May 15, 2007

Les quatre cents mots: Querla a l'envers

Walter Laqueur has a new book coming out, The Last Days of Europe: Epitaph for an Old Continent; and an essay based on it, "So Much for the New European Century", appears as the cover story in section B of The Chronicle of Higher Education (known as The Chronicle Review) for May 11, 2007.

The premise of this essay is a familiar one for those who have read Bat Yeor's Eurabia, or Melanie Philips Londonistan, or Oriana Fallaci's The Force of Reason, or have listened to the echoes of these women's work through the intellectual world, most recently in Christopher Hitchen's essay for the latest issue of Vanity Fair, "Londonistan Calling": Europe is being transformed by Muslim immigration and by its own cultural and political exhaustion.

Laqueur's conclusion:

Europe as we once knew it is bound to change, probably out of recognition, for a number of reasons, partly demographic and cultural, but also political and social. Even if Europe should unite and solve the various domestic crises facing it, its predominant place in the world and predominant role in world affairs is a thing of the past.

The argument is a familiar one, but I was shocked to see it featured on the front page of The Chronicle Review.

I'll leave it to those who know more about demography, politics and Europe to evaluate the premise and the conclusions of this argument. What got my attention was a strange bit of linguistic misinformation that Laqueur drops in along the way.

Today, if our friend really wanted to see the future, a short walk or bus ride would do in order to get a preview of the shape of things to come. An excellent starting point would be Neukölln or Cottbusser Tor in Berlin, or Saint-Denis or Evry in the Paris banlieues. In some ways, moving about European cities has become much easier. There are fewer language difficulties; the argot of the outlying areas of major cities populated by immigrants, the banlieues (verlan), we are told by Le Monde, consists of 400 words.

Overall, the business about "fewer language difficulties" is just a joke, for obvious reasons. Immigrants in Europe generally learn the languages of their host countries, and to the extent that they don't, they add more languages to the European mix, not fewer. But the claim about the impoverished lexicon of "verlan" is not a joke. Rather, it's somewhere between false and meaningless.

The first thing to say about verlan is that it's really not an argot or slang, as such, but rather a language game, something like Pig Latin or Ubbi Dubbi, a way of transforming ordinary French words by re-arranging their sounds. The basic technique is to put things backwards: verlan itself is verlan for l'envers, "reverse". Thus fou become ouf, pourri becomes ripou, vérité becomes tévéri, and so on. But as Marc Plénat explains ( "Une approche prosodique de la morphologie du verlan", Lingua 95 (1-3), March 1995, 97-129),

Ce retournement, cependant, n'a rien de méchanique. Les inititiés insistent souvent sur le fait qu'un même mot peut être codé de plusieurs façons et que l'acceptabilité d'une forme se juge 'à l'oreille'...

This reversal, however, is not at all mechanical. The initiates often stress the fact that the same word can be encoded in several ways, and that the acceptability of a forme is judged 'by ear'...

Some particular verlanized words have become common replacements for their originals. In some cases, there's a shift in meaning, so that beur, which is verlan for arabe, is used to refer second- or third-generation North African immigrants. But most common verlanized words are (as I understand it) just slangy and thus informal, like meuf for femme ("woman"), or keum for mec ("guy").

Given all this, to refer to verlan as an "argot of the outlying areas of major cities populated by immigrants", which "consists of 400 words", is a preposterous misunderstanding. Verlan is an open-ended process, that can be applied to any French word. The fraction of words in the speech of banlieue residents that is verlanized is in any case small. And I'm prepared to wager a year's salary that the overall vocabulary of "the outlying areas of major cities populated by immigrants", whether in France or elsewhere in Europe, is many times greater than 400 words.

The Le Monde article that Laqueur cites appears to be Frédéric Potet, "Vivre avec 400 mots", 3/19/2005. It quotes Alain Bentolila in support of the "400 words" number:

Pas simple de chercher du travail, d’ouvrir un compte en banque ou de s’inscrire à la Sécurité sociale quand on ne possède que "350 à 400 mots, alors que nous en utilisons, nous, 2 500", estime ainsi le linguiste Alain Bentolila, pour qui cette langue est d’une "pauvreté" absolue.

Not so easy to look for work, to open a bank account or to sign up for social security when you have no more than "350 to 400 words, while we use 2,500", estimate the linguist Alain Bentolila, for whom this language is radically "impoverished".

Bentolila was the author of the "Rapport de Mission sur L'enseignment de la Grammaire" that we discussed last winter (see "Back to Bentolila", 12/27/2006; "Cultural specificity and universal values", 12/22/2006; "French report: It's lucky Copernicus had grammar", 12/18/2006). Either Potet is radically misquoting Bentolila -- always a possibility in mainstream journalism -- or Bentolila is talking through his hat, as we can see by comparing Bentolila in a 2002 interview in L'Express: "today, a certain number of citizens are less capable than others of expressing their thoughts accurately: 10% of children entering elementary school have the use of less than 500 words, instead of 1,200 on average for the others".

Can it be that between 2002 and 2006, we went from a characteristic of 10% of children entering first grade (some of whom may not have spoken French at home), to a characteristic of the language of the entire population of the banlieues? It seems more likely that all of these numbers are exaggerations or fabrications, created by Bentolila to impress his interviewers, or by his interviewers to stir up their readers.

