The Japanese Writing System: Three Scripts

Japanese is written with three scripts at once, woven into every sentence with no spaces between words. This is the single most disorienting fact for a newcomer coming from English, where one alphabet does all the work. Once you see how the three scripts divide the labor, though, written Japanese stops looking like chaos and starts looking like a remarkably well-organized machine — you can often tell a word's job in a sentence just from which script it is written in.

The three scripts at a glance

ScriptJapanese nameCharactersWhat it writes
Hiraganaひらがな46 basic, rounded and cursiveNative grammatical words: particles, verb and adjective endings, and native words with no common kanji
Katakanaカタカナ46 basic, angular and sharpLoanwords, foreign names, onomatopoeia, and emphasis
Kanji漢字Thousands, borrowed from ChineseThe meaning-carrying roots of content words: nouns, and the stems of verbs and adjectives

Hiragana and katakana together are called the kana — two phonetic syllabaries that represent exactly the same set of sounds, one used for native material and one for foreign. Kanji, by contrast, are logographs: each carries a meaning, not just a sound.

One sentence, three scripts

Here is an utterly ordinary sentence. Watch how all three scripts appear inside it.

私はコーヒーを飲みます。

watashi wa kōhī o nomimasu

I drink coffee.

Now let's take it apart character by character:

PieceScriptReadingJob in the sentence
kanjiwatashithe noun "I" — meaning
hiraganawatopic particle — grammar
コーヒーkatakanakōhī"coffee," a loanword — foreignness
hiraganaoobject particle — grammar
kanjinothe root of the verb 飲む "to drink" — meaning
みますhiragana-mimasuthe polite verb ending — grammar

Notice the pattern: the kanji carry the meaning (私 = I, 飲 = drink), the hiragana carry the grammar (the particles は and を, the inflection みます), and the katakana flags the one word that came from a foreign language (コーヒー). This is not an accident of this particular sentence — it is the governing logic of the entire writing system.

The key insight: layers, not alternatives

Almost every beginner asks the wrong question first: "When do I choose to write in hiragana instead of kanji?" The word choose is the trap. In a normal sentence you do not choose — the three scripts are functional layers used simultaneously, and each word lands in the script its job demands.

  • Kanji is the meaning layer. A verb's or adjective's root sits in kanji so the eye can grab the concept instantly.
  • Hiragana is the grammar layer. The particles that mark who-does-what, and the endings that tense and negate a verb, are almost always hiragana. When you see hiragana wrapped around a kanji, you are looking at inflection.
  • Katakana is the foreignness layer. It says "this word is not originally Japanese" — a loanword, a foreign name, a brand — or occasionally "read this word with special emphasis," a bit like italics.
💡
A quick reading shortcut: the trailing hiragana on a word (called okurigana) almost always tells you its grammar. 飲みます, 飲まない, 飲んだ — the kanji 飲 stays put and carries "drink," while the changing hiragana tail carries "polite," "negative," "past." Learn to read the tail and you can parse a verb you have never seen.

Here are three more everyday sentences. In each, see if you can spot the meaning layer (kanji), the grammar layer (hiragana), and the foreignness layer (katakana) before you read the breakdown.

私はテレビでニュースを見ます。

watashi wa terebi de nyūsu o mimasu

I watch the news on TV.

Here テレビ (television) and ニュース (news) are katakana loanwords; 見 is the kanji root of 見る "to see"; and は・で・を plus the ending ます are the hiragana grammar.

明日、友達とレストランに行きます。

ashita, tomodachi to resutoran ni ikimasu

Tomorrow I'm going to a restaurant with a friend.

明日 (tomorrow), 友達 (friend), and 行 (the root of 行く "to go") are kanji; レストラン (restaurant) is a katakana loanword; と・に plus きます are hiragana.

外は雨がザーザー降っている。

soto wa ame ga zāzā futte iru

Outside, the rain is pouring down.

This one shows katakana's other main job. ザーザー is onomatopoeia — the sound of heavy rain — and Japanese frequently spells its rich stock of sound-and-mood words in katakana for visual punch.

No spaces — so how do you know where words end?

