alt Hacker NewsDon’t want to sound negative, just want to share my observations as someone who’s been actively learning Japanese for a couple of years. I much prefer how Japanese people actually teach verbs: 一段 (ichidan), literally "1-step"; 五段 (godan), literally "5 steps", plus a few exceptions sometimes called irregular verbs. It's not hard; it makes sense, just some textbooks (e.g. Genki) teach it really bad.
Learning Japanese is about learning a new way of thinking and structuring your thoughts. The more you learn, the more you realize it just doesn't fit into the English world. You can't really translate Japanese into English without losing nuance — and sometimes that nuance is important. So start early and start training your brain to think in the language, instead of trying to translate it and force it into English or some other language brackets. It won’t work; it won’t make sense; you will get stressed and confused.
Replies
> I much prefer how Japanese people actually teach verbs: 一段 (ichidan), literally "1-step"; 五段 (godan), literally "5 steps", plus a few exceptions sometimes called irregular verbs
This is literally what I teach in the article, including these translations. Quoting it:
> in ichidan ("one-row") verbs like taberu, the last syllable of the stem is fixed. it's always going to be be, no matter the suffix:
>
> (table)
>
> it stays on a single row in the hiragana table, hence "one-row".
>
>on the other hand, in godan ("five-row") verbs like nomu, the final syllable of the stem alternates between ma, mi, mu, me, and mo:
> (table)
> it spans all the five rows, which is why it's godan ("five-row"). the m* "wildcard" represents the entire ma/mi/mu/me/mo column.
You’re also mischaracterising my approach. I am not teaching to “think in English”. Quoting from the article:
> i'm using romaji as a convenient way to refer to phonetics in text. however, your "mental algebra" should match the hiragana table. so this is a reminder to not think in romaji when you do calculations. when we conjugate godan verbs, we literally go up and down the column. (maybe all these textbooks that used hiragana had a point!)
If you have objections, please engage with the article’s actual content, not with what you assume it is based on a glance (“oh he’s using romaji, this is thinking in English”). I’m using romaji for specific reasons that are motivated and explained in text, and I show every single pitfall of that choice as well.
> Learning Japanese is about learning a new way of thinking and structuring your thoughts.
I suspect this is true with most languages that are not in the same language family. The indoeuropean languages are all pretty similarly structured (don't @ me for not knowing the eighteenth tense of lithuanian), but it still takes time and effort for an english speaker to think in french, even if for the most part concepts translate directly. But bantu languages, eg kirundi or zulu, lean heavily on verbs with an entirely different conception of sentence structure and morphology, where you can stick entire clauses into one word, and you realize that your brain is picking up on patterns to decipher grammar that wouldn't make much sense in indoeuropean languages like dahl's law (except apparently in greek and sanskrit, where it's called grassman's law? Huh... now I know). Hausa is different still where you need to think about tense in an entirely different way—the pronoun is conjugated and the verb remains unchanged, and sometimes it feels like there are more irregular words than regular. Mandarin is beautiful, and actually quite simple to speak a little conversationally, if probably as difficult as english (or maybe more so) to master.
Learning foreign languages really makes you realize how central language is to basic cognition. You see the world in a different way, with different values and relations, depending on which language you speak/think in.
Ironically, I think this makes people extra susceptible to thinking that chatbots are intelligent (or even conscious, mr dawkins), even though it fails basic tests of memory and learning necessary to convincingly mimic understanding of, idk, time passing. Or motivation. Or emotion. But you can tell when it's awkwardly translating from another language in a way that a human would likely catch quite quickly if exposed to billions of documents.