SLA Theories – An Overview

1950s–60s

Contrastive analysis hypothesis

Lado (1957) · Behaviourist learning theory

Proposed that where the learner’s L1 and the target L2 are structurally similar, acquisition will be easy; where they differ, acquisition will be difficult. A systematic comparison of two languages should therefore predict which features will cause difficulty, what errors will be produced, and how persistent those errors will be.

Rooted in behaviourist psychology: language learning is habit formation, and L1 habits interfere with new L2 habits.

1970s

Creative construction & interlanguage

Selinker (1972) · Corder · Nemser

L2 acquisition is a creative construction process: the learner analyses input, forms hypotheses about structure, tests them, and revises. The learner’s developing linguistic system — distinct from both the L1 and the target L2 — is called the interlanguage. Errors are not failures but evidence of an active grammar.

Three strands of evidence drove this shift away from behaviourism:

Error data — "He cans" / "teached" Silent period — learning without producing Morpheme studies (Dulay & Burt 1974) — fixed acquisition order

Innatism & universal grammar

Chomsky · UG-based SLA theories

Nativist approaches argue that input is too impoverished to account for the rich grammatical knowledge learners ultimately display (“poverty of the stimulus”). Certain aspects of language must therefore be innate. Input functions only as a trigger that activates an internal language faculty, not as the source of that knowledge.

A range of UG-based SLA theories developed in the 1970s–80s to explore how much of universal grammar remains accessible to L2 learners, and how L1 parameter settings transfer into L2 acquisition.

1980s

Krashen’s theory of SLA

Krashen (1981, 1985) · Five-hypothesis model

The most comprehensive SLA model of the period. Krashen occupies a middle ground — the weak innate hypothesis — and attempts to explain the entire acquisition process through five interrelated claims.

Input (i+1)

Acquisition requires comprehensible input one step beyond the learner’s current level.

Natural order

Grammatical features are acquired in a predictable sequence, unaffected by teaching order.

Acquisition vs. learning

Subconscious acquisition and conscious learning are entirely separate; learning cannot become acquisition.

Monitor

The learned system acts only as a conscious editor, requiring time, rule knowledge, and a focus on form.

Affective filter

Anxiety, low motivation, or negative attitudes block acquisition even when input is available.

Long’s interaction hypothesis

Long (1983) · Modified input hypothesis

Extends Krashen’s input hypothesis by asking how input becomes comprehensible. Native speakers consistently modify their speech in conversation with low-proficiency learners, and non-native speakers actively negotiate for comprehensible input.

Linguistic simplification Slower rate of delivery Comprehension checks Repetition & paraphrase

Interaction ensures input is dynamically calibrated to the learner’s developing competence — something exposure alone cannot guarantee.

1990s

Noticing hypothesis & input→intake

Schmidt (1990) · Schmidt & Frota (1986) · Swain (1998)

Comprehensible input is necessary but not sufficient. Learners must consciously attend to linguistic form for acquisition to take place — converting input into intake. Mere exposure, however extensive, does not guarantee learning.

Noticing the gap

Schmidt & Frota (1986). Learner perceives a mismatch between a native-speaker form in the input and their own interlanguage.

Noticing the hole

Swain (1998). Learner realises during production that they lack the resources to express what they want to say.

Both lead to awareness of a deficiency in the interlanguage — the trigger for potential revision of the learner’s internal grammar.