A person who is severely prescriptive in matters of English usage is often called a “grammar Nazi.” I must say I dislike this term. Real-life fascists and racists do not amuse me, and I am resistant to trivializing them. We wouldn’t think it reasonable to call a person a “grammar rapist,” after all, would we?

Various replacements have been suggested, among which I like “grammar numpty” – a numpty being a stupid person (it’s a Britishism). But that’s not quite optimal either; ‘stupid’ is arguably different from ‘obnoxiously purblind’. I find myself more partial to “grammar Nazgûl” (you can leave off the circumflex on the u if you want).

If you’re a Tolkien (Lord of the Rings) fan, I don’t need to tell you what a Nazgûl is, so off you go. You can spend the next few minutes writing a piece explaining something like how Gollum would not fall into lava with the ring; lava, being molten rock, is still so dense that a person falling on it would fall onto it, not into it, and would immediately combust on the surface (and the ring would of course melt). Facts, after all.

For the rest of you: ring? Yes, ring. Sauron, the great sorceror, forged rings of power, which gave their wearers great powers but gave him greater powers over them, for he had the one ring that had power over all of them, a ring that bore this inscription:

Ash nazg durbatulûk, ash nazg gimbatul,
Ash nazg thrakatulûk agh burzum-ishi krimpatul.

Which means

One ring to rule them all, one ring to find them,
One ring to bring them all and in the darkness bind them.

It’s a tidy little exercise in morphology – frankly, I’m surprised it’s not more often used in introductory linguistics classes. You can see readily enough that ash nazg must mean ‘one ring’, that a verb root plus –atul means the verb is done to ‘them’ and a verb root plus –atulûk means it is done to ‘them all’ (meaning –ûk must mean ‘all’, more or less). Other details that are known, though not inevitable from the text, are that burzum means ‘darkness’ and ishi is a postposition, and agh means ‘and’, and in –atul the –at is a verbal ending (meaning something like ‘to’) and –ul means ‘them’. Also, ash means ‘one’ and nazg means ‘ring’, not vice-versa (nothing in the sample itself makes that necessarily so).

So is nazgûl ‘ring them’? No, because of the difference between ul and ûl – the circumflex means it’s a long vowel (literally: hold it longer, just like you hold the /k/ in bookkeeper longer than the one in bookie). That makes it a different sound. Also, nazg isn’t a verb. Nazgûl is from nazg plus gûl, which means ‘wraith’. What’s a wraith? Kind of like a ghoul. Hmm.

Well, it is a constructed language. ‘Bind’ uses the root krimp, like English crimp, for instance. Whether Nazi influenced Tolkien’s invention of nazg here I don’t know (the Nazis did like Wagner’s Ring Cycle), but he was writing this in the 1940s.

What is a Nazgûl, then? A ring-wraith, yes, but what are those? Men who took up rings of power forged by Sauron, came under Sauron’s control, and, after Sauron lost the One Ring, were tasked with seeking it out mercilessly. If anyone was carrying it, and especially if anyone was wearing it, they could sense it, and they would swoop down with the aim of destroying the wearer and restoring the ring to its rightful condition. They descended as if from nowhere, uttering unbearable sounds.

I’m not saying that the parallel is exact. But consider: A person learns a set of tidy rules about English that give a sense of knowledge and control. They believe that in enforcing them they are serving the betterment of English and their own expression, but the rules become ends in themselves, and their enforcement an exercise in condemnation and destruction. They dedicate themselves to seeking out transgressions. A hapless person uses a bit of grammar that falls within their set of magical rules – or, rather, strays against it – and they swoop in as if from nowhere, making sounds terrible to the hearer.

Does that seem a bit of an extreme caricature? Not all that extreme. Look, I was one in my younger days, so I have some idea.

That does mean that, unlike a Nazgûl, one may always stop being one. Nonetheless grammar Nazgûl is at least as viable as any other similar term. And it has a good ring to it.

conversate, incent

Hmm. I wonder what will incentivize me to conversate today.



