Everyone hits this wall. You’re cruising through katakana, feeling good, and then you meet シ and ツ in the same word and your brain just quietly gives up. They’re the same character. They have to be the same character. Nobody could reasonably expect you to tell them apart.
You’re not imagining it. シ (shi) and ツ (tsu) really are almost identical — two little marks and a longer stroke, arranged in what looks like the exact same way. But there is a real, consistent difference, and once you see it you can’t un-see it. Here’s the trick that finally makes it stick.
It’s not the dots. It’s the direction.
The instinct is to count the marks or measure the gaps. Don’t. Both characters have the same pieces. What actually separates them is the direction you write the strokes — and, as a result, the angle the marks sit at.
- シ (shi) — the two short marks sit stacked vertically, near-horizontal, and the long stroke sweeps up from the bottom, like an underhand throw. Think: shi → the marks are side by side going down, and you finish sweeping upward.
- ツ (tsu) — the two short marks sit horizontally across the top, more vertical themselves, and the long stroke comes down from the top, like an overhand throw.
The cleanest mental hook: シ smiles up, ツ points down. シ’s long stroke curves up at you. ツ’s comes down. If you learned to write them by hand — even badly, even once — the muscle memory of “up sweep” vs “down sweep” locks the distinction in far better than staring ever will.
The dots reinforce it (once you know what to look for)
The angle of the two little marks follows from the stroke order:
- シ — marks lie flatter, more horizontal, hugging the left side.
- ツ — marks stand more upright, more vertical, sitting along the top.
You don’t need to measure anything. Learn the up-sweep-vs-down-sweep and the dots fall into place on their own.
And they travel in packs
シ and ツ are the famous pair, but katakana has a whole crew of lookalikes that ambush beginners. The other big two:
- ソ (so) and ン (n) — same story, smaller scale. ソ’s stroke comes down (like ツ, its cousin); ン’s sweeps up (like シ). The up/down logic is the same. Once you’ve got シ/ツ, you basically get this pair for free.
- ワ (wa), ク (ku), and ケ (ke) — the frame is similar, the internal angles differ. Less brutal than the first two, but worth a deliberate look so they don’t blur.
The reason katakana does this to you and hiragana mostly doesn’t: katakana is built from short, sharp, angular strokes with far less rounding to distinguish them. Fewer curves means fewer visual handles, which means more collisions. It’s not you. It’s the alphabet.
Why this matters more than it seems
Here’s the thing about katakana specifically: it’s where a lot of the words you already half-know live. Loanwords — コーヒー (koohii — coffee), ビール (biiru — beer), コンビニ (konbini — the convenience store), バイト (baito — part-time job) — are written in katakana, and they’re everywhere in real life. A 26-year-old in Tokyo runs into katakana constantly, because modern casual Japanese is full of borrowed words. Getting shaky on シ vs ツ isn’t a cute beginner stumble; it’s the thing standing between you and reading half the signs in a convenience store.
So it’s worth the ten minutes to lock it down properly.
シ smiles up. ツ points down. That’s the whole secret, and now it’s yours.