Swain's output hypothesis

Theory · All theories
Evidence · 12 claims (6 for, 6 mixed) · 6 studies (4 qualitative, 1 causal, 1 review), q1–q3 · 0 of 6 report an effect size · 12 claims rest on one study

Description

The articles date the output hypothesis to Swain (1985) and present it as an answer to Krashen's input hypothesis. It grew from studies of French immersion in Canada. Students there had received years of comprehensible input, were fairly fluent, and still showed less grammatical and syntactic proficiency than their native-speaking peers (eric-ej1095572, eric-ej1195386). Swain explained this partly by how little they produced. As Yu Liming (eric-ej420159) reports it, there was "little social or cognitive pressure" on them to say exactly what they meant, because their current output already got the message across. The hypothesis states that the learner must be "pushed toward the delivery of a message that is not only conveyed, but that is conveyed precisely, coherently, and appropriately." In Ellis's summary (eric-ed338037), output forces precision, obliges learners to process language syntactically rather than only semantically, and lets them revise their hypotheses.

Later work, which the articles attribute to Swain (1995), names three functions of output. In the noticing or triggering function, producing language can make learners aware of a gap "between what they want to say and what they can say". In the hypothesis-testing function, learners try out their interlanguage against the feedback they get from interlocutors. In the metalinguistic function, learners reflect on their own language use, which helps them internalise forms and rules. Geist (eric-ej1155586) adds that noticing in output can be triggered by an interlocutor's feedback or by the learner. Learners then either produce modified output or go back to the input to look for a solution. Yu Liming's diary study extends the idea to one self-directed adult learner, with intake moving in a circle from input to output and back to input.

The articles do not agree on how well the hypothesis is supported. Ellis (eric-ed338037) called the evidence for it "circumstantial", with no direct evidence linking pushed output to acquisition. Geist reports that studies since then (Izumi and colleagues among them) have shown that producing output can lead to noticing. She adds that "much research still has to be done to prove" that noticing in output leads to acquisition. VanPatten (eric-ej912912) says of whether speaking triggers grammatical development that "the jury is out". Ponniah (eric-ej1071042), writing from Krashen's side, argues that the three functions concern conscious learning, not acquisition. Wei (eric-ej1195386) reports Zhao Ping's classroom writing study, which found evidence for the noticing and hypothesis-testing functions but "no apparent" metalinguistic function. Wen's "output-driven, input-enabled" hypothesis (eric-ej1144345) builds on the output hypothesis but is a theory of teaching, not of acquisition: it makes output both the driver and the goal of tertiary EFL teaching in mainland China.

Implications

Context

• Output works "under some circumstances" (Swain 1995, quoted in eric-ej1095572). Learners must actually be pushed: tasks where a vague message still succeeds leave them no reason to reprocess the grammar.
• Written output with note-taking, comparison with a model and rewriting is how the fetched studies of the noticing function set up the push (eric-ej1095572, eric-ej1155586).
• In both writing studies, the noticing that output triggered was mostly about vocabulary rather than grammar.

Target Learners

• Learners who already understand a lot but produce inaccurate language: immersion students, and advanced adult learners (eric-ej420159).
• Adult and university EFL writers, in the classroom applications from Saudi Arabia and China (eric-ej1095572, eric-ej1195386, eric-ej1144345).

Target Learning Objectives

• Grammatical accuracy and sociolinguistic appropriateness in production, beyond comprehension and fluency.
• L2 writing, including lexical and grammatical problem-solving while writing.

Claims

• Comprehensible input alone is insufficient for acquisition of high levels of grammatical proficiency [+M]
• Mere exposure to the target language does not automatically lead to native-like communicative competence in adult learners [+W]
• Pushing for comprehensible output drives understanding and assimilation of comprehensible input [+W]
• Native speakers rarely supply negative input, so learners must infer or actively seek it by pushing for comprehensible output [+W]
• Syntactic rather than semantic processing of input is essential for correcting incomprehensible output [+W]
• Review-reported experiment found output performed noticing and hypothesis-testing functions but no apparent metalinguistic function in L2 writing [~M]
• While producing L2 paragraphs, an EFL learner's self-noticed problems were overwhelmingly lexical (about 79%) rather than grammatical (about 21%) [~M]
• Rewriting after noticing gaps and comparing to models produced modified output that solved more lexical gaps than grammatical problems [~M]
• Comparing original texts to model paragraphs prompted noticing that was again focused on lexical features [~W]
• Notetaking as a feedback tool facilitates resolution of lexical problems but leaves grammatical problems unresolved, suggesting reformulation as an alternative [~W]
• Lexical LREs dominate grammar-related LREs in young learners' L2 writing noticing [~M]

