Hiragana vs Katakana: What's the Difference?

Hiragana (ひらがな) and katakana (カタカナ) are the two Japanese syllabaries. Both spell the same 46 basic sounds, so ka is か in one and カ in the other. Hiragana carries native Japanese words and all the grammar; katakana marks foreign loanwords, foreign names, onomatopoeia, and emphasis. Hiragana is curved, katakana is angular.

That is the whole answer, and it surprises most learners. The two scripts are not competing alphabets or regional variants. They are the same sound inventory written twice, kept separate so that a Japanese sentence can tell you something extra: not just what a word sounds like, but what kind of word it is.

What Each Script Is For

Hiragana: native words and all the grammar

Hiragana is the default. It does four distinct jobs.

Native Japanese words. Words with no assigned kanji, or whose kanji is uncommon, are simply written in hiragana. さくら (sakura, cherry blossom), きれい (kirei, pretty), ありがとう (arigatou, thank you).

Grammar particles. Japanese marks the role of every phrase with small function words, and these are always hiragana: は (wa, topic), を (o, object), に (ni, direction), の (no, possession), が (ga, subject). A sentence’s skeleton is hiragana even when its content words are kanji.

Okurigana. These are the hiragana endings that conjugate verbs and adjectives. The kanji 食 carries the meaning “eat,” and the hiragana tail tells you the tense and politeness: 食べる (taberu, eat), 食べた (tabeta, ate), 食べません (tabemasen, do not eat). The kanji never changes. The hiragana does all the work.

Furigana. Small hiragana printed above or beside a kanji to tell you how to read it. This is how children’s books, manga aimed at young readers, and any text using a rare character stay readable.

So hiragana is not merely one option among two. It is the connective tissue of written Japanese. You cannot read a single sentence without it.

Katakana: foreign words, sound effects, and emphasis

Katakana handles everything that sits slightly outside native Japanese.

Loanwords. Words borrowed from other languages, called gairaigo (外来語), are written in katakana: コーヒー (koohii, coffee), コンピューター (konpyuutaa, computer), パン (pan, bread, from Portuguese pão), アルバイト (arubaito, part-time job, from German Arbeit). Loanwords are commonly estimated at around 10 percent of the modern Japanese vocabulary, so katakana appears constantly in ordinary text.

Foreign names and places. Personal names, countries, cities. アメリカ (America), マリア (Maria), ロンドン (London). If you write your name in Japanese, it goes here.

Onomatopoeia and mimetic words. Japanese has an unusually rich stock of sound words, and katakana is where many of them live. ドキドキ (dokidoki, a pounding heart), ザーザー (zaazaa, heavy rain), ゴロゴロ (gorogoro, rumbling or lounging around).

Emphasis. Katakana does the job italics do in English. A native word normally written in kanji or hiragana can be flipped into katakana to make it jump: ダメ (dame, no good), スゴイ (sugoi, amazing). Advertising leans on this heavily.

Technical and scientific terms. Species names, chemical names, and medical vocabulary are conventionally katakana even when native words exist. Plant and animal names in a field guide are almost all katakana.

Non-human speech in manga. A robot, an alien, an artificial intelligence, or a character speaking in a flat mechanical register will often be lettered entirely in katakana. Readers register it instantly as not quite a human voice. It is one of the quiet pleasures of Japanese typography.

Hiragana tells you what a sentence is doing. Katakana tells you where a word came from.

How They Look Different

Hiragana is curved. Loops, hooks, single continuous strokes that double back on themselves. It looks handwritten because it is descended from handwriting.

Katakana is angular. Short, straight, separated strokes with sharp corners and very few curves. It looks stamped rather than written.

Once you have seen the contrast, you can identify the script without knowing a single character. Compare の and ノ, or ぬ and ヌ. Hiragana rolls; katakana cuts.

The visual difference is not decorative. It exists because the two scripts were invented for different purposes by different people, at roughly the same moment in history.

Where They Came From

Both scripts descend from kanji, and both crystallized during the Heian period (794 to 1185), broadly in the ninth century.

Before them, Japanese was written with man’yougana (万葉仮名), Chinese characters used purely for their sound values with the meaning discarded. It worked, but it was cumbersome: every syllable required a full, complex character.

Hiragana simplified whole characters. Scribes wrote man’yougana in flowing cursive script (草書, sousho) until the shapes wore smooth. 安 (an) became あ. 加 (ka) became か. 乃 (no) became の. Each hiragana is a whole kanji that melted.

