Why is it standard practice to type "ha" when you are translating は{wa} using a QWERTY keyboard? This has always confused me. これは... is pronounced "Kore wa" but when you type it out you type "Kore ha". Hiragana は{ha} when used as a particle is pronounced "wa" so why do we not type "wa"?

  • That is because on computers, we are using romaji-like key strokes to specify kana letters. The usual romaji rules are based more on pronunciation. I am not sure if this is really a question about Japanese or a question about computers. – Tsuyoshi Ito Sep 3 '12 at 17:43
  • I am not sure I am understanding your comment. This is not when using a Japanese keyboard and transcribing it into Romaji, this is using an actual "English" QWERTY keyboard and still using "ha". I have seen it is handwritten notes as well, so no, not a computer question. – BillyNair Sep 3 '12 at 17:50
  • Are you asking why can't you can't type "k-o-r-e-w-a" and have it come out as これは? – silvermaple Sep 3 '12 at 17:51
  • 1
    answer- just because it is standard practice. All my jp friends when they text in jp on an american phone, they write は as ha, because this is how one would type it in Japan, even though it is not the same sound. I type/ write it like that because i became used to it as well. – yadokari Sep 3 '12 at 18:34
  • 3
    You should "think" in Japanese. We write the particle as は. It is defiantly は, not わ at all. It is a kind of spelling rule. Japanese children lean that it should be は. If they write わ, it will be corrected. I would say that English spelling confuses us much more. – Gradius Sep 3 '12 at 21:12

Because computers cannot read your mind. When you type "wa", the computer cannot decipher whether you mean わ or は, so it was decided that わ would be the only way. You could argue that you could develop a system to perform 変換 based on context, but I would imagine that any attempt would fail. Also, if you don't like this system, you can always use かな入力 (mainly used by people who don't understand ローマ字), so when you hit the は key, you will always get what you want.

  • And even if such an system failed only 0.1% of the time, that's a 0.1% which is avoided with the current system. There has to be a way to type the hiragana "ha" directly anyway, so users might as well use that and skip the potentially flawed AI-guided conversion step. (Although, even right now, some processing still has to be done to figure out whether は is the topic particle or 刃, 葉, etc...) – Hyperworm Sep 4 '12 at 0:28
  • 1
    @Jesse Good, Seeing that you have the most votes shows that the way I asked the question made people this this was a computer question causing the question to be closed. I was asking why they use HA even when they are not using a computer that can do hiragana. It was absolutely a Japanese question and not at all a computer question. You answered the way most people read my question to mean, and I thank you for trying to help. If you read yadokari's comment on the OP you will see the answer that I was trying to get. – BillyNair Sep 5 '12 at 3:29
  • 2
    @BillyNair: After reading your comments, etc. I understand what you were trying to ask now. Putting QWERTY keyboard in the title of your question is quite misleading. However, the reason why they would use HA is because they would use whatever romanization method they learned in school (which is probably hepburn or kunrei-shiki). – Jesse Good Sep 5 '12 at 3:45
  • Thank you for this response that is what I am looking for, that makes sense, you get used to a WordProcessor and you will type that way even if it is a normal English only computer/keyboard. I tried re-reading my question strictly from the POV of someone who had no idea what I was talking about, and yes, it is confusing. If you can think of a better way to re-word it and get it re-opened, use that answer and I will chose it ;-) – BillyNair Sep 5 '12 at 3:53

Because of the way IMEs are designed. All Japanese IMEs based on romaji that I know of (roughly) divide the keyboard-to-script transformation into two conceptual parts:

  1. romaji-to-kana. This part is usually deterministic, i.e. no romaji combination gives more than one option of kana.
  2. kana-to-(kanji or kana or whatever). This part is not deterministic, and requires the user to choose between different renderings.

There is no reason why the two steps couldn't be combined into one non-deterministic step, i.e. directly from romaji-to-(kanji or kana or whatever), but given that Japanese already has many homophones, this would probably only add to the confusion when choosing between renderings.

Japanese doesn't have one single universal romanization scheme which everyone agrees upon. And even if they did, that scheme might not be the best to use for input. Some romanization schemes try to be (somewhat) true to pronunciation, some try to be true to phonemes, and some try to be true to (kana) orthography. I think it's fairly natural that the ones used for input are somewhat true to orthography. For example, there's no pronunciation which renders as "ッァ", but that doesn't mean that you cannot have reasons to type this.

As for why you saw someone write the particle は as "ha" in handwritten notes, my best guess would be that that person frequently uses IMEs with romanized input, and therefore that came more natural to them.


Every kana has a "set" romaji in the qwerty layout, even if it doesn't always make sense from an English perspective. For example づ is "d-u", even thou it's pronounced closer to "zu".

The instances where a kana is not pronounced how it looks (as far as I know) are limited to は when it acts as a particle. If you're just starting out, you might find this confusing, but it really become second nature to read or pronounce は properly. Compare that to English's spelling/pronunciation!

If one kana had multiple ways to type it, it would become complicated. For example, to say 3時半, and you could type, "s-a-n-j-i-w-a-n" you'd just be wrong and confused.

Also, I can't tell if this is causing you confusion, but Japanese people use qwerty keyboards just like in America. There is a layout where each key is a different kana, but I've never met anyone who uses it.

  • But 半 doesn't have the reading of "wan". If you read は it could be "ha" or "wa". When I was learning Japanese (back in 1987), the romaji used to teach us how to read used "wa", most books do, but actual Japanese people use "ha". I can read hiragana pretty quickly, especially は either being "ha" or "wa". When I was in Japan, I saw that they used "wa-puro" (if that is the proper spelling) and that is what I was trying to steer clear from. I was wondering when they are NOT using a Japanese specific computer/keyboard that they still use "ha". Your "du/zu" analogy is more in line with my question. – BillyNair Sep 3 '12 at 18:25
  • I'm sorry, it wasn't really clear where your confusion was coming from. I know 半 isn't read "wan", that's why I used it. How does the computer know if when you type "w-a" do you mean は or わ...and if you type "w-a-n" do you mean half (半) or bowl (椀). Yes you could use the drop down function, but that just complicates matters. は=h+a is simple. I do no know what "wa-puro" is. When I learned, "は" was also "wa", but I remember our teacher telling us not to get used to it that way...using romaji didn't last long in my class long enough for us to get used to it anyway... – silvermaple Sep 3 '12 at 18:40
  • "Wāpuro" - bit.ly/NbJcSp (Word Processor). Where you type in romaji then hit a letter and it changes it to kanji or whatever kana you need. Thanks for trying to help, but obviously I worded the question weird because everyone was thinking I was asking why the computer changes the letters and not why Japanese people will type HA even when they are not using a computer than will change it. It was a Language question not a computer question, I am still not sure how it could be asked differently to emphasize that, but thanks again. – BillyNair Sep 5 '12 at 3:41
  • "へ" when used as a particle is pronounced 'e' instead of 'he'. – jastako Feb 24 at 11:55

Or, to put Jesse Good's answer another way: When the romaji input was fixed it was probably decided it was more efficient to make human's input h-a rather than further complicate the programming and avoid additional time to perfom 変換.

Even if this could be done with modern programming I don't expect it is more likely to change than the QWERTY keyboard layout, which was designed to slowdown the input to a mechanical type-writer and reduce problem of the letter-hammers jamming.

(A better question would be why is the "wa-particle" the hiragana は and not わ?)

  • 1
    The answer to your question is sound change. See here or here. – Zhen Lin Sep 4 '12 at 0:30

Not the answer you're looking for? Browse other questions tagged or ask your own question.