« Open Thread on LTG Lute | Main | Obfuscation in the media and IO - Mahmoudiya/Baqouba »

18 May 2007


Feed You can follow this conversation by subscribing to the comment feed for this post.

john in the boro

Abu Sinan reminded me of my own frustration with Arabic. Twenty years with brief and intensive periods as an Arabic linguist describe my Army service. And, I was lucky enough to get basic and intermediate Arabic, two and a half years at a DoD school. At the end of my time, I felt functionally illiterate with twelve years at the 3/3 level. On those occasions when I made commo with the locals, my dialect mixture was truly inspired.

Language maintenance was a sore point with me throughout my time in the Army. Once upon a time, a civilian DoD employee commented that he just couldn’t understand why more linguists didn’t max out their Foreign Language Proficiency Pay (FLPP). At the time, that was an extra $100 a month and a new program. I looked at him and told him it wouldn’t happen as long as language maintenance remained an individual responsibility that was done on individual time. Besides, we got $110 a month for falling out of airplanes on Army time. By the time retirement rolled around the Army had doubled FLPP and held commanders responsible for their soldiers’ staying qualified. Wow, what a difference that made; language maintenance got locked in the training schedules. Despite the new emphasis, keeping soldiers qualified in Arabic remained a challenge.

Arabic was hard to learn, hard to maintain, and harder to get good at. Aside from a few geniuses, Pat and Abu Sinan are right; the military has a tough job turning out Arabic linguists who can hit the ground running. This is particularly true in the case of interrogators. Likewise, as Abu Sinan notes, the defense contractors that participate in what I like to call Operation Loot and Scoot pay better. Stop loss only works for so long. I watched in disgust as the United States Armed Services paid for the training of linguists, nurtured them, and sent them to defense contractors. Smedley Butler might comment, “What a racket.”

I am currently collecting Uncle Sugar’s dime at the local university and know several students in the Arabic program. It’s a new program, and the students are just beginning to realize the magnitude of learning Arabic. I hope they do better than I did, especially the student in ROTC.

robt willmann

Abu Sinan,

Could you please explain what the different "levels" of ability to function in Arabic mean?

You referred to levels 3 and 4. What do they and other levels entail?


My experience is that language acquisition more than anything has to do with age. I learned my 2nd language with 40 and still have difficulty with the correct turn of phrase. For my Swiss wife English was her 7th language, she was speaking passably in 2 months and fluently in 8.

Since moving to Switzerland I better understand the issue of American mono-culturalism. Americans are at a tremendous disadvantage especially when it comes to language. For example, one frequently sees grade school age children playing word games on the train, a kind of a game of telegraph where one child states a compound sentence with the first and second parts in German and French respectively. The second kid will elaborate in French and Italian, and so on. These are pre-teen children. My eight year old Niece has a new boy in her class who does not speak the local language of Rumantsch so she is writing out flash cards for him.

It just really seems that because language is an innate cognitive skill it has to be developed early (research indicates before puberty), and most likely plateaus at the pre-puberty level.

I have a distant relation that married into a Lebanese family and had no difficulty acquiring Arabic as her 3rd or 4th language, even though it was non-latin.

You can't make the brain young again, the ship of hope for 'creating' Arabic speakers sailed a generation ago.


All right, this is all true. Yes, Arabic is hard.

But Americans are not stupider than Arabs. If there were enough cultural desire to learn Arabic in America, folks would have done it. That's all I'm saying.

And regarding not enough time or money to create on the ground fluent Arabic speakers - well, maybe so, but shouldn't our leaders have thought of that going in? That's all I'm saying.

I don't think my opinion on this subject is "anti-American". I think Americans are perfectly capable of learning Arabic.

Cold War Zoomie

Some thoughts from my own foreign language experience.

1. After five years in the UK, I still ran into words and phrases that needed translation. Yes, you read that right...after five years in England!

