The Gojūon: Reading the Hiragana Grid

The gojūon (五十音, "fifty sounds") is the master grid that organizes every kana. It is the Japanese equivalent of the alphabet's A-B-C ordering: dictionaries sort by it, charts are laid out by it, and — as you will see — verb conjugation is secretly built on it. Learn to read the grid fluently and you can navigate any kana chart, look up any word, and get a head start on grammar you have not even met yet.

How the grid is read

The grid has five vowel columns — あ い う え お, the a-i-u-e-o order that runs through all of Japanese — and a stack of consonant rows. To read any cell, take the row's consonant and add the column's vowel. The k-row plus the u-column gives く (ku); the n-row plus the e-column gives ね (ne). Reading order is always row first (the consonant family), then column (the vowel).

aiueo
(vowels)あ aい iう uえ eお o
kか kaき kiく kuけ keこ ko
sさ sashiす suせ seそ so
tた tachitsuて teと to
nな naに niぬ nuね neの no
hは haひ hifuへ heほ ho
mま maみ miむ muめ meも mo
yや yaゆ yuyo
rら raり riる ruれ reろ ro
wwaを o
nん n (stands alone, outside the grid)

That is the entire basic syllabary — 46 characters. Everything else in kana is built on top of this grid with small marks (covered on the dakuten and handakuten page) or small companion kana (the yōon page).

Start with the five vowels

The vowels あ い う え お are the backbone; every other kana ends in one of these sounds. Nail them first, because a mispronounced vowel colors every mora built on it.

KanaReadingRoughly like
athe a in "father" (short)
ithe i in "machine" (short)
uthe u in "flute," but with lips relaxed, not rounded
ethe e in "bed"
othe o in "or" (short)

Japanese vowels are pure and short — they do not glide the way English long vowels do (English "oh" drifts toward a "w" at the end; Japanese お does not). Keep them clipped and steady.

いえ

i e

house — two clean vowels, two beats.

あお

a o

blue — again, two pure vowels held separately.

The consonant rows, with the readings that trip everyone up

Most of the grid is beautifully regular: か ki-ku-ke-ko, ま mi-mu-me-mo. But four cells break the pattern, and they are the exact four that English speakers get wrong. Do not read across the grid on autopilot — these are not "si," "ti," "tu," "hu."

KanaRegular pattern would predictActual reading
sishi (as in "she")
tichi (as in "cheese")
tutsu (the "ts" of "cats" + a relaxed u)
hufu (a soft f made with both lips, not teeth)

Here are those four rows in real words so the sounds stick:

しま

shi ma

island — し is 'shi,' never 'si.'

ちから

chi ka ra

strength — ち is 'chi,' never 'ti.'

つくえ

tsu ku e

desk — つ opens with a real 'ts' cluster.

ふゆ

fu yu

winter — ふ is a gentle bilabial 'fu.'

And a few from the tidy, regular rows, so you can feel how predictable most of the grid is:

かさ

ka sa

umbrella.

たけ

ta ke

bamboo.

はな

ha na

flower (or nose).

やま

ya ma

mountain.

The gaps: missing cells in the y- and w-rows

The grid is not a complete rectangle. Some cells are simply empty in modern Japanese, and knowing which ones saves you from hunting for kana that do not exist.

  • The y-row has only three kana: や (ya), ゆ (yu), よ (yo). There is no "yi" and no "ye" — those slots are blank. (They existed in ancient Japanese and then merged away.)
  • The w-row has collapsed to two: わ (wa) and を. The characters for "wi" (ゐ) and "we" (ゑ) are (archaic) — you will only meet them in prewar texts, some proper names, and stylized signage; they were officially retired in 1946. There is no "wu" at all.
  • を is a special case. It is historically "wo," but in modern speech it is pronounced simply "o." Its one and only job today is as the object particle, and it is written を rather than お purely by convention. You will meet it constantly; see the particles は, へ, を.
💡
The most useful mnemonic for を: it is "the o you never write anywhere else." If an "o" sound is marking the direct object of a verb, it is を. If an "o" sound is inside a normal word (お茶, おと), it is お. There is no overlap, so the rule is airtight.

