A while back I mentioned that I’d no longer be using the “X-thousand Sentences” type books because my language exchange partner said they weren’t natural-sounding sentences. In retrospect, I’ve been wondering if that wasn’t a mistake.

Here’s the thing about those books. They may not be the exact sentence that a Taiwanese person would choose in whichever situation they appear in, but the thing is, I’m not memorizing the sentences to use, I’m just repeating them. So it doesn’t matter all that much that they’re not all perfectly idiomatic. They’re still acceptable Chinese, both syntactically and semantically, because they’re written for native speakers, by native speakers. And there are thousands of sentences, with MP3 recordings, just sitting there on my shelf waiting for me to shadow.

I can shadow 250 sentences in half an hour using these books. And I don’t have to put any effort or time into finding or preparing them. Sentence mining, by comparison, is extremely time-consuming. I might find and input 20 good sentences in half an hour of mining TV shows or movies, and that doesn’t even include actually reviewing them. For the purposes of developing muscle memory and strengthening the neural pathways in the brain required for immediate, non-thinking production of fluent speech, it’s quite obvious which approach ought to be more effective.

I’ve also spent some time recently (which probably should have been spent on preparing MA applications) looking into methods used by high-level interpreters, military personnel, field linguists, and other people who have a real need to become fluent in a short time. I’ve found some really interesting, useful stuff in the process. For instance, there’s a paper here by Christopher Guichot de Fortis, a senior staff interpreter at NATO, describing how he works on developing his ‘C’ languages into ‘B’ languages. ‘C’ languages are ones which he can interpret from (that is, languages in which he has a strong receptive ability), as opposed to ‘B’ languages that he can interpret into (languages in which he has a strong productive ability), and his ‘A’ language or mother tongue. It’s a really good read, and well worth your time.

What I’ve been doing recently is watching a TV show or part of a movie, selecting sentences from it to type into Anki, and then reading the sentences aloud while reviewing my Anki deck. Like I said, using the “sentence-mining” model I might add 20 sentences in half an hour. I think on average, I probably added 30 sentences per day, weekdays only, and usually reviewed about 50. To reach the magic number of sentence repetitions Mike Campbell often talks about (which is 100,000), it would take just under 5 years at this rate.

My plan going forward with this is to do 250 new sentences three times per week, and repeat a total of 750 sentences (whether new or review) 6 days per week. And at 750 new sentences and 4500 total repetitions per week, it would take just over 22 weeks to reach 100,000 repetitions. I’m not sure I’ll be able to sustain that volume with everything else I’ve got going on (after all, repeating 750 sentences takes about 90 minutes each day), but I’m sure going to try to.

After reading the paper I linked above, I also decided to take another pass through Thought and Society 思想與社會, choosing one article/lesson per week that I’ll shadow daily. After that, if I’m seeing positive results from it, I’ll choose some lectures from ICLP’s Speech Series: Real World 社會大學 to work with, but of course that will be a couple of months down the road.

So after all this reading and researching and such, this past Friday I tried something new for my morning routine. I shadowed a lesson of Thought and Society, which took about 15 minutes. After that, I did about 250 sentences from one of my books. That’s fairly intense, tiring work, but it paid off when I met my tutor that afternoon. “What happened to your Chinese? It seems so much easier for you today than it usually does.” She does not say things like this lightly (read: ever), so I take that as a pretty solid preliminary indication.

This is not especially pleasant work. It’s actually fairly fatiguing, but if the results are there I’m going to stick with it. I’m aiming for ILR4-level by the end of this year, and while I feel like my receptive skills (reading and listening) would get there fairly easily with what I’m doing, that’s a serious jump in speaking ability. I really don’t think “more of the same” is what I need here, especially considering “the same” is exactly what got me into this situation in the first place. I need to focus on results-driven approaches, regardless of how much fun they may or may not be, and I feel like this is what I need to be doing right now for my speech. The real fun for me lies in being able to use the language easily.

I know a while back several people got excited about Glossika’s method, and I wonder if anyone stuck with it. I’d be interested in hearing about how it’s going and what kind of results you’ve seen with it in the comments.