"England and America are two countries separated by a common language."
-George Bernard Shaw

2. It took 12 months in Central America, speaking Spanish every single day, before I was able to carry on a conversation with a taxi driver. That was the ultimate litmus test! (It didn't take as long to get around otherwise - I was functioning in about 3 months, and actually conversing a little bit in about six). I had no training whatsoever. Apon arrival, my only tools were a dictionary and an old Defense Language Institute book I had picked up somewhere along the way. Later, I got a discarded State Department language book from the embassy. It didn't take long to realize in university Spanish classes that three semesters were not even close to equalling my 14 months in Latin America. Nope, being thrown into the streets with a dictionary, phrase book and a hankering for a few beers and a meal is definitely the best way to learn a language. Well, plus some one-on-one instruction from las muchachitas, por supuesto. But even after 14 months of immersion and three semesters in college, my skills are only a notch or two about a tourist.

And Spanish is about as easy as it gets for English speakers.


Colonel Lang:

Excellent post, as usual.

I'm a native Arabic speaker who has grown up pretty much bilingual all my life. Despite speaking Arabic at home and attending Arabic school for a part of my childhood, my cousins would definitely say that 'my tongue is heavy', especially after living in the U.S. for the last 15 years.

Arabic is not only a difficult language to master, even for a native speaker, it takes constant practice to maintain even passable fluency. When you factor in the relatively huge differences between formal Arabic that is used on newspapers and tv and that of the hugely varying informal dialect, you are talking about the equivalent differences of Shakespearean English and Jamaican patois.

It is very difficult. I have to will myself to practice the language by reading (very slowly) the Arabic dailies and following (with some frustration) some of the Al-Jazeera talk shows. And God help me if a Hizbullah guy is on the air- I never have a clue what those guys are talking about.

The British where able to produce people like T.E. Lawrence because Britain had accepted its fate as an Empire and viewed its involvement in the Middle East as a generational commitment. We can't expect our men and women to have the same outlook unless we as a nation are prepared to accept our involvement in Iraq on similar terms.

I think it's also illuminating to look at the relative distance between English and Arabic speakers from the other side of the coin. While it is true that there are many Arab emigrants who are able to speak English on a day to day basis, there are few who can speak both languages without 'giving themselves away' in terms of their accent.


I can back up PL's comments and Abu Sinan's analysis, and add something from the tech side. I've been involved in the development of most speech synthesis technologies and languages, including Arabic. The "Long War" will have to last much longer before technology solves military communication/intelligence issues with automatic Arabic translation.

Arabic, both written and spoken, poses first-order challenges for machine processing, higher than any other major language. Because it effectively has no vowels and instead relies upon vestigial (i.e., often missing) diacritical marks, to make a good synthesizer for Arabic you first have to build a program to contextually interpret word abbreviations in sentences and make good guesses about what phonetic morphologies were intended.

It's like trying to determine whether the letters "br" should represent bare, beer, bore, burr, bar, bear, or something else. While you can build valid statistical models to reduce error rates, you can't always read someone's mind. Humans are far better than machines at that, but it's no picnic for us, either.

For this reason there are no sophisticated enterprise- or military-grade synthesizers yet available for Arabic. But there have been some recent first steps on a long journey.

DARPA has been chewing on this problem for some time, as has NSA. DARPA just deployed 35 IBM-developed translators in Iraq, called MASTORs (Multilingual Automatic Speech-to-Speech Translator). They're not ready for mobile deployment, and I have yet to hear positive feedback about them. Just experimental at this point, probably slightly better than nothing. http://www.digitalworldtokyo.com/2006/10/us_military_to_use_ibm_arabic.php

Because pronunciation distinctions between Arabic dialects tend towards severe, building a one-size-fits-all recognizer (or synthesizer) is of questionable value to begin with. What's really needed are fifteen localized versions. A sizeable task. There has been much more NSA work on Arabic speech technology, but it's not shared with the military and the applications differ. Sharing data would still require considerable work, years of it, before automated Arabic were truly "fluent" in field translation conditions.

(Oh, and if you ever want to have some fun, ask a group of Arabic academic linguists if they can agree upon what dialect is most representative of Modern Standard Arabic. Then grab your best girl, dinner, and a movie. The Nicene Creed was written in less time.)


