Argh! Megidduh!

My Globe column today, on snowmanteaus, discusses the relative merits of snowpocalypse and snowmageddon as descriptions of this month's mid-Atlantic weather events. But in it, I made a dopey mistake, as reader Ed Berman, of Marblehead, Mass., points out, and compounded the error by attributing it to the Oxford English Dictionary: I said the original Armageddon was a site in Jerusalem.

Writes Berman:

Armageddon is a corruption of Har Megiddo, or Mount Megiddo (more accurately Tel Megiddo, a tel being a mound built up by successive communities built one upon the ruins of the previous one). Megiddo is a site in the southern Jezreel valley in northern Israel where there had been many battles in ancient times. We visited there in l965 at which time they were excavating ruins of fortifications built during the reign of King Solomon.

I had read up on Har Megiddo while researching the column, and I didn't at any point believe that the site was in the city of Jerusalem, so I can only surmise that I meant to write "Israel" and somehow substituted city for country. Is this what linguist Geoff Nunberg calls a thinko, or something even less explicable, just a dumb-o? Apologies to all, especially the OED.

[Update, 12:33 p.m.: Online version now fixed, thanks to my colleagues at boston.com. I love the 21st century.]

Just follow our lead ...

Headline over a group of letters about changes to No Child Left Behind in the Feb. 8 Globe:

ABANDONING THE 'PROFICIENY' GOAL

(Yes, it's still there on the web version.)

Word: Fleur-de-lis

I was amazed to read in yesterday's NYT that the French pronunciation of fleur-de-lis (or fleur-de-lys) sounds the final s: "fluhr duh LEES," in the TImes's transcription. Amazed not because the pronunciation is odd, but because I've never heard that there was an alternative to the usual American "fluhr duh LEE." I mean, the world is full of people reminding each other that vichyssoise and coup de grace (sorry, no time to find accent marks) do sound those final consonants. And just last week, I was reading that Haiti's capital, Port-au-Prince, would not (in French) sound the t in Port: "pore oh prahns," it would be.  So why, in all the decades since I studied Little Lessons in French in an antiquated Book of Knowledge encyclopedia, have I never heard the pronunciation police address fleur-de-lys?

I think it must be because the Anglicized pronunciation has been grandfathered (or great-grandfathered) into our vocabulary. Charles Harrington Elster, the pronunciation maven, doesn't list the word in his "Big Book of Beastly Mispronunciations." The 1889 Century Dictionary doesn't have an entry for fleur-de-lis, but it has the verb fleurdelised, "adorned or ornamented with fleurs-de-lis," and the sole pronunciation given is (in Times style) "fluhr-duh-LEED." My 1969 American Heritage and my circa 1963 Funk & Wagnalls give LEE as the pronunciation of the singular, but oddly, both sound the s in the plural: fleurs-de-lis is "fluhr duh LEES."

So maybe I've never even heard the French-style fleur-de-lis; certainly I've never heard anyone's pronunciation corrected. But if there are people out there trying to re-Frenchify the word, do let me know how the campaign is going.

Google Books: Searching in vain

Sometime in 2007, Google Books added a wonderful option for users: We could put books into a collection labeled "My library" and then search only within that list. 

A week or so ago, the "Search my library" button disappeared. The new Google Books page offers all sorts of ways to waste time sorting and labeling books, but no longer can I just click and search just the books in my own library. 

Whose bright idea was that? Someone at Google who doesn't appreciate the true sublimity of Google Books, which resides not in the contentious copyrighted works but in the huge collection of older books and periodicals available for full-text searching. When I researched my annotated edition of Ambrose Bierce's 1909 usage book, I amassed a collection of grammar, vocabulary, reference, and usage books dating from the mid-1700s to the early 1900s, all copyright-free and searchable and clippable. (I could have bought all the books and searched them by hand, but I'm not sure I would have managed to publish in time for the book's 100th anniversary.)

Google's book search was far from perfect, as researchers well know. The same search may bring different results on different attempts. I might search my library for a term I knew was in Bierce's own book -- "expectorate," say -- and find that citation unaccountably missing from the results. But with repeated searches and cross-reference checks, I think (and hope) I approximated the level of accuracy a manual search would have yielded. (As for Google's laughable metadata -- dates off by centuries, crazy categorization, misattribution -- Geoff Nunberg is all over the case.)

OK, this was all a free service, and I actually got lucky -- it was working just at the time I needed it. So I shouldn't be complaining, right? Besides, Google has explained that there's a workaround: To search your library, just plug your parameters into this sample formula:

http://books.google.com/books?hl=en&as_coll=0&start=0&num=10&uid=44483708056731249&q=mad

* The "as_coll=0" parameter defines which bookshelf you search within. In this example, the search is within my "favorites" bookshelf. You can change the parameter to other bookshelf IDs, which are visible in the URL for each of your bookshelves.

* To search within your accounts, you will need to change the value in the uid= parameter to your Google account ID. You will see this account ID in the URL when you visit My Library.

* The q= parameter is what specifies what word or term to search for. So in the example above, I am searching for the word "mad." You can replace this parameter with your search term, Please note that you will need to include "+" between multiple search terms.

Yes, I can do that. Thanks, I guess. But would it really be that hard to just restore the "Search my library" button?