Where Hiragana and Katakana Come From (and Why Japanese Has Two Alphabets)

Why does Japanese have two complete sets of characters for exactly the same 46 sounds? And why do some kana faintly resemble kanji? The two questions share one answer: both scripts are kanji, transformed — one by flowing handwriting, the other by radical trimming. The story is a thousand years old, genuinely charming, and (for a learner) surprisingly practical: shapes with a history are shapes you remember.

First, there was borrowed kanji

Japan had a rich spoken language but no writing until Chinese characters arrived around the 5th century. Kanji worked beautifully for meaning, but Japanese grammar — particles, verb endings, all the connective tissue — had no characters of its own. So scribes began using certain kanji purely for their sound, ignoring the meaning: a system now called man’yōgana, after the Man’yōshū poetry anthology written in it.

Writing every syllable as a full, many-stroked kanji was exhausting. Two shortcuts emerged — and became the two kana scripts.

Hiragana: kanji written fast and flowing

Court writers — famously including the women of the Heian court, who produced masterpieces like The Tale of Genji — wrote man’yōgana in ever more cursive, connected brush strokes. Written fast enough, often enough, each character melted into a simple flowing form that kept only the sound:

• 安 (“peace”) flowed into あ (a) — the roof of 安 melts into あ’s curve
• 加 (“to add”) flowed into か (ka)
• 女 (“woman”) flowed into め (me) — the crossing strokes are still visible
• 世 (“world”) flowed into せ (se)

That cursive birth is why hiragana feels round and continuous, and why its strokes flow into each other: each character genuinely is a small piece of fast calligraphy. (It’s also why stroke order matters so much for writing it well — more here.)

Katakana: kanji fragments

Meanwhile Buddhist monks, annotating Chinese texts in cramped margins, took the opposite shortcut: instead of flowing the whole kanji, they broke off a piece of it — one corner, a few strokes — as sound shorthand:

• 加 lost everything but its left half → カ (ka)
• 多 (“many”) is two stacked 夕 — keep just the top one → タ (ta)
• 止 (“stop”) contributed its top-right corner → ト (to)
• 乃 was reduced to its single opening stroke → ノ (no)

Fragments, not flows — which is why katakana is all sharp angles and straight lines. Same parent writing system, opposite haircut.

Notice 加 appears in both lists. Several kanji parented a child in each script: 加 became both か and カ, 女 both め and メ, 世 both せ and セ, 乃 both の and ノ. Seeing those cross-script pairs side by side is one of the quiet delights of learning kana — two siblings, one flowing and one angular, from the same ancestor.

Why both survived

Over centuries the two scripts settled into the division of labor they hold today: hiragana carries native words and all the grammar; katakana marks loanwords, foreign names, onomatopoeia, and emphasis — functioning much like our italics. Modern Japanese uses kanji, hiragana, and katakana together in nearly every sentence, each doing the job it’s best at. (For the full beginner’s map of the three systems, see The Japanese “Alphabet,” Explained.)

Why this history is a learning tool, not trivia

Here’s the practical payoff. Language apps usually help you remember kana with invented picture-mnemonics — “き looks like a key!” Those work for some people, but they’re fiction, and they add a translation step (picture → story → sound) you eventually have to shed.

The etymology does the same memory work while being true:

• あ hides 安, “peace” — the very first kana has a calm heart.
• ツ standing tall comes from 川, “river”; シ lying flat comes from winding 之 — which is exactly how to tell those look-alikes apart.
• ハ is 八, “eight,” barely changed — the monks hardly needed to trim it.

A shape with a story attaches to something; a shape without one is just strokes. And as a bonus, every origin you absorb is a tiny head start on kanji — you’ll already “know” 安, 加, 女, 川 and friends by sight when formal kanji study begins.

This is exactly how KanaLearn uses the history. Twenty-six characters — the ones whose lineage is visually obvious to a complete beginner — carry short origin stories, told on the character’s introduction card by Sumi, the app’s ink-drop spirit. Each derivation is verified against the scholarly man’yōgana tables (where the resemblance is too weak to help, the app deliberately stays quiet — a strained story damages trust in the good ones), and cross-script pairs get their reunion moment in both entries. History as memory glue, applied exactly where it holds.

The bottom line

Japanese has two “alphabets” because two different groups of writers, solving the same problem a millennium ago, simplified kanji in two different directions: courtiers flowed it into hiragana, monks fragmented it into katakana. The sounds stayed identical; the jobs diverged. And for you as a learner, that history isn’t background reading — it’s one of the best mnemonic systems available, because it happens to be real.