(Clears throat.) Hmm. I wonder what will incent me to converse today.



Would you prefer I orate? Would that make you ovate? Would you be less disorientated? Or disoriented?

Look, I’m not sure why you’re fixated on something that’s not fixed. I will be happy to notate these usages if you will note them.

Conversate is, according to many people, “not a word.” Of course that’s not true; it’s a distinct lexical item, established in usage, with a clear meaning. But it’s generally dispreferred in many a person’s idea of the prestige standard version of English. It’s not new, though of course age doesn’t automatically make a word part of the prestige standard (ain’t is very old indeed); it’s attested since 1811, mainly in American colloquial usage. The Oxford English Dictionary notes that it is “In later use associated esp. with African-American usage.”

But the verb converse has been around far longer. And why have that extra –ate when you don’t need it, right? You’d think that logic would incent people to accept incent. Instead, many get incensed by it. “Give incentive to!” some insist. Others allow incentivize. But before the mid-1800s, there was no verb form for incentive, and between the 1840s and the 1960s the only available verb for it was incent. Finally someone added those extra syllables to make incentivize – so much more acceptable, right?

Well, yes, incent is a backformation from incentive. But if you want to edit it out, remember that edit is a backformation from editor. And if, like some excitable word-warriors, you would like to get a syringe and euthanize anyone who uses incent, you might pause to consider that syringe is backformed from the plural syringes – the original singular is syrinx – and euthanize is backformed from euthanasia. And orate is backformed from oration, and ovate is from ovation, and yet, although those two words have similar ages and traditions of use (both tracing back to the 1600s), I’ll bet orate sounds more acceptable to you than ovate does (though, to be fair, some people dislike orate too – not so much now as a century ago, however).

And then of course there are fix and fixate, and note and notate, which have different meanings. And the verbs orient and orientate, which mean exactly the same thing except one says you’re American or Canadian and the other says you’re from Britain or Australia or New Zealand or…

Meanwhile, incent is generally associated with business-speak, that buzzword-laden argot that seems far too impressed with itself and not nearly thoughtful enough. And yet it’s short and effective. Like orient.

Conversate, of course, is the converse: longer than it needs to be. Just like orientate. But it’s not really about length, is it. Not when incent is just as ardently dispreferred. When people inveigh against “abuses” and “barbarisms,” if you listen for a bit, you find that what exercises them is often that they attribute the words to people who don’t know how stupid they sound. Who think too highly of themselves. Who lack educational status and don’t know their place. Who are, in short, uppity.

Hmm. Almost makes you wonder if the word-peevers are compensing for something.

Say what?

Oh. Yeah. The tidy verb compense, directly formed from Latin compensare, was current from the 1300s to the 1700s but, starting in the 1600s, came to be displaced by compensateCompense can’t be used as a verb anymore. What a botheration.

We can’t magically instantly change which words are associated with which variety of English, of course, and we are not obliged – or even obligated – to use words that we dislike if words we like are available. Skillful writers should be aware of how their audiences will receive and react to the words they choose. But we should stop to consider how we react to words we dislike, and ask ourselves why.

Well, nice conversating with you. So to speak.

Beatboxing: the podcast

A couple of years ago, I did an article on beatboxing for The Week – how much of it is made of tweaked-up speech sounds. We’ve dug it up and turned it into a podcast now. If you have seven minutes and are curious, give it a listen:

A phonological description of beatbox noises

when all is said and done

No word is an island, when all is said and Donne.

No phrase is an island either. Everything comes from somewhere and goes to somewhere else. Even if it’s an isolated utterance, it draws on established meanings and references and a history of uses. Just try saying “when all is done and said.” And everything you say implies a future, even if just a near one. If you’re saying something to someone else, it’s to have some effect on them.

And everything you say is something you do. Speech is a kind of action, like all communication. If you’re on an island and you see a boat in the distance, you may wave at it, which is something you do. But you do it to communicate, just like shouting “Hey!” Communicating is moving muscles to produce gestures intended to be seen and/or heard to produce some effect (usually multiple effects) on the seers and hearers.

