Chapter 2 — Hiragana

This chapter presents the complete hiragana writing system for recognition. By the end, you should be able to look at any string of hiragana and read it aloud — slowly at first, then with increasing fluency as you practice.

This is the last chapter in the book that includes romaji. From Chapter 3 onward, all Japanese text appears in kana and kanji only. If you find yourself relying on the romaji in this chapter rather than reading the hiragana directly, stop and force your eyes to the kana. The romaji is here as a reference, not as a primary reading path.

2.1 The Basic 46 Characters

Hiragana is organized in a grid called the 五十音図 (gojuuon-zu, "fifty-sound chart"), though the actual number of distinct characters is 46. The chart is arranged by consonant row (行, gyou) and vowel column (段, dan). Each character represents exactly one mora.

* を is used only as the object particle in modern Japanese. Its pronunciation is [o], identical to お. The historical reading "wo" survives in the romanization but not in standard speech.

The three blank positions in the や行 (yi, ye) and the three in the わ行 (wi, wu, we) represent sounds that existed in older Japanese but have merged with other vowels. They are no longer part of the writing system.

Irregular romanizations: し, ち, つ, ふ

Four characters have romanizations that break the consonant-plus-vowel pattern of their rows:

• し is romanized "shi," not "si" — the さ行 consonant before い became a palatalized [ɕ] sound (similar to English "sh"), not a plain [s].
• ち is romanized "chi," not "ti" — the た行 consonant before い became an affricate [tɕ] (similar to English "ch").
• つ is romanized "tsu," not "tu" — the た行 consonant before う became an affricate [ts].
• ふ is romanized "fu," not "hu" — the は行 consonant before う became a bilabial fricative [ɸ] rather than a glottal [h].

These are not exceptions to be memorized as oddities. They are the natural result of historical sound changes: certain consonants shifted their pronunciation before specific vowels. The irregularity is in the romanization system, not in the Japanese — these are simply the sounds the language uses.

2.2 Dakuten and Handakuten

The 46 base characters are expanded by two diacritical marks that modify the consonant.

Dakuten (゛) — voicing mark

The dakuten (two short strokes at the upper right of a character) voices the consonant. "Voicing" means the vocal cords vibrate during the consonant — compare the English pairs k/g, s/z, t/d, where the second sound in each pair is the voiced version.

か行 → が行 (k → g)

さ行 → ざ行 (s → z)

た行 → だ行 (t → d)

は行 → ば行 (h → b)

Handakuten (゜) — semi-voicing mark

The handakuten (a small circle at the upper right) applies only to the は行 and changes the consonant to [p].

は行 → ぱ行 (h → p)

Note on じ/ぢ and ず/づ

In modern standard Japanese, じ and ぢ are pronounced identically — both are [dʑi] (romanized "ji"). Similarly, ず and づ are both [dzɯ] (romanized "zu"). The distinction is purely orthographic: which character you use depends on the word's etymology and established spelling conventions, not on pronunciation.

In practice, じ and ず are far more common. You will encounter ぢ and づ mainly in two situations: compound words where voicing occurs at a morpheme boundary (e.g., はなぢ hanaji, "nosebleed," from はな + ち), and a handful of individual words where づ is the standard spelling (e.g., つづく tsuzuku, "to continue"). For recognition purposes, simply know that じ = ぢ in sound and ず = づ in sound.

2.3 Combination Characters (拗音)

Japanese has a set of combination sounds called 拗音 (youon), formed by pairing a consonant-い kana with a small ゃ, ゅ, or ょ. The small size of the second character is meaningful — it signals that the two kana together represent a single mora, not two.

Compare:

• きや (ki-ya) = two morae
• きゃ (kya) = one mora

The difference matters. きゃく (kyaku, "guest") is two morae: きゃ・く. If you pronounce it as three morae (ki-ya-ku), you will be misunderstood.

