Author: Geoffrey K. Pullum

Six of us — three philosophers, two linguists, and a mathematician — were having dinner the Café Noir in Providence on Friday night, and when three of us decided on the excellent boeuf bourguignon, someone at the table told a story of a colleague who tried to include the phrase boeuf bourguignon in a word-processed file and found that the spell-checker recommended correcting the spelling to Beowulf Burlington.

We all giggled happily, and referred to the dish as Beowulf Burlington for the rest of the evening, but later (cursed as I am with the habits of a scholar) I checked with Microsoft Word on a Mac, and it proposes no such thing (though it does, rather in a rather desperate list, suggest correction boeuf to either beau, beef, beefs, beefy, bogus, bough, bout, or pouf). Google doesn’t offer the alleged correction either, of course (there are hundreds of recipes for the beef dish in question).

I sometimes wonder if we aren’t doomed to a future of suspiciously implausible and uncheckable cupertino effect stories forever — an endless series of newly invented spell-checking chestnuts, false but enjoyed too much to be checked up on or rejected, like Eskimo words for snow, or the old story about the preposition-stranding correction that Churchill almost certainly never made.

I’ve simply had it with all the people who keep telling me that they revere The Elements of Style because it’s such a nice little book and helped them so much with their writing when they were college that they carry it everywhere they go and give it to all their students or hand a copy to each new employee that they hire for their company yadda yadda yadda… I have decided that my campaign against Strunk and White’s toxic little compendium of unfollowable dumb advice, bungled grammar claims, and outright mendacity must be taken directly to America’s colleges, starting with the great universities of the East Coast. For the opening event I have chosen Brown University in Providence, Rhode Island. I will speak on the Brown campus at 7:30 p.m. on Tuesday night next week, April 13, in the Metcalf Chemistry Building Auditorium at 190-194 Thayer Street. Admission is free, and Language Log readers get a 30% discount off that. Be there.

The lecture is designed to be accessible and enjoyable for almost anyone with an interest in the English language — except perhaps for people who want to tell me that they revere The Elements of Style because it’s such a nice little book and helped them so much with their writing when they were college that they carry it everywhere they go and give it to all their students or hand a copy to each new employee that they hire for their company yadda yadda yadda.

I hoped originally to stoke up the crowd to a huge outburst of anger against The Elements of Style and end the evening with a cathartic mass burning of copies of the book, everyone setting fire to copies they had brought with them and holding them aloft like a candles, waving them so that embers of flaming nonsense like “Write with nouns and verbs” scatter through the air like precious twinkling sparks of wisdom and truth… But I am told that the fire marshals for the Metcalf Chemistry Building have ruled that out in no uncertain terms. They are not happy about the courtyard bonfire idea, either.

Another idea I had was that everyone would strip naked in a symbolic throwing off of prescriptivist strictures, a return to the naked innocence of using the language as it is rather than as a pair of misinformed old authoritarians once wanted it to be… But it turns out that they already do that at Brown in the various undergraduate libraries, during the celebrated naked donut run every semester, so this would be absolutely nothing new.

The event now planned, therefore, is likely to be quite safe despite the huge crowds expected, and fit for family TV news viewing. (The networks are a little disappointed; their location crews had been making plans for coverage of either a major blaze or a lot of titillating pixelized shots of public nudity.)

Brown University, the 7th to be founded in America, was the first to accept students regardless of their religious faith. That element of tolerance is important when you are going to speak out against a book that many people seem to regard as a sacred text. Brown is also the university where it was first proposed that computer-searchable corpora of text would one day be important for linguistic investigation — for finding out what the language is actually like so you don’t have to go by the unfiltered prejudices that dopey old coots stuff into their cockamamie usage books.

The Longman group now own the rights to the current (4th) edition of The Elements of Style. I hope the rumors I have heard about them capitalizing on my campaign by setting up a book display in the lobby of the Metcalf Building are untrue. You wouldn’t believe how tiring it is kicking over tables of books. I want to reserve my energy for giving the lecture. See you there.

A Language Log reader named metanea points out to us that the Urban Dictionary claims aibohphobia is a technical term for the irrational fear of palindromicity. The etymology will raise a smile. Just stare at the word for a few seconds, and it will reveal itself to you.

What is the connection between (a) successfully executing something tricky that not everyone could get away with, like an escape or an acrobatic maneuver or a daring sartorial fashion statement, and (b) removing by tugging, stripping, or peeling?

Rather than set it as a quiz and waiting for the answers to come in, let me just supply the following sentence that Daniel Deutsch saw in the New York Times:

With flowing black hair and a remarkable ability to pull off form-fitting black leather pants, Ms. Chang is a particularly glamorous ambassador of an art form not necessarily associated with lipstick and glitz.