One reason to be suspicious of these numbers is the fact that they are so different in the two interviews. But another reason is that none of them really makes sense as an estimate of anyone's vocabulary size. A credible estimate of the passive vocabulary of average American high-school graduates -- the number of words they know, on plausible definitions of "word" and "know" -- is about 40,000 (See M. Graves, "Vocabulary Learning and Instruction", Review of Research in Education, 13 49-89, 1986; W.E. Nagy & R.C. Anderson, "How many words are there in printed school English?" Reading Research Quarterly 19, 304-330, 1984.). As for the number of words that someone actively uses in speaking or writing, that depends on how long you track them (see "An apology to our readers, 12/28/2006). I'm sure that Prof. Bentolila uses far more than 1,200 distinct words, or 2,500 either, and that you wouldn't need to transcribe more than a couple of hours of his lectures to prove it. In general, less well educated people display fewer distinct words per unit time than an intellectual like Prof. Bentolila does, but again, we would not have to transcribe very many hours of the conversation of typical banlieue residents in order to see more than 2,500 (or for that matter 10,000) distinct words.

If you know any references to actual vocabulary measurements in various sectors of the French population, please let me know. For now, I'll just observe that this kind of limited-vocabulary complaint is a staple of "kids today" hand-wringing (see "Vicky Pollard's revenge", 1/2/2007), and that among dozens of journalistic plaints on this topic, I've never seen any credible empirical support for the idea that the population's word stock is declining, whether in France or in Britain or anywhere else.

The ironic thing is that Potet leads his 2005 article with an example of vocabulary impoverishment -- caused by imposition of standard French! A young woman in a remediation program in Grenoble gets something wrong in class, and exclaims "Je suis trop une Celte!" ("I'm such a Celt!"), meaning "I'm such an imbecile!". Potet asks her how and why this word "has been diverted from its [proper] meaning?" She doesn't know, but

L’adolescente sait seulement qu’elle ne prononce plus beaucoup cette expression, en tout cas plus en classe. Elle veut "réussir dans la vie et avoir un métier" et espère reprendre bientôt une scolarité normale, commencer une formation, faire des stages. "Pour cela, il faut que j’apprenne à bien parler", reconnaît-elle.

The young woman only knows that she won't use this expression any more, at least not in class. She wants "to succeed in life and to have a career" and hopes soon to resume normal schooling, begin an education, do internships. "For that, I have to learn to speak properly", she realises.

But in any case, it's hard to believe that someone who knows French well enough to read Le Monde could possibly misunderstand the meaning of verlan as drastically as Laqueur does -- it's like thinking that hiphop is the name of a district in New York, or that jazz is an instrument. So I suspect that Laqueur got the "400 words" from a tertiary source, that is, some English-language article quoting Potet quoting Bentolila.

More generally, the fact that Laqueur didn't even manage to get his slang terminology right doesn't inspire much confidence that he has some better evidence of vocabulary impoverishment in his footnotes.

While we're here, let's look into verlan a bit further. It's been studied under that name at least since the mid-1980s (e.g. C. Bachmann. and L. Basier, "Le verlan: Argot d'école ou langue des Keums?". Mots 8, 1984; N.J. Lefkowitz, "Talking backwards and looking forwards. The French language game Verlan" Ph.D. Dissertation, University of Washington, 1987).

It surely has roots in a much older tradition of language games in France and elsewhere. See "Noi Lai and contrepets", 1/8/2005, for some more literary examples; and Natalie Lefkowitz ("Verlan: Talking Backwards in French", The French Review, 63(2), 1989), cites a 1985 TF1 interview with then-president François Mitterand, in which a bit of verlan came up (chébran, from branché, "plugged in"):

Q: Vous savez ce que c'est, le 'chébran'?
A: Vous savez, quand j'etais enfant, on renversait l'ordre des mots; ce n'est pas très nouveau ça! Ça veut dire 'branché, bien entendu. Je ne veux pas faire le mâlin, je ne suis pas trè informé, mais c'est déjà un peu dépassé. Vous auriez dû dire 'câblé'.

Q: You know what it is, "chébran"?
A: You know, when I was a kid, we reversed the order of words; it's not exactly new! It means 'branché', of course. I don't want to be a smart-ass, I'm not that well informed, but it's already a bit out of date. You should have said 'câblé'.

By the mid 1980s, as Lefkowitz explains, verlan already pervaded "both Standard French and French society":

From the president of the Republic, to the enormously popular French singer Renaud and his album Laisse Béton, to the filmmaker Claude Zidi and his flim Les Ripoux, to the novelist San Antonio, to the large movement of second generation North African Arabs in France commonly known as Les Beurs, to large advertising campaigns, to the Parisian intellectuals whose children I taught [at the prestigious Lycée Henri IV], the widespread influence of Verlan and its penetration of both Standard French and French society is evident on all levels of the popular media.

There may well be good reasons to be concerned about vocabulary development, and especially standard-language instruction, among residents of the banlieues in France, or of the inner cities in the U.S. (For example, see the discussion of Martha J. Farah, et al., "Childhood poverty: Specific associations with neurocognitive development", Brain Research 1110(1) 166-174, September 2006, at the end of this post.)

But it doesn't advance the discussion to proliferate ignorant repetitions of misleading quotations of exaggerated or fabricated statistics about vocabulary size.

[Joseph Ruby writes to point out some similarities between verlan and the back slang of 19th-century London. There are some differences as well -- back slang seems to have been much more strongly influenced by spelling, reversing letter-strings rather than phoneme-strings, and also to have reversed polysyllables letter-by-letter rather than syllable-by-syllable. ]

Posted by Mark Liberman at May 15, 2007 08:40 AM