Psycholinguistic Studies on the Acquisition of French as a Second Language: The 'Learner Variety' Approach

MARZENA WATOREK AND CLIVE PERDUE

Introduction

In this chapter we propose to discuss some studies on adult language acquisition undertaken within the learner variety approach, concentrating on French as a target language (TL). The studies range from an analysis of the initial stages of acquisition (Benazzo, 2000; Starren, 2001), around the so-called 'basic variety' (Klein & Perdue, 1997), right up to the advanced, quasi-bilingual stage (cf. Carroll & von Stutterheim, 1993, 1997; Lambert, 1997; Watorek & Perdue, 1999).

We are interested in the recurrent phenomena attested in these studies, namely the 'idiosyncratic' productions (cf. Corder, 1967, 1971) that characterise early stages on the way to the TL, and the 'grammatical' but nonetheless inappropriate (or 'unfluent') productions characterising the very advanced stages (cf. Bartning, 1997). We start by outlining the particularities of the 'learner variety' approach, before summarising some results from the studies cited above.

The 'Learner Variety' Approach

We draw on some results from a large body of empirical work undertaken from a functional, longitudinal and cross-linguistic perspective, which takes into account both communicative factors 'pushing' acquisition and structural factors 'shaping' it, in an attempt to explain the process of acquisition. In general terms, the cognitive and linguistic predisposition of the learner interacts with the formal characteristics of the L2 input in shaping the acquisition process, but a further set of factors – communicative factors – intervenes in pushing the learner to acquire the L2.

Two recurrent phenomena will be of interest in the following section: the type of utterance the learner constructs in order to convey her/his meaning, and the orders of acquisition reported. First, we have a fresh look at an old acquisitional chestnut, 'aspect before tense', and then at learners' utterance patterns and use of anaphoric devices in the construction of descriptive discourse.

Communicative factors

The type of communicative factors of interest here are those intervening in the learner/user's need to express recurrent relational meanings between items of vocabulary that languages grammaticalise to a greater or lesser extent – for brevity we will call these '(grammatical) functions' – relations such as assertion, temporal reference and determination. Such functions are numerous (but not unlimited), and the ones mentioned involve the interaction of sentence grammar, discourse grammar and context-relating rules. There is little reason to assume that they are all equally important for the learner when communicating. The relative communicative importance of expressing such functions is thus held to be a determining factor for acquisition. Another communicative factor of relevance to the examples below is Levelt's (1981) 'linearisation problem', that of arranging the information for production in temporal order, between utterances and within each utterance. Some of the principles underlying the speaker's linearisation of information will be discussed in more detail in the section below.

Formal factors

Languages develop devices to express grammatical functions to different degrees of specification – one speaks for example of 'aspect-prominent languages' as opposed to 'tense-prominent languages'. Different languages therefore give different formal priorities to functions which are nevertheless shared (temporal reference is accomplished in aspect-prominent languages, and vice versa). The learner who by virtue of her SL competence understands how to apply these functions, has to find some means of expressing them in the TL. It is therefore necessary to understand which are the linguistic means used at first, and how the means used for expressing a particular function change – and possibly grow more complex – over time. Note that the grammatical organisation of the SL, or characteristics of the TL input, individually or in tandem, may make certain aspects of the input highly salient, and others less so.

The very mention of 'communicative factors' reveals that we are interested in the learners' real-time communicative activity (restricted to language production in this chapter). The analyst attempts to retrace over time how the learner succeeds, or does not succeed, in the communicative task s/he is engaged in, and this reconstruction allows one to identify what the learning problem was at any given time during the acquisition process. The acquisitionist therefore looks first at the way the learner's linguistic repertoire is organised at a given moment, how this repertoire is put to use in particular communicative tasks, and how the repertoire changes over time in respect to the same tasks. Acquisition and use – or rather, use and acquisition – are therefore not dissociated. The object of investigation is the L2 learner/user. We further assume, for argument's sake, that real-time communicative activity forms part of the goals of any L2 learner, be s/he an adult economic immigrant (2.1), or a university student of a foreign language (2.2).

'Idiosyncratic' Utterances

We look in this section at an old debate in acquisitional studies: whether temporal or aspectual distinctions are acquired first – but with the difference (from some published studies) that the expressive means analysed (the 'alternatives of expression') are not limited to verbal morphology. By 'aspect' we mean grammatical aspect, which we define following Klein's (1994) model.

Klein defines tense and aspect by appealing to a semantic function of finiteness. Finiteness is traditionally associated with the morphosyntactic categories of person and tense. However, Klein distinguishes between the concept of finiteness and the way...