Could the glamorous Ms. Chang really be noted for her prowess at tearing off her black leather pants? One assumes that only her most intimate friends would know how good she is at getting undressed; even a glamorous art ambassador — even Lady Gaga — doesn’t customarily do that in public. (And Ms. Chang is not in the same area of the arts as Lady Gaga; she is a poet — Tina Chang, the poet laureate of Brooklyn.) So intended claim was surely the less dramatic one, illustrating not meaning (b) but merely meaning (a): that while you or I or the average poet might look silly in tight black leather pants, Ms. Chang succeeds in looking good in them.

But how remarkable it is that the writer chose the pull ___ off verb-preposition combination here without (one assumes) noticing the double entendre, and without the editors noticing it either. As I believe I may have said before, it is really remarkable how much polysemy and ambiguity we tolerate during every single minute we listen to our language — and how little regard utterers pay to the business of ensuring that the addressee will not misunderstand.

Keith Allan has bravely outed himself as editor of the journal from which I recently received a thoroughly discourteous message sequence. I thank him for responding to the discussion, and for confirming that it was not about him pressing the buttons in the wrong order. The reason his fine journal (the Australian Journal of Linguistics) sent me a message sequence I found annoying and presumptuous is the design of the stupid ScholarOne Manuscript software. Let me explain a little more about the nature of my life (perhaps my experiences will find an echo in yours), the part that involves those arbitrary strings of letters and digits we are all supposed to carry around in our heads like mental sets of keys.

I have to keep a secret laptop file (it had better be secret — I can only hope I have hidden it well enough, since laptops do get stolen) containing more than a hundred triples consisting of URL, login name, and password or PIN that I have been issued with in pursuance of my many duties. Many of these accounts are for duties absolutely required of me in my university roles. Often there is no option to change either the login name or the password. (One of these also involved a complicated and entirely numerical login name. I was expected to use it the other day to fix a problem with student records entry that arose in 2005 at a university that I left in 2007. Apparently people expect you to keep in perpetuity these records of the unmemorizable passwords they give you.)

I simply hate being issued with new account names and passwords, to add to this burden. Yet these days a new one is obligatory for every association membership, email service, records database, blogger identity, banking arrangement, credit card, cellphone service, online purchase, loyalty program, travel agent, grant application, computer account, encryption system, or reviewing chore.

For many of them, I know, it is true that I would in principle be able to pick my way slowly through their “Edit my profile” page and figure out how to change the password to a standard one (not that they recommend it: security experts say you should have different passwords for every account!); but with over a hundred to work through, it would be hours of work, and I would still need to keep records of which ones I had changed so far, and what I had changed them to. There are widely differing rules regarding password composition: some (with no financial interest to protect) are ridiculously slack, and would accept “abcdef” (I even encountered one idiot organization that sent me an email confirming what password I’d chosen, and repeated it in their message, in plain text!), but others insist on something like “aQz&g9#B”, with at least 8 characters and including case distinctions and non-alphabetics and embedded numerals, and repeatedly reject passwords until they get one they consider properly secure. I haven’t got the five minutes to fiddle around with each such account and choose which memorized password to try and make it accept. I just hide the passwords in a file and look them up.

But this is not optimal, and I want to minimize the problem. So one thing I want to insist on is that no new accounts are set up for me without my permission. The ScholarOne/ManuscriptCentral software violates that tenet, and it then spams me to tell me it has done so! I find it infuriating. But the editor, I now learn, has no power to use the software without it behaving in this way. Thomson Reuters is to blame for what is apparently an uncustomizable piece of garbageware that editors everywhere are increasingly taking up in order to ease the intolerable burdens of their virtually unpaid work. But it is not the fault of the hard-working editors. I sympathize with Keith, and with others like him all over the world.

Said the story in the Ottawa Citizen:

The woman was trapped in her car unconscious for about 20 minutes while firefighters performed an extraction, he said.

And alert Language Log reader Diane commented: “I had no idea our firefighters were also trained at dentistry!” She also asked me whether the misleading phrase an extraction was a dangler (an analog of the dangling modifier that prescriptivists warn against).

It isn’t, despite the fact that we are not directly given a logical subject for the implicit reference to extracting. An extraction is a noun phrase (NP) functioning as direct object of the verb performed, and a direct object never needs a predicand (or “logical subject”, i.e., an NP it can act as a predicate of). It’s only predicative adjuncts that can be danglers.

