The Modifier-Before-Noun Principle (No Relative Pronouns)

English builds noun phrases in two directions: one-word adjectives go before the noun ("a red car"), but anything longer trails after it ("the book that I read yesterday," "the man standing over there"). Korean has no such split. Every modifier — a single adjective, a possessor, or a whole multi-clause sentence — comes before the noun, the noun lands last, and there is no relative pronoun (no "who," "which," or "that") anywhere. This is the single biggest reordering an English speaker has to master, and it's the gateway to every relative-clause pattern in the language.

The principle: modifiers point right, at a noun that comes last

Korean is head-final: the head (the noun being described) sits at the end, and everything modifying it stacks up in front. A plain adjective shows this in miniature.

비싼 ("expensive") sits before 가방 ("bag") — so far, just like English. The revolution comes when the modifier is longer than one word, because Korean does not flip it behind the noun the way English does. It keeps pointing right, at a noun that always comes last.

A whole clause fits in the adjective slot

Take the English phrase "the book that I read yesterday." English puts "book" first and hangs the clause behind it with "that." Korean packs the entire clause in front of 책 ("book") and ends that clause with a special attributive verb ending — here -은 on 읽다 ("to read").

Read the modifier: 어제 내가 읽은 = "yesterday I read." The whole thing sits before 책, exactly where the one-word 비싼 sat before 가방. There is no word for "that" — the ending -은 on 읽은 does all the connecting work. That is the trade English speakers must accept: the linking job English gives to "that / which / who" is done in Korean by the ending on the final verb of the modifying clause, and by nothing else.

Subject-gap and object-gap relatives, both head-final

English distinguishes "the teacher who teaches Korean" (the noun is the subject inside the clause) from "the person whom I met" (the noun is the object inside the clause), using different relative pronouns. Korean uses no pronoun for either — the same head-final packing, and the "gap" (the missing slot inside the clause) is simply left empty.

Subject-gap — the noun is the doer inside the clause:

한국어를 가르치는 ("teaches Korean") modifies 선생님 ("teacher"); the teacher is understood as the subject inside that clause, but there is no "who" — just the present attributive ending -는 on 가르치다.

Object-gap — the noun is the thing acted on inside the clause:

내가 만난 ("I met") modifies 사람 ("person"); the person is the object of 만나다 inside the clause, yet again no "whom" — the past attributive ending -ㄴ on 만난 links it. The bare phrase is the classic example: 내가 만난 사람, "the person I met."

The four attributive endings (preview)

The verb at the end of a modifying clause doesn't take a normal sentence ending (like 읽어요); it takes an attributive ending that marks tense and aspect while turning the clause into a noun-modifier. There are four, and each gets its own page:

For descriptive verbs (adjectives), the present attributive is -(으)ㄴ, not -는 — so 비싼 가방 ("expensive bag") uses the same -ㄴ that marks the past on action verbs. The details are covered under the present -는, the past -(으)ㄴ, and the prospective -(으)ㄹ. For now, the structural point is enough: the modifying clause always ends in one of these endings and always comes before the noun.

Possessors go in front too

Head-finality isn't just about verbs. Possessors — the "of / 's" relation — also precede their noun, consistent with everything else.

친구의 ("friend's") comes before 가방 ("bag"), and 제 ("my") comes before 친구 — modifiers all the way down, each one to the left of what it modifies. The noun being described is, as always, the last thing you say.

The reframe: build the description, land the noun last

The mental move for English speakers is to resist the urge to say the noun first and then describe it. In Korean you do the opposite: you assemble the entire description — including full clauses with their own subjects, objects, times, and places — and only then release the noun it all points at. And you never look for a "that / which / who" word to bolt the clause on; that word does not exist. The attributive ending on the clause's final verb is the join. Once this clicks, you can decode a long Korean noun phrase by finding the noun at the very end and reading everything in front of it as its description.

Everything from 제가 to 산 describes 신발 ("shoes") and sits in front of it — a five-word clause in the adjective slot, no relative pronoun in sight. For clauses headed by the general noun 것 ("thing / the one that…"), see 것 head-noun clauses.

Common Mistakes

Mistake 1 — Placing the clause after the noun, English-style. The modifier must come before its noun; a trailing clause is ungrammatical.

Mistake 2 — Hunting for a "that / which / who" word to insert. There is no relative pronoun; the attributive ending is the only link.

Mistake 3 — Leaving the clause verb in its sentence form instead of the attributive form. Inside a modifier, the final verb must switch to an attributive ending, not the polite 요-ending.

Mistake 4 — Putting a plain adjective after its noun. Even one-word describers precede the noun, and they take the attributive ending (-ㄴ/은), not the predicative 요-form.

Key Takeaways

• Every Korean modifier — adjective, possessor, or a whole clause — comes before the noun, and the described noun lands last.
• There are no relative pronouns; the linking work of English "that / which / who" is done by an attributive ending on the modifying clause's final verb.
• The four attributive endings — -는, -(으)ㄴ, -(으)ㄹ, -던 — mark tense/aspect and turn a clause into a noun-modifier.
• Decode a long noun phrase by finding the noun at the end and reading everything in front of it as its description; never look for a "that/which" word to insert.