic; and though the technique may be rhapsodic – even euphuistic – the technical terms are rock-hard canonical rhetoric, classed more in the classical canon than in hard rock.

Can you sense their sense? Are the above paragraphs sensible or nonsensical? Well, never mind, I’ll ease your mind – or I’ll remind you if you were once mindful of these terms: they refer to related figures in speech and writing.

Antanaclasis comes from Greek ἀντανάκλασις, from ἀντανακλᾶν antanaklan ‘reflect, bend back’, from ἀντί anti ‘against, in the opposite direction’ and ἀνακλᾶν anaklan ‘bend back, break’ (from ἀνα ‘back’ and κλᾶν ‘break’), and it refers to use of a word in multiple meanings: not to find the mean, nor to be mean, but just to mean in more than one way along the way.

Polyptoton comes from Greek πολύπτωτος, which comes from πολυ polu ‘many’ and πτωτος ptótos ‘falling’, and it refers to use of many cases or derived forms of a word: you derive forms by forming derivations to inform your readers formally.

Both of these have been used judiciously by great writers for subtle effect – they are certainly most effective when used subtly. Mind you, antanaclasis is really a way of punning; when Pistol in Shakespeare’s Henry V says “To England will I steal, and there I’ll steal,” he’s using just the same kind of figure as in the joke “Time flies like an arrow; fruit flies like a banana.” But polyptoton sounds more rhetorical, more speechy: “The Greeks are strong, and skillful to their strength, Fierce to their skill, and to their fierceness valiant” (Troilus and Cressida, by Shakespeare again).

Anyway, you can figure out whether and how they will figure in to your writing. You may enjoy writing their figures – their repeating loops of a’s and o’s and p’s – or you may find them disfiguring; you may like playing with the play and interplay of that their senses denote, or you may find it a senseless display. It’s up to you.

This late loopy type foray is for a type IVa who has lately closed another loop.

The sounds of historical English

A couple of weeks ago, I did an “English language time machine” piece for The Week. This week, it’s up as a podcast, for those who prefer to listen:

What the English of Shakespeare, Beowulf, and King Arthur actually sounded like