It is possible, though, for a predicative adjunct to consist of nothing but an NP. That occurs in A skilled fireman, Dan didn’t take long to get the woman out of the vehicle, where Dan is the predicand so we understand that Dan is being described as a skilled fireman. But such an NP can also be a dangler, as in ^??A skilled fireman, things didn’t take long, which sounds deeply weird because you aren’t told who is a skilled fireman.

Diane is of course quite right that the sentence in the Citizen sounds extremely silly. I actually think it’s a case of nerdview: perform an extraction must be how fire station record books and regulation manuals would refer to getting trapped women out of crushed vehicles. A reporter with sense would have translated the nerviewese and said “The woman was trapped in her car unconscious for about 20 minutes while firefighters got her out.”

Picture this: that you receive two unexpected emails from me in quick succession. The first is a boilerplate pre-packaged message informing you that I have entered your address on my website as my temporary address for two or three days later this month, and I have let my employers know that people can call me or fax me at your house. I’m a complete stranger to you, except that you know my name from Language Log; I have obtained your email address from public sources, and pre-emptively set up arrangements to that assume I’ll be staying with you.

The second of the two emails is personally addressed, and says that I’ll be in your area later this month to give a lecture, and since I’m on a tight budget, would it be all right if I came to stay for two nights?

I take it you’d be somewhere between insulted and shocked, despite the fact that it is sort of flattering that a famous Language Log writer has singled you out as a person he would like to stay with. Well the equivalent not only happened to me today; it happens to me every couple of months.

Out of the blue comes an email telling me that my name has been added to a database of manuscript referees — academics who can be called upon to supply donated time reviewing papers submitted for publication. Then shortly after that comes a personal message from an editor (often a total stranger to me), asking me if I’d be so kind as to do a favor by reviewing a manuscript that has been submitted on some topic that I know about.

Every time my involuntary reaction is the same: repulsion, even anger, at the sheer rudeness of it. Despite the fact that a famous journal has singled me out as an expert they would like an opinion from.

Here’s the latest example, with journal, location, and editor’s name disguised to protect the not-particularly innocent. First, message number 1:

From [email protected] Sat Mar 20 07:26:27 2010
Date: Sat, 20 Mar 2010 03:26:20 -0400 (EDT)
From: [email protected]
Subject: Journal of Wwwww – Account Created in Manuscript Central

20-Mar-2010

Dear Professor Geoffrey Pullum:

Welcome to Journal of Wwwww – Manuscript Central site for online manuscript submission and review. Your name has been added to our reviewer database in the hopes that you will be willing and able to review manuscripts for the Journal which fall within your area of expertise.

The site URL and your USER ID for your account is as follows…

When you log in for the first time, you will be asked to complete your full postal address, telephone, and fax number. You will also be asked to select a number of keywords describing your particular area(s) of expertise…

And now for message number 2:

From [email protected] Sat Mar 20 07:27:26 2010
Date: Sat, 20 Mar 2010 03:27:19 -0400 (EDT)
From: [email protected]
Subject: Journal of Wwwww – Invitation to Review Manuscript ID JWWW-2010-0041

20-Mar-2010

Dear Geoff (if I may):

The above manuscript, entitled “On the snrdpql vbrh of frueqbd sjhdpbc” has been submitted to Journal of Wwwww.

We would be grateful if you would kindly agree to act as a reviewer for this paper. The abstract appears at the end of this letter…

Notice that the first message was sent off 59 seconds before the second.

The culprit is not necessarily the well-meaning editor Professor Xxxxxx, who I have heard of but not met, or any of the staff of the Journal of Wwwww, which I have seen but am not antecedently involved with. Quite probably it is a suite of standard editorial software, owned by the huge Thomson Reuters global publishing empire, once called Manuscript Central and apparently now renamed ScholarOne Manuscripts. It is “the proven industry leader” in editorial discourtesy, designed as

an innovative, web-based, submission and peer review workflow solution for scholarly publishers. Easy-to-configure, it allows for streamlined administrative, editing and reviewing capabilities.

ScholarOne serves more than 365 societies and publishers, over 3,400 books and journals, and 13 million registered users.

ScholarOne Manuscripts reduces time to decision, eliminates paper distribution costs, decreases administrative overhead and increases submissions.

So there are 13 million of us exploited reviewers! And in almost all cases, it seems, we were first informed that we had been press-ganged and entered into naval personnel records and issued with a sailor’s uniform, and only then, a minute later, politely asked by the captain of the ship if we would be prepared to serve the navy as an ordinary seaman for zero pay. If hardly anyone else in academia has ever been offended by this, then I guess there must be way over 12.99 million people out there who are much more tolerant than I am.

