In Icelandic, part of the complex reciprocal hvor annar matches in case with the reciprocal’s antecedent. In structures where the reciprocal is embedded in a PP, the preposition intervenes between the two parts. A recent analysis of these data suggests that part of the reciprocal overtly moves to the base position of the antecedent by an operation termed e-raising . We show that such an analysis makes a number of wrong predictions about the constituencies of such structures and also about the behavior of reciprocals in coordination. We show that this is also the case for other languages that display case-agreeing reciprocals. We instead argue that matching in case between antecedent and reciprocal can occur with the reciprocal staying in situ. Instances with PPs do involve movement but only to the edge of PP and no farther. This analysis is in line with a number of recent approaches that advocate for a morphosyntactic feature-matching relation between antecedent and locally bound anaphors.

Within the typology of embedded pronouns, there are languages that allow for non–first person pronouns to apparently control first person agreement morphology when in certain embedded contexts. This type of agreement displays some degree of optionality: it is also possible for the pronoun to control the expected agreement morphology given the pronoun’s own overt morphological features. This squib provides new data from the Dravidian language Telugu showing that when the embedded pronoun controls agreement on two separate targets, agreement may be uniform across the two targets or the two targets can mismatch in one direction, but crucially not the other. I show how this paradigm can be accounted for using the assumptions that the pronouns in question are similar to so-called hybrid nouns and that agreement features are restricted in principled ways.

On the basis of certain semantic intuitions, Barros (2012) argues that ellipsis does not require structural isomorphism between elided structure and its antecedent. We tackle this claim. Semantic intuitions cannot be a pointer to the analysis of silent structure. We provide empirical evidence that raises the question of to what extent semantic intuitions about plausible articulable syntax must inform one’s analysis of silent structure. We conclude that the answer to this question must be crosslinguistically informed. We conjecture that ellipsis introduces ellipsis-specific interpretive mechanisms, so that intuitions about “how the unelided structure would be interpreted” are not empirically relevant.

A number of works have attempted to account for the interaction between movement and ellipsis in terms of an economy condition Max- Elide. We show that the elimination of MaxElide leads to an empirically superior account of these interactions. We show that a number of the core effects attributed to MaxElide can be accounted for with a parallelism condition on ellipsis. The remaining cases are then treated with a generalized economy condition that favors shorter derivations over longer ones. The resulting analysis has no need for the ellipsisspecific economy constraint MaxElide.