English uses spaces to separate words. Japanese uses none. At first this seems impossible to read, but the script changes themselves do much of the work spaces do in English. The eye learns to treat the switch from kanji to hiragana, or from hiragana to katakana, as a soft boundary.

In 私はコーヒーを飲みます, the jump from 私 (kanji) to は (hiragana) tells a Japanese reader that a new grammatical unit has begun, and the jump from を (hiragana) to 飲 (kanji) signals the start of the verb. The scripts are not just labels of meaning — they are also built-in punctuation. This is one more reason the three scripts are a system, not a nuisance.

Compared to English

For an English speaker, three things are genuinely new here:

  1. One alphabet becomes three scripts. English writes every word — native (house), borrowed (rendezvous), and imitative (bang) — in the same 26 letters. Japanese sorts these into three visually distinct scripts, so a word's origin and role are visible at a glance.
  2. Letters become syllables. English letters map (roughly) to individual sounds; you assemble c-a-t. Each kana instead stands for a whole mora — a consonant-plus-vowel beat like か (ka) or ね (ne). You will meet the mora in depth on the hiragana overview page.
  3. Spaces disappear. Word boundaries are signaled by script changes and grammar, not by whitespace.

Romaji is a crutch, not "real" Japanese

You will often see Japanese written in the Latin alphabet — watashi wa kōhī o nomimasu. This is romaji (ローマ字), and it is the single biggest source of a misconception that quietly sabotages English speakers: the belief that romaji is a legitimate way to read Japanese.

It is not. Romaji exists to help absolute beginners and to type on a keyboard; no adult Japanese text is written in it. Menus, novels, street signs, text messages, subtitles — all of it is kana and kanji. If you lean on romaji past the first week or two, you build fluency in a script that Japanese people never actually read, and you stall permanently at the door of the real language.

That is why this guide does not romanize its running text. Every example sentence is written in real kana and kanji, exactly as a native would write it. We provide a romanization line beneath each example purely as a temporary reading aid — say it as a helping hand you are meant to outgrow, not as the sentence itself. Your job over the coming pages is to make that helping hand unnecessary.

💡
Treat the romanization under each example the way a pianist treats the note names penciled above a staff: useful on day one, embarrassing by month three. Cover it with your thumb as soon as you can, and read the kana directly.

How this guide teaches the scripts

The path is deliberate and paved in order:

  • Hiragana first — the core syllabary that writes all the grammar. It is non-negotiable and comes before everything else.
  • Katakana next — the foreign-word syllabary, the same sounds in an angular skin.
  • Kanji, continuously — the meaning characters, learned steadily over years, always supported by the kana you already command.

Common mistakes

❌ Watashi wa gakusei desu.

Incorrect — this is romaji, not Japanese writing. No native text looks like this.

✅ 私は学生です。

watashi wa gakusei desu

I am a student. — real kana and kanji, the way it is actually written.

❌ 私 は 学生 です 。

Incorrect — Japanese does not put spaces between words.

✅ 私は学生です。

watashi wa gakusei desu

I am a student. — no spaces; script changes mark the boundaries.

❌ オハヨウございます。

Incorrect — a native greeting written in katakana. Katakana is for foreign words, not native ones.

✅ おはようございます。

ohayō gozaimasu

Good morning. — a native word, so it takes hiragana.

❌ こーひーを飲みます。

Incorrect — a loanword written in hiragana. Foreign words belong in katakana.

✅ コーヒーを飲みます。

kōhī o nomimasu

I drink coffee. — the loanword takes katakana.

Key takeaways

  • Written Japanese uses three scripts at once: kanji for meaning, hiragana for grammar, katakana for foreignness.
  • There are no spaces; script changes signal word boundaries.
  • The scripts are layers, not choices — a word lands in the script its job requires.
  • Romaji is a temporary crutch, never real Japanese text — this guide reads in kana and kanji from day one.

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.
  • Katakana: The Second SyllabaryN5Katakana is hiragana's phonetic twin — the same 46 sounds in angular form — used for loanwords, names, and onomatopoeia, and beginners meet it on day one, not 'later.'
  • Kanji: Meaning-Carrying CharactersN5What kanji are — characters borrowed from Chinese that carry meaning rather than sound — why each is a morpheme with several readings, and how beginners grow from a few dozen to literacy.