ん — the one kana outside the grid

ん (n) sits apart from the grid because it is unique: it is the only kana that is a single consonant with no vowel, and it can never begin a word. It attaches to the mora before it and counts as a full mora of its own — にほん is three beats, ni-ho-n, not two. Its exact sound shifts with what follows (it can lean toward "m" or "ng"), but you will always write it ん.

Why the grid pays off twice: it is the engine of conjugation

Here is the insight that competing kana charts never tell you, and the reason to truly own the あ/い/う/え/お order now rather than just memorizing shapes. The five vowel columns are the skeleton of Japanese verb conjugation.

Take the verb 書く (かく, kaku, "to write"). It is a k-row verb, and it shifts across the k-row to build its forms:

ColumnStemBuilds
あ (a)書か kaka-the negative: 書かない (won't write)
い (i)書き kaki-the polite stem: 書きます (write, polite)
う (u)書く kakuthe dictionary form (write)
え (e)書け kake-the command / potential: 書け, 書ける
お (o)書こ kako-the volitional: 書こう (let's write)

This is exactly why the verb class you will meet later is called godan (五段, "five rows"): the stem literally steps through the five vowel rows of the grid. The order you are drilling right now — a, i, u, e, o — is the same order that will generate every form of thousands of verbs. Learn the grid well now and you learn half of the verb system for free. The full treatment is on Godan across the あ/い/う/え/お rows.

Common mistakes

❌ し read as 'si'

Incorrect — し is 'shi.' There is no 'si' sound in this row.

✅ しお

shi o

salt — the し is 'shi.'

❌ つ read as English 'too'

Incorrect — つ is 'tsu,' an initial ts-cluster, not a plain 't'.

✅ つき

tsu ki

moon — feel the 'ts' at the very start.

❌ ち read as 'tee/ti'

Incorrect — ち is 'chi,' as in 'cheese'.

✅ いち

i chi

one — the ち is 'chi.'

❌ hunting for a 'yi' or 'wu' kana

Incorrect — those cells are empty; the y-row has only や・ゆ・よ and the w-row only わ・を.

✅ ゆき

yu ki

snow — the y-row gives you や, ゆ, よ and nothing between them.

❌ ふ read as a hard English 'foo'

Incorrect — ふ is a soft f formed with both lips, close to a blown-out 'hu'.

✅ ふね

fu ne

boat — keep the ふ light and breathy.

Key takeaways

  • The gojūon is five vowel columns × consonant rows; read row (consonant) then column (vowel).
  • Four cells are irregular: し shi, ち chi, つ tsu, ふ fu — never si/ti/tu/hu.
  • The grid has gaps: no yi/ye, only わ and を in the w-row (ゐ/ゑ are archaic), and no wu; ん stands alone.
  • The あ/い/う/え/お order is the engine of godan verb conjugation — learning it now pays off twice.

Now practice Japanese

Reading grammar gets you part of the way. The exercises are where it sticks — free, no signup needed.

Start learning Japanese

Related Topics

  • Hiragana: The Core SyllabaryN5Why hiragana is the non-negotiable first script — the phonetic syllabary that writes all of Japanese grammar — plus the mora, the gojūon ordering, and the look-alike kana to watch.
  • Dakuten and Handakuten: Voicing MarksN5The two small diacritics that expand hiragana — the dakuten (゛) that voices k→g, s→z, t→d, h→b, and the handakuten (゜) that turns the h-row into p — plus the じ/ぢ, ず/づ tangle.
  • Yōon: Combined Kana (きゃ, しゅ, ちょ)N5How an i-column kana plus a small ゃ, ゅ, or ょ fuses into a single palatalized mora — きゃ kya, しゅ shu, ちょ cho — and why the whole combination counts as one beat, not two.