Katakana chopped characters up. Buddhist monks annotating Chinese scriptures needed something fast and small enough to squeeze between lines, so they took a recognizable fragment of a kanji and used that. 阿 gave its left element to make ア. 加 gave its left side to make カ. 呂 became ロ. The name says so: kata (片) means partial or one-sided. Katakana is fragment kana.

Hiragana carried a social label too. It was called onnade (女手), “women’s hand,” because aristocratic men were expected to write in Chinese while women wrote in Japanese. The label was meant as a limitation and produced the opposite. The Tale of Genji (源氏物語), written around 1008 by Murasaki Shikibu and widely considered the world’s first novel, is composed largely in hiragana. So is most of the era’s greatest poetry and diary literature. Japan’s classical canon exists because of the script that was considered the lesser one.

Both sets stayed messy for another thousand years. Multiple competing forms per sound, called hentaigana (変体仮名), remained in use until the Elementary School Order of 1900 fixed a single standard shape for each syllable. A further reform in 1946 tidied spelling into the modern kana usage still taught today. If you have ever tried to read a pre-1900 manuscript, or looked at shodo calligraphy in an older hand, you have met the variants that got retired.

Hiragana vs Katakana at a Glance

Both scripts also take the same two diacritics: dakuten (゛) turns か into が (ka into ga), and handakuten (゜) turns は into ぱ (ha into pa). With those marks plus small combining characters, each script covers roughly 100 distinct syllables from its 46 base shapes.

Where Kanji Fits In

Written Japanese uses three scripts at once, and a normal sentence contains all of them.

Kanji (漢字) are the borrowed Chinese characters that carry meaning. There are 2,136 in the jouyou kanji list, the set of characters taught through Japanese schooling, last revised in 2010. Kanji handle nouns and the roots of verbs and adjectives. They compress meaning: 愛, the kanji for love, packs a whole word into thirteen strokes.

The division of labour looks like this in a single sentence:

私はコーヒーを飲みます。 (Watashi wa koohii o nomimasu. I drink coffee.)

• 私 and 飲 are kanji, carrying “I” and “drink”
• は, を, and みます are hiragana, marking grammar and conjugation
• コーヒー is katakana, flagging a borrowed word

Three scripts, one sentence, each doing a job the others do not. This is also why Japanese gets away without spaces between words: the script changes tell your eye where the boundaries are. Our kanji library covers the meaning side of the system in detail.

Which Should You Learn First?

Hiragana. This is close to unanimous advice, and the reasoning is practical.

Hiragana appears in every sentence, so learning it unlocks reading immediately. It is the script used for furigana, which means hiragana is the key to reading kanji you have not learned yet. And beginner textbooks and children’s material are written in it.

Katakana second, and expect it to feel harder despite being simpler in shape. Learners consistently find katakana stickier to memorise, for two reasons: the characters are visually plainer and therefore easier to confuse, and you encounter them less often in beginner text, so they get less practice. The payoff is high, though. Once katakana clicks, a large slice of Japanese vocabulary becomes guessable, because it is English or German or Portuguese wearing a Japanese sound system.

Kanji third, and continuously, for years. Japanese schoolchildren learn both kana sets in their first year of elementary school and then work through the jouyou kanji across the following nine.

A realistic pace for an adult learner: one to two weeks per kana set with daily practice, writing them by hand rather than only recognising them.

How Names Work

Foreign names go in katakana. Always, without exception, in ordinary use.

The process is transcription by sound, not by letter. Japanese syllables are almost all consonant-plus-vowel, so foreign sounds get bent to fit. “Chris” becomes クリス (Kurisu), because Japanese cannot end a syllable on s. “Smith” becomes スミス (Sumisu). “Brad” becomes ブラッド (Buraddo). Long vowels get a bar: マーク (Maaku, Mark).

Japanese names, by contrast, are written in kanji, with the reading often unguessable from the characters alone, which is why Japanese business cards print the kana reading alongside.

Two things people frequently want, and the honest version of each:

Your name in Japanese. What you actually want is the katakana transcription. Our name in Japanese converter handles the sound mapping for you, including the awkward consonant clusters.

