Overview

This chapter reviews research since the mid 1980s on the acquisition of reflexives (especially zibun) by L2 learners of Japanese. A first wave of generative-inspired work assumed parameterization of the binding principles. Empirical studies indicated that L2 learners' grammars are largely consistent with Universal Grammar-based constraints on crosslinguistic variation in the grammar of reflexives. Subsequent research abandoned parameterization in favor of raising reflexives in LF. Studies in this vein produced equivocal results, despite benefiting from methodological advances and a well-articulated research hypothesis about L2 learners' knowledge of zibun. Recent work has explored child L2 acquisition of zibun, and whether L2 learners control relevant pragmatic and discourse-based constraints.

I begin with some background on each of the three topics that converge in this research: Japanese, reflexives, and L2 acquisition.

Japanese

While a strong tradition of generative analyses of Japanese has been sustained since Kuroda (1965), there also exists a long-standing commitment to functional linguistics (e.g. Kuno, 1987). Sometimes generative formalism is viewed as complementing functionalism, sometimes as competing with it. Both approaches are in evidence in research on L2 acquisition of Japanese reflexives.

It also is worth mentioning that within western linguistics Japanese is often taken as emblematic of “otherness,” reinforcing native speakers' investment in a deep-rooted “myth” (Miller, 1977) of the uniqueness of the language. One can only speculate about the extent to which a predisposition to magnify the distinctiveness of Japanese has influenced linguistic analysis.