2.2 Katakana
Key Takeaways
- Katakana maps to the same 46 sounds as hiragana but is used for loanwords, foreign names, onomatopoeia, and emphasis.
- Long vowels in katakana use the bar mark (chouonpu), not a doubled vowel: coffee is コーヒー, not コオヒイ.
- The look-alikes shi/tsu and so/n are distinguished by stroke direction and are the most common katakana reading traps.
- Loanwords add vowels to lone consonants (bus -> basu) and merge English L and R into the r-row (hotel -> hoteru).
- Dakuten, handakuten, small ya/yu/yo, and the small tsu all work in katakana exactly as they do in hiragana.
The Second Syllabary
Katakana (カタカナ) is the second of Japan's two phonetic syllabaries. It maps to exactly the same 46 sounds as hiragana, using the same gojuon grid, the same vowels and consonant rows, but with sharper, more angular character shapes. Anything you can write in hiragana you can transcribe in katakana and vice versa, and the two are pronounced identically. What differs is when each is used. Hiragana handles native Japanese words and grammar; katakana is reserved for a specific set of jobs, and recognising those jobs is half the battle when reading it on the N5 test.
When Katakana Is Used
Katakana signals that a word comes from outside the native vocabulary. Its four main uses are:
- Loanwords (外来語, gairaigo): words borrowed from English and other languages, such as コーヒー (koohii, coffee), テレビ (terebi, television), and パン (pan, bread, from Portuguese).
- Foreign names and place names: アメリカ (Amerika, America), スミス (Sumisu, Smith), フランス (Furansu, France).
- Onomatopoeia and sound effects: ワンワン (wanwan, a dog's bark), ゴロゴロ (gorogoro, rumbling).
- Emphasis and technical or scientific terms: used the way italics are used in English.
On the N5 exam, most katakana words you meet are everyday loanwords for food, technology, and places, so building a loanword vocabulary pays off directly.
The Katakana Chart
| a | i | u | e | o | |
|---|---|---|---|---|---|
| — | ア a | イ i | ウ u | エ e | オ o |
| k | カ ka | キ ki | ク ku | ケ ke | コ ko |
| s | サ sa | シ shi | ス su | セ se | ソ so |
| t | タ ta | チ chi | ツ tsu | テ te | ト to |
| n | ナ na | ニ ni | ヌ nu | ネ ne | ノ no |
| h | ハ ha | ヒ hi | フ fu | ヘ he | ホ ho |
| m | マ ma | ミ mi | ム mu | メ me | モ mo |
| y | ヤ ya | ユ yu | ヨ yo | ||
| r | ラ ra | リ ri | ル ru | レ re | ロ ro |
| w | ワ wa | ヲ wo | |||
| n | ン n |
The same modifiers apply as in hiragana. Dakuten and handakuten work identically: カ (ka) becomes ガ (ga), and ハ (ha) becomes バ (ba) or パ (pa). Yoon use a small ャ/ュ/ョ, as in キャ (kya), シャ (sha), and チョ (cho), seen in ニュース (nyuusu, news). A small ッ marks a doubled consonant just like っ, as in ベッド (beddo, bed) and コップ (koppu, cup).
The Long-Vowel Bar
Here is the one structural rule unique to katakana. Long vowels are not written by doubling a vowel kana; instead a single horizontal bar, the chouonpu (長音符) ー, extends whatever vowel precedes it. So 'coffee' is コーヒー (koo-hii), not コオヒイ. This bar is everywhere in loanwords because English is full of long vowels: ケーキ (keeki, cake), スーパー (suupaa, supermarket), テーブル (teeburu, table), ノート (nooto, notebook), エレベーター (erebeetaa, elevator). When reading a katakana word aloud, treat ー as 'hold the previous vowel one extra beat'. The bar is horizontal in normal writing but rotates to vertical in vertically-set text.
The Classic Reading Traps
Four katakana are notorious look-alikes, and the JLPT deliberately uses them as distractors:
| Pair | Sound | How to tell them apart |
|---|---|---|
| シ vs ツ | shi vs tsu | シ's short strokes sweep up from the lower-left (near-horizontal); ツ's strokes come down from the top (near-vertical). |
| ソ vs ン | so vs n | ソ's second stroke drops down; ン's second stroke sweeps up from below. |
| ク vs ワ vs タ | ku / wa / ta | Check whether the shape is open (ワ), has one inner stroke (ク), or has two (タ). |
| ノ vs メ vs ヌ | no / me / nu | ノ is one stroke; メ crosses; ヌ adds a tail. |
A reliable habit is to judge シ/ツ and ソ/ン by stroke direction and angle rather than by the dots alone. Because these appear inside common words such as タクシー (takushii, taxi), テニス (tenisu, tennis), and パソコン (pasokon, personal computer), a single misread character can flip an answer.
Why Katakana Trips Up Beginners
Most learners find katakana harder than hiragana even though both have the same 46 sounds, and the reason is exposure. Hiragana appears in every sentence, so it becomes automatic fast; katakana shows up only in scattered loanwords, so the shapes stay unfamiliar longer. The fix is deliberate practice: spend extra drilling time on katakana specifically, and read menus, product labels, and station signs, which are packed with it. On the exam, katakana appears in Vocabulary items, in Reading passages about shopping and travel, and in the answer choices for Listening, so weak katakana quietly costs points across the whole test. Treat it as equal in priority to hiragana, not as an afterthought.
Reading Loanwords Back to English
The hardest N5 katakana skill is 'un-Japanising' a loanword to recognise the original. Japanese phonology inserts vowels into consonant clusters and swaps sounds, so English words arrive reshaped:
- Final consonants gain a vowel: 'milk' becomes ミルク (miruku), 'bus' becomes バス (basu).
- English 'L' and 'R' both become the r-row: 'hotel' becomes ホテル (hoteru), 'salad' becomes サラダ (sarada).
- English 'V' usually becomes b: 'television' becomes テレビ (terebi).
- Long English vowels take the bar: 'card' becomes カード (kaado), 'game' becomes ゲーム (geemu).
Practise by reading the katakana slowly and restoring the likely English: ボールペン is 'boorupen', a ball-pen (ballpoint pen); アイスクリーム is aisukuriimu, ice cream; レストラン is resutoran, restaurant; デパート is depaato, department store; コンビニ is konbini, convenience store. If you can read all 46 katakana, apply dakuten and the long-vowel bar, and separate the シ/ツ and ソ/ン pairs by stroke direction, you can decode virtually every katakana word on the N5 paper.
On the JLPT N5, katakana is normally used to write which of the following?
The loanword コーヒー (coffee) contains the mark ー. What does it indicate?
Which katakana pair is a classic reading trap because the two characters look almost identical?