How to approach nanori?

My knowledge of kanji is starting to feel somewhat okay when I try to read Japanese in the wild, but there remains a very significant difficulty: Japanese proper nouns.

Whenever I seen a Japanese person’s name I feel like I’m back to WaniKani level 1, when they were just some inscrutable moon runes. Even if I know the kanji I can be sure that I won’t be able to guess the correct reading, and indeed I almost never do.

I wonder how I should remedy this. Should I just start learning the names of significant people and use that to bootstrap my nanori knowledge? Or should I just learn the nanori of individual kanji wanikani-style? Or maybe some hybrid approach where I would memorize the nanori of kanji very commonly used in names and then switch to a more organic approach later?


Even if you know all the nanori, it doesn’t necessarily mean that you’d be able to read any given person’s name. It would still have to be confirmed with them in many cases, though obviously you’d be better positioned to guess it. If you see someone has the name 英治, possible readings include えいじ or ひではる. Knowing those readings means you could guess more easily if it’s important… but if you just stumbled across the name with no other information, you still have to simply accept that it’s essentially impossible to be certain.

So… I would just say, learn names as you need to know them, and don’t worry about names that don’t really matter… like the name of some company president in a news article you just read on a whim.


Very warily, from downwind and as silently as you can. When you get close enough, pounce.


Even Japanese have the same issue. When meeting someone for the first time they often need to ask. In a business setting you can often “cheat”, since usually you will exchange business cards up front and generally you can see their name written in romaji in their email address, if you did not catch them saying it when they introduced themselves (which can be tricky on first meeting). For a while one of my Japanese co-workers thought I was some sort of master of knowing or guessing Japanese name readings, til I told him. His response was “Of course. Why did I never think of that. It is so obvious”. Of course, he was just conditioned to ask.

On pretty much every form a Japanese person will ever fill out, there are two spaces for your name. Your name in Kanji and your name in katakana.

It is really no different than (for me in English) often needing to ask someone how their name is properly pronounced. I had an online meeting just today with a team from America and started off the meeting clarifying how to pronounce the names of 2 of the people (and their company name) to make sure I was addressing them correctly.

With my name (English and with roots from England), I would say that it is pronounced correctly about 20% of the time if someone has not seen it before (except when I am in England where it is a bit more common). Whenever I am paged somewhere or addressed by someone that has seen and read my name but not heard it, it is mostly incorrectly pronounced.

Interestingly, living here in Japan it is always pronounced correctly (well as close as Japanese can get - which BTW is much closer than all the mistaken variants I am used to). But then they are reading it in katakana, which is a directly phonetic spelling :grinning:


names of places and people are impossible for me at a first moment.

I have to see the furigana first.


I have heard that it is in fact possible for parents to effectively choose a somewhat arbitrary reading for whatever kanji they decide to use for their child’s name, but I would expect that generally people are a bit more conservative?

I guess what I’m trying to say is that it may be worth memorizing a couple hundred common names like I already know 田中 and 山田 for instance?

I think the issue is that I just don’t know how to bootstrap this knowledge. Right now when I see the name of a person I either first see it in roumaji/kana and then look up the kanji, or vice versa, but it never sticks because it feels so arbitrary. Like yesterday I read the name of Nobuo Uematsu (Final Fantasy composer) in kanji but I’ve since completely forgotten it.

Apparently it’s 植松 伸夫, out of which I could only have correctly guessed the correct pronunciation of 松(まつ)…

Just as it does for Japanese people. Given that, is it a realistic expectation that it will, or should, be easier for you?

For family names it is not a matter of them choosing it. But there are names that have more than one reading/pronunciation and you (or anyone) will not know which one that person’s family uses until someone tells you (or you hear someone else using it with them). Likewise, when conversing and hearing someone’s name, it is quite common for them be asked which kanji they use. More so with given names, since, as you mentioned, parents can choose what they like (and some do get creative - I guess because they want their child to go through the rest of their life having to explain their name to every single person they ever interact with).

At least there are no middle names here (which actually has an impact on those of us that do), so you have 1/3 less work to do…

1 Like

Oh I’m not expecting it to be easy at all, but obviously for Japanese people growing up in a Japanese environment it’s probably easier to build some intuition about the basics, such as the name of famous people or their schoolmates for instance that they will see written in kanji all the time.

I have no such background, all the Japanese people names I know are in roumaji, not kanji, so I just don’t know where to start and so far my approach has been to effectively ignore the issue altogether. Therefore I was curious to know how others foreign students here had overcome this hurdle.

But apparently the answer is “they haven’t”, unfortunately.

The guess here comes with vocabularies; or sometimes vocabularies bearing that Kanji meaning, albeit not using the Kanji. Vocabularies will aid memorization when time comes.

Historical things and place names might need memorization, but I don’t think it’s a priority.

You put place and people names into Anki. Then you grind :grin:


Well, family names are kun’yomi about eight times out of ten. It’s given names that are the really tricky ones. Of the top 100 most-common family names, 81 are pure kun’yomi, 10 are pure on’yomi, and the remaining nine are either a mixture of kun and nanori, or on and nanori.

(Fun fact: all of the pure on’yomi names, and only those names, have 藤 as their second kanji. That rule only works in the top 100 list, however - it breaks down before you hit the 200 mark.)


Oh that’s extremely good to know, I had no idea!

1 Like

Names are a massive pain in the ass, but I have noticed I tend to learn the names of characters a lot easier for various reasons (they’re more memorable, they often come up a lot in dialogue, etc) so that’s how I’m doing it - just learning the names of people I come across through my reading (and adding them to an SRS, because I’m obsessed with them).

I actually like names since you only really need to connect kanji + pronounciation and don’t need any of those pesky ‘meanings’ attached, lol.


I guess it doesn’t help that most of my reading at the moment is either from fiction where the overwhelming majority of the time names are full katakana or newspapers where it’s just a free-for-all and there are tons of names thrown at you that may or may not come back later.

I should probably try to consume fiction that contains proper Japanese names.

It’s definitely the names that come up over and over again that are easier to remember. Obviously there’s fiction, but just from NHK easy I picked up 岸田、安倍晋三, etc. So if you were reading a lot about 植松さん that’s probably a good starter as well.

きらきらネーム I think lmao. I don’t think I ever ran into any in the wild.


Just watch out for the fiction that never bothers to furigana its character names at all…


There’s this programming language called Malbolge that was designed specifically for being as hard to program as possible:

Malbolge was very difficult to understand when it arrived, taking two years for the first Malbolge program to appear. The author himself has never written a Malbolge program.

Sometimes I feel like written Japenese is the human language version of that.

1 Like

Aren’t some of those names of relevant people like prime ministers, famous politicians, etc.? I actually used these names for mining name vocab and it proved very successful :slight_smile: .

Yeah for sure, I just need to be selective. I guess I was hoping to approach Japanese names much the same way I approached kanji here with wanikani: learning individual kanji readings and then graduating to full names, but after reading this thread it’s clear that it’s probably a dead end.

1 Like

Oh yeah, that wouldn’t be very practical unfortunately :frowning: . But it’s also true for regular vocabulary - it’s “words are spelled with kanji and pronounced such and such” and not “kanji become words” I think. Not dabbling into precise etymology here, of course.

1 Like