Full combination table

Voiced combinations (with dakuten):

Semi-voiced combinations (with handakuten):

Note that し, ち, and じ produce irregular romanizations in their combinations: しゃ is "sha" (not "shya"), ちゃ is "cha" (not "chya"), and じゃ is "ja" (not "jya"). This follows the same pattern as the base characters — the romanization reflects the actual pronunciation rather than a mechanical consonant+y+vowel formula.

Example words with 拗音:

2.4 Small っ (促音)

The small っ (called 促音, sokuon) represents a geminate consonant — a doubled consonant that occupies one mora of time. During that mora, your mouth moves into position for the following consonant and holds. No sound comes out. Then the consonant is released.

In romaji, the geminate is written by doubling the consonant that follows:

The small っ is visually smaller than the regular つ. This size difference is the only way to distinguish them in hiragana text — there is no other marker. When reading, pay attention to character size: つ is a full-sized kana representing the mora "tsu," while っ is a reduced-size kana representing a beat of silence before a doubled consonant.

As discussed in Chapter 1, the っ is not optional. Dropping it changes the word:

• きて (kite, "come") vs. きって (kitte, "stamp" or "cut")
• かた (kata, "shoulder") vs. かった (katta, "won / bought")
• もと (moto, "origin") vs. もっと (motto, "more")

2.5 Long Vowels in Hiragana

Japanese distinguishes short and long vowels. A long vowel is held for two morae instead of one. In hiragana, long vowels are written by adding a second vowel kana after the first. The rules are mostly straightforward, with two important exceptions.

General rule: add the same vowel

For the あ, い, and う columns, a long vowel is written by repeating the vowel kana:

え-column exception: usually えい, not ええ

A long え sound is most commonly written えい, not ええ:

True ええ spellings exist but are uncommon. The most frequent is the informal affirmative ええ (ee, "yes").

Strictly speaking, せんせい could be analyzed as ending in the diphthong [ei] rather than a pure long [eː]. In practice, many speakers — particularly in Tokyo dialect — produce something closer to [eː], especially in casual speech. For recognition, treat えい as a long え sound.

お-column exception: usually おう, not おお

A long お sound is most commonly written おう, not おお:

As with えい, the combination おう is pronounced as a sustained [oː] in natural speech, not as two distinct vowels [o] then [u].

True おお spellings do exist in a small number of words. The most common ones you will encounter at this level:

There is no shortcut for determining whether a word uses おう or おお — it depends on the word's etymology. In practice, おう is far more common. When in doubt for a new word, おう is the safer assumption, but always verify with a dictionary.

2.6 Particle Exceptions

Three hiragana characters have special readings when they function as grammatical particles. These are the only cases where a hiragana character is not read according to its standard value.

は → "wa" as the topic particle

The character は is normally read "ha." When it marks the topic of a sentence, it is read "wa."

• わたしは がくせいです。(Watashi wa gakusei desu.) — "I am a student."
• きょうは いいてんきですね。(Kyou wa ii tenki desu ne.) — "Nice weather today, isn't it."

In these sentences, は is the topic-marking particle and is read "wa." In all other positions — at the beginning or middle of a word, for instance — は is read "ha" as expected: はな (hana, "flower"), はは (haha, "mother," where the first は is "ha" and the second は is also "ha" because it is part of the word, not a particle).

へ → "e" as the direction particle

The character へ is normally read "he." When it indicates direction of movement (meaning "to" or "toward"), it is read "e."

• がっこうへ いきます。(Gakkou e ikimasu.) — "I go to school."
• にほんへ きました。(Nihon e kimashita.) — "I came to Japan."

In all other contexts, へ retains its standard reading "he": へた (heta, "unskillful"), へや (heya, "room").

を → "o" as the object particle

The character を is read "o" (identical to お). In modern Japanese, を exists solely as the object-marking particle — you will not find it inside any word.

• みずを のみます。(Mizu o nomimasu.) — "I drink water."
• ほんを よみます。(Hon o yomimasu.) — "I read a book."