All the software would have to do is to ensure that the default behavior is to send the polite request first, and send out the login name and password only later, after receiving a reply. That wouldn’t seem presumptuous and annoying at all. I’m prepared to believe that it just might have been Professor Xxxxxx’s fault: he could have had two tasks to execute and pressed the buttons in the wrong order. The reason I suspect the software design is that this has been done to me so often: the defaults must be such that this is the behavior resulting from the most natural way of using the program.

How could anyone design software with defaults so stupid? How could anyone (let alone a linguist) not notice this gross violation of polite discourse? You don’t tell someone first that you have already been put in the database and given an account name and assigned some password that they didn’t choose, and append a whole lot of terse instructions about what their duties will be in their new non-paying job, and then ask them to agree to do this favor!

So my policy now (since I really have too much to do, and some things have to go) is that I refuse refereeing requests when they arrive in this way. And from now on I will do it by sending the editor a link to this post. I’m sorry if this makes me seem unpleasantly grouchy, but I find these you-have-been-added messages unpleasantly rude. There may be 13 million people out there who tolerate this kind of discourtesy, but they aren’t going to include me.

[If you would like to comment below, please do so. Notice that if you have never commented before, you have not already been entered into our database of commenters, and you will choose your own identifying name and supply your email address. (Don’t forget that email address, because I may need to get in touch about coming to stay with you for a few days.)]

Added two days later:

For those of you whose reaction is “Oh, what’s the big deal, why not just ignore the email?”, let me say that (as I expected) I have now had further unwanted correspondence:

From [email protected] Sat Mar 20 07:26:27 2010
Date: Sat, 20 Mar 2010 03:26:20 -0400 (EDT)
From: [email protected]
Subject: Reminder: Manuscript ID JWWW-2010-0041

20-Mar-2010

Dear Professor Geoffrey Pullum:

Recently, we invited you to review the above manuscript, entitled “On the snrdpql vbrh of frueoabd sjhdpbc.” we have yet to hear from you about this.

This e-mail requests that you respond to the invitation to review. We very much appreciate your help in accomplishing our goal of having an expedited reviewing process. Thank you for your time and trouble.

Agreed: http://mc.manuscriptcentral.com/wwww?URL_MASK=3D5h78kQ359×64RHR22Z4

Declined: http://mc.manuscriptcentral.com/wwww?URL_MASK=3DFc8jFQ2H8cf3RX5bJ

Unavailable: http://mc.manuscriptcentral.com/wwww?URL_MASK=3DrTSkNBQ6h34q8h

I will now get emails for the rest of my life bugging me about a job I was never even asked if I would consider taking on. It’s like a collection agency starting to bug you about an unpaid bill for a purchase you were never even invited to make. Sigh.

A really good crash blossom slows down even a fast reader who is a professional grammarian with a lot of experience in rapid reprocessing of garden-path ambiguities. And this one (New York Times, March 8; thanks to Helenmary Sheridan) is a really good one, which they might better have rephrased:

Google’s Computer Might Betters Translation Tool

Took me a full extra second or two (and that is a long time in sentence processing) to find the main clause verb.

The Apple is a site “where teachers meet and learn”. It has a page where teachers can supposedly learn from “11 Grammar Mistakes to Avoid“. And guess what: as Steve Jones has pointed out to Language Log, not a single one of these alleged grammar mistakes is both (a) genuinely relevant to English grammar and (b) actually a mistake. It is truly extraordinary what garbage teachers are exposed to when it comes to matters of how to describe what is and what is not grammatical in Standard English.

I suppose I have to go through all eleven of the misbegotten peeves and fumbled explanations that this execrable site provides. Here goes:

1. The page presenting the first alleged mistake is headed “Constipated Clauses“, and the advice is that you should never use “it goes without saying” or the adverb “obviously”. This has nothing to do with grammar; it is about trying to direct people in the matter of what they should say, not of what form of words is permissible for saying it.

2. The second is headed “Comma Vomit“, and recommends against certain uses of commas. In fact it recommends against the comma use in my previous sentence, and against the comma in this one. There’s a lot of variation of choice in comma use among expert users of Standard English, and it certainly cannot be claimed that “Commas should only precede and, but, for, or, nor, so, or yet when they introduce an independent clause”. Do not trust this page.

3. The third is headed “The Death of Adverbs“, and says that adjectives modify nouns and adverbs modify verbs. As it happens, a new paper (by John Payne, Rodney Huddleston and Geoffrey K. Pullum, forthcoming in the journal Word Structure) shows this is actually not always true. But never mind the wider picture: what the page is telling you is that I can do that easy is wrong and should be corrected to I can do that easily. This is a style difference. Most American speakers will say they can do it easy when speaking in relaxed and casual mode, and most will agree that “do it easily” sounds more careful and formal. As an observation about formality levels, this might be worth making; but it doesn’t amount to an error in syntax.

