Negative Evidence

Definition:

Negative evidence in second language acquisition refers to any information a learner receives indicating that a particular form is ungrammatical, inappropriate, or incorrect in the target language—as opposed to positive evidence, which consists of grammatical examples of target-language use. The question of whether negative evidence is necessary for successful SLA, or whether positive evidence (comprehensible input) alone is sufficient, has been one of the most debated issues in the field.

In-Depth Explanation

Positive vs. negative evidence:

Positive evidence: Grammatical utterances in the input that show learners what IS possible in the language (comprehensible input, authentic materials, native-speaker conversation).
Negative evidence: Information showing what is NOT grammatical—explicit error correction, recasts, metalinguistic feedback, grammaticality judgments, grammar rules.

Types of negative evidence:

Explicit negative evidence: Direct correction (“No, that’s wrong; you should say X”); metalinguistic explanation (“In Japanese, the particle は marks the topic, not the object”); red-marking of errors.

Implicit negative evidence: Recasts (teacher rephrases a learner’s erroneous utterance correctly without explicitly saying it was wrong); clarification requests (“Sorry, what did you mean?”) that indirectly flag trouble without explicit correction.

The poverty of the stimulus and negative evidence:

Chomsky’s poverty of the stimulus argument (extended to SLA by White, 1987) holds that the positive evidence available to learners is insufficient to explain the overhypothesization problem: if learners overgeneralize (e.g., “I goed”), they need negative evidence to retreat from the overgeneration. Without negative evidence, they cannot in principle learn that something is impossible in the TL.

Krashen (1982) famously argued that negative evidence (error correction) is “of limited value” for acquisition; only comprehensible input matters. Research has since substantially challenged this: Long (1996), Doughty & Williams (1998), and many others have demonstrated that corrective feedback—especially focusing on specific forms below the learner’s current awareness—can accelerate acquisition of grammatical targets that are not reliably deducible from positive evidence alone.

Recasts as implicit negative evidence:

Recasts (the teacher restates the learner’s erroneous utterance in correct target-like form) occupy a theoretically uncertain position: are they negative evidence? Studies disagree on whether learners consistently perceive recasts as corrections or as confirmations. Lyster & Ranta (1997) found recasts in French immersion classes rarely produced uptake (learner incorporation of the correct form), suggesting their value as negative evidence is limited when they are not clearly marked as corrections.

When negative evidence matters most:

Long (2007) and others argue negative evidence is particularly important for:

Features absent from the L1 that are therefore not reliably noticed in positive evidence (e.g., English articles for Japanese learners—Japanese has no articles, so learners systematically miss article occurrences in input)
Features that are low-frequency in authentic input but grammatically obligatory
Features where overgeneralization produces quasi-grammatical forms not distinguishable by meaning from correct forms

Japanese SLA context:

Japanese learners of English need negative evidence for:

Article use (a/the/zero; not deducible from input without massive exposure)
Subject pronouns (Japanese pro-drop; English non-pro-drop features require negative feedback for retreating from dropped subject)
Plural -s morpheme (absent in Japanese; frequently absent in learner English without systematic correction)

History

1965: Chomsky’s poverty of stimulus argument introduces the theoretical stakes of negative evidence.
1982: Krashen’s Input Hypothesis argues corrective feedback is ineffective; positive input suffices.
1987: White’s access to Universal Grammar paper raises the specific claim that negative evidence may be needed for parameter resetting.
1990s: Recasts debate (Lyster, Long, Nicholas et al.) empirically examines implicit negative evidence.
1998: Doughty & Williams (eds.) Focus on Form in Classroom Second Language Acquisition synthesizes research showing targeted corrective feedback benefits acquisition.
2006: Mackey & Goo’s meta-analysis confirms feedback effects in SLA.

Common Misconceptions

“Error correction disrupts fluency and harms learners.” Research shows appropriately timed, explicit negative evidence accelerates accuracy without harming fluency when pedagogically managed.

“Recasts are effective error correction.” Recasts are the least salience-marked form of negative evidence; learners often do not perceive them as corrections, limiting their efficacy.

“Positive input alone is always sufficient for L2 acquisition.” For some features (especially those absent from the L1 or low-frequency in input), positive evidence alone is demonstrably insufficient; negative evidence accelerates acquisition.

Criticisms

Whether learners need negative evidence to “retreat” from overgeneration is debated; some argue statistical learning from input patterns is sufficient without explicit correction.
The effects of negative evidence on long-term retention are underresearched.

Social Media Sentiment

The error correction debate is lively in language teacher communities. Krashen-influenced voices argue against explicit correction; communicative teachers worry about damaging student confidence. Japanese learner communities debate whether native-speaker conversation partners should correct errors or not—a direct experience of the negative evidence question. Most experienced learners conclude that targeted, judicious correction from a good tutor helps more than it harms.

Last updated: 2026-04

Practical Application

Request explicit feedback: When working with tutors or language partners, explicitly ask for correction of specific target features (e.g., “Please correct my article use every time”).
Written corrective feedback: For writing, explicit metalinguistic correction on specific L2 features the learner is ready for (Bitchener & Knoch, 2010) demonstrably improves accuracy.
Recasts vs. metalinguistic feedback: In one-on-one tutoring, metalinguistic feedback (“Actually in Japanese, this should be て form because…”) is more reliably perceived than implicit recasts.

Related Terms

Research

Krashen, S. D. (1982). Principles and Practice in Second Language Acquisition. Pergamon. [