I want to say I can — or can't — tell something that isn't on the surface.
序 · The move
1Pick the perception verb: 看, 聽, 感覺, 想, 認, 猜.Is this about discerning something, not moving a body out of a place?
2Wedge 得 (can) or 不 (cannot) between the verb and 出來: 看得出來 / 看不出來.Am I claiming reachability of the recognition, not a finished fact? For a finished fact use 看出來了, no 得/不.
3Name what is discerned after it, often a whole clause: 看不出來他在生氣.Does 出來 mark the thing surfacing into notice, not a physical exit?
例 · Examples
1從他說話的樣子the way he speaks來看，感覺不出來can't sense/detect — the perception never surfaces他很緊張nervous。
From the way he talks, you can't tell that he's nervous.
可能補語 is the bare V得/不+result machine for any outcome (看得懂). 出來 is one specific result fixed into that slot: the thing coming out into recognition. 看得懂 = grasp the meaning; 看得出來 = detect a hidden state.
V出1 is the literal exit — a body leaving an enclosure (走出餐廳). Here 出來 is figurative: a perception surfacing, nothing physically moves. 拿出來 (take out) is literal; 感覺出來 (sense) is this point.
A resultative states the outcome reached (看懂了). The 得/不 here gates whether it can be reached at all — possibility, not accomplishment. 看出來了 (told it) vs 看不出來 (can't tell).
✗ 我不感覺出來他緊張。 → ✓ 我感覺不出來他緊張。 (不 wedges between verb and 出來, never before the verb.)
✗ 我看出來得他在說謊。 → ✓ 我看得出來他在說謊。 (得 sits right after the verb, before 出來.)
✗ 這個字我看得出來懂。 → ✓ 這個字我看不出來 (是什麼)。 (出來 is the detect result; don't stack a second result like 懂 behind it.)
English 'I can tell' / 'I can't tell' puts the modal can before one verb; Chinese splits ability inside the verb as 得/不 and pins the result with 出來, so learners reach for 能/會 and drop 出來.