Who and What Do Who and What Range Over Cross-Linguistically?

Singular simplex wh phrases in Greek, Hungarian and Spanish lack the uniqueness presupposition of singular which phrases, yet plural ones keep an anti-singleton inference. Dayal's account and the weak plural survive if a simplex phrase can also range over upward-entailing quantifiers, which presuppose nothing. Strengthening is then obligatory wherever an equivalent alternative is defined with it, so without a plural antecedent quién is usable in any context and quiénes only where antiuniqueness holds, and the answer and context judgments follow. Heim's catfish shows what the principle yields at a context: the weak implicated presupposition.

Implementation notes

• A world of a distributive nucleus is the plurality it holds of, so the member for x is Set.Ici x. The paper's undefined members are left out of its Hamblin sets, which does not affect Dayal's presupposition since they are never true.
• A higher-order phrase ranges over the upward-entailing quantifiers living on its domain, §4.3: upwardGQ, positive combinations and the trivial quantifiers. hamblinPlus is the fragment (32) the paper focuses on for presentational purposes.
• EXH is Presupposition.strengthen, which excludes only strictly stronger alternatives where (40) excludes every non-weaker one; they agree on comparable alternatives, as on the scale (53).
• AvoidsIneffability is (45) for one occurrence of EXH, all the Spanish data needs. Clause (c) allows the competitor no deactivated EXH, so it competes with EXH active, as (46) and (47) do.
• A context is the set of worlds compatible with what the relevant individual believes. Under the Matrix K axiom they adopt from Meyer, n. 14, the same set is the doxastic state K quantifies over, as in the catfish theorems.
• The paper makes quién_e an alternative to both plural forms, p. 564, quién_(et)t one to quiénes_e, p. 571, and the plural one to the singular forms where the context supplies it, p. 570, the proposition pl. Without a plural antecedent the higher-order forms are both usable where existence is not assumed, and Brevity chooses between them, n. 19, Form.simpler.
• (24c) has a plural antecedent. With one quiénes_e can also drop EXH where existence is presupposed, which the paper does not test. (68) has none, and the paper leaves it open: both phrases are predicted usable, against the judgment on quiénes.

Errata

Table 1 prints a cross for the anti-singleton inference of plural which phrases, and (15) tags a plural which question as carrying uniqueness, against (3), (16), (17) and the surrounding text. The proposition said to answer (76) prints B ⊕ C where (76) and the sentences before it have A ⊕ C. The text before (55) calls quién_e acceptable in ≥ 1 contexts, against (51); it should be the ambiguous quién.

TODO

• Percus's (41), which needs EXH under a quantifier and the projection of its presupposition.
• (79), (80): quién with collective predicates, unexpected on (78). The paper suggests moving the non-atomicity presupposition to the event and leaves the matter open.
• Context as a source of alternatives is the proposition pl, not computed from the antecedent.

References

• Elliott et al. 2022, Who and What Do Who and What Range Over Cross-Linguistically?
• Dayal 1996, Locality in WH Quantification: Questions and Relative Clauses in Hindi
• Maldonado 2020, Plural Marking and d-Linking in Spanish Interrogatives
• Spector 2007, Scalar Implicatures: Exhaustivity and Gricean Reasoning formalized
• Fox 2018, Partition by Exhaustification: Comments on Dayal 1996 formalized
• Heim 1991, Artikel und Definitheit
• Percus 2006, Antipresuppositions
• Meyer 2014, Deriving Hurford’s Constraint