Doubling Clitics are Pronouns: Reduce and Interpret

Controversy has plagued the question of whether verbal “object markers” (OMs) are object pronouns cliticized to the verb or realizations of object agreement. Beginning with data from Amharic, we develop a reliable and consistent diagnostic for resolving this question. Specifically, we claim that OMs should be analyzed as clitic pronouns if and only if they are unable to double nominals that are quantified, anaphoric, or contain a variable bound by a quantifier. These restrictions can be derived from familiar principles of grammar—the Crossover condition and the Binding theory—once the OM is taken to be a pronoun at LF. These OMs are formed within the syntactic derivation by first moving an object DP to Spec vP (object shift) and then applying a novel syntactic operation Reduce, which replaces DP with its D head. We also explain why the same OMs can double nonreferential DPs when they appear in experiencer subject constructions in terms of the syntactic structure. Finally, we show that previous diagnostics for clitics versus agreement give conflicting results, and we consider some typological implications of our proposal. We confirm that both object clitics and object agreement exist cross-linguistically, and discuss why clitic doubling seems to be more common with objects than subjects.