Hepburn for the win!

I’m often critical of the Japanese authorities (and rightly so), so I must also underline when they do the right thing. And for once, they are. Japan is officially dropping the Kunrei-shiki way of transcribing their language into the Roman alphabet in order to officially adopt the Hepburn system.

Let me explain if you’re not familiar with the whole thing.

 

In the 19th century, when Japan opened up to the world, it soon became necessary to transcribe the Japanese language into the Roman alphabet for the increasing number of foreigners coming to Japan (not many tourists yet, but many officials, scientists, and so on). Many systems were developed, but two became dominant:

The Hepburn system, designed by Mr. Hepburn, which is what you and I use when writing Japanese words, and the Kunrei-shiki system (not designed by Mr. Kunrei-shiki 😉 ), which you may not have heard of unless you live in Japan.

Their difference?

The Hepburn system is designed to be as close as possible to English pronunciation. It’s not perfect, some Japanese sounds don’t exist in English, but it’s as accurate as possible (but “ramen” is not pronounced in Japanese the way you might think, for example).

On the other hand, the Kunrei-shiki system is designed to be as logical as possible from a Japanese language perspective. However, it doesn’t care much about how a word written with this system would be pronounced in a foreign language. So, in short:

The Hepburn system is better for English speakers and other Westerners.
The Kunrei Shiki is better for Japanese people.
The latter was adopted as the official way to romanize Japanese decades ago and has been taught to all Japanese people ever since. Of course, it’s the one that was adopted, it was designed by a Japanese national for Japanese people, so it must be the best, right? (That’s Japanese nationalism for you).

One thing the government hasn’t thought about is who actually needs to use romaji and when?

This may surprise you, but it’s usually not Japanese people who need to use romaji (Japanese written in Roman characters), but foreigners.
However, in typical Japanese fashion, this fact doesn’t seem to have occurred to any officials. A romanization system exists, it works, what is there to complain about? Right?

But before I go any further, I might need to explain how they’re different.

First, let me explain how Hepburn makes more sense to English speakers and Kunrei-shiki makes more sense to Japanese speakers.

Take this set of kana (Japanese phonological units): た, ち, つ, て, と. They are usually grouped together in Japanese, you’ll understand why right away:
In Kunrei-shiki, this set it becomes: ta, ti, tu, te, to. Logical, right?
Well, in Hepburn it will be: ta, chi, tsu, te, to.
Why? Because the pronunciation of ち is much closer to “chi” than to “ti”, and つ is actually pronounced “tsu,” not “tu”.
So you can see how one is designed to help foreigners pronounce Japanese, while the other is designed for Japanese to use the Roman alphabet without really taking phonology into account.

Here is another concrete example:

There’s a beautiful island where I live called 小豆島. Can’t read it? No problem, let’s romanize it.

In Kunrei-shiki it’s Syoudosima.
In Hepburn it’s Shodoshima.

Neither is a perfect transcription, the first “o” is a long “o” (and it’s sometimes transcribed as “ō” but I’ll keep it simple and spare you the macrons) , the “sh” is not 100% the English “sh,” but the way you read the Hepburn is actually much closer to the actual pronunciation than the way you read the Kunrei-shiki. But the Kunrei-shiki is more logical for Japanese people.

Except that… No Japanese people will ever write Syoudosima, they will write 小豆島. Unless they write to foreigners, and in that case, if they write Syoudosima, it won’t help at all.

So Kunrei-shiki is basically useless, right?

Okay, there’s one caveat. Because all Japanese people learn Kunrei-shiki in school, when computers became a thing, Japanese people started using romaji a lot more than before, because that’s what they had to do to type on a computer. I don’t know about the past, but nowadays a computer will accept both Kunrei-shiki and Hepburn when typing, but remember that most Japanese people hardly know Hepburn.

So that’s the mess the romanization of Japanese has been in for many decades. It got even worse when more and more Japanese people realized that foreigners needed Hepburn, but most Japanese still had no idea because they had little to no contact with foreigners in their daily lives.

Nowadays you can find things romanized one way or the other, which causes a lot of confusion.

Well, as I’m typing these lines (in March 2025) it’s almost over. The Japanese authorities have finally understood (after 150 years or so) that the goal of romaji was to make Japanese readable for foreigners, so a system used only by Japanese people was pointless.

Better late than never.

 

If you liked what you read, why don’t you subscribe to the blog?