The evidence by question

• Does writing in the second language make learners notice gaps in their own language, which gaps do they notice, and does comparing with a model help them close them? Review-reported experiment found output performed noticing and hypothesis-testing functions but no apparent metalinguistic function in L2 writing [~M]
• Does pushing learners to produce output, and to modify it after feedback, develop their second-language grammar more than input or a supplied correction alone? Producing modified output involving developmentally advanced question forms in response to negative feedback is the only significant predictor of ESL question development [+M]

Related Theories

• Swain's output hypothesis with three functions of output in second-language acquisition
• Swain's Comprehensible Output Hypothesis with a circular input-output-intake model
• Output-driven, input-enabled hypothesis: output as both driving force and objective of EFL teaching — a teaching hypothesis built on this one
• Krashen's input hypothesis and Monitor Model of second language acquisition — the account it was proposed against
• Long's interaction hypothesis
• Schmidt's noticing hypothesis — the noticing function carries noticing from input to output
• Language related episodes (LREs) as the operationalization of noticing in L2 writing
• The Comprehension Hypothesis: comprehensible input as the crucial ingredient of second language acquisition
• Interaction hypothesis (Long): negotiation of meaning during interaction generates comprehensible input that drives L2 acquisition
• Structure-building trigger model: adult L2 learners scan input for cues but use different triggers than children

Examples

• Give students more chances to produce output and communicate during writing, with varied assignment types and clear sufficient feedback
• Three-stage writing task: write with problem notes, compare to a model paragraph, then rewrite
• Diary recording with recall, reflection and delayed verification as a self-directed learning strategy
• Start grammar instruction from the problems learners notice in their own written output
• Increase Student Language Production
• Modeling feedback: an original native-speaker text used as a comparison model
• Design interaction so learners modify their utterances in response to negative feedback

Key Sources

• Swain, M. (1985). Communicative competence: Some roles of comprehensible input and comprehensible output in its development. In S. M. Gass & C. G. Madden (Eds.), Input in second language acquisition (pp. 235–253). Newbury House.
• Swain, M. (1993). The output hypothesis: Just speaking and writing aren't enough. The Canadian Modern Language Review, 50(1), 158–164. doi:10.3138/cmlr.50.1.158
• Swain, M. (1995). Three functions of output in second language learning. In G. Cook & B. Seidlhofer (Eds.), Principle and practice in applied linguistics (pp. 125–144). Oxford University Press.
• Izumi, S., Bigelow, M., Fujiwara, M., & Fearnow, S. (1999). Testing the output hypothesis: Effects of output on noticing and second language acquisition. Studies in Second Language Acquisition, 21(3), 421–452. doi:10.1017/S0272263199003034
• Izumi, S. (2003). Comprehension and production processes in second language learning: In search of the psycholinguistic rationale of the output hypothesis. Applied Linguistics, 24(2), 168–196. doi:10.1093/applin/24.2.168
• Krashen, S. D. (1998). Comprehensible output? System, 26(2), 175–182. doi:10.1016/S0346-251X(98)00002-5

Page metadata

Field	Value
type	theory
title	Swain's output hypothesis
description	Swain's claim that comprehensible input is not enough for second language acquisition: learners must be pushed to produce precise, coherent and appropriate output, which serves noticing, hypothesis-testing and metalinguistic functions.
status	review
canonical	True
generated	by claude/unspecified on 2026-09-30
trust	unverified — no one has confirmed this page's content

Frontmatter is the page's machine-readable record — see the Schema & Guide.