Japanese Homophones: 箸 / 橋 / 端 and How to Tell Them Apart
Say はし and you could mean chopsticks (箸), a bridge (橋), or an edge (端). Japanese is full of these homophones (同音異義語), and they worry beginners far more than they should. In practice, context does almost all the work, and pitch quietly takes care of the rest.
Why Japanese Has So Many
Japanese has a fairly small set of sounds, so plenty of words end up pronounced the same. The bigger source, though, is borrowing. Japan imported huge numbers of Chinese compounds that were kept apart by tones in Chinese, and those tones didn’t survive the trip. That’s how a reading like こうしょう ended up with a dozen different written forms.
It’s also a good answer to “why bother with kanji at all?” The writing system exists partly to keep these meanings apart on the page, even when the spoken word is identical.
Common Homophone Sets
Context and Pitch
Context carries most of the load. Nobody confuses 「はしを渡る」 (cross the bridge) with 「はしで食べる」 (eat with chopsticks), because the verb already gives it away. You rarely even notice the ambiguity in a real sentence.
Pitch handles the edge cases. In standard Tokyo speech 箸 (chopsticks) is high-low (HA-shi), while 橋 (bridge) is low-high and 端 (edge) stays flat. You don’t need to sit down and drill pitch charts, but knowing the difference is there explains why a native speaker never hesitates between them.
When Even Native Speakers Clarify
Context does not solve every case, and Japanese has its own workarounds for the pairs it does not. 化学 (chemistry) and 科学 (science) are both read かがく, and in a spoken sentence they can genuinely be ambiguous, since both are the kind of abstract noun that shows up in similar contexts. Native speakers sometimes say 化学 as ばけがく instead, using an alternate reading purely to keep it distinct from 科学 in conversation.
Seeing this exist should be reassuring, not worrying. If native speakers occasionally need a workaround for the hardest pairs, an ordinary amount of context-based confusion on your part is not a sign you are behind. It is the normal experience of using the language.
Where This Shows Up on the JLPT
Homophones rarely appear as a dedicated question type, but they shape kanji-reading and kanji-writing questions directly: examiners often build wrong answer choices from a real homophone of the correct word, betting that a learner who has only memorized sounds, not kanji, will pick the wrong one.
In the listening section, homophones matter less than you would expect, precisely because the surrounding dialogue almost always disambiguates them the same way it does in everyday conversation. Your reading and writing accuracy on homophone pairs is the more reliable predictor of exam performance here.
How to Handle Them
- Trust the sentence. In real listening, the surrounding words remove the ambiguity instantly.
- Learn the kanji that separate them. For reading, the character is the disambiguator: 雨 vs 飴 is obvious on the page.
- Pick up pitch passively. Shadow native audio; the high/low pattern sinks in without memorizing accent numbers.
- Don’t panic at compound homophones. こうしょう has many forms, but context narrows it to one every time.
Teacher's Note
Beginners worry they’ll mishear homophones constantly. In practice you almost never do, since Japanese speakers rely on context so heavily that the “ambiguity” only exists on paper. Focus on the kanji for reading, and let context carry your listening.
See the kanji, know the meaning
Nihongo Pass shows every word in kanji with audio and context sentences, so homophones sort themselves out.