Excerpt
Contents
Introduction
Krashen’s Input Hypothesis
Long’s Interaction Hypothesis
Swain’s Comprehensible Output Hypothesis
Foreigner Discourse modifications and SLA
(Un)modified input-SLA relationships – a selection of studies
Conclusion
Bibliography
Introduction
This essay addresses a question which is central to input research, namely to what extent the quality of input has an effect on SLA, particularly on the learner’s ability to comprehend. Thereby the term ‘quality’ shall be interpreted as referring to any impact of modified input and negotiated interaction that is not related to its amount or frequency. After a brief review of Krashen’s Input Hypothesis, Long’s Interaction Hypothesis and Swain’s Comprehensible Output Hypothesis, I will then proceed to suggest ways in which Foreigner Discourse modifications might facilitate SLA. I shall also review a number of studies dealing with (un)modified input-SLA relationships, among which are Long’s (1985) and Gass & Varonis’ (1994) study. Due to the constraints imposed upon this essay, I will of course be drawing upon a very small selection of the pertinent literature.
Krashen’s Input Hypothesis
In many approaches to SLA, input is seen as a highly important factor in acquisition. Krashen’s Input Hypothesis, which largely predated Universal Grammar approaches to SLA research, continues to make the strongest claim about the role of the linguistic environment in SLA. It states that L2 input must both be comprehensible and be at a stage slightly beyond the learner’s current competence (i + 1) in order to be acquired (Krashen 1982: 16). An added stipulation is that the learner be emotionally receptive to the input. Thus, comprehensible input, delivered in a low (affective) filter situation, is held to be not only a necessary condition for acquisition, but actually causing acquisition[1] (1982: 33). Within Krashen’s framework, if input is not comprehensible, it cannot serve acquisition at all. Along with this is the claim that modifications in speech addressed to learners, such as caretaker speech in L1 acquisition, or foreigner-talk and teacher-talk in L2 acquisition, provide ideal input for i + 1, because they increase input comprehensibility.[2]
Long’s Interaction Hypothesis
Particularly over the past two decades, related work on input comprehensibility has focused on the nature of the linguistic environment available to learners and its relationship to L2 acquisition. Recent work has taken as basic the notion that conversational interaction in a second language forms the basis for L2 development rather than being only a forum for practice of specific language features (Wagner-Gouch & Hatch 1975). Long (1985: 378) argues deductively that modifications to discourse structure, such as modified input and negotiated interaction, promote comprehension of input, and comprehensible input promotes acquisition; therefore, interaction promotes acquisition. Thus, while in Krashen’s framework, comprehensible input itself remains the main causal variable in the language acquisition process, Long (1996), in his updated version of the Interaction Hypthesis, claims that modified interaction, or “negotiation for meaning”, is a crucial element in the SLA process. In particular he argues that negotiation work that triggers adjustments by the native speaker or more competent interlocutor facilitates acquisition “because it connects input, internal learner capacities, particularly selective attention, and output in productive ways.” (451-2)
Swain’s Comprehensible Output Hypothesis
As we can see then, both Krashen and Long regard comprehensible input as a source of acquisition. However, some researchers such as Sharwood-Smith (1986) have suggested that comprehensible input obtained through interaction forms part of an overlapping but not identical set with input which leads to acquisition. In Swain’s (1995) Comprehensible Output Hypothesis, output is also crucial in SLA. Her claim was based initially on observations of a lack of target-like productive skills of French immersion students in Canada (Swain 1985), whose grammatical performance, in spite of seven years of putatively comprehensible input in the target language, was not equivalent to that of native speakers. However, they performed similarly to native speakers on those aspects of discourse and sociolinguistic competence which do not rely heavily on grammar for their realisation. Swain (1985: 247) suggests that the diet of non-interactive, extended discourse received by the immersion students may account for their strong performance in this domain relative to native speakers. She argues that language production allows learners not only to create awareness of language knowledge gaps, experiment with language forms and structures, and receive feedback about language use, but also moves them from a primarily semantic analysis of language (as takes place in comprehension) to a syntactic use; thus, output would seem to have a potentially significant role in the development of syntax and morphology (1995: 128). What the immersion data suggest is that comprehensible input will contribute differentially to SLA depending on the nature of that input and the aspect of SLA one is concerned with.
[...]
[1] Corder (1967) made an important distinction between input and intake: “presenting a certain linguistic form to a learner […] does not necessarily qualify it for the status of input, for the reason that input is ‘what goes in’ not what is available for going in, and we may reasonably suppose that it is the learner who controls this input, or more properly his intake.” (165)
[2] However, as Mitchell & Myles (1998: 126) point out, “it is not clear how the learner’s present state of knowledge (‘i’) is to be characterised, or indeed whether the ‘i+1’ formulation is intended to apply to all aspects of language, from lexis to phonology and syntax.” Thus, the resulting impossibility of finding indisputable evidence for this is a severe shortcoming to the hypothesis’ usefulness in SLA theory building.