The price of monolingualism

A few months ago on my other blog, I made a point about how the costs of multilingualism have to be set against the costs of monolingualism. It seems certain quarters of the CIA and the American Republican party agree with me, according to today’s New York Times.

C.I.A. Needs to Learn Arabic, House Committee Leader Says

The Republican chairman of the House Intelligence Committee [Rep. Porter Goss] said Tuesday that prewar American intelligence about Iraq had been hampered by significant shortcomings, including what he called the C.I.A.’s unsatisfactory response to Congressional directives to improve its foreign language capacity. […]

“Our capabilities were not what they should have been,” Mr. Goss said in an hourlong interview. He said there had been “way too many gaps” in American intelligence gathering, including information about Iraq’s conventional military power and any illicit weapons programs.

Congressional officials have long expressed concern that intelligence agencies do not have nearly enough officers who speak Arabic, Persian or Pashto, languages needed to gain access to information in Arab nations, Iran and Afghanistan. […]

On the issue of language training for intelligence officers, a senior Republican Congressional official said a significant amount of money allocated by Congress for the foreign language training of C.I.A. officers, particularly in Arabic, Persian and Pashto, had been redirected by the agency for other purposes during the last fiscal year.

An agency official who spoke on condition of anonymity said he understood that some of the money had been spent on computer-driven document translation rather than on training for individual officers.

“Our view is that we need both,” the official said, but he defended the computerized capacity as one that would prove useful, for example, in translating the reams of Arabic-language documents being accumulated by the American investigators in Iraq who are working under David Kay, a special adviser to Mr. Tenet.

“We’ve been working on language capability for a number of years,” said the C.I.A. official, who added that the agency had increased hiring bonuses and other inducements.

But Mr. Goss was sharply critical, saying the agency sometimes seemed hamstrung by uncertainty over which languages it might need most in the future, when “the answer is we need them all.”

First, I’d like to take this opportunity to remind our pro-war readers that this is why you needed to get France and Russia on board, and why you still do, even if the price is completely disenfranchising Iraq’s American admistrators. Having had a colonial empire can mean having to deal with headaches like Gibraltar, but it has distinct advantages. I note that the DGSE doesn’t seem to have Arabic language recruiting issues.

The troubles the US is having are a part of the cost of undermining immigrant languages. When schools focus on English fluency to the exclusion of native language skills, you get a second generation with poor literacy in their parent’s languages, and a third generation with no skills at all. Furthermore, by demanding that everyone speak the same language, students learning foreign languages are denied the real communication opportunities they need to become genuinely fluent.

And, as this Mr Goss points out, you don’t know in advance which languages you’re going to wish you’d planned on having. The long lead time it takes to gain adequate literacy – a couple years at best, a decade at worst – means you can’t decide to just support some small set of languages and hope for the best. Hiring bonuses and other inducements will not bring you employees who don’t exist.

Americans tend to place a great deal of faith in their collective ability to solve problems with engineering. I know the American intelligence community has been throwing a lot of money at Arabic machine translation lately. Since I probably stand to make money off of this situation, albeit indirectly, I suppose I shouldn’t complain. However, the best automated solution likely to actually be developped will still not improve matters very much. This problem can’t be made to go away by tossing money at it.