I am a PhD linguist and language teacher with a couple of years of college MSA and could not agree with you more, both of you. These points are all right on the money.

Col. Lang, how would you compare your experiences with the Vietnamese language on these same points? If anything, VN seems to me even more difficult than Arabic.


I'm glad to see a real discussion of this issue--I agree with the points made about learning Arabic, espcially that regional dialects make a huge difference, and that 'textbook' Arabic will get you looked at like somebody trying to use Shakespearean English in today's America.

I'd also like to add that it's easier to speak than to write--my wife is Lebanese, and fluent in spoken Arabic, however, as she grew up in France, she is not able to read or write Arabic beyond a basic level.

Also, similar to revelations of official ignorance of the differences between people of the Mideast, I wonder how many Farsi speaking translators there are?

Michael Murry

As a graduate of the Defense Language Institute: in my case, the 32-week intensive Vietnamese Southern Dialect course, I have some understanding of, and experience with, this subject of language learning out of military/civilian necessity. In my opinion, the discussion in this forum thus far tends toward the formally academic and fails to address the truly effective practical language-learning successes that our government and military have at times produced in the past. Please allow me to elaborate.

As a Navy enlisted electrician with a public high school diploma (circa 1965) and one year of college, I managed to successfully complete the Navy's one-year intensive nuclear power plant operator school curriculum. However, since cutbacks in the nation's nuclear ship-building programs left too few billets for many of us recent graduates, I then got orders to attend foreign language school in Monterey, California, since someone with access to my personnel records discovered that -- in addition to Algebra II -- I had taken two years of high-school French. Thousands of us enlisted men, from all branches of the military services found ourselves in similar circumstances in 1969 when the doomed "Vietnamization" (or "Yellowing the Corpses") stall for more time, blood, and money got underway. At any rate, it only took eight months of language work and three months of "counter-insurgency" training before off to South Vietnam I went along with those thousands of other enlisted men whose only appointed task involved "standing up" (in today's jaded parlance) the demoralized and unmotivated South Vietnamese military in the technical aspects of our various lowly military occupation specialties. Nothing about us or our mission involved rocket-science or high strategic policy in any way, shape, or form. We just considered ourselves ordinary.

Once "in country," however, those of us with a good, practical, rapidly-acquired working knowledge of Vietnamese -- as applicable to our own technical specialties -- quickly separated out into two general types: (1) those who got to reinforce and add to their basic language competence through daily work with the Vietnamese and (2) those who pretty much remained in all-American units with little or no interest or incentive to do anything but forget, serve out their indentured time, and go home. Most of us, at least initially, fell into the second group because both the American and Vietnamese political/military establishments, each for their own self-serving reasons, did not want "their men" "fraternizing" with the inscrutable and untrustworthy (at least at the officer level) "other." In my case, though, I got "percussively sublimated" (i.e., "kicked upstairs") into the first group of active language learners when I grew a beard (as authorized fleet-wide by Navy CNO, Admiral Elmo Zumwalt) only to have my commanding officer retaliate against me for doing so by exiling me to the furthest, most remote outpost he could locate: "Solid Anchor," a river outpost support base two kilometers from the very southern tip of the country. For those who wish to know the precise location of the world's asshole, I can still clearly locate it for them on a map (although, some will now say that it has moved to Baghdad, right down the street from the increasingly-besieged Green Zone Castle).

Upside, though: I got to spend the next year actually using and improving upon my Vietnamese since I had no choice but to acually work and live beside Vietnamese in a situation of mutual dependence.

I have since gone on to obtain a second bachelor's degree in Japanese (which I even taught in an American high school for one semester), and I can function quite well in Mandarin Chinese, too. So, I understand full well the difference between formal high school and college approaches to language learning through reading and writing (for a few hours a week), as opposed to more natural, short-duration, intensive "aural-oral" (listening and speaking) methods of language acquisition as implemented (at least in former days) by the outstanding and highly effective Defense Language School programs. Everything I have heard or read about to date concerning the ludicrous and haphazard American "advisory" program in Iraq (i.e., "Iraqification," or "Browning the Bodies") convinces me that the bureaucratic bungling I alluded to in my third paragraph above still characterizes the Lunatic Leviathan: namely, Parkinson's Law meeting the Peter Principle in an atmosphere of rampant military careerism and political schizophrenia. None of this has anything to do with -- except to make impossible -- effective language learning by our military and civilian government employees.

