The ㄹ Drop: 살다 → 삽니다 / 사세요 / 사는

Korean textbooks list the ㄹ irregular (ㄹ 불규칙) among the "irregular verbs," and that framing scares learners more than it should. Stems ending in ㄹ — 살다 ("to live"), 알다 ("to know"), 만들다 ("to make"), 팔다 ("to sell") — do lose their ㄹ before certain endings, which is why they land in the irregular chapter. But of all the classes here, this is the one you can master by rule alone: the ㄹ drops before a fixed, tiny set of sounds, every single time, on every ㄹ-stem, with no lexical exceptions to memorize. Learn the trigger set once and the whole class is yours. (For the same rule presented as a regular stem alternation, see ㄹ-stems; this page treats it as the odd member of the irregular family.)

The rule: ㄹ drops before ㄴ, ㅂ, ㅅ, 오

The stem-final ㄹ elides when the ending begins with one of four sounds. The classic Korean mnemonic strings them together: ㄴ · ㅂ · ㅅ · 오.

There is a second half to the rule, just as important: ㄹ-stems never take the 으 buffer. Where a normal consonant stem inserts 으 before -세요, -니까, -면, -ㄴ, a ㄹ-stem does not. So the "(으)" simply never materializes on these verbs.

That 삽니다 is pronounced [삼니다] — the ㅂ nasalizes to [m] before the ㄴ, which is why the romanization is samnida. That is ordinary nasalization riding on top of the drop; the spelling rule is just "ㄹ out, ㅂ in."

The full 살다 paradigm

Seeing the drop-vs-keep split laid out end to end is the fastest way to internalize it:

The one row to slow down on is -(으)면. Its consonant is ㅁ, which is not a trigger, so the ㄹ stays — but the 으 is still blocked because it's a ㄹ-stem. Result: 살면 (not ×살으면, and not ×사면). Contrast that with -(으)니까, whose consonant is ㄴ once the 으 is stripped, so there the ㄹ does fall: 사니까.

Same rule on every member

The whole class behaves uniformly. Once 살다 has shown you the pattern, every other ㄹ-stem does the identical thing — verbs and descriptive verbs alike:

Descriptive verbs (adjectives) do it too — they just take the attributive -(으)ㄴ instead of a -는: 길다 → 긴, 멀다 → 먼, 달다 → 단, 힘들다 → 힘든. The ㄴ triggers the drop.

Why it acts "batchim-less"

The cleanest way to hold the whole class is this: for the purpose of choosing 으-endings, a ㄹ-stem behaves as if it had no batchim at all — like a vowel-final stem. Compare it directly with a normal consonant stem such as 먹다:

먹다 takes the "heavy" allomorphs meant for stems that end in a consonant (-습니다, -으세요, -으니까); 살다 takes the "light" ones meant for stems that end in a vowel (-ㅂ니다, -세요, -니까). The ㄹ is there in the spelling, but for suffix selection the language treats the stem as open. Internalize that single idea and you no longer have to remember which forms drop the ㄹ — you just reach for the vowel-stem endings and the ㄹ falls out of the way on its own.

The "오" trigger and the ㄹ+ㄹ merger

The fourth trigger, 오, shows up mainly in the formal-archaic 하오체 ending -(으)오: 살다 → 사오, 알다 → 아오. You will meet it in older novels, historical dramas, and set formal phrases rather than everyday speech (archaic / literary). Separately, when a ㄹ-stem meets an ending that itself begins with ㄹ — the prospective -(으)ㄹ, as in 살 거예요 ("will live") — the two ㄹ's simply merge into one; there is no doubling. So the future of 살다 is 살 거예요, spelled with a single ㄹ.

Reframing for English speakers

English has no consonant that appears and vanishes according to the next suffix, so the instinct is to see randomness. There is none. Picture the ㄹ as a light that switches off in front of exactly four things — ㄴ, ㅂ, ㅅ, 오 — and stays on everywhere else, while the stem quietly refuses the 으 buffer throughout. Because the trigger set is fixed, small, and applies to the entire class without a single exception, the ㄹ irregular is arguably the most rule-governed member of the whole irregular family — the opposite of the ㅂ or ㄷ classes, where you must learn membership word by word. Learn the trigger set, not a list of forms.

Common Mistakes

1. Inserting the 으 buffer. ㄹ-stems never take 으; the ㄹ drops instead.

2. Keeping the ㄹ in the formal form (and using -습니다). The ㄹ drops and the ending is -ㅂ니다.

3. Keeping the ㄹ before the attributive -는. ㄴ is a trigger, so the ㄹ falls.

4. Dropping the ㄹ before -(으)면. Here the ㄹ stays; only the 으 is blocked.

Key Takeaways

• A stem-final ㄹ drops before endings beginning with ㄴ · ㅂ · ㅅ · 오: 살다 → 사는, 삽니다, 사세요, 사오.
• ㄹ-stems never insert 으 — they take the vowel-stem allomorphs (-ㅂ니다, -세요, -니까), behaving as if they had no batchim.
• The ㄹ stays before -아/어, -고, and -(으)면: 살아요, 살고, 살면.
• Trap: -(으)면 keeps the ㄹ but drops the 으 → 살면 (not ×살으면, not ×사면).
• Adjectives too: 길다 → 긴, 멀다 → 먼 (ㄹ drops before the attributive ㄴ).
• It sits with the irregulars but is fully rule-governed — no membership list, unlike the ㅂ/ㄷ/ㅅ classes.