4. The fourth is headed “Less vs. Fewer“, and warns against substituting less for fewer. It is claimed that the latter “describes finite, listable items”. Strictly that would imply that it’s ungrammatical to say There are fewer rational numbers than reals, because neither the rationals nor the reals are finite in number, and the reals are not even listable. But never mind the math. The page recommends saying “fewer brains”, as in “He has fewer brains than I thought”, which is ludicrous (how many more does he need, if he has one?). It’s an old, old usage quibble, and here it’s very badly presented and described.

5. The fifth says you shouldn’t reduplicate “etc.” — it doesn’t say why, but merely alleges that if you write “etc. etc.” it will show that you don’t know what you’re talking about. This is just a style peeve. It also says you shouldn’t confuse it with “et al.”, which is a purely lexical point.

6. The sixth (oddly headed Prevarication Junction) tells you not to use “I think” or “studies show”. Again, this is being bossy about what you should say, and has nothing at all to do with what it is grammatical to say.

7. Number 7 distinguishes the words affect and effect in their noun and verb uses. You can look them up in a dictionary. Purely lexical information, no grammar.

8. Number 8 is also purely lexical, or even just a spelling point: it distinguishes than and then.

9. Number 9 actually is about grammar, but what the page says is not true. It asserts that none “is always singular” for purposes of verb agreement. This just isn’t true for Standard English. When none is a subject, the agreement is often plural (are, for instance). None of us are perfect, says the Reverend Dr. Chasuble in Oscar Wilde’s The Importance of Being Earnest. Wilde was not intending to portray Dr. Chasuble as incapable of speaking correct English. The myth that none takes only singular agreement on the verb lives on despite many refutations. Serious handbooks of grammar and style don’t represent it as ungrammatical. (Of course, the idiots Strunk and White do in their clueless book The Elements of Style; but they get almost everything wrong.)

10. With number 10 we are back to Latin abbreviations: “i.e.”, it tells us, is different from “e.g.” — and so it is, but this is not grammar.

11. Finally, alleged mistake number 11: writing could of for could’ve or could have. A spelling point, really; nothing to do with grammar. A guy who writes I could of been a contender can’t spell, sure, he knows how to say grammatically that he could’ve been a contender (and who knows, perhaps in some realm other than writing he could’ve been).

That’s it. There is nothing more. The Apple has nothing for you but unmotivated content bossiness, unacknowledged style preferences, familiar lexical distinctions, and inaccurate punctuation guidance. There is only one clearly grammatical point, and it makes a clearly false claim about verb agreement in English and misinforms the user badly.

If this pathetic parade really the best that The Apple could provide for the teachers who turn to it, then I can only say that my previous blasts at the idiocies of the burgeoning industry of uninformed grammar punditry must have been far too understated or far too little noticed.

I have seen repellently bad poetry on various subjects (mortgage services and sewage disposal, to name but two); but my horror at the poem publicized by National Grammar Day was not evoked solely by the poetic standard, low though it is:

I love the King of Ing
He makes me want to sing
Add him to an action word
And it’s a gerund… now a thing!

Nor was it that the poet, Nancy Wright, won a prize for it. What makes me shudder is that it does that noun/thing confusion again (the one that underlies Jon Stewart’s terror error). Even under the traditional (but incorrect) notion that if you add -ing to a verb stem you get a “gerund” or verbal noun, it is not claimed that you get a thing. What is claimed is that you get a word of the syntactic category Noun, the category that includes (among other words) all of our most basic one-word ways of making reference to things. National Grammar Day is celebrating, rather than condemning, one of the worst and most elementary popular confusions about grammar.

Incidentally, the reason it’s a bad idea to use the Latin term “gerund” for words like talking, and a mistake to think that they are nouns, is that the form in question (The Cambridge Grammar calls it a gerund-participle) functions in several different ways:

• In Talking is not allowed the underlined word is a Subject, and you could say it acts rather like a non-count noun.
• In They were talking the same form is the Head of a catenative complement, and most definitely a verb.
• In I bought a talking doll it is an attributive Modifier, and thus functions in one of the ways that are common for adjectives.

So in one seven-word line (“And it’s a gerund… now a thing!”) the poem perpetuates not one but two long-standing and troublesome blunders. The National Grammar Day organization (if “organization” is the word I’m looking for) should be ashamed of itself.

