On the exponence of gender in the Irish DP

For a language with two gender values, Irish has a surprising amount of morphological variation and instability, which emerges when looking closely at the dialects which collectively make up the language. We owe to Ó Siadhail (1984) an early formulation of the problem, which identified the key aspects of this irregularity: some nouns vary in gender value across dialects, some have alternative values in the same dialect, some display genitive endings that are characteristic of one gender value but trigger a mutation on following adjectives that expresses the other value, and some have different values (evidenced by the form of the article) in the nominative and in the genitive. In addition, the choice of the gender value for pronouns anaphoric to a DP is often not dictated by morphological agreement with the antecedent but determined on semantic grounds. The following sections will illustrate these categories with several examples; however, since more recent research has considerably sharpened the picture, my main goal will not be to describe the phenomenon. I will rather address the question of what these data can tell us about the competence underlying such puzzling behaviour. A truly satisfactory theory would model the Irish competence in such a way as to predict the boundaries of non-deterministic variation: where gender may fail to determine a certain spell-out, where it may not, and above all, why this is so. As a contribution towards that goal, this paper aims to show that the instability in the exponence of gender in the Irish DP coexists with a significant core of systematicity. This can only be appreciated if we draw a clear distinction between the two types of exponence in question, namely initial mutation (mainly lenition) and phonologically overt exponents. The latter, namely articles and nominalizing suffixes, act as overt exponents which directly spell out a gender value. Initial mutation, on the other hand, is a piece of the Irish morphological system (a morphome, in the sense introduced by Aronoff 1994; see Luís and Bermúdez-Otero 2016) which has several functions, only one of which is the marking of a configuration of gender agreement inside DP. Its realization is subject to a number of constraints, particularly complex in the case of the complement of a lexical noun. It is this relation between mutation and gender agreement that is subject to a significant weakening; when gender has a different realization, its systematic morphological realizations are stable. An empirically successful theory must account for this state of affairs.