August 25, 2006

Who done it?

It's August, and the world's news media are apparently being managed by the night-shift cleaning staff. In today's Guardian, there's an article under the headline "A tale of many tongues" that presents an astonishing statement:

The four most often spoken languages in the world are, in order, Mandarin, English, Hindustani and Spanish. Spanish is fast rising in importance and there are now more Spanish speakers in the United States than English. [emphasis added]

Now, the facts in this matter are not hard to find. The U.S. Census publishes a data set on Language Spoken at Home, which is one reasonable way to define who is a "Spanish speaker" as opposed to an "English speaker", and according to the data from the 2000 census, 10.71% of households use Spanish, as opposed to 82.105% who use English.

In fact the Census bureau asks a number of other questions about Language Use, and publishes the results. So we also know that of the 28.1 million people who spoke Spanish at home, 14.3 million speak English "very well". If we subtract these from the Spanish-speaking group, the proportion of Spanish-speakers in the U.S. drops to under 5%. (The overall proportion of the U.S. population who speak English "very well" -- or English only -- was 92%.)

These are easy numbers to find -- it took me one Google search and a couple of clicks on resulting links. But before doing the research, I already knew that the statement in the Guardian -- "[T]here are now more Spanish speakers in the United States than English" -- had to be a preposterous falsehood. What I learned from my 30 seconds of research was just the specific numbers -- 82 to 11, or 92 to 5, or whatever, depending on how you define your terms. Anyone with a working brain and basic knowledge of the modern world must have had the same reaction. You certainly don't need to live in the U.S. to have this basic common-sense understanding of how things probably are -- I got a link to the Guardian article in email from John Wells, who lives in London.

So how did this spectacular piece of nonsense get into the pages of the Guardian, which generally attempts to pass itself off as a serious publication, rather than what in the U.S. we call a "supermarket tabloid"?

Frankly, I'm baffled. We can't directly blame the (admittedly often slipshod and credulous) research practices of journalists, because the author of the article, Alan Smithers, is "director of the centre for education and employment research at the University of Buckingham", and thus not a journalist at all. On the other hand, we can't be sure that this is just one of the (often careless and even dishonest) talking points of public intellectuals, because the article was edited at the Guardian, and might well have been changed substantially from the text that Prof. Smithers submitted.

It's that old problem of attributional abduction. My best guess is the one I started with -- the Guardian's entire editorial staff is on vacation, and has delegated its duties to the night office-cleaning crew, who are having a little competition among themselves to see who can slip the most extravagant falsehoods into print.

[No disreprect to janitors is intended. When I was in college, I worked for about a year as a night-shift office cleaner. None of the clients was a newspaper, worse luck -- I would have enjoyed the opportunity. Let's see: "Paris is closer to Istanbul than it is to London"; "Americans spend more on coffee than they do on gasoline"; "Women speak twice as fast as men do"; "More British teenagers play World of Warcraft than cricket". I coulda been a contender! ]

[Update -- Brett Reynolds writes:

Perhaps 'English' was supposed to refer to people from England rather than speakers of English, although clearly the former is not the salient interpretation.

Maybe so. Then the statement would only be wrong by a factor of 2 or so. But I think that an editing glitch is more likely: perhaps Prof. Smithers wrote something like "there are now more Spanish speakers in some U.S. cities than...", and a rushed and undercaffeinated sub-editor, trying to make a word-count target, cut carelessly.]

[Update 2 -- John Wells wrote to Alan Smithers, who responded:

Many thanks. The statement as it appears is ludicrous. The fault is entirely mine and The Guardian is blameless.

The thought that was in my mind when I wrote that part of the sentence was `there are now more Spanish speakers in some of the United States than English', and I didn't notice in the read-through that I hadn't written it this way. (I was taking native speaker to be implied by the context.)

The Guardian will be printing a correction.

John later sent him a link to this post, and he responded:

Many thanks for this also. Mark Liberman is too kind to me. It was me reading what I thought I had written rather than what was on the page that led to the false statement. The Guardian newspaper, especially its night office-cleaning crew, are blameless on this occasion.

Prof. Smithers is courageously taking the blame on himself, but I submit that he's being too kind to the Guardian's editorial staff. We often hear about the "multiple layers of checks and balances" in the editorial process of the big-time media, which ought to catch such simple slips of the pen.

With respect to the facts of the case, I'm not convinced that there are any U.S. states where native speakers of Spanish now outnumber native speakers of English. I'll ask Prof. Smithers what measures he's relying on, but I can't see how to square this with (for example) the census bureau's table of Language Spoken at Home and Ability to Speak English by Nativity for the Population 5 Years and Over by State. ]

[Update: more here, and the Guardian's correction is here, three days later and not linked from the original article... ]

Posted by Mark Liberman at August 25, 2006 06:58 AM