Some speakers, particularly in western dialects, do produce a faint [w] before the [o], but in standard Tokyo Japanese, を is simply [o].

Summary

These three cases are the complete list of particle exceptions. No other hiragana character changes its reading based on grammatical function. Memorize these three and you have the full picture.

2.7 Recognition Practice Passages

The following passages use only hiragana. Romaji is provided beneath each one. This is the last time romaji will appear in this book — from the next chapter onward, you will read Japanese script without it. Use these passages to practice reading the hiragana directly. Cover the romaji with your hand or a piece of paper and try to read the hiragana first. Check the romaji only when you get stuck.

Passage 1 — Self-introduction

はじめまして。わたしは たなかです。にほんじんです。とうきょうに すんでいます。だいがくせいです。にほんごと えいごを はなします。どうぞ よろしく おねがいします。

Hajimemashite. Watashi wa Tanaka desu. Nihonjin desu. Toukyou ni sunde imasu. Daigakusei desu. Nihongo to eigo o hanashimasu. Douzo yoroshiku onegaishimasu.

"How do you do. I am Tanaka. I am Japanese. I live in Tokyo. I am a university student. I speak Japanese and English. Pleased to meet you."

Passage 2 — Greeting exchange

A: おはようございます。きょうは いいてんきですね。

B: そうですね。きのうは あめでしたが、きょうは はれですね。

A: そうですね。きょうは なにを しますか。

B: ともだちと こうえんに いきます。

A: Ohayou gozaimasu. Kyou wa ii tenki desu ne.

B: Sou desu ne. Kinou wa ame deshita ga, kyou wa hare desu ne.

A: Sou desu ne. Kyou wa nani o shimasu ka.

B: Tomodachi to kouen ni ikimasu.

"A: Good morning. Nice weather today, isn't it."

"B: It is, isn't it. Yesterday was rainy, but today is sunny."

"A: That's right. What will you do today?"

"B: I'm going to the park with a friend."

Passage 3 — Daily routine

まいにち しちじに おきます。あさごはんを たべて、はちじに いえを でます。でんしゃで かいしゃに いきます。ごじに かえります。ばんごはんを たべてから、すこし ほんを よみます。じゅういちじごろ ねます。

Mainichi shichiji ni okimasu. Asagohan o tabete, hachiji ni ie o demasu. Densha de kaisha ni ikimasu. Goji ni kaerimasu. Bangohan o tabete kara, sukoshi hon o yomimasu. Juuichiji goro nemasu.

"Every day I wake up at seven. I eat breakfast and leave the house at eight. I go to work by train. I come home at five. After eating dinner, I read a little. I go to sleep around eleven."

Passage 4 — Weekend plans

こんどの どようびに、ともだちと いっしょに しぶやに いきます。まず、ひるごはんを たべます。それから、えいがを みます。えいがの あとで、すこし かいものを します。たのしみです。

Kondo no doyoubi ni, tomodachi to issho ni Shibuya ni ikimasu. Mazu, hirugohan o tabemasu. Sore kara, eiga o mimasu. Eiga no ato de, sukoshi kaimono o shimasu. Tanoshimi desu.

"This Saturday, I'm going to Shibuya together with a friend. First, we'll eat lunch. After that, we'll watch a movie. After the movie, we'll do a little shopping. I'm looking forward to it."

Vocabulary List

All words that appeared in this chapter's examples, listed with pitch accent notation. Pitch accent numbers follow the system introduced in Chapter 1: the circled number indicates the mora after which pitch drops, and ⓪ indicates a flat (heiban) pattern with no drop.

From the next chapter onward, romaji will not appear. Every Japanese word, sentence, and passage will be written in kana and kanji only. If you have not yet internalized the hiragana chart, return to this chapter and practice until you can read each character without hesitation. Speed will come with exposure — what matters now is accuracy. You should be able to look at any hiragana character and identify it without consulting a table. That is the foundation everything else builds on.