For those who prefer learning grammar using English/Romaji

10 comments

1. i doubt this is optimal. i don’t speak japanese, but i speak chinese, and i can’t even imagine trying to just use their equivalent of romaji to learn, but it’s really up to you. if it’s doing what you want it to then that’s good. more power to ya, blacksmith!

2. For what it’s worth, the single best Japanese grammar resource I have ever found, the [Dictionary of Japanese Grammar](https://omgjapan.com/products/a-dictionary-of-basic-japanese-grammar?_pos=1&_sid=672cbbdd4&_ss=r), comes with romaji (instead of furigana). So there is that. I still hate romaji though and you should at least do kana first (that’s <= 3 days), before doing anything else.

3. If learning with romaji’s your thing, I *highly* recommend tracking down a copy of Japanese Verbs and Essentials of Grammar by Rita Lampkin. It is a great beginner reference for learning verb/adjective conjugation and overall structure. Its one glaring flaw is that it’s in romaji, but since you seem to want that, you should give it a try.

4. It’s totally okay! I learned only with romaji for the first couple of years, I learned how to read later. No rush.

   It took me a couple of months to learn kana, and then a couple of years to get to any decent reading speed.

   Memrise has some romaji decks.

   Duolingo has the option for romaji over the kanji/kana in the app now.

   Language Reactor can turn Japanese netflix subs into romaji.

   jisho.org is romaji friendly.

   Erin’s Challenge has romaji as an option for their activities.

   Maggie Sensei uses romaji.

   I think that’s all my semi-current romaji friendly resources.

5. It seems clear that you are probably not willing to have your mind changed about this no matter what, but I’ll throw my hat in the ring quickly…

   kana only take a few days to learn, and even if you spend like 15 mins each day it shouldn’t take more than 15 minutes. I would not suggest starting grammar on romaji because sooner or later you will have to switch to kana anyway, so you are simply kicking the can down the road.

   it also will mess up your speaking, because the way that Japanese is spoken reflects the mora system, not the English vowel system. For example, try saying konnichiwa. You probably said it ko-nee–chee-wa, four syllables. But it’s actually こんにちは, 5 mora, which is obvious if you write it in kana but not if you write it in romaji.

   of course, you can always fall back on the infallible defense of “this is just the way that I do things,” which I think is an incredibly poor mentality to have when you learn something. A) How could you possibly know this is more effective for you if you’re literally a beginner and B) This won’t change the fact that the method is objectively worse. There are people with 2000 day duolingo streaks that can barely read a basic passage and there are people who’ve studied 300 days who’ve passed JLPT N1 with perfect marks, so yes some methods are objectively better than others.

6. > I’m not forgoing Kana or Kanji

   The comments in this thread have been pretty heated, so I’m trying to focus on the fact that you said you are **not** foregoing kana and kanji. I.e., you do plan to learn them, and I hope you are spending at least some time studying them already, especially kana. Don’t delay too long.

   > but I can’t code switch fast enough for learning grammar and retaining it, I’m too task-loaded, as we say in scuba. I can’t retain anything when I’m trying to do too much at once.

   Here we have the crux of it, and I both sympathize and partially agree. I’ll give an example. When I started using one of the Tango N5 anki decks, I was overwhelmed by each new card. Even though each card was “i+1”, meaning it only had one new word, I still struggled. I had to figure out each kanji / kana character, and then use the combination and order of kanji/kana to come up with a meaning, and I also had to remember the reading (in terms of kana), and I had to remember how to sound it all out (roughly equivalent to converting to romaji), which was slow for me at the time. And I tried to read the context sentence and understand it, including readings and saying it out loud, all before clicking the show answer button, at which point I had to listen carefully to the audio to make sure I had read it correctly (including not only pronunciation, but pitch accent). I would replay the audio until I could more or less shadow it. And I was still learning about the little idiosyncrasies, such as the way that certain consonants are delayed in order to blend into the next kana. There was just so much going on, that my learning the words and context sentences for 10-15 new cards per day, plus reviews, was as exhausting as studying a grammar section in Genki. (Indeed, I stopped studying Genki for months, and just did the Tango N5 deck, and made huge progress in reading and understanding. I did eventually go back and finish Genki 1 and 2.)

   So I get it. Rather than overwhelm yourself with 5-10 different things, you’re taking a divide a conquer approach. Study grammar in a setting where all your focus is on the grammar. And that approach has its advantages, even if it has its disadvantages. Just make sure you have time in your weekly schedule, if not daily, to work on kana and kanji. Don’t let that skill fall behind, or you’ll hit a wall.

   > Plus, learning words/sounds/kana/kanji without any clue how to put them together is just too difficult to motivate myself.

   Again, unless you never plan to read Japanese, you’re going to need to tackle this someday. And the longer you wait, the longer you’ll be unable to utilize the vast majority of learning resources out there.

7. I’m also bad at grammar but found this app has helped me commit it to memory better: https://twitter.com/kaku_app/status/1531959776350916608?s=21&t=VvrZjVHkTFwUNOptvKEwMQ

8. By using Hepburn it defeats the entire advantage of it.

   Even Japanese Wikipedia uses romanization to explain Japanese grammar, because it’s, frankly, a better way to explain Japanese grammar than 仮名, and of course all linguistic papers use it too, but they don’t use Hepburn because the biggest criticism on it is that it obfuscates the grammatical patterns and suggests stem changes occur when they don’t.

   form|待つ|書く|話す
   –:|:–:|:–:|:–:
   conclusive|mat-**u**|kak-**u**|hanas-**u**
   negative |mat-**anai**|kak-**anai**|hanas-**anai**
   polite|mat-**i**-**mas-**|kak-**i**-**mas-**|hanas-**i**-**mas-**

   The system is perfectly logical, and one can see that these three classes of consonant verbs receive the same endings after the final consonant of the stem:

   form|待つ|書く|話す
   –:|:–:|:–:|:–:
   conclusive|mats-**u**|kak-**u**|hanas-**u**
   negative |mat-**anai**|kak-**anai**|hanas-**anai**
   polite|mach-**i**-**mas-**|kak-**i**-**mas-**|hanash-**i**-**mas-**

   Hepburn offers the illusion of stem consonant mutations by being based on English, rather than on Japanese phonology, where none occur.

   Also, how Hepburn explains sequential voicing:

   voiced|unvoiced
   :-:|:-:
   t|d
   ts|z
   ch|j
   k|g
   h|b
   f|b
   s|z
   sh|j

   While it could simply be:

   voiced|unvoiced
   :-:|:-:
   t|d
   k|g
   h|b
   s|z

   Which does not offer the impression that it’s more complex than it is.

   Despite the common aversion to romanization to explain grammar, there are good reasons to do it, which is why it’s even found in linguistic literature in Japanese about Japanese, but one loses all these advantages, and introduces far more problems than the native script, by using Hepburn for it

9. FWIW, the textbook [Japanese – The Spoken Language](https://en.m.wikipedia.org/wiki/Japanese:_The_Spoken_Language) is entirely in Romaji.

10. No thanks. No one prefers this bullshit way of learning.