"The problem with people who want to impose their linguistic tastes on others," says David Crystal, "is that they never do so consistently." I'm not so sure I agree that's the problem. Consistency wouldn't be quite enough to excuse grammar fascism. I'd say the problem with people who want to impose their linguistic tastes on others by writing books on how to write is that they are so bad at it: though often they are good enough at writing (I have never said that E. B. White or George Orwell couldn't write), they actually don't know how they do what they do, and they are clueless about the grammar of the language in which they do it, and they offer recommendations on how you should write that are unfollowed, unfollowable, or utterly insane.

Both Crystal and I have been suffering the same painful experience — reviewing the same ghastly, insufferable, obnoxious, appallingly incompetent book. It is by Simon Heffer, the associate editor of the UK newspaper The Daily Telegraph, who imagined that he could improve the world by offering 350 pages of his thoughts on grammatical usage, uninformed by any work since he was in college thirty years ago — in fact pretty much innocent of acquaintance with any work on English grammar published in more than half a century.

One of the most interesting papers at an interesting conference was Michael Newman, "Identifying native English speaking Pacific Asian Americans by voice".

I have often stressed the point to my students: it is not your ignorance that interferes with your education in this subject; it's the very opposite. It's the fact that you are a highly intelligent human being and you know many things deeply and thoroughly that can prevent your learning. Of the things I teach, it is in phonetics that this comes out most vividly: the reason you can't learn to hear and produce the difference between Hindi dental [t] and retroflex [&#x0288], I tell them, is not that you are no good at this practical phonetics stuff, but that you have had twenty years of training in ignoring this contrast (so as to become an expert speaker of English or some other language), and you have done brilliantly at it. Well, there was an echo of the same line that popped up today in some news about the phishing industry. Dr Emily Finch, a University of Surrey criminologist, said:

The general public is more internet security-aware than it was five years ago. Malicious anti-virus scams are an indication that criminals are now tapping into this.

Rather than exploiting our ignorance – the basic premise of common scams such as phishing – they are actively using our knowledge and fear of online threats to their advantage.

A painfully sad health story in today's news media. For some time now there have been suspicions that isotretinoin (= Roaccutane = Accutane = Amnesteem = Claravis = Clarus = Decutan — drugs have more names than the devil) tended to increase the risk of depression and suicide in its users. But it wasn't the drug. It was the acute acne (and of course the social consequences thereof). For once the familiar cliché is reversed: it turns out the disease was worse than the cure.

This is funny, though unfair:

From a letter to the editor in the Nov. 8 New Yorker:

Such rhetoric then [by left-wing critics of George W. Bush] was hardly less corrosive, or less supported by scholarly reasoning, than the crackpot vitriol now spewed by Beck and his ilk.

As we've noted many times, combinations of negation and scalar predicates are hard for our poor monkey brains to process.

What in the hell motivates the tweeting craze? Twitter seems insane to me. If all my Language Log posts had to be 140 chars I wouldnt be abl

And people wreck their lives tweeting. A UK politician's "joke" suggesting a muslim writer should be stoned to death got him arrested by th

The Linguistic Data Consortium will change its network IP address between 4PM and 6PM EST on Friday, November 12th. During that time, Language Log will be unavailable, since its server is on the LDC network.

It may take up to 72 hours for external networks to propagate the new IP addresses, so Language Log readers may have trouble accessing the site until Monday, November 15th. The new IP address, if you're able to make use of this information, should be 128.91.252.31.

In addition, I'm now in Groningen for ExAPP 2010, and will be traveling back to the U.S. on Saturday, so I may be an even worse correspondent than usual this weekend for independent reasons.

From the AAAS (the American Association for the Advancement of Science), a holiday t-shirt:

The Nov. 10 Subnormality , featuring a tour of the Museum of the Theoretical, has a nice example of dude used as an interjection:

The crash blossom of the day comes to us from Rebekah Macdonald via Twitter. This headline appeared on the New Zealand news site Stuff.co.nz:

Police chase driver in hospital

Of course, the police didn't chase a driver in a hospital, like some wacky action movie sequence. The subject of the headline is "police chase driver," a compound noun pileup typical of headlinese in the UK and other countries.  The driver had "led police in a 150 kmh chase in Lower Hutt" and landed in the hospital after crashing (!) into a power pole. We await the inevitable followup headline, "Police chase driver out of hospital."

Everybody acquainted with colloquial English knows that Yes has alternations in pronunciation: it may lose its final [s] and add a centralizing offglide to become Yeah, and it may pick up an alternative final consonant, an unreleased [p] (simulating the sudden closure of the lips at the end of the utterance), to make Yep. No also gets a final unreleased [p] sometimes, hence the spelling Nope (notice that in each case there is a conventional spelling of the [p]-final pronunciation for use when direct reporting speech, e.g. in novels). But my colleague Heinz Giegerich just pointed out to me a surprising constraint on the final-[p] pronunciations: for a long time those pronunciations have been current only as single word utterances.

In particular, he noted (on receiving an email from a Chinese student who agreed to a meeting by writing "Oh, yep", and noticing that it seemed odd) that the [p]-final pronunciations don't seem to occur when preceded by the interjection oh.

I recently noticed that the category of English autoantonyms now includes a derivational suffix.