I owe nearly all of my lifelong interest and competence in foreign languages to two principal sources: (1) the practical-intensive "listening-speaking" (as opposed to the usual "book translating" classroom stuff) course in Vietnamese that I took -- along with thousands of other ordinary enlisted men -- at DLIWC; and (2) the petty, vindictive American officer who found some authorized hair on my face so offensive that he banished me to a year of crushing boredom (with occasional terrifying interludes of bloody mayhem) so that I could and would learn Vietnamese from the Vietnamese in about one year's time.

In summary, then, no human language (at least as spoken by fellow humans) presents any insurmountable obstacle to learning by ordinary American enlisted men with a high-school diploma, about a year's worth of intensive/practical language study, and one year in-country working in a situation of mutual dependence with the "natives." Two years, tops. If America cannot afford that minimal commitment of time and resources to training our foreign legion -- at least its working, enlisted component -- then America ought not to have one.


My experience with trying to learn Arabic matches Abu Sinan's statements completely. I took 4 semesters of MSA in college and hired a tutor for Jordanian/Palestinian colliquial but I just didn't make much progress. I was starting to think that I had some sort of language learning disability when I spent a couple months taking German language classes in Germany and ended up learning more German in that time than I'd learned Arabic in 2 years.

Arabic is just so different from English and so hard that it requires not only much more effort but much more sustained and constant effort. If you devote just an hour a day or spend just a year, you will never learn it.

I definitely recommend spending significant time in an Arab country, preferably not Cairo, Beirut, or Morocco though. There are actually some very cheap programs in Yemen if you're willing to rough it a little.


My wife is taking Arabic. She is an excellent student. The goal is to be able to read classic Arabic texts.

She is no more learning to speak Arabic than I am learning how to play quarterback for the New England Patriots.

They speak English. We don't speak Arabic. Who's going to have the better intelligence?

john in the boro


This website has good info about language proficiency levels and language categories (how hard a time an English speaker might have in learning a particular language)


Duncan Kinder

Two years, tops. If America cannot afford that minimal commitment of time and resources to training our foreign legion -- at least its working, enlisted component -- then America ought not to have one.

This presupposes that - even for counter-insurgency purposes - Arabic speaking skills would be the only language skills needed.

Obviously Arabic is of limited benefit in Afghanistan while Al Qaeda is involved in Africa - and Latin America, for other reasons, is potentially very interesting. So there are lots of languages our soldiers would need to learn.

The question arises: if we train our soldiers in Arabic would be thereby be handicapping its ability to function anywhere other than the Middle East?

This line of questioning leads to a Reductio ad absurdumi> to the effect that the United States should limit its counter insurgency operations to English-speaking countries.

That reductio, in turn, bolsters my fundamental Middle East strategy, which is "Develop solar power and get the hell out of there!".

If you cannot solve a problem, then it is best to avoid it.

David Habakkuk

How does Farsi compare in difficulty with Arabic?

How well equipped are the American military and in particular intelligence agencies with people fluent in the language?


As someone with no arabic or foreign language skills, this has been an enlightening. I've been puzzled as to why seemingly simple things - like names - constantly end up with multiple spellings, as if there were no consistent guidelines. Even al-qaeda, or al-qaida, or al-qaydah or ...

One of the stories from interrogations at GITMO involves a young boy who was asked by his interrogator where the boy would go to get cash. The story goes that in the boy's dialect, the word the interrogator used meant tomato. So the boy mentioned place after place he could go to get tomoatoes and they decided he was involved in pretty high scale terrorist financing activities.

