Why is "konnichi wa" in hiragana with は and not わ？
There are some lingering question marks in the initial post and in the other answers regarding why は is read as
/wa/ in こんにちは.
Other posters have already noted that this は is the topic particle -- which is always read as
The remaining question is, why is this particle は read as
The short(ish) answer
The answer lies in history. According to historical linguists, all of the modern "H" kana (and "F" kana), はひふへほ, started out as "P" kana, read as
/pa pi pu pe po/. Over time, this initial
/p/ sound lenited (i.e. softened) to become more of an
/f/ sound in most places, and a
/w/ sound in some other places. Still more recently, the
/f/ sound lenited further to
/h/ at the beginning of a word, and in the middle of a word, it either became a
/w/ (for は) or vanished to leave just the vowel (for ひふへほ).
/p/ sounds may be due to gemination (when a consonant sound is doubled), or to a preceding ん. Word-initial
/p/ sounds contrastive with
/f/ sounds appear to be a development in perhaps the 1300s-1400s, used for imitative words and borrowings.)
The longer story
For the details, let's look backwards. We'll start with the Japanese of the early Edo period, when the language was in its Early Modern stage. There was a very important Japanese-Portuguese dictionary compiled in 1603, the Vocabulario da Lingoa de Iapam, a.k.a. 日葡辞書 (Nippo Jisho), which gives us a window into the Japanese phonology (sounds) of the time. We find things like 日本 with a reading of Nifon, or 春 with a reading of faru, showing that many of the はひふへほ sounds of today were read instead with an
/f/ sound. But as a particle, は was already
Further back, even earlier during the stage called Late Middle Japanese (roughly the 1100s through the 1500s), linguistic research suggests that many of the "H" kana within a word (not at the beginning) underwent a shift, probably going from the
/f/ we see at the beginnings of words in the early 1600s to an even-softer
/w/ sound. From there, many of these
/w/ sounds just disappeared -- becoming just the vowel sounds -- except for the は, which became
/wa/. This is possibly because of the biomechanics of Japanese pronunciation (the shapes the mouth makes), perhaps making sound combinations like
/wo/ difficult, combined with a lack of any strong need to distinguish these
/w/ sounds from the plain vowels.
/wu/ doesn't seem to have ever existed as a meaningful sound contrast to just plain
/wo/ is only very rarely encountered for を when people are deliberately overpronouncing. Meanwhile,
/we/ apparently became just
/i/ some time in the 1200s.)
We see remnants of this sound shift in modern speech. For instance, the everyday verb 言う【いう】 "to say" has that
/w/ sound that seems to appear out of nowhere in the negative conjugation iwanai. Historically, いう was formerly spelled in kana as いふ, which you can still see in many Japanese-Japanese dictionaries. This is the old kana spelling from before the spelling reforms of the early and mid-1900s, more formally known as historical kana orthography (spelling), or 歴史的【れきしてき】仮名【かな】遣【づか】い in Japanese. Starting from いふ, we find that the conjugated stems would be:
- いふ - plain form
- いは - incomplete stem, most often seen in the negative
- いひ - continuative stem, often seen followed by ます
- いへ - conditional and imperative
But in modern Japanese, the conjugated stems are:
- いう - plain form - the "F" disappeared
- いわ - incomplete stem - the は became わ
- いい - continuative stem - the "H" disappeared
- いえ - conditional and imperative - the "H" disappeared
- いおう - volitional
(Side note: the modern volitional ending
/oː/ (long ō) came from the incomplete stem plus う, and the
/au/ sound (like English "ow!") evolved into
/ɔː/ (like English "awwww") and then to
/oː/. For volitionals where the ending
/oː/ would start with a
/w/, like for いふ, the
/w/ disappeared just as it mostly has for modern を.)
The result: our modern Japanese particles with funny readings
Particles are treated phonologically as suffixes to some extent, so this pattern of historical mid-word "H" kana being read with a
/w/ sound for は also effectively applies to the particle は. Similarly, the "H" kana read as just the vowel sound also applies to the particle へ, and that same vanishing-"W" process that results in the
/e/ reading for へ also applies to the particle を. The spelling reforms of the 1900s adjusted official kana spellings to match the expected pronunciations in almost all cases -- but the particles were apparently deemed special, and the older spellings were maintained, even though they don't quite match the expected pronunciations: leaving us with modern は (
/wa/), へ (
/e/), and を (
This is a matter of grammar and orthography. は is the transcription for the sound "wa" when used as a grammatical particle (namely, the topic marker). わ is used in virtually all other cases (I say virtually because older orthographic conventions don't as neatly follow this rule).
In saying, こんにちは, you have two parts こんにち which means "this day" and は which marks こんにち as the topic.
There's another particle which is pronounced as "wa". It's the sentence ending particle わ, but this is transcribed as expected.
Incidentally, I'd saying writing the sound "wa" as は is not really a matter of translation but transcription or transliteration (though transliteration sounds a bit odd for referring to kana).
Incidentally, you could write こんにちは using katakana. In that case, it would be written as コンニチハ.