Abstract for Donna Lardiere's talk

Feature Reassembly in Second Language Acquisition
Donna Lardiere (Georgetown University)
October 16, 2008 (Thursday)
4:15 PM - ; Room 6417, The CUNY Graduate Center

In this talk I consider prevailing notions of "parameter-resetting" in adult second language acquisition (SLA) and argue that these notions are problematic. Within a principles-and-parameters framework, parameters are hypothesized to constitute a limited set of highly restrictive options, or points of variation, between languages. Over the past two decades within formal linguistic approaches to SLA, the failure of many adult language learners to reach nativelike grammatical proficiency has been descriptively modeled in terms of an inability to reset one or more parameters from the L1 value to that of the L2. For example, Haegeman (1988: 255) outlines a basic strategy for modeling L2 syntactic development in terms of parameter-resetting:

To go from the L1 to the L2, learners will often have to reset existing parameters or reassign values to them. Failure to do so will mean that the learner does not attain the L2. The latter possibility seems to be what negative transfer is about.
More recently, essentially the same view has been updated in terms of "parametric feature selection" in which certain features that are morphologically expressed (or "selected") in the L2 but not in the learner's L1 (or any language learned prior to a hypothesized critical period), are claimed to be no longer available and thus unacquirable, resulting in the phenomenon of incomplete L2 acquisition known as "fossilization" or "impairment" (e.g., Hawkins 2005; Hawkins & Hattori 2006; Tsimpli & Dimitrakopolou 2007).

Using findings from a longitudinal case study (Lardiere, 2007) and a cross-sectional study (Choi & Lardiere, 2006) as well as linguistic examples from English, Mandarin Chinese, and Korean, I show how the formal task facing a second language learner is actually much more complex than the parametric "selecting" of a new feature such as [+past] or [+plural] in the target language. Under the view that grammatical categories are bundles of morphosyntactic features, it is clear that these features can be combined and permuted in various configurations cross-linguistically. So, among the difficulties confronting any learner in figuring out how to express the morphological categories of any language are the following:

- What are the particular factors that condition the realization of a certain form (such as an inflection) and are these phonological, morphosyntactic, semantic or discourse-linked?
- Are certain forms optional or obligatory, and what constitutes an obligatory context?
- In which functional categories are various features expressed, clustered in combination with what other features? To what extent are such categorial-feature correspondences and clusters universally invariant?

I will illustrate just a few of the types of interesting learning problems confronting an adult native speaker of one (or more) language(s) who is trying to learn the grammatical features of another. In sum, I will argue that acquiring the L2 involves determining how to reconfigure or remap features from the way these are organized in the L1 into new formal configurations on possibly quite different types of lexical items in the L2. This is a formidable learning task that goes far beyond the simple "switch-setting" or "selecting" metaphors often used to characterize the acquisition of a secondlanguage grammar.

You can listen to the talk here in WMA format.
You can download the handout here in PDF format.