History and evolution
Katakana developed in the 9th century alongside hiragana, derived from shortened fragments of Chinese characters. Originally used by Buddhist monks as pronunciation aids for Chinese sutras (via the kunten reading system), the script was a scholarly shorthand, monks would write a small katakana character next to a kanji to mark how to read it aloud. Over the medieval period, katakana's role shifted toward foreign words entering Japan: Portuguese and Dutch loanwords during the 16th and 17th centuries (パン from Portuguese pão, ガラス from Dutch glas), then English loanwords from the Meiji era onward. The post-war 1946 kana usage reform (現代仮名遣い) fixed the modern 46-character set alongside hiragana. Today katakana marks three distinct registers in Japanese text: foreign vocabulary (コーヒー, coffee), scientific names (ネコ for the biological species cat, even though the common noun 猫 is usually kanji), and onomatopoeia or emphasis (like italics in English).
Where the shapes come from
Each katakana character is a fragment cut from a specific kanji, the monk shorthand preserved as a national script. A few anchors: ア from 阿 (left element), イ from 伊 (left side), ウ from 宇 (top), カ from 加 (left), ケ from 介, フ from 不 (first strokes), ヌ from 奴 (right side). Compare with hiragana: hiragana derives from the cursive full-character simplification, while katakana derives from cutting out a piece. Same parent kanji, different daughters, for example, か (hiragana from 加) and カ (katakana from 加) share the same ancestor.
How Katakana fits in written Japonés
Katakana marks anything foreign, technical, or emphatic in Japanese text. Loanwords dominate: computer is コンピューター, restaurant is レストラン, coffee is コーヒー. Scientific and biological names use katakana (ネコ, cat). Onomatopoeia and sound effects, the constant ドキドキ of a pounding heart, the ガシャン of breaking glass, go in katakana. The long-vowel mark is ー (a dash), specific to katakana; hiragana uses a second vowel instead. When you see angular shapes in Japanese text, the word is almost always a loanword or an emphasis cue.
Common pitfalls
- シ (shi) vs ツ (tsu)
- The classic confusion pair. The key tell is stroke count, not direction: シ has one stroke, ツ has two. The dashes above are stacked vertically in シ and horizontally in ツ once you internalize the stroke order.
- ン (n) vs ソ (so)
- Similar shape confusion. ン is one stroke; ソ is two. The dash in ン enters from the upper-left and curves down; ソ's dash enters from the upper-right.
- The ー long-vowel mark is not a hyphen
- コーヒー reads "kōhī" (coffee) with extended vowels, not "ko-hi-". ー lengthens the preceding vowel in katakana. It only appears in katakana; hiragana uses a second vowel (えい, ou) for the same effect.
- English loanwords get heavily adapted
- Japanese syllable structure inserts vowels into consonant clusters. Strike becomes ストライク (su-to-ra-i-ku); McDonald's becomes マクドナルド (ma-ku-do-na-ru-do). Reading katakana fluently requires recognizing the stretched shape of English words.
- Double-check katakana in scientific contexts
- Biology papers and scientific articles use katakana for species names even for native Japanese organisms, イヌ for dog, ネコ for cat. Spotting this register prevents confusion with loanwords.
How to learn Katakana
- Skip ahead from hiragana, the sound map is identical. If you know か (ka), you already know カ (ka). Learning katakana is mostly about memorizing new visual shapes for sounds you already pronounce.
- Group the look-alikes first. シ/ツ and ン/ソ confuse everyone; learn them as pairs and drill the stroke-count difference until it's automatic.
- Use spaced repetition (Karpicke & Roediger, 2008 on testing effect). Ten minutes of daily recall for one week is enough for most hiragana-solid learners.
- Read katakana-heavy text: foreign menus, product packaging, brand names, tech articles. Recognizing stretched English words in katakana is its own skill that only improves with volume.
- Practice stroke order if you plan to handwrite; for screen-only learners, recognition matters more than production.