Navajo Code Talkers

January 22, 2003 6:54 AM

You've probably heard of the WWII Navajo "code talkers" who managed to baffle crack Japanese cryptanalysts and were credited with enabling US success at Iwo Jima. Civil engineer, journalist and photographer Philip Johnston was the determined mind behind the "windtalkers". The son of missionaries, Johnston grew up on a Navajo reservation and was one of only a handful of outsiders fluent in the Navajo language. A bit of his background is included this article, and you can read a complete history of his plan, view an archive of photos by Johnston, and see copies of his enlistment application letter to the Marine Corps commandant, as well as a recommendation letter from the Commanding General. (more inside...)
posted by taz (13 comments total)

Called "the simplest, fastest, and most reliable means available to transmit secret orders by radio and telephone circuits exposed to enemy wire-tapping", the Navajo code also proved the most inscrutable; according to the Japanese chief of intelligence, his code breakers were able to decipher the codes used by the U.S. Army and Army Air Corps, but never cracked the Marines' Navajo code. There is an online dictionary of the code, and a brief explanation of its implementation and a sample page from the actual code talkers' manual here. For more information, Harrison Lapahie, son of one of the code talkers, has an in-depth tribute site, and there are code talker photo galleries here and here.

We have films, books, documentaries and commemorative medals memorializing the contributions of these men, but they all pale to insignificance next to the ultimate gesture of recognition. I wonder if Philip Johnston would have seen this as the bel-dil-khon ut-zah for his code talkers, or would he have said "di-giss-yahzie do-ya-sho-da"?
posted by taz at 6:55 AM on January 22, 2003

From the online dictionary.

AFRICA - ZHIN-NI - BLACKIES

I guess in times of war political correctness goes out the window...
posted by PenDevil at 7:07 AM on January 22, 2003

Irony: Nearly all the American Indian Code Talkers were forbidden to speak their Native languages when they were students in federal boarding schools.

One who was beaten for speaking his Comanche language was Charles Chibitty, the last survivor of the 17-member Comanche Code Talkers unit of the 4th Signal Company. The Comanche Code Talkers were recognized in 1999 with the Knowlton Award for distinction in Army intelligence. Chibitty received the honor in a ceremony in the Pentagon, where he kept repeating a question, "Why did they wait so long to recognize us?"

If you're even in Kayenta, Arizona - probably on your way to or from Monument Valley - the Burger King there is owned by a former Code Talker, and has an interesting exhibit on them.
posted by gottabefunky at 7:15 AM on January 22, 2003

BTW, good post Taz.
posted by gottabefunky at 7:16 AM on January 22, 2003

For all the depth about what was said in Navajo, what the reports failed to capture is the reason why linguistic analysis of the Navajo tongue failed so spectacularly:

Apparently, Navajo uses phonetic distinctions nobody else even dreams of.

My linguistics professor in college was an interesting character. Not sure exactly where he came from, but among his many quirks, he absolutely could not hear the difference between the words "pen" and "pin".

Now, on the one hand, the difference is obvious. On the other...if you think about it hard...your mouth does almost the exact same thing with both words. They really are pretty similar...

Turns out, when we're babies we babble all possible sounds our mouths can make. As we get older, the set of babbled phonemes coalesces into what we're hearing from the language around us, and eventually *becomes* the language around us. But while we gain the ability to speak the language, we lose the ability to recognize the phonemes that weren't around when we were young. Eventually, this loss becomes near-permanent.

(Thus, the absolute idiocy of starting foreign language acquisition in high school and college. But that's another story.)

The Japanese are famous for simply not hearing the difference between the liquids -- L and R. They can get by even with when they fail to relearn the distinction, due to the relatively recognizable form of the rest of the syllables. Navajo engenders no such flexibility. There are so many unique discriminators, that are so different from what the "traditional" phonetic selections are, that listeners likely couldn't even transcribe what they were hearing, let alone attempt to translate it.

This has cryptographic significance: A simple symbol code, even amplified by having common letters posessing multiple forms, is pretty useless -- you can train a grunt to use "a-kah-foo" instead of J. But the grunt will use an easily parseable American accent. The Navajo will use something else entirely -- and every time, it'll be slightly different, because the mood is different, the tone of voice is different, the position in the sentence is off. And all those subtle shifts, combined with the inability to segment out the speech, lead to an undecipherable codec.

It wasn't the strangeness of the words. It was the oddness of the phonemes. In the interests of simplifying the explanation, it's a bit sad they missed this one critical point.
posted by effugas at 8:14 AM on January 22, 2003

thanks, effugas. i guess that means that this kind of code isn't likely to work in the future - presumably we either can now, or soon will be able to (in say the next ten or so years), use statistical analysis to fragment a (digitised) recording into phonemes, even if those phonemes have don't exist in our native tongue. or does the shift in intonation between messages due to context imply that we need a (culturally-based) model to correlate between messages? i would guess that's not a problem, given sufficient data.
posted by andrew cooke at 8:38 AM on January 22, 2003

Navajo is apparently full of subtlety, nuance, and puns. (Thanks Tony Hillerman)

And even a Navajo soldier captured by the Japanese had no idea what his fellow tribe members were saying.
posted by gottabefunky at 9:05 AM on January 22, 2003

On effugas'stopic raised - I've always wondered about a link between music and language acquisition. ("musical ear" = "ear for language"?)

I play a few instruments, and while my Spanish vocabulary and grammar (learned in college) are atrocious, my accent sometimes gets compliments. Then I say something like "I wanting go bus now please?", and that stops.
posted by gottabefunky at 9:13 AM on January 22, 2003

fascinating link and discussion. thanks.
posted by mokey at 9:16 AM on January 22, 2003

There's a little bit about the language itself here, including this comment:

Navajo verb is "like a tiny imagist poem." na'il-dil means "You are accustomed to eat plural separable objects one at a time." This linguistic and phonetic complexity makes the language not only difficult for non-Navajos to understand but almost impossible to counterfeit.

So... poetic, and economic, too!
posted by taz at 9:31 AM on January 22, 2003

gottabefunky - i've heard there's a link between perfect pitch and certain languages. iirc some languages (cantonese?) the (absolute?) pitch carries information, so native speakers tend not to lose the ability which (pushing my vague memory too hard) everyone is born with. or something.
posted by andrew cooke at 9:32 AM on January 22, 2003

It wasn't just the Navajo; the Apaches and the Commanche were all involved in the code talking program. It's so ironic that the code talkers were so honored for their unique language when they were beaten severely for using their native language in schools in Arizona and New Mexico. There are people in New Mexico my age (in their 30's) who don't know their tribal language or heritage because their parents were forbidden from speaking it in schools and deliberately avoided teaching their children.
posted by answergrape at 9:52 AM on January 22, 2003

Yay, taz is back with another great post. Thanks!

If anyone is curious, here's a sound clip of the Navajo language.
And this site isn't too pretty, but I found the content interesting, especially the tribute to one of the Code Talkers, Harrison Lapahie made by his son. I found some of the individual profiles of the Code Talkers interesting.
posted by madamjujujive at 7:28 PM on January 22, 2003