Who knows if the story is true, but even when I've seen it, I've seen the "word" that created the difficulties get spellings ranging from salata to zalati.

To an outside observer, it certainly doesn't seem that we are well geared to get accurate information. Thank you for some explanation of the difficulties involved.


I have to say that I wonder what definitions are being used here. (see Arbogast)

W. Patrick Lang


There is no generally accepted way of transcribing arabic into English text. There are many systems. Some prefer one others another. Complicating this is the former colonial heritage of many places in the Arab World. Arabic transcribed for a Francophone is necessarily different than if it is transcribed for an Anglophone. Thus - fouad for Francophones and fuad for Anglophones. They are the same name in Arabic. In addition, the Arabic lexicon is rich and varied by region. There are so many words for the same things that vocabulary selection in dialects is huge. Lastly, there are letters in the Arabic alphabet that simply do not exist in English or French although thre may be similar sounds. i.e., aspirated "h." Example, the name "ahmad." This is pronounced with a "hard" "h" sound. (There is also a "soft "h" sound which is a different letter. People who don't understand Arabic usually try to say "ahmad" as though the "ah" is like "ah" in "ah, yes." That is incorrect. Alternatively, they want to say the "ah" as though the "h" is like the "ch" in "loch." That is also incorrect. For some reason Hebrew speakers are particularly prone to the latter mistake. It is quite comic to hear Israeli spokesmen make mistakes like that. What it shows is that they have no Arabic at all. The mutilation of the name of Usama bin Laden's group is instructive. It is not "al qayda." The word has a consonant in the middle which is calle "ayn." It has the sound of a growl. The name minus the "al" has three syllables, not two. pl

Babak Makkinejad

David Habakkuk:

"Persian" is the name of the name of the language in English language - in German it is "Persisch" etc.

Using "Farsi" when referring to the official language of Iran is akin to saying that Italtians speak "Italiano" and Germans speak "Deutsch".

The historical name of the language has been "Farsi-ye Dari", i.e. Court Persian. And in fact, at times, the words "Farsi" and "Dari" where used interchangeably.

After WWII, Afghans started to refer to the Persian language in Afghanistan as "Dari". The Soviets, since 1920s, were using the word "Tadjiki" to refer to the Persian language spoken in the Tadjik Soviet Socialist Republic. In Iran, we had been using the word "Farsi" for centuries - having dropped "Dari" from it long time before.

So, all of a sudden, in the 20-th century, we were faced with 3 languages, "Farsi", "Dari", and "Tadjiki"! Which, of course, was just a political lie serving certain interests such as those of the Russian communists and the misguided Afghan nationalists!

I think that Persian is easier to learn for an Indo-European speaker than is Arabic.

On the other hand, as some one who studied Arabic for 2 years, I found that language's grammar confusing. Also, I did not care much about learning Arabic because I did not think that that knowledge could help me in my life; knowledge of the languages of the dominant civilization on Earth seemed to be more useful – English and French.

There was an Iranian fellow who had studied Arabic (Classical) for years and was in UAE with his parents an looking for the men’s room. And he did not know how to ask for it in (colloquial) Arabic, even after years of instruction in Arabic. He had to fall back onto the Quran; he started reciting, from memory, the story of Yusuf (Joseph) and Butifar to a security guard who burst into laughter and directed him to the men’s room!

Which brings me to my final point: one learns a language because one is going to get some benefit from it. And that benefit should be equal or greater than the time & effort spend on acquiring that language skill.

In case of Arabic, once the current US war in Iraq winds down what benefit will that knowledge bring? There might be opportunities in teaching it, in doing scholarship in it (if there is any funding for that – a very very big “if”), and doing business with Arabs. But for the vast majority of people who have learnt that language there is really no opportunity to exercise their skills.


Again, I can't judge the strategic weight of the American engagement in the Islamic regions, but it seems ponderous. Col. Lang's concluding statement in the pre-amble is too instructive to leave tangentially addressed.

"To think that young soldiers can be given any more than a passive "listening" ability in Arabic in a year or so of study is just illusory." pl

If plastic language acquisition skill is critical shouldn't the American Forces field children, say 10-12 year olds? Hearts and Minds indeed!