[Thanks to: Steve Jones.]

At this page in the Daily WTF you may find a verbatim reproduction of an email in which an office worker told her colleagues:

Please be advised- I will be bouncing Nude in 5 minutes. Please let me know if this presents an issue.

Presents an issue? It sure does! Does this woman have no conception of workplace manners? I find it hard enough to concentrate when co-workers are just sitting around nude in the common room. When they start bouncing around, I really feel I have to draw the line.

One minute later, however, came a second email explaining that the word “Nude” had been — can you guess? — a cupertino. OK, everybody, false alarm. Debbie will not be bouncing nude after all. It’s just some server called NewDev that will be bounced (i.e., taken down and quickly rebooted). Nothing to see here, folks; back to your desks.

[Thanks to: Jens Fiederer and Urban Garlic.]

The crash blossom of the day, at least here in the part of Scotland known as the Lothians, must surely be “Number of Lothian patients made ill by drinking rockets“, in the Edinburgh Evening News today. Would you drink a rocket? I’m sure you would sensibly say it depends what the ingredients are. You wouldn’t just down a rocket if I fixed one for you in the cocktail shaker, would you?

Only slowly, as one ploughs through the article looking for more details of these rocket beverages that have wrecked the health of so many in the Lothians, does it dawn on you that you have made a major mistake in syntactic analysis. Try making rocket the main verb instead.

[Hat tip: sharp-eyed Language Log reader Kenneth MacKenzie.]

My preliminary experiments with dictionary searching suggest that English has absolutely no words with roots of forms like *bobbib, *papoop, *tettit, *doded, *keckick, *gaggig, *mimmom, *naneen, *faffiff, *sussis, etc. These are simple CVCVC shapes that do not seem to contain any un-English sequence. They aren’t hard to say. In fact there is an example of a verb with the shape dVd that has a regular preterite tense: deed has the preterite deeded (as in The farmer deeded back his farm to the bank [WSJ w7_016]). But the pronounceability of deeded only makes the puzzle more acute: why are there no roots with the phonological form C_iVC_iVC_i (where C_i is some specific consononant sounds and the V positions are filled with vowel sounds). Why? Or is the generalization perhaps wrong? Have I missed some words of the shape in question?

See Plethoric Pundigrions¹ for screen shots showing a version of Microsoft Word (I don’t know which one) that for levelheaded suggests correcting it to level-headed and for level-headed suggests correcting it to levelheaded. That should give rise to a frustrating morning of trying to finalize the draft, shouldn’t it?

¹ Hat tip to Bob Ladd.

You will probably want to know what The Cambridge Grammar of the English Language says about what the right answer is; and those who yearn not just for authority but for actual authoritarianism will be disappointed that it reports, “It is an area where we find a great deal of variation” (p. 1760, in the section on lexical hyphens).

If you think that nonetheless an answer should be stipulated, then go ahead and make up a stipulation. What The Cambridge Grammar is telling you is that you won’t have any basis for it. You might just as well have stipulated the opposite. Educated usage will not always match your stipulation (thus showing it to be a good one), and it won’t always fail to match.

There are two general tendencies, though. (1) The longer a compound has been in use, the more likely it is to have started being written without a hyphen. (2) American English is a bit less likely to favor hyphenating than British English is. Apart from those two rules of thumb, you are out there working with no net, trying to follow the shifting tendencies in the usage of other people. I think I would recommend simply finding a recent use of the term in the writing of an author whose work you really like to read, and following that. If Stephen King describes someone as levelheaded, and you like reading Stephen King, then write levelheaded. Nothing much will hang on it. Not everyone will agree with you (and Word may even disagree with itself), but hey, it’s a free country.

Does that make me sound like an anarchist? I hope not. I believe there are thousands of quite strict constraints on Standard English, constraints such that you would be ill-advised to violate, because you would look like a gormless illiterate. All I’m saying is that whether or not to hyphenate a compound like level(-)headed is not one of the areas of English in which a strict and widely respected constraint holds.

With some misgivings (the above will not make me popular with those who long for stern grammatical parenting) I have left comments open below. But please hyphenate correctly.

What do loads, accumulations, obligations, and (idiomatic) kicks have in common with management, custody, people in care, sets of instructions, expenditures, liabilities, prices, loan records, and allegations?

You know I hate it when everyone shouts at once. Form an orderly line and enter your answers below. Reload in a different window before submitting to make sure somebody didn’t just give the answer you were going to give: Language Log quizzes tend to get answers submitted within about nine minutes of appearance.

