Sluicing and the Inquisitive Potential of Appositives

This paper investigates experimentally the generalizations made in AnderBois (2010, 2011, 2014) that a sluice may never take an appositive clause as its antecedent. We find that experimental participants rated sentences with sluice-antecedents in appositives as acceptable. We highlight two factors which influence the acceptability of appositive antecedents for sluices: whether the indefinite NP antecedent and the stranded wh-item include descriptive content (e.g., a man, which man), and whether the appositive clause engages with an issue raised in the preceding context. We argue that AnderBois’s claim that appositive clauses are conventionally unable to antecede sluices is too restrictive and suggest that any theory of sluicing must allow appositive clauses to antecede sluices.