Wednesday, August 30, 2006

The Din of the Night

A while back I wrote about "din", which I equate to getting language "stuck in your head" like a pop song. Common sense suggests that one way to experience this would be to listen to repetitive input for many hours in one day. The most Thai listening I've been able to squeeze into a day since then is about two and a half hours, which isn't much.

So I may not have experienced exactly what Krashen would call "din", but I have noticed something interesting. When I'm falling asleep at night or waking up in the morning, I'll usually hear something from my Thai mp3s in my head. Maybe this is some kind of mental "practice", similar to din.

It's not always the case that what I hear when falling asleep comes from something the same day. Often it's from audio which I haven't listened to in a few days or more. It's interesting how the mind works.

Monday, August 28, 2006


I almost never watch westerns, but I have been enjoying the HBO series "Deadwood" on DVD. I'm pondering the use of English in the show and comparing it to the Thai movies and television I sometimes watch.

The vocabulary in Deadwood consists of common words, but for an ESL learner the level of English would be quite advanced. The characters speak in outdated idioms and turns of phrase. Important plot points are communicated by indirect dialogue. There's an implicit assumption that the viewer is familiar with American history, culture, and geography. Even as a native speaker, I have to watch and listen carefully.

The closest Thai equivalent I can think of is a movie that uses a regional dialect (like parts of Ong Bak, Thai Warrior) or a royal dialect (like Legend of Suriyothai).

It's interesting to me that the "advanced" nature of the English in Deadwood is due to cultural, stylistic, and idiomatic features, rather than vocabulary.

Monday, August 21, 2006

Training for Thai

135 hours:

A few weeks ago, AJ wrote a great motivational post entitled, "Learn a Language, Get in Shape!", pointing out how L2 input and physical exercise can be done at the same time.

I recently started training for the Seattle half-marathon, which will take place in November. My times in previous half-marathons have been slow, but I'm quite content to get some exercise, enjoy the race, and finish without injury. If I'm lucky, I may even take a few minutes off of my personal record.

Running provides a great opportunity to listen to Thai language mp3s. I'm running for a half-hour a few times per week, plus a longer run on the weekend, so the hours accumulate quickly.

I've had to solve a few issues. The main one is that ambient noise like wind or traffic is a big obstacle to comprehension. Since I value my hearing, I don't want to compensate for ambient noise by turning up the volume. I never noticed a problem with ambient noise while listening to NPR podcasts or English talking books, so I must be much better at unconsciously filling in gaps in English.

After a number of experiments, I'm now using a pair of Sony earbuds which cut down considerably on ambient noise. It also helps to listen to short, 1-5 minute segments over and over, rather than trying to listen to an entire newscast. This enables me to follow the thread even when a few words are lost.

Thursday, August 17, 2006

Din in the Head

Stephen Krashen has written about a phenomenon he calls "din". It's something like getting a song stuck in your head. Din is an "involuntary mental rehearsal" which people sometimes experience after a lot of exposure to L2 input. Krashen quotes a Russian language learner describing the experience this way:

"By the third day, the linguist in me was noticing a rising din of Russian in my head: words, sounds, intonations, phrases, all swimming about in the voices of people I talked with. This din blocked out all my other languages."*

I've personally experienced short Thai phrases and sentences getting stuck in my head, but never anything quite as dramatic as what's described here. It would be an interesting experiment to try to induce "din" by listening to comprehensible input for a long time at once. It seems like short, interesting, repetitive input might be the most likely to induce this phenomenon.

Unfortunately, this experiment would require that I find a few consecutive hours during which I can listen to Thai audio, and the only time that is available is in the evening after my son goes to sleep. It could be hard to stay awake and alert. Maybe a cup of coffee would help.

(*Elizabeth Barber, quoted in Krashen 1983, found in this pdf file which takes a long time to load. The original paper does not appear to be archived on the internet.)

Tuesday, August 15, 2006

130 hours:

For a change of pace, I'm listening to audio from a Thai cartoon called Pang Pond (flash site). Some of the silly dialogue is getting stuck in my head, which is perfect.

To say the least, the vocabulary is different from that of internet news broadcasts. I haven't heard one mention of President Bush, Lebanon, or Bird Flu. The word "jao naa tii" (เจ้าหน้าที่ , meaning "police" or "authorities"), which I learned from the news, does appear in an episode about UFOs. Many words which haven't been reinforced for a long time also appear.

I suppose this doesn't do much to fine-tune my pronunciation, unless I want to talk like a cartoon. If an ESL learner only ever listened to Bugs Bunny, they might end up with an unusual way of talking. But a little of this keeps things fun.

Monday, August 07, 2006

117 hours:

I've spent about 15 hours so far listening to news stories. One nice thing about news stories is that the domain is somewhat restricted, so new words get repeated a lot.

Examples of words which have recently become very familiar are words for "report", "news", and "company". Many geographical and political terms have become very familiar. Sadly, words dealing with war and loss of life are also becoming very familiar.

Examples of words which have recently become somewhat familiar are words for "government", "president", "prime minister", "science", and "environment".

Words for "crude oil", "atomic", and "fuel" have appeared, but they have not yet become familiar.

Listening to the news is a lot like narrow listening. The domain of discussion is restricted, subjects change slowly from day to day, and the subjects are interesting. An entire news broadcast is longer than the 1-3 minutes recommended for narrow listening, but broadcasts can be edited in Audacity, and individual stories are usually 1-3 minutes, give or take.

Saturday, August 05, 2006

115 hours :

Earlier this year, I was surprised by the very rapid way my vocabulary was growing. I think this was partly because I had such a small vocabulary to begin with. The old AUA books and tapes where I started learning Thai deemphasize vocabulary almost completely, teaching just enough to support grammar patterns, simple dialogue, and pronunciation. When I started attending the new AUA and focusing on input, my vocabulary increased very rapidly.

Recently, my vocabulary acquisition has noticeably slowed. This is expected -I have already acquired the very high-frequency words. As I get closer and closer to the long tail of the Zipf distribution, new words occur and recur less frequently. So of course, they take longer to acquire.

I spend most of my practice time these days listening to the news. Despite my previously mentioned concerns about propaganda, VOA is the most comprehensible Thai news I've found on the web. Almost every segment has a mix of completely new vocabulary, somewhat familiar vocabulary, and very familiar vocabulary, so it's never too far from i+1.

One of my goals for the future is to understand Radio Japan news broadcasts in Thai. Radio Japan is much more demanding than VOA. The announcers speak faster, the vocabulary is harder, and the subject matter is less familiar. When I can understand Radio Japan, I'll know that I've made a lot of progress.

Wednesday, August 02, 2006

AUA uses a student's cumulative number of classroom hours as a guideline for expected level of acquisition. Every 200 classroom hours or so, most native English speakers are ready to advance to a new level of Thai. Classroom time at AUA is primarily spent actively listening to native speakers.

Inspired by AUA, I started keeping track of how many hours I listen to Thai. I took 32 hours of AUA classes in April. Since then, I've listened to 80 hours of other content, so I'm currently averaging around 27 hours a month, for a total of 112 hours.

I only count quality time. Mp3s count, but only if I'm paying attention. Movies count, but only if I watch without subtitles or significant distraction. Thai radio playing in the background does not count. Little snippets here and there don't count. In other words, my total is a conservative measure of high-quality input since April, but it's not anywhere close to the number of hours I've actually been exposed to Thai.

For the last three months, I've blogged mainly about my ongoing epiphany that people can acquire languages naturally through input. Now I think it might be interesting to reflect on changes as I listen more and more, using this cumulative total as a measure.