So in theory you could just say “when all is done,” since saying is doing. And yet we do divide them in our minds. We tend to use done to mean that some physical change has been accomplished – a change that has usually been started and guided by saying, but not usually fully accomplished by saying. Getting married, getting fired, getting commended, getting cussed out, sure, saying will do for those. But not doing the dishes, doing the laundry, or even doing the taxes. Talk may be a kind of action, but we can still say someone is “all talk and no action” because we see a divide. Fine words butter no bread, and “I’m doing it right now!” doesn’t get it done right now. And yet it still has an effect on the hearer – even if it’s not the effect you wanted.

So. At the end of the day uses a metaphor of time. The bottom line uses a metaphor of accounting. When all is said and done uses a metaphor of speech and action. Actually, it doesn’t really use a metaphor at all. When you’re talking about the situation at the end of everything, it may or may not be the end of the day, and it’s probably not on a spreadsheet, but it is after all the words and other actions… within the scope of the particular arbitrarily delimited event you have in mind.

And at the end of the day connects us to the world of work; the bottom line connects us to the world of money; but when all is said and done connects us to the world of things that happen between people. Things said are things said to other people. You may be more likely to use when all is said and done than the others in reference to interpersonal relationships. Like this ABBA song, which was written about the divorce of two of the members of the group:

(The other two members had gotten divorced a couple of years earlier; that divorce is said to have produced “The Winner Takes It All.”)

What I particularly like about that song, though, is the cover for the single. The B-side was a song called “Should I Laugh or Cry.” So the cover appears to say “When all is said and done, should I laugh or cry”:

Which, really, is a viable question in many circumstances.

There’s just one thing about when all is said and done. It’s a fine phrase, with good rhythm (iambic trimeter), the consonants riding nicely on the tip of the tongue, one liquid /l/ to expand the all, then apical fricatives /z/ /s/, and then just the /n/ and /d/ dancing together. And the turn of phrase has been in the language since the 1500s. But it’s a bit long. After all this time, you’d think we could have come up with a shorter way of referring to the state that exists after all other events and states.

Which we did, within a decade or two at most. English can be very efficient when we want it to be, after all.

the bottom line

OK, so now we know the bottom line on the end of the day. But at the end of the day, what’s the bottom line?

They seem to be roughly the same thing, right? A summative discourse marker? But one refers to the situation after everything has happened; the other refers to the situation when everything is tallied up. They tend to be used in generally interchangeable situations, but can we think of a place where we would use one and not the other, or at least would incline more to one than to the other? Or even where the tone or sense might be a little different?

They make great cakes, but at the end of the day, I’d still rather eat at home.
They make great cakes, but the bottom line is I’d still rather eat at home.

He’s capricious and demanding, but the bottom line is that he makes great cakes.
He’s capricious and demanding, but at the end of the day, he makes great cakes.

You can have all the cosmetic surgery you want, but at the end of the day you’re still older.
You can have all the cosmetic surgery you want, but the bottom line is that you’re still older.

These are pretty clothes and they’re inexpensive, but the bottom line is that they just don’t suit me.
These are pretty clothes and they’re inexpensive, but at the end of the day they just don’t suit me.

At the end of the day, while there are differences in tone and general suitability, the bottom line is that I can’t think of an example where one or the other would be too awkward to use. Unintentionally funny, perhaps:

It may have fancy expensive ingredients, but at the end of the day, it’s breakfast.
Your bass and tenor and alto parts are all singable, but the bottom line is that the soprano part is too high.

And in some cases, you really should ask: Why this clichéd metaphor and not another? (Why any clichéd metaphor is sometimes a good question too.)