17 thoughts on “The price of monolingualism

  1. Geoffrey Nunberg also wrote a good article on the subject. However, no amount of intelligence will be sufficient to win the war. Clearly, the decisive factor here is social engineering. Prior to the war Iraq had 20% unemployment. Immediately after the occupation, this rose to 30%. Now we read numbers like 50 and even 75%. We read that Bechtel takes a lot of cash, but doesn?t deliver the goods. One military unit was supposed to hand out a 23$ million contract for a job that ultimately was done by the Iraqis themselves – for as little as 10000$ in U.S. taxpayers? money. Maybe the U.S. military should divide the 87$ billion budget by a divisor of 2300 – ah well, let?s say, a divisor of ten – and cut out companies like Bechtel? Iraq was once – prior to 1975 – a nation whose GDP grew twice as fast as world GDP. It is not at all obvious that it needs to import massive numbers of foreign experts. In fact, this may well be seen as a display of colonialist attitudes (I think that perception nails it). The Iraqis will certainly not want to be saddled with unpayable amounts of foreign debt. In this respect, they cannot pin their hopes on anybody but themselves – as evidenced by the Democratic iniative in the Senate to make the Iraq budget repayable by the Iraqis. The Iraqis are already burdened with what the neocons rightly termed Saddam?s odious debts to the French and Russians. Putin has indicated tentative concessions on that front, but I haven?t heard anything from the French in this matter.
    When reading how Westerners calculate the costs of renovating Iraq?s schools, I simply don?t wonder that many Iraqis want all foreigners out. They may be increasingly under the influence of fundamentalists, but that doesn?t necessarily imply that their views of their own domestic economic affairs are irrational. The neocon story about the need for more money is just plain bullshit. If you are an engineer, you draw up a to-do-list, start allocating the jobs to the people at hand and note how many hours every worker has put in. You then distribute goods and money according to the records you made. The U.S. occupation, however, seems to be working along the lines of a Soviet-style command economy – bloated, slow, inefficient, overpriced. It?s not all about contracts, stupid!
    A friend of mine once traveled through West Africa by car. The car broke down completely. In Germany nobody would have touched it anymore. West African mechanics, however, rebuilt the engine using old parts from many different manufacturers. Though the end result was neither aesthetically perfect nor ISO-compliant, it was dirt cheap, robust and functional – a textbook example for what Levi-Strauss calls bricolage. If the West cannot align its interests with those of the bricoleurs in the Arab world but insists on getting in their way, it might just as well pack the bags right now. The yearning for economic self-determination was what launched the American War of Independence. If this lesson has been forgotten, then there is no hope for a Western-style open society in Iraq in the immediate future (if at all…)

  2. who’s “we” exactly ?
    While I agree with him on a superficial level, I find his arguments mostly specious (sorry, but I do).

    for example,
    Iraqis were burdened with unpayable debts (France, USSR, and perhaps posssibly the U.S.A too ?) long before the U.S. invaded. But by virtue of its invasion, and its role as occupier, it is now the U.S. which has responsibility for those debts.

    The Iraqi economy seems worse managed than the Soviet Union’s ever was. (Overpricing was not a soviet problem, underpricing and scarcity were more to the point) A truer analogy might be to the post-soviet economy.

    There are signs that the CPA is starving for money (e.g. the Iraqi police riots protesting ~60% pay cuts). remember the 67/20 split of the $87B request that Bush made and got. that’s $67Billion to maintain a force of 130 000, and $20Billion to run a country of ~ 26 000 000.

  3. “However, no amount of intelligence will be sufficient to win the war.”

    So? Having good intelligence is a worthwhile end in itself. Doing something useful with it would be even more worthwhile. And it’s patently obvious that the people making the decisions about investment in NLP have very unrealistic expectations of it.

  4. Scott,
    “The troubles the US is having are a part of the cost of undermining immigrant languages. When schools focus on English fluency to the exclusion of native language skills, you get a second generation with poor literacy in their parent’s languages, and a third generation with no skills at all. Furthermore, by demanding that everyone speak the same language, students learning foreign languages are denied the real communication opportunities they need to become genuinely fluent.”

    The three-generation transformation is a direct consequence of assimilation. I would be more concerned about groups that did not undergo this transformation, in case it might reflect social marginalization.

    A significant number of immigrant groups in the U.S. go through an almost absurd amount of trouble to keep their culture/language alive in their children,…with limited success.

    Even with Latinos in the U.S., growing up with plenty of spanish communication opportunities, does not mean that they don’t struggle in high school spanish classes like the anglos (Vocabulary: real easy, Grammar: not so easy).

    Can you point to a country that handles this better/differently than the U.S. that could serve as an example of what you have in mind ?

  5. Scott:
    This article by Randall Parker will surely be of interest:
    http://www.parapundit.com/archives/001764.html#001764
    Patrick (G):
    Canada? Why is absurb for immigrants to preserve tehir language/culture? I’m fluent in Catalan thanks to my parents and I live in Canada. Also are all cultural values that the immigrants bring with them so bad? Would America be less democratic, less open, less innovative if the Hispanics give toys to their kids on Jan 6 instead of 24 Dec?
    xavier
    xavier
    xavier

  6. Examples of bilingualism are Finland (which has become an important reference point in European debates on educational policy, due to high scores Finnish pupils generally achieve in tests comparing abilities and achievement among European nations), Switzerland, southern China, Estonia, the Brussels region, Wales, Koreans in Japan, Quebec, Catalonia, parts of Africa and India. Edward?s Indian contacts would probably know whether the impression the European elements in this list create – i.e., that bilingualism is more likely to be correlated with economic success of the countries/societies embracing it than otherwise – also applies to the Indian subcontinent.