[Later]
OK, as you see, it took less than half an hour on a slow Sunday night for someone to spot that the astonishingly polysemous noun charge bears all of the above senses, and Tim Silverman is the winner, beating catsidhe by less than a minute. You can look the word up on Webster’s here. A charge may be the load of explosive in a bullet or shell or an accumulation of electricity or some metaphorically similar force (a poem can carry an emotional charge); a charge may be placed on you in the sense of an obligation or duty; you can get a charge out of doing something exciting; being in charge of something means managing it and being in someone’s charge is being in their custody; if you are the guardian of some young people they are described as your charges; the judge gives a charge to a jury; you can notice illicit charges on your credit card bill; there may be a charge for some service; a library has a charge on a book when its records show that it has been lent out; and if you commit a crime you may find a criminal charge brought against you.

How the hell do we manage with a word that has this many meanings?
I didn’t even get to all its senses; I was going to include “onrushing military assault, especially of infantry or cavalry”, but I simply forgot that one. There are several others that could have been added. And then a whole lot more meanings for the related verb.

I really don’t do lexical semantics, but I really am struck by the astonishing degree of polysemy in English: words that have multiple meanings, sometimes recognizably if distantly related (charge has an etymology going back to the same Latin root as the word car), but sometimes apparently a thousand miles away from each other conceptually. Prescriptivists get so red-faced furious about the idea that a word might develop a new meaning or function (that disinterested might pick up a second meaning “bored” alongside “unbiased”, for example); but they never say a word about most of the cases of rampant polysemy in the dictionary.

Charge is not just ambiguous, having two separable meanings; it is multifariously, outrageously, promiscuously polysemous. What it suggests is that human languages do not strive to avoid ambiguity. They do not try to align words with meanings one to one. It follows (since things don’t fall apart just because we have thousands of words like charge) they are not in danger of anarchy when a new word sense evolves. People don’t just tolerate languages with multiply polysemous words, they seem to love them; they thrive on multiplicity of meaning. There are thousands of examples that show this. It is only the prescriptivist thickheads who cannot see what that means…

But wait a minute; I seem to have said some of this before.

At the very hour when, a few days ago, Victor Mair was posting his piece about Valentine’s Day in Japan (I Tiger You), I was at ground zero for the event: the candy section of the biggest department store in Tokyo’s Ginza district. I have never seen anything like it. Excited young women by the thousand buying up all the chocolate and other candy that industry could pack into pink and white heart-bedecked boxes and bags. What an incredible coup the candy manufacturers have made out of this celebration of girlfriendhood and boyfriendhood. The ratio of refined sugar and teenage girls to oxygen had reached danger level in the confined space of the department store basement, and I fled from this stampede of candy lust, escaping into the cold afternoon air. I’ll tell you a secret: I simply cannot bear Tokyo.

Sorry, but it’s not my kinda town. I am not ready to {heart} it, {tiger} it, or {chocolate} it. It is a concrete nightmare turned movie. (Ridiculous to say this when in fact we have had the privilege of staying in the wonderful Four Seasons hotel in the gardens at Chinzan-so, I know that; but the cross-town freeways and skyscrapers began to oppress me after the first few days.) And as regular Language Log readers will know, in addition I have a horror of becoming illiterate, and I become illiterate the moment I arrive in Japan.

My schedule here is a busy one, and the Twitter format seemed like a sensible one for the very few linguistic and cultural insights and observations that impress themselves on a simple and illiterate man like me. So the snippets that follow are all 144 characters or less. [Prescriptivists: don’t tell me that should be “144 characters or fewer”. When it comes to my native language, I’ll be the grammarian here, OK? But I accept that I got the limit wrong: texts are 144 but for Twitter it’s actually only 140.]

• Tokyo indescribable in human language: Atlanta concrete and freeway density, super-Manhattan people density, only Blade Runner gives hint.

• Could disastrous Japanese writing system be left here by aliens to make sure human development is held back by a few centuries?

• Friends in Kyoto report counting system also unsuited to human cognitive powers. Numerals vary by what is being counted, and with location of use.

• Neat phonology evidence from monolingual Japanese: “Yahoo!” comes out [yafu] (bilabial [f], unrounded [u]) due to well known allophonic rule.

• Speech act distribution very different here. US sports stars & politicians reluctant to apologise; Japanese do it every few minutes.

• Disaster: took wrong 11:29 Shinkansen train. There were 2. And me illiterate. Went 100 miles in wrong direction. Missed appointment.

• Shinkansen (“bullet train”) quiet, smooth, incredibly fast. Best train on planet. Shame about platform signs written mainly in Klingon.

• Am beginning to BLAME the Japanese for brain-damaged 3-layer unlearnable excuse for writing system. Turning into an orthographic racist.

