The Japanese Alphabet: What It Actually Is

Three scripts, and none of them is an alphabet

An alphabet breaks a word into separate consonant and vowel letters, so cat needs three signs for three sounds. Japanese does not work that way. Its basic unit is the mora, a short beat that is usually a consonant plus a vowel, and one kana character stands for one of those beats. か is not k plus a, it is ka, one indivisible sign for one beat. That is why the correct word for hiragana and katakana is syllabary rather than alphabet, and why counting letters gives you a strange answer.

Written Japanese then runs three systems at once. Hiragana handles grammar, particles, and verb endings. Katakana handles imported words, foreign names, and anything the writer wants to set apart. Kanji carry the meaning of nouns and the stems of verbs and adjectives. A separate guide on this blog covers which script a given word takes; here the job is simply to show you the two kana grids in full and explain how they are organised.

How the gojūon grid is built

The kana chart is called the gojūon, the fifty sounds. It is a table with five vowel columns in the fixed order a, i, u, e, o, and consonant rows in the fixed order k, s, t, n, h, m, y, r, w, with the single character ん sitting outside the grid at the end. Every position combines its row consonant with its column vowel, so the t row reads ta, chi, tsu, te, to. Three of those cells are pronounced irregularly, which is worth marking now.

The irregular cells are the ones learners misread for months: し is shi rather than si, ち is chi rather than ti, つ is tsu rather than tu, and ふ is closer to fu than to hu, made with both lips rather than the front teeth. Everything else in the grid is exactly what the row and column predict. Learn the five vowels first, then take the rows in the order above, because that order is also the order Japanese uses for filing anything alphabetically.

The full hiragana chart

Read the chart down each column as well as across each row. Reading down forces you to hold one vowel while the consonant changes, which is how the grid becomes navigable rather than memorised as a list. を is written in the w row for historical reasons but is pronounced o and appears in modern Japanese only as the object particle.

The full katakana chart

The sounds and the order are identical to hiragana, so you are relearning shapes rather than a system. Four katakana pairs cause almost all early errors: シ shi against ツ tsu, and ソ so against ン n. The clue is stroke direction. シ and ン are written with strokes coming in from the left and finishing upward, while ツ and ソ are written from the top downward.

Voiced and semi-voiced marks

Two small marks extend the base grid without adding new shapes. The dakuten, a pair of short strokes at the top right, voices a consonant: k becomes g, s becomes z, t becomes d, and h becomes b. The handakuten, a small circle, applies only to the h row and turns it into p. This is why the chart does not need separate characters for ga or pa, and why a learner who knows the base forty-six can already read several hundred more syllables.

Two cells in the z and d rows overlap in sound. じ and ぢ are both read ji, and ず and づ are both read zu. Modern spelling uses じ and ず almost everywhere; ぢ and づ survive mainly inside compound words where the original character is still visible, as in 続く tsuzuku. Do not treat them as free alternatives when you write.

Small ya, yu, and yo make one beat, not two

Any kana in the i column can be followed by a small ゃ, ゅ, or ょ to make a single glided beat. きゃ is one mora, kya, not ki plus ya. This matters for rhythm as much as for spelling, because reading きょう as ki-yo-u gives you three beats where Japanese has two. The small kana are written at roughly half height, and in vertical text they sit to the right of the line rather than centred.

The small tsu and the moraic n

Two characters do not carry a syllable of their own. The small っ or ッ marks a doubled consonant: you close your mouth into the following sound and hold one silent beat before releasing it. It is never pronounced tsu. The character ん is the other one, a nasal sound that occupies a full beat by itself and changes colour depending on what follows, sounding closer to m before p and b, and closer to ng before k and g. Both are timing, and both are where beginners lose the rhythm of a word.

Say kitte with a hand tap on each beat: ki, silent tap, te. If the word has the same number of taps as the kana, your timing is right. The apostrophe in Kin'yōbi exists only in romaji, to show that the ん is a beat on its own and the word is not read kinyō with a glide.

Long vowels are two beats, not a stress

A long vowel in Japanese is the same vowel held for two beats. It is not louder and it does not change quality, and getting it wrong changes words. Hiragana writes the length by adding a vowel kana: おかあさん adds あ, おにいさん adds い, くうき adds う, and the o and e long vowels usually add う and い, which is why とうきょう is read Tōkyō and せんせい is read sensei with a long e. Katakana takes a simpler route and writes almost every long vowel with the bar ー.

Why the chart order runs your dictionary

The gojūon order is not just a teaching layout, it is the sorting order of the language. A dictionary, a class register, a list of stations on a signboard, an index at the back of a book, and the tabs on a form all run a, i, u, e, o and then k, s, t, n, h, m, y, r, w, n. Once the grid is automatic you can find a word in a paper dictionary and read a list without hunting, which is a practical reason to learn the order rather than only the shapes.

A few filing conventions come with it. Marked kana are filed with their base kana, so が sorts beside か rather than in a separate g section. Small kana sort with their full-size equivalents. Long vowels written with the katakana bar are usually filed as the vowel they extend. And the ordering works on readings, not on kanji, which is why dictionaries of kanji words are indexed by their kana pronunciation.

• Say the row heads in order every morning: a, ka, sa, ta, na, ha, ma, ya, ra, wa, n.
• Sort a short list of five Japanese words yourself before checking against a dictionary.
• When you meet a new word, note where it would fall in the grid, not just what it means.
• Practise finding a station name on a route map by its first kana rather than by shape.

Read these kana sentences aloud

These lines are written in kana only, the way material for beginners and young children is written. Read each one slowly without looking at the romaji, then check yourself. Slow and correct beats fast and guessed, because guessing trains you to recognise word shapes rather than characters.

Notice three spelling habits in those lines. は is read wa when it is the topic particle, へ is read e when it marks direction, and を is read o as the object particle. They are the only common places where a kana is read differently from its chart value, and they are frequent enough that you will absorb them within a week.

A first week with the chart

Learn the sets in the order the grid gives you, one row at a time, and always in both directions: shape to sound and sound to shape. A separate guide on this blog lays out a fourteen-day method with memory techniques and handwriting drills; what follows is the shorter version for someone who only wants to be able to read the chart and use it.

• Day 1: the vowels and the k row in hiragana, then read every two-kana combination you can make from them.
• Day 2: the s, t, and n rows, saying shi, chi, and tsu aloud each time so the irregulars never settle wrong.
• Day 3: the h, m, y, r, and w rows plus ん, and read the ten kana sentences above.
• Day 4: dakuten and handakuten, then reread the same sentences hunting for marked kana.
• Day 5: small ya, yu, yo, the small tsu, and long vowels, tapping the beats with your hand.
• Day 6: start katakana at the vowels and work down, testing the four look-alikes every session.
• Day 7: read menus, product labels, and station names, and write your own name in katakana.