Kevin Rooney

I only studied a tiny bit of it, but Farsi should not be particularly difficult. About the same as Hindi I would guess.
First because it is an Indo-European language. Second, because it is one national language, like French or German or Japanese.
Even though we say Arabic as though it is a language, it is more like a language family. "Arabic" is not analogous to French. It is analagous to all the Romance languages but with all the mass media being in a modernized Latin that no one actually speaks. And with the Church still using the original Latin.
When I look at the world, I notice that the places that were functioning nations and had a unified language before colonialism did well after colonialism. Even if they were flattened by devastating war (Korea, Vietnam). The places that were tribal not national have suffered. (Africa) The Arabs are somewhere in between. Not quite fully separate nations, but clearly not one nation.
The difficulties of learning Arabic are thus a mirror of the difficulties the Arabic people are having finding their place in the modern world.

Kevin Rooney

I would like to see some program analogous to ROTC for critical languages: pay for folks university in return for some years of service. Regardless of what happens in the next few years in Iraq, we are going to need a sizeable number of American Arabists for decades to come.
I would also extend that to any language that we might need personnel for over the next couple of decades. Obvious candidates include Farsi, Dari, Pushto, Mandarin, Urdu, Russian, Korean.
Part of the long-term intellectual infrastructure for a nation that benefits so much from global trade.

Kevin Rooney

Yohan: I definitely recommend spending significant time in an Arab country, preferably not Cairo, Beirut, or Morocco though. There are actually some very cheap programs in Yemen if you're willing to rough it a little.

Yohan, I would be curious why you recommend against Cairo, Beirut, or Morocco. How receptive are Yemenis to Americans?


Many years ago there was an independent film called, "The Ballad of Gregorio Cortez," about a Texas manhunt in 1906, if memory serves. The entire story took place because of a faulty translation. A Sheriff, his deputy and an Anglo translator were questioning a Spanish-speaking horse dealer about a stolen horse they had been told he had bought.

The translator asked, "Did you buy a horse?" And Cortez answered, "No, I didn't buy a horse." The Sheriff became angry and drew his gun and then all hell broke loose--resulting in the deaths of the Sheriff and the brother of Cortez.

When Cortez was questioned in prison by his lawyer, using a bi-lingual Latina translator, it turned out that the whole tragedy had occurred over a word. The Anglo had used the word "caballo," thinking it was a gender neutral word for horse. But to Cortez a caballo was a stallion, and he truthfully answered that he hadn't bought a stallion, but a "yegua," a mare. Since that word was unfamiliar to the incompetent translator, he lazily told the Sheriff that Cortez was flat out denying what they knew he had done. But Cortez had innocently bought a horse that he hadn't known was stolen and was trying to be helpful!

You can judge from this how difficult it must be for U.S. troops to gain accurate intelligence from the locals.

Bob Haskell

I wonder if, in addition to the difficult and time consuming job of creating a relatively few truly fluent Arabic linguists, the military should not also attempt to teach a large number of soldiers (one per company?) 'enough' Arabic?

I pose the question based on my experience as a Peace Corps volunteer in the mid-1970s, first in Afghanistan and later in Morocco. In each country I took a six week, intensive language course that was entirely speaking-listening, with carefully selected vocabulary. In Afghanistan, I learned Dari, the local variant of Persian, which is admittedly far easier for an English-speaker than Arabic. Following the course and 4-5 months living and working soley with Afghans, I tested 2+ on the FSI. In Morocco, I studied Moroccan dialect and, though I used French professionally and (usually) socially, I was still able to get along quite well in Arabic in the market, ask and understand directions, engage in superficial social banter and understand a great deal of what was being said around me.

I suspect that having even a similar knowledge of Arabic available on a company level would prove extremely useful, and wonder why the army and marines have not attempted something along these lines.

The comments to this entry are closed.

My Photo

July 2020

Sun Mon Tue Wed Thu Fri Sat
      1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31  
Blog powered by Typepad