What, anyway, is the bottom line? Is it the bass part in a score? The end of a novel? The weighted lower edge of a fishing net? Well, that last one is the first definition for bottom line in the Oxford English Dictionary, but it’s not the one that’s the basis for the sense ‘final analysis’ or ‘crux of the argument’ – though this cliché does seem to be a rather wide net, dragging the lexical ocean to corral coral and filter in flotsam. No, it’s the summation of an accounting: the last line of a bill or ledger, showing the balance after all has been added and subtracted. It’s what a restaurant gives you at the end of the evening.

The bottom line has been used in this figurative sense since at least the 1830s, but its use rose rapidly starting in the 1970s. At the same time, the phrases “the bottom line is” and, more narrowly, “the bottom line is that” first appeared in print. Have a look. Before then, its use really was mostly literal (and sometimes referring to geometric drawing). In the early 1970s we start seeing it in fiction and plays but also in political writing and speeches. It seems to have become popular among a certain set. By 1979 there was a book called The Bottom Line: Communicating in the Organization by T. Harrell Allen. And once the 1980s came, it was a current phrase used to convey a pragmatic, hard-edged, business-minded sensibility. Not the weariness or nostalgia of at the end of the day. Just whatever’s right on the money.

But at the end of the day, is the bottom line really the last word?

at the end of the day

At the end of the day, in the final analysis, when all is said and done, the bottom line is…


It’s not as though discourse markers are some weird excrescence in English. We may not quite joint our sentences with them as much as Italian does with its dunque, comunque, and quindi (Italian, contrary to some stereotypes, is a language with discursive habits that are optimally suited to intellectual discourse – much more engagingly so than German, in my experience), but we still have lots of but, then, so, as a result, which means, and on up the formality scale to therefore and hence and in sum… and beyond into Latin.

And it’s not as though hackneyed metaphors are foreign to the language either. An enormous amount of our daily-use vocabulary traces back to physical references used for abstract concepts. Hackneyed originally referred to a horse worn out from being rented out all day, for instance – rental horses were hackneys, named after a town in England. Trace was first a literal reference to a path. Then there’s marker, scale, stereotype (a printing reference), joint, and all the terms we borrowed as abstract from Greek and Latin, where they began as literal references (metaphor meaning something carried beneath, for instance).

But hackneyed phrases aren’t just metaphorical discourse markers. They’re long metaphorical discourse markers, and they still flaunt their literal reference. We’ve generally forgotten the literal sense of hackneyed and stereotype, but we can’t miss the literal reference of a longer phrase made of common words. It’s in your face (so to speak). So people tire of them.

People have been tiring of them for a long time. George Orwell, in his essay “Politics and the English Language” (a hypocritical, xenophobic, classist clarion call for cranks, but not without its points) inveighs against this kind of bombast; two examples of “flyblown metaphors” he cites are “explore every avenue” and “leave no stone unturned.” Both of them are indeed flyblown, as it were. So is flyblown, but it’s a single word and so slips in like a quiet party crasher in a decent suit. (Bombast has lost its original reference entirely, but that’s not Orwell’s word here, it’s mine.)

Orwell didn’t mention “at the end of the day,” and it’s likely he didn’t have it in mind. Although the phrase in its literal use has been around for a long time, its summative use wasn’t especially common by his time. Before that, it slid in (hmm) very gradually, sometimes still having literal reference but also perhaps a more metaphorical sense – here’s a quote from William Cobbett, writing in 1806 but paraphrasing a speech from 1773: “that common fame, it was true, might set the enquiry on foot, but could never have sufficient ground for accusation; that is might be a very good breakfast, but at the end of the day would prove to be a very bad supper.” Other written texts using the phrase refer to the parable of the labourers, in Matthew 20, where Jesus talks of day labourers being hired at different times of the day but all getting the same pay at the end of the day. This no doubt had some influence on its occasional use in a metaphorical or at least partly metaphorical sense.

But the tired, sun-bleached (hmm) use that irritates some people so much first started spreading in the 1970s, it seems, and really took off (hmm) in the 1980s. Somehow some people – possibly more at first in England, and certainly among the moneyed business and law set – came to have the phrase stuck in their heads in such a way that it offered itself up when they reached for something equivalent to “when all is said and done.”