• School visit to see English class. Not allowed in with shoes. Given slippers 65% length of my feet. Shuffled around like a mental patient.

• Went to Korean restaurant for dinner last night. Menu in Japanese, monolingual staff, no pictures. Could not order anything. Left.

• Got dinner in backstreet ramen/donburi shop. Pictures of food to point to. One guy spoke a small amount of English! Turned out he was Korean.

• Being here is one long sequence of difficulties, mistakes, and small humiliations. NOT THEIR FAULT. Must volunteer for adult literacy.

The linguistic point that is so interesting about the PartiallyClips cartoon strip that Mark just pointed you to is that the “suffixes” involved are not all suffixes. The endings of the words are -like, -esque, -ward, -proof, -(a)thon, -riffic, -master, -go-round, -ism, -kabob, -(o)phile, -(i)licious, and -gasm. Of these, I think I’d say (it is a theoretical judgment) that only -like, -esque, -ward, and -ism should be called suffixes.

I think words ending in -proof, -master, and -kabob are best treated as compounds (formed of two roots, like treehouse, where tree isn’t a prefix and house isn’t a suffix). The element spelled -phile or -ophile is a Greek-derived combining form (neither a suffix nor a word, but a separate word-formation element nonetheless). And the rest, most interestingly, represent cases of what Arnold Zwicky and I have called playful or expressive word formation. The endings aren’t really separate elements at all in the word formation system; they are salient pieces of words reattached where they don’t belong in a way that represents monkeying with the language system and having fun with it, not simply employing it. English has a suffix of the form -ism, certainly; but (Reginald in the strip is wrong) it doesn’t have a suffix of the form -gasm. At least, not yet (serious morphology from little jokes can grow). We discuss the distinction between plain and expressive derivational morphology, and lay out a few of its characteristics, in the paper that you can find here (PDF copy of a paper published in the proceedings of the Berkeley Linguistic Society in 1987). We have taken some flak from people who think we are making an invidious and untenable distinction between “proper” language and mere messing about; but we have given our reasons, and criteria, and we haven’t changed our mind.

Working on a paper today, my partner Barbara found that Microsoft Word objected to her use of the word relativizing as nonexistent or misspelled, and suggested firmly that she should change it to the most plausible nearly similar word: gelatinizing. But she is wise to the extraordinarily bad advice Word gives on spelling and grammar, and firmly resisted what could have been one of the worst cupertinos in the history of philosophy.

Sticking a label on a manila file of household papers this morning I noticed that the instructions on the sheet of labels said “Insert opposite end into typewriter.” It wasn’t so much the ridiculous controllingness that made me smile (the labels had no header strip, so they were symmetrical, and it would make absolutely no difference if you used the sheet one way up rather than the other); it was the quaint old lexical item typewriter. I wonder what young people would think of that advice, if they ever read the instructions on anything (they don’t, of course; they learn the operating systems of their new cellphones by intuition). A typewriter? When did I last even see one? It was like coming upon a word like “spats” or “snuffbox” or “inkwell” in a modern business context. I wonder if the wording will survive unnoticed on every sheet of labels manufactured by that company until the phrase has become a sort of dead metaphor or incomprehensible incantation.

Thinking about the pace of change in office products (anyone want a box of unused floppy diskettes?) has reminded me that in revising the third (1979) edition of Strunk and White’s obnoxious little volume of vapid guidance and grammatical ignorance The Elements of Style to make the fourth edition (long after White’s death), they spotted the phrase “ink erasers” in some of the irrelevant babbling in White’s chapter 5, and changed it to the phrase “toner cartridges” in a desperate attempt to hide the fact that this zombie of a book is showing its age. (As I may occasionally have remarked here before, my opinion is that Strunk’s little book was never much good, and the horrible E. B. White update of it is now half a century beyond its use-by date; if you want a detailed linguistic analysis of why the book is such crap, I give one here, and for a shorter volley of abuse see the Chronicle of Higher Education article here).

‘Definitely’ is always spelled with an ‘a’ —’definitely’. I don’t know why,” says Paul Budra, an English professor and associate dean of arts and science at Simon Fraser.

So reports CNews in Canada here.

But I think what they meant was that Professor Budra (who is talking about the disastrous state of the spelling and grammar skills of students in Canada’s universities today) said (or rather, emailed) ‘Definitely’ is always spelled with an ‘a’ —’definately’. The in-house automatic spelling checks, I conjecture, flagged definately as an error (which it is: undergraduates take note), and they incorrectly corrected it to the correct spelling, which here was incorrect!