    Geoffrey Nunberg – The answer is on the tip of our many tongues:
    http://www-csli.stanford.edu/%7Enunberg/tongues.html

    South Korea provides an example of the fears fueling the rejection of biligualism:
    http://itre.cis.upenn.edu/~myl/languagelog/archives/000085.html

  7. Patrick – my mother’s native language survived in Canada for over a century, and there are still large communities where Ukranian is spoken in Western Canada, and Italian and Yiddish in the east from immigration waves over a century ago. France has a sizeable Arabic speaking population, much of which has been present for over three generations. In those areas in the US that have traditionally had large Spanish speaking populations – New Mexico especialy – there are indigenous communities dating back to before the Mexican War where Spanish is still spoken.

    Across Europe, the norm until the 20th century was that communities rarely lost their languages. France was the major exception, and it took until WWII before its minority language communities (who at the time of the revolution represented a majority of the French population) became genuinely threatened despite a stated policy of suppressing minority languages.

    In addition to Finland’s stable Swedish community, there are the Frisians in the Netherlands, Catalonia, the Basque country, Gallicia (although language policy there is beginning to fail), Wales (where things are mostly stable now), Roumanch in Switzerland, Finns in Sweden… these are not recent immigrant communities though.

    In Latin America, German survives in Mexico, Paraguay, Argentina and Uruguay. Japanese held out for most of century in Brazil. English is still spoken in Panama and Nicaragua by the decendents of Jamaican immigrants from over a hundred years ago. In Asia, there are Russians in China, Jawi in Thailand and Arabs in Iran who retain their languages.

    Language disappearance through integration does not happen in a three generation pattern in most places. It often takes quite a lot longer. In most places, the school system and the army are the major tools of linguistic integration.

    Xavier – that NAS study was a good deal more circumspect in its findings than the anti-immigrant lobby claims. However, their model – unless I’m misreading it – considers only differences in tax payment and in the value of government service received. By looking at immigration from an exclusively tax-driven standpoint, they are completely ignoring value created through the reduced costs of service provided by immigrants, services that would cost more if they had to be performed by a smaller, better paid workforce. Comparing the price of strawberries in Belgium to the price in Berkeley would make this point quite easily.

    In short, the study ignores the very reason that there are immigrants in the US: there is work for them at wages they are willing to accept.

    If we were to use a purely public services based fiscal analysis, we would conclude that the United States would be a vastly better off expelling every state in the union that voted for George W Bush and keeping the states with the largest immigrant populations. If tax reciepts and estimated expenditures were the mechanism for voting people off the show, it’s rural white US citizens and Alaskans who would be the first to go.

    I haven’t the time right now to get into a fight on some right-wing blog, but feel free to point this out to the folks at ParaPundit.

  8. Xavier,
    I didn’t say “Why is absurb for immigrants to preserve their language/culture?”

    I said that some go to “an almost absurd amount of trouble” to do so. The difference is important.

    My statement is based largely on the feedback from a 2nd generation Lithuanian-American friend whose parents were very active in the Lithuanian-American community, after visiting Lithuania for the first time as an adult.

    The Lithuanian culture that’s being preserved by Lithuanian-Americans is in divergence of actual Lithuanian culture and is in some respects a romanticization of prior Lithuanian culture. Because it serves a need in the immigrant Lithuanian American community which never had a counterpart in Lithuania.

    The same could be said for the Polish-American community, the Philipino-American community, the Korean-American, to name just a few.

    That said, just because their children are steeped (willingly and unwillingly) in their traditions (remembered and invented) does not mean that they are comfortable or even feel capable of returning to their parents homeland, even if they do fluently speak the language.

    Speaking for myself as a child immigrant to the U.S., I couldn’t go back to France, at least not directly (though I’ve considered emigrating to a DOM/TOM as an intermediate step).

    I may be fluent, but there are large gaps in my vocabulary and I lack the life experiences that would have transparently, naturally given me that vocabulary and the appropriate cultural instincts. My prior exposure is a handicap in that I don’t have the excuse of being an ignorant foreigner.

