Background

Sometimes an accidental decision has interesting second-order consequences that you really don’t see coming. One of them was being dropped from my intended class during my first-year fall and having to take JAPN 101 instead. Unexpectedly, I found that I quite liked the challenge, and soon I found myself increasingly invested in something I had never originally intended doing:

• Fall of 2023: JAPN 101, starting to quite like the grammar challenges, investing increasingly more time into the language.
• Winter of 2023/2024: JAPN 99 (Winter Study) course, applying to Japanese summer intensive programs, getting into KCJS!
• Spring of 2024: APN 102, begin assembling my own Anki deck (Zipf’s Law sorted collection of coursework vocabulary), finishing Genki 1 and 2 in my free time to be ready for the summer.
• Summer of 2024: Complete KCJS, which is effectively equivalent to completing JAPN 201-202 at my school.
• Fall of 2024: Then JAPN 301, which is the single hardest humanities course I have ever taken.

Then something shifted in my thinking around the winter of 2024. By that point, Japanese study was so engrossing that I had to make the call to prioritize what mattered most: my Geosciences studies. What drove me to make this choice was because I realized that a class I really wanted to take in Geosciences conflicted with the time offerings for Japanese, and that taking both would effectively be an overwhelming workload that would prevent success in either course.

That didn’t mean I was giving up on the language, though. I had come far too long to suddenly quit now. So I knew I would need a new method…

Breakthrough

The SRS Idea

Most of my time spent studying was with Anki anyways, where the process is a nice iterative loop. For those of you unfamiliar with Anki:

• Bad at a card? You’ll be seeing it a lot
• Good at a card? You’ll see it less

So you spend most of your study process working a lot on the things you’re bad at and doing only the minimum needed maintenance on the things you’re already good at. The trick is what’s called spaced repetition: seeing cards at the moment you’re about to forget them helps you remember them better (and figure out what you’re bad at).

This is quite nice, but what about if we used SRS for every single part of Japanese? After my final exams were over, that’s exactly what I did.

WaniKani (鰐蟹)

Japanese has a lot of common characters: 2,136 常用漢字, to be exact. It’s difficult to remember them because they often can be read multiple ways, and don’t mean very much on their own anyways. Fortunately, WaniKani exists and is very good. Its SRS system is a bit more coarse than Anki’s, and has longer time scales, but otherwise it is quite good at its job and is a functional SRS.

In WaniKani, all the kanji in the common set are organized into levels, of which 60 is the final level (all kanji completed). At the time of writing, I’m at Level 20, and I expect to hit Level 30 before the end of the year. Assuming I can reach that deadline, I can realistically expect to reach the end of WaniKani (and thus functional kanji fluency) by the end of 2027. That said, this is a long process that I do not expect to be able to rush. If it takes two years, it takes two years; if it takes three, it takes three.

Bunpro (文プロ)

Bunpro is basically the same thing as WaniKani but for grammar. In my experience it’s quite good at the lower level I’m at (N3), which I will complete by the end of the year, but the way it designs questions is often frustrating (many grammar points are often valid for the same sentence, and disambiguation can often be difficult at higher complexity grammar). At the high grammar levels, it seems that Bunpro will still be a useful reference and way to learn new grammar, but it will not be useful in remembering it long-term.

Future Direction

This year has felt like trying to climb through the Intermediate Plateau that all language learners face; beginner materials are too easy and native materials are too hard. I think that completing WaniKani and finishing the Bunpro grammar series up to at least N2 will give me a strong starting point to try and get out of this trap by having a strong enough set of background information to read most any text.

My daily routine at this point still consists of just doing Bunpro, WaniKani, and Anki in a big loop, as my studies have me too down to end up working on anything else. Like this summer, next summer I will have a lot of time to spend on my studies. Instead of just repeating my SRS study habits in a loop, I think I want to aim for doing more output and media input, as understanding where I’m going wrong will be a crucial input for assessing my progress once SRS stops being helpful and one of the few remaining sources of easy language skill growth.

In the very long term, I would like to work up to reading a “real” novel in Japanese, and I also need to work more on my pitch accent. Natives report that I sound intelligible but a bit off, which is good… but also bad. Still better than where I was at the start, that being unable to speak any Japanese at all!