Long-distance agreement limits

Hindi agreement is normally local: a finite predicate finds its controller inside its own clause. A limited apparent exception occurs in tightly integrated infinitival complements, especially with predicates such as चाहना “want.” An unmarked object inside the infinitive can sometimes determine the form of both the infinitive and the higher perfective verb: मीरा ने चिट्ठी लिखनी चाही. This is traditionally called long-distance agreement, but it cannot cross any arbitrary string of words or a finite कि boundary.

The ordinary local baseline

In a simple perfective clause, a ने-marked subject is unavailable and an unmarked object controls agreement.

When a closely integrated infinitive is added under चाहना, the feminine object can be reflected on the infinitive and the matrix perfective:

The agreement chain is चिट्ठी — लिखनी — चाही. Semantically, चिट्ठी is the object of लिखना, not of चाहना. Its features become visible higher in the predicate because the two verbs form a tightly integrated non-finite structure.

The internal agreement of such complements is introduced in infinitive subjects and agreement; the present page asks how far those features can remain visible in a higher predicate.

A controlled gender-number contrast

Keeping the subject and verbs constant shows the embedded object’s effect.

The masculine plural infinitive is पढ़ने, and standalone feminine plural matrix perfective is चाहीं. This pattern is grammatically available in Standard Hindi, though speakers differ in how freely they use it and often choose a nonagreeing infinitive-plus-masculine-singular matrix alternative, especially in conversation.

The nonagreeing alternative

Modern usage also permits or favors an invariable masculine singular infinitive followed by default masculine singular matrix morphology in many contexts: कहानी पढ़ना चाहा. The choice is sensitive to predicate, dialect, register, information structure, and speaker grammar; it is not honest to claim that every infinitival complement must show long-distance agreement.

Both patterns occur. The second makes the agreement chain overt; the first packages the infinitival activity more like an invariable proposition. The full range of speaker variation is more complex than a single classroom rule.

को blocks the embedded object

An object marked with को cannot launch long-distance agreement. With matrix ने also blocking the higher subject, masculine singular default appears on both the infinitive and matrix verb.

Feminine singular and plural features remain invisible because लड़की/लड़कियों को is case-marked. Forms such as बुलानी चाही would wrongly treat that phrase as eligible.

A finite कि clause is a hard boundary

Long-distance agreement does not cross a finite complementizer. Compare a shared-subject infinitive with a clause whose embedded verb has its own mood and subject.

In the first, Sara is understood as the memorizer and feminine कविता can support agreement through the non-finite predicate. In the second, कि introduces a finite clause with its own subject बेटी and subjunctive करे. Matrix चाहा is masculine singular default; embedded कविता cannot control it.

This contrast also explains why an English finite complement cannot simply be copied into Hindi without कि. Clause structure, not semantic relatedness, determines the agreement domain.

Each finite clause calculates independently

Matrix कहा is default because it takes a proposition. Embedded योजना controls बनाई है after सीमा ने. The feminine object is important to what Rajat said, but importance is not an agreement relation.

Matrix बताया and embedded पढ़ रही हैं belong to separate finite domains. The embedded progressive agrees with feminine plural लड़कियाँ, not with matrix अध्यापिका.

Not every verb sequence is a long-distance construction

Compound verbs such as पढ़ लिया are a single complex predicate: the vector carries local agreement with the object. Conjunctive participles such as पढ़कर link events without carrying gender-number agreement. Neither should be confused with the agreeing infinitive pattern.

In the compound, feminine किताब controls vector ली locally. In the converb chain, पढ़कर is invariant, while intransitive main predicate सो गई agrees with Noor. Word adjacency alone cannot identify an agreement path.

Negation and focus stay inside the structural limits

Negation can scope over wanting or over the infinitival event, but it does not open a path through को or कि. Focus can make the embedded object prominent without changing its case eligibility.

Feminine unmarked किताब supports the agreement chain in both. The location of नहीं affects scope—denial of desire versus contrast within the desired action—but not controller eligibility.

An embedded or multi-clause diagnostic

Matrix कहा is default. Inside the कि clause, feminine plural फ़ाइलें supports देखनी चाहीं. The final clause has its own unmarked masculine subject समय, controlling मिला. One sentence therefore contains three independent agreement calculations.

Common Mistakes

Key Takeaways

Hindi agreement normally stays within one finite clause. Restricted long-distance agreement can occur through a tightly integrated, shared-subject infinitive, especially with चाहना: an unmarked object may control the infinitive and higher perfective. A को-marked object blocks the pattern, and finite कि creates a hard boundary. Nonagreeing infinitive-plus-default variants are common and speaker-sensitive. Compound verbs, converbs, and finite complements have different structures and must not be diagnosed from word order alone.