They always sell themselves that way and it is true. They are also negligent in their refusal to properly tailor their curriculum, or at least some courses, to the operating forces.

Whenever you pay people to do language all day long, five days a week for six to eighteen months and provide the greatest collection of native language speakers in one place AND don't give them anything else (college classes, other burdens, jobs to pay college bills) then you are going to have a great capacity to teach languages. That is why DLI is the best in the world and that is also why it is criminal that they cannot do better with what they've got.

One problem is that the school is only aimed at producing professional linguists in one mold and it is assumed that they will be back for following intermediate and advanced courses, dialect courses, etc. Thus, it focuses on learning a language clasically, stressing grammar, writing etc, to the detriment of being able to actually use it to communicate with people on the street. It is also very focused on listening and document reading rather than conversation and spends a lot of time on creating skills, such as detailed transcription, translation, and other listening skills, that not all linguists will need. I do think that we need to have people who have such a good basis, but the school should have stood up a program more focused on interacting with native speakers shortly after 9/11 to run alongside their main courses. So, you've got HUMINT guys and FAOs who are going to do a lot more face-to-face interaction studying alongside folks who are going to spend their careers listening to recordings and the skill sets are different. If you need operational language capability now, you can't wait to train up a linguist through three separate schools (basic, intermed, advanced). You need to focus the training.

With regard to Arabic, a major stumbling block is that despite the pleas of operational linguists, highly experienced military language instructors, some native instructors that worked as terps or in their own militaries, and even language academics at other premier institutions, DLI institutionally refuses to move away from the complete Modern Standard Arabic model they've run for years and move toward what many call "Educated Spoken Arabic." Basically, all literate Arabs know how to read MSA and understand it spoken, so it is the language of the press, official forums, etc. If you can speak MSA, almost anyone will understand you. Problem is, most people will respond to you in some mix of dialect. DLI says we can't teach all the dialects, so we won't try at all. However, other schools and agencies recognize that there is a core of common words that a lot of the dialects share and that you can teach a "standard dialect." DLI will have none of it. So, if a DLI grad from the Arabic basic course is stellar and attains a 3/3, which is on the order of less than 10% of the graduating population, from my limited experience, they still won't understand when an Arab in any city says "What are you doing here" because all the words in that sentence differ from MSA to dialect, but they are relatively common between dialects. To give an indication of the problem, the words that vary between MSA and dialect are basic, critical words: to do, to see, to look, to go, question words, negation, now, today, tomorrow, left, go straight, man, woman, etc. If one learns the standard dialect, Arabs will still be able to go deeper into their local dialect and not be understood, but if they want to communicate with you, they will be able to. Not all Arabs can easily speak to you in MSA though, or will try to.

If the right pressure was brought to bear, the school could be training at least some of your linguists, intel types, and FAOs to speak this standard dialect and could emphasize speaking conversationally over transcription, translation, and other more technical linguist skills, but to date the bureaucracy has successfully resisted.