‘Real’ and apparent optionality in second language grammars: finiteness and pronouns in null operator structures

Abstract

The existence of optionality in acquisitional data presents a problem to the view that developing grammars do not include more than one grammatical system at any stage. In this article data from Greek and Spanish, on the one hand, and English, on the other, constitute the background for the discussion of the existence of ‘true’ optionality in second language (L2) grammars and its potential persistence at advanced proficiency levels. We also discuss the relation between optionality and finiteness features in L2 grammars as well as how morphology interacts with the development of null operator structures (NOS). Specifically the article deals with the use of pronouns or empty categories in NOS and their relation to finiteness. We discuss the role of clitics in adult second language acquisition when the first language (L1) and the L2 differ in the choices of the pronominal system and in their choice for a gap or clitic pronoun in NOS. The subjects studied are speakers of Greek and of Spanish, languages with clitics, learning English, a language without clitics, as well as speakers of English learning Greek or Spanish. The data collected support the claim that optionality is found in developing grammars but not randomly. First, there is a difference in the degree and nature of optionality found developmentally in advanced as opposed to intermediate learners; secondly, the degree of optionality depends on the morphological richness characterizing L1 and L2 in relation to the phenomena studied. Thus, English learners of Spanish or Greek show more optionality in the use of clitic pronouns and less evidence for a correlation between finiteness and clitics in NOS. On the other hand, Spanish/Greek learners of English show constrained optionality in the use of empty categories or pronouns in NOS.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 62 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.