Lesson 2 · Day 2 notes
Completing the hiragana: the last twenty-one
Twenty-one characters finish the chart. Most of them are the regular pattern you already know, running on for another five rows — and then four of them refuse to behave, which is why the bottom of the chart looks broken.
1.What is left of the grid
Day 1 gave you the vowels and four consonants riding on them: k, s, t, n. Today adds the remaining four — h, m, y, r, w — and one lone character that belongs to no row at all. Twenty-five plus twenty-one is forty-six, and forty-six is the whole of hiragana.
| row | a | i | u | e | o |
|---|---|---|---|---|---|
| h | は | ひ | ふ | へ | ほ |
| m | ま | み | む | め | も |
| y | や | — | ゆ | — | よ |
| r | ら | り | る | れ | ろ |
| w | わ | — | — | — | を |
| n | ん — a row of one, and no vowel anywhere in it | ||||
Two of these rows are almost free. The m row and the r row are perfectly regular: five characters each, consonant plus vowel, nothing to memorise beyond the shapes. The h row is regular in four places out of five. Compared with day 1 — where three characters out of twenty-five broke the pattern — the arithmetic today is friendlier than it looks.
The r is not the r you are expecting
Japanese r is a single tap of the tongue against the ridge behind the teeth — somewhere between an English r, an l, and the d in "ladder". It is much closer to the Arabic ر in رَجُلٌ than to anything in English, except that it is tapped once and never rolled. If you already say ر cleanly, you are most of the way there.
2.Write them, hear them, use them
Same chart as yesterday, showing today's twenty-one. Tap any character: it draws itself stroke by stroke in writing order, says itself aloud, and gives you three real words — each one in Japanese, English and Arabic, all three with a play button.
は ひ ふ へ ほ ま み む め も や ゆ よ ら り る れ ろ わ を ん
The interactive chart needs JavaScript. The 21 characters are listed above in row order.
Three shapes are worth extra minutes
ふ is four separate marks, not one flowing squiggle, and it is the character beginners most often draw as an unrecognisable blob. を and ん are rare enough on the page that they get the least practice and stay the shakiest longest. Give those three twice the repetitions of anything else in the chart.
3.Four things that don't behave
Everything else in today's chart is consonant plus vowel, exactly as advertised. These four are the exceptions, and each one is a different kind of exception.
ふ is neither hu nor fu. The sound is made with both lips brought close together and air pushed between them — as if blowing out a candle. English f and Arabic ف both put the top teeth on the bottom lip; Japanese never does. Romaji writes it fu because that is the nearer of two wrong answers. Take the teeth out of your ف and you have it.
を is a character with exactly one job. It is pronounced o — identical to お, no difference at all. It never appears inside a word. Its only use is to sit immediately after a noun and mark that noun as the thing the verb is being done to. And this is a place where the two languages are doing the same work with different tools:
| "read a book" | what marks the object | |
|---|---|---|
| Japanese | ほんをよむ | を — a separate character, parked after the noun |
| Arabic | يَقْرَأُ كِتَابًا | the accusative ‑an ending, written on the last letter of the noun itself |
Japanese hangs a whole character off the back of the word; Arabic changes the word's final vowel and leaves the letters alone. Same grammatical idea — mark the object — solved from opposite directions.
ん is the only kana that is not a syllable. Every other character in the chart is a consonant and a vowel together. ん is a consonant on its own. But — and this is the part worth slowing down for — it still takes a full beat. Japanese counts in mora, and ん is one mora, the same length as か or ま.
| word | reads | beats | |
|---|---|---|---|
| Japanese | にほん | ni · ho · n | three |
| Arabic | مِنْ | min | one |
Sukūn is not ん
A consonant carrying sukūn in Arabic — the نْ in مِنْ — closes the syllable it is attached to and takes no time of its own. ん looks like the same thing on paper and is not: it is its own beat. にほん is ni-ho-n, three even beats, not ni-hon in two. Arabic reading habits will make you rush it, and rushing it is the single most audible beginner accent in Japanese.
Two rows have real holes in them. The y row is missing yi and ye; the w row is missing three of its five. These are not omissions in the chart. The syllables yi and ye never existed in Japanese, and the characters ゐ (wi) and ゑ (we) were abolished in the 1946 spelling reform — you will meet them only in pre-war text. There are forty-six characters rather than fifty because the language genuinely has forty-six sounds here, not because five were left out.
4.Confusion pairs, now that both halves are taught
Yesterday's reference sheet listed some of these with a note that the partner had not been taught yet. Both halves are now yours, so the comparisons are worth making properly. Every one of these is a pair that gets mixed up in the first month, and none of them is fixed by staring — write both, side by side, and the difference lands.
| This | vs | Tell them apart by |
|---|---|---|
| ね ne | れ re / わ wa | all three share an upright: ね closes into a full loop, れ ends in a hook, わ in an open curve |
| は ha | ほ ho / ま ma | ほ is は with a second crossbar; ま has no upright on the left at all |
| る ru | ろ ro | る closes its loop, ろ leaves it open — that is the only difference |
| さ sa | ち chi | mirror images: さ opens to the left, ち opens to the right |
| ぬ nu | め me | ぬ finishes with a loop, め stops without one |
| り ri | い i | り leans in and the right stroke runs long; い stands straight and even |
| を o | お o | same sound, different job — を is only ever a particle, お is a letter in words |
5.What you can read now
Yesterday's classroom phrases had gaps in them — は, ま, り, わ and ん were not yet yours, so several sentences were half-lit. They are yours now. Everything highlighted below is a character you can read.
| Japanese | reads | English |
|---|---|---|
| はい、わかりました | hai, wakarimashita | Yes, I understood. |
| こんにちは | konnichiwa | Hello. |
| ほんをよみます | hon o yomimasu | I read a book. |
| やまとうみ | yama to umi | mountain and sea |
| わたしのなまえ | watashi no namae | my name |
| ひらがな | hiragana | hiragana |
That is most of the hiragana in the table above, which is as complete as it gets from this chart alone. The one character left dark is が — か wearing two small marks in the top right. Those marks are dakuten, and they turn a whole row of sounds voiced: k becomes g, s becomes z, t becomes d, h becomes b. No new shapes to learn — the same forty-six characters with a diacritic, exactly as ب and ت and ث are one shape with different dots. That is a later day's work, and it is a short one.
6.Sources
- Teacher's lecture notes, Day 2 — the second half of the chart. Authoritative for what is in scope.
- KanjiVG — stroke order and direction for all 21 characters. © Ulrich Apel, licensed CC BY-SA 3.0; the stroke data reused here stays under that licence.
- Tofugu, "Learn Hiragana" — mnemonics and the standard confusion pairs.
- Jisho.org — checked every example word for readings and for whether it is actually common.
- NHK, "Easy Japanese" — recorded speech for the mora timing of ん, which synthesised audio renders only approximately.
- Madinah Arabic and Desert Sky Arabic grammar — the MSA glosses, the accusative ending, and sukūn.