Your name in kanji. This is a different and much looser exercise: choosing kanji whose sounds approximate your name and whose meanings you like. It is not your name translated, it is a nickname built from characters, and Japanese people will read it as exactly that. Fine for a stamp or a tattoo, not for a form. If a tattoo is the goal, the Japanese tattoo generator shows both the kana and the kanji options so you can see the difference before anything is permanent.

Lookalike Traps

A short list of the pairs that catch everyone, worth learning as pairs rather than separately.

シ (shi) and ツ (tsu). Identical structure, different stroke direction. シ has strokes that come off the left, running roughly horizontal, and its long stroke sweeps up from the bottom left. ツ has strokes off the top, running vertical, and sweeps down. Mnemonic: ツ has its marks on top like tsunami spray.

ソ (so) and ン (n). Both are two strokes. ソ starts with a short stroke angled downward; ン starts with a short stroke angled more horizontally, and the second stroke sweeps up. The difference is a few degrees of angle, and it is the single most common katakana misreading.

ノ (no), メ (me), and ヌ (nu). A progression. ノ is one stroke, メ adds a crossing stroke, ヌ adds a tail to that.

ク (ku) and タ (ta). タ is ク with an extra stroke through it.

In hiragana: ね, れ, and わ. All three share the same left-hand vertical with a loop. The right side differs: ね loops closed, れ hooks outward, わ curves in without the loop.

Also in hiragana: る and ろ. る has a small loop at the bottom; ろ does not.

Handwriting these rather than only reading them is what separates the pairs permanently. The muscle memory of the stroke order fixes the distinction in a way visual drilling does not.

A Note on Register

Script choice carries tone, which matters once you start writing rather than reading.

Writing something in hiragana that would normally be kanji makes it softer, younger, gentler. Brands do this deliberately: a shop name in all hiragana reads as warm and approachable.

Writing something in katakana that would normally be hiragana or kanji makes it sharper, more modern, or slightly ironic. It is the same word with a different edge.

This is a cousin of what happens in spoken Japanese with keigo, the honorific speech system, and with honorific suffixes like san and chan. Japanese encodes social texture in choices that English handles through tone of voice. In writing, the script is one of those choices.

FAQ

What is the difference between hiragana and katakana?

They are two scripts that spell the same 46 Japanese sounds but are used for different kinds of words. Hiragana is used for native Japanese words, grammatical particles, and verb and adjective endings, which makes it the script that holds a sentence together. Katakana is used for foreign loanwords, foreign names, onomatopoeia, scientific terms, and emphasis. Hiragana characters are curved and flowing; katakana characters are angular and made of short straight strokes.

Which should I learn first, hiragana or katakana?

Hiragana, without much debate. It appears in every Japanese sentence because it carries the grammar, and it is the script used for furigana, the small reading aids printed above unfamiliar kanji. Learn hiragana and you can start reading beginner material immediately. Katakana comes second and typically takes a similar amount of time, one to two weeks of daily practice, though many learners find it harder to retain because they see it less often at first.

Why does Japanese have two alphabets?

Because they were invented separately, for different jobs, and both turned out useful enough to keep. Around the ninth century, court writers simplified whole Chinese characters into flowing cursive shapes, producing hiragana. Around the same time, Buddhist monks annotating Chinese texts needed compact marks that fit between lines and clipped fragments out of characters, producing katakana. Rather than one replacing the other, Japanese assigned them complementary roles. The result is a writing system that signals the origin and register of a word through its script, information English has to convey some other way.

Is my name written in hiragana or katakana?

Katakana, if it is a non-Japanese name. Katakana is the standard script for foreign personal names, and using hiragana for one would look wrong to a Japanese reader in the way lowercase-only writing looks wrong in English. The transcription follows sound rather than spelling, so consonant clusters get vowels inserted and final consonants get padded: “Kate” becomes ケイト (Keito), “Alex” becomes アレックス (Arekkusu). Japanese names, by contrast, are written in kanji, with hiragana or katakana readings supplied when the reading is unclear.

How many characters does each one have?

Each script has 46 basic characters in the modern standard set. Both then take the same two diacritic marks: dakuten, two small strokes that voice a consonant, and handakuten, a small circle that turns the h row into the p row. Adding those marks plus small combining characters brings each script to roughly 100 writable syllables. Kanji is a separate matter: the jouyou kanji list, the set covered by Japanese compulsory education, contains 2,136 characters as of the 2010 revision, and a full dictionary holds many thousands more.