In Practice
Have you noticed that if you want to improve basically any skill beyond a certain point, the only way to do that is to do the thing a lot? You can learn fingerings and practice scales and embouchure or what have you till you're blue in the face, but that Beethoven sonata still won't learn itself. There's a lot of learning to do calculations and toy proofs and unwinding definitions and working examples that goes into learning what it is to do mathematics, but at the end of it, there's just mathematics to do. Memorizing the ways to conjugate a verb or decline a noun or which prepositions imply which cases in the end won't make you freeze up less when it's time to speak as much as speaking will.
Actually, if you notice, music, math, and language have some of the largest amount of peripheral skills one can learn and hone rather than just doing the thing. Since these are some of the things I love most in life, it's easy for me to confuse the preparatory work needed for those activities with there being that same amount of secret sauce or eat your vegetables before other activities. Seemingly ironically, all three activities also have many stories of people jumping right over the prep work and doing the thing successfully. I remember being about 12 and envying my little sister's ability to just work out something she wanted to play on the piano by ear. I explained it to myself that she would need to learn "the proper way" like I had been for several years at that point. In the end she did, but now I wonder a little bit whether she needed to.
At the end of next month I'll run in my second marathon. This week I ran 36 miles split across four runs as 5 miles, 8 miles, 5 miles and 18 miles. Last week my long run was 17 miles and it felt great. This week 18 miles felt a lot harder, maybe because I pushed harder on my shorter runs than last week. 36 miles is the greatest weekly distance I'll run in the training program I'm following, although there are a 19 and a 20 mile run coming up in the next few weeks.
Reading the description of the training program, I was sort of shocked that there wasn't really any time devoted to technique. Sure, we're urged to run slow and to take opportunities to walk—which is fantastic advice in ways that a beginner like me was totally unable to see—but I was expecting technical insights into the mystery of The Gait or something. I guess it turns out that to learn how to run, you run.
This week I reached the end of the kanji-learning app I started last August. Over those 380+ days I reviewed nearly 160,000 flashcards in order to learn how to read 2089 kanji and 6,600-some vocabulary words.
It's undeniable that the app made a big difference in my ability to read Japanese: I remember spending whole days last year stumbling through the first couple pages of the light novel series ようこそ実力至上主義の教室へ (Welcome to the Classroom of the Elite) that I wanted to be able to read. Yesterday I started the third book in the series and read about 21 pages during my night out. I even encountered a vocabulary word from the final level (桟橋, "pier") in those pages.
Back in January, partway through the kanji training, I complained that there wasn't much I was able to find to read in the middle ground between the graded readers and easy material that I'd found on the one hand, and native content like novels, articles and manga on the other. I was hoping that one of the fellow travelers of the app would have a silver bullet for me, but actually instead what I heard from several people in different ways was: it's time to start.
It seems to me like there are several AI-powered language learning apps out there with amateur designers hoping to smooth the ladder between graded readers and native content. I wish them the best, especially because I do understand the feeling that would cause someone to want to reach for such a product. (Actually in fact, marketing to people who are wanting to have done the thing but not wanting to do it seems to me to be exactly AI's niche at the moment. But let me not rant.) In a world where the ramp up into learning a new language has literally never been smoother, I'm sure it's easy to see more of the little bumps along the way.
But on some level the only way to learn to understand the diction of a novel is to read the novel. I haven't made it through Ulysses or Finnegan's Wake yet, but I remember the way that reading Antigone or Romeo and Juliet involved a certain amount of brain calibration such that by the end of the piece, the quirks of usage felt natural and beautiful, and I imagine a similar process happens with Joyce where understanding is more of a frisson thing than a conscious calculation. And some of the bits that I needed to work at understanding (どうやら〜らしい for example, or what is going on when people end their utterances with ねぇ instead of ない) have nothing to do with kanji.
That is to say, starting ようこそ実力至上主義の教室へ in earnest in April, it still sucked at first. Following some Japanese posters on Bluesky in August and starting to read their posts about current events and politics still sucked at first. The first time you read a really long book in English probably sucked at first, right?
Over the last year I've taken classes and mentoring programs for Japanese and music. I also attended a math conference recently and got a ride from the conference to my relative's house where I was staying the weekend from a kind grad student who really grilled me most of the drive about why she should want to get a research postdoc.
It was a really interesting conversation: she understands the appeal of research very well, but given that research faculty like me (I'm at an R2 institution that is aiming for R1 status) still teach, she wondered why get a job which makes research a necessity for advancement rather than, say, a teaching position and doing research as a leisure activity. In other words, it might seem like more work and pain than it's worth. I'm not sure I gave particularly good reasons, but it was fun to think about, and useful for me to get a better sense of what it's like to be a particular graduate student rather than just rely on a projection of my fading memories of graduate school.
The thing she described as the point of grad school which I'm not sure I ever consciously articulated to myself but wholeheartedly agree with is this: her goal is to not need her advisor to access and do math.
In Japanese class on Zoom last fall, I couldn't help but wonder a little what I and the other students were really doing. Unkindly, all of us were for one reason or another still not yet airborne, and I wonder a bit about what had caused us to fail to launch. The one student, for example, who was able to really prattle in Japanese better than me; why was she in class? I guess my reading ability was maybe better than hers?
In some ways, I wonder if taking a class feels like what it means to try? Certainly that was the equation that lead me to sign up to take a class. But—maybe since I was in other ways actually trying—what I wound up taking away from the class is that I didn't need another class.
With music too; in the mentoring programs I did recently, there were some bits of technique that I had twisted up that were diagnosable in about an hour. And then hours and hours of playing student.
It is kind of sad to "graduate" in this way! It's often more comfortable to be a student in the way I was in these classes, which is to say, practicing with a certain amount of helplessness. It's less comfortable to go out and do the thing, especially if you put it down for a while or are picking it up for the first time. But in the end: if music sounds good, it is good. If graphic design looks good it is good. If you run far you can run far. If it sounds like Japanese to the person you're speaking with, it is Japanese. If the logic is sound, the proof holds water. If your goal is to improve at something, you should give doing it another try.