    As it is for me, so it is even more so for 2nd and 3rd generation immigrants.

  9. Scott,
    you completely ignored my point about social marginalization, which I believe explain quite a few of your counter-examples.

    Your counter-example of Arabs in France is a perfect example of what I mean. Franco-Arabs do not yet have socio-economic parity in France.

    Your point about them having been in France more than three generations is misleading; the Franco-Arab population has exploded in the last twenty years. Most are immigrants and 1st generation. we’d need another 30-60 years to see if their language cohesion lasts, and/or if they still remain as second-class citizens.

    Otherwise, Language integration efforts began in France with Fran?ois Ier, not with the french revolution. And yes, the hold-outs were rural peasants. But the British phrase ‘speak the king’s english’, is not from the twentieth century, either.

    Up until the 60s I believe, Quebec was run by an english elite with the french-speaking majority essentially marginalized. The political backlash, once loosed, lasted well over thirty years.

    And many of the same steps that Canada took to pacify the Quebecois, are being adopted in the U.S. with regards to spanish-speaking latinos, albeit with a twenty-ish year lag. But more importantly, without the threat of secession by a major chunk of the country.

  10. Joerg Wenck: While I agree with you that it would be more sensible to hire local talent in Iraq to bring down the costs of reconstruction, I do think that it is silly to assume that the conditions are ideal for that kind of an approach.

    Anyone who has ever worked on a contracting job, or on a film shoot, will know how important it is to have a dependable, pre-budgeted crew on hand. There is no room for bricolage in large undertakings where time is an important factor.

    Bricolage works well exactly in cases you’ve cited: an unexpected need solved ingeniously by local talent bringing in unusual skills. But I can cite countless cases where I’ve been stuck in production because someone just assumed things would work out by themselves in the end. I’d much rather had paid extra to make sure things indeed did work out.

  11. Patrick (G):
    I agree with you that there’s a danger of romanticizing one’s cultural roots which as nothing to do with the reality ion the country of origin. However, the Internet will disabuse anyone from romantizing what goes on. I follow rather closely wh at happens in Spain and I’m under no illusions about the country’s flaws or foolishness that occurs there.
    I think that the communications revolution has sobered the immigrants as to what to expect (or not) from the countries of origins.

    xavier

  12. Xavier,
    “However, the Internet will disabuse anyone from romantizing what goes on. I follow rather closely what happens in Spain and I’m under no illusions about the country’s flaws or foolishness that occurs there.
    I think that the communications revolution has sobered the immigrants as to what to expect (or not) from the countries of origins.”

    I wish I could agree.
    we, as immigrants, have to work at exposing yourself at what our fellow countrymen/women would take in by osmosis.

    Yes, it’s easier to do than it ever has been, thanks to the Internet. But you still have to do triage, because you can’t completely focus on following your home culture to the exclusion of the culture that you’re living in (well, you can, but then you’d be better off repatriating).

    That’s work that the subsequent generations may not be able or willing to make, as they build themselves a cultural identity that’s less tied to their parent’s remote home culture than it is to the one that they’re acquiring by osmosis. And they will lack the life experiences that shaped the framework that you use to evaluate events at home.

  13. “If we were to use a purely public services based fiscal analysis, we would conclude that the United States would be a vastly better off expelling every state in the union that voted for George W Bush and keeping the states with the largest immigrant populations. If tax reciepts and estimated expenditures were the mechanism for voting people off the show, it’s rural white US citizens and Alaskans who would be the first to go…”

    Scott,

    Would the same hold true of electoral returns on a county level within a given US State? Federal and State expendature and tax revenue stats are available to the public. If one were to consider only urban versus rural numbers, and neglect suburban counties, wouldn’t that suggest something similar in terms of “public service based fiscal analysis” to your own claim?

    Furthermore, I’d suggest that not ALL migrants fit quite as neatly into the lowest segment of the economic food chain often presumed. South and East Asia migrants form a large enough portion of the applicant pool competing for higher education that they’ve found themselves in the dubious category of “over-represented minorities” at many academic institutions (UCal Berkeley and UofMichigan to name two)… And this includes bi- and tri-lingual Muslim migrants who, like the majority of the worlds Muslim population, neither read nor write Arabic, and don’t speak Arabic as their native language.

Comments are closed.