It may not be a coincidence that Les Miserables was a West End (and, after that, Broadway) hit starting in 1985, given that it included this song:

It didn’t invent the phrase. It didn’t even invent the figurative use of the phrase; it wouldn’t have been very effective if the phrase hadn’t already been established – it would have been just confusing. (The French original, by the way, was “Quand un jour est passé” – same rhythm, similar sense, but literally ‘when a day is done’.) But it may well have served as a vector for it, amplifying its popularity.

That’s not the song that comes to my mind first when I hear “at the end of the day,” though. I first hear this one, by the Canadian group Great Big Sea:

It’s not an originator and it’s not as widely influential. But, hey, at the end of the day, it’s all your state of mind. What’s just vivid enough to stick? And if metaphors aren’t vivid and sticky, well, then vivid and sticky might as well not be metaphors.


I had a panini for lunch today, which, as always, set me thinking about grammar.

You’re probably thinking “Oh! Because panini comes from Italian, where it’s a plural, and panino is the singular!” You may also be thinking “He used panini as a singular. What an ignoramus.”

In fact, panini makes me think about grammar because of Panini – which is more properly written Pāṇini (which means the “ah” takes twice as long to say as it otherwise would, and the first n is said with the tongue tip farther back in the mouth; also, since it’s not written with a Ph, the P is closer to an English “b”). He was a Sanskrit grammarian; he lived in India sometime before the Buddha was born (and thus also sometime before Socrates and everyone after that), probably around the 6th century BCE. You could almost say he was the Sanskrit grammarian, though others came after. Panini wrote the authoritative manual on Sanskrit grammar. It is a concise work, effectively an algorithm. It’s an exercise in figuring out a natural phenomenon, and at the same time it’s what computer dorks might call an API (basically a set of instructions on how to make a certain kind of thing work). He observed something, figured out as best he could how it worked, and set down as elegant a description of it as he could, which thereby became a means of standardizing its production in formal contexts. I don’t want to go on too long here; this article on him is worth your 5 minutes to learn more.

Do I think of him every time I see panini because I’m a pretentious self-regarding twerp who is mighty pleased with himself for knowing something to do with Sanskrit? Of course not. I mean, I am a pretentious etc., but the reason I think of him every time is that I knew Panini as his name for years before I ever saw it as a name of a food item. I learned about him in university in the mid-1980s; paninis (or panini, if you prefer) didn’t encroach on my sphere of existence until the late 1980s or early 1990s. Our firstborn impressions of a lexeme have birthright: they get the full baby albums and all the brand new toys and clothes. The later impressions get the hand-me-downs.

So. First the Sanskrit, then the sandwich. When it showed up in North America, the average Anglophone saw panini and took it for the singular. People who know some Italian say “No, panino is the singular,” but they might as well be saying “No, it’s Panini’s monster. Panini is the one who created it.” Ask yourself how often you see biscotto or graffito. Even I, who know enough Italian to pass a graduate proficiency test in it (it was one of my two for my PhD, the other being French), seldom make a point of using the Italian singular. It would almost be like asking for a wedgie instead of a sandwich.

Look, Panini saw grammar as a means to understanding the divine, and thus perhaps good grammar as next to godliness, but he still worked with the data he had before him in the state it was in. He didn’t, for example, try to reverse sandhi. And I won’t try to reverse the sandwich. In Italian, after all, panino just means ‘small bread’ or ‘bread-ette’ and that’s often all they mean when they say it (though they can mean the sandwich too). If you’re going to be a purist, get that meat and cheese out of it.

And if you think someone who takes a word that is one thing grammatically in the source language and makes it another thing grammatically in English is an ignoramus, allow me to remind you that ignoramus is, in Latin, a verb in the first-person plural indicative, meaning ‘we don’t know’ (it comes to us by way of a character named Ignoramus in a 17th-century play of the same name). And you have just used it as a singular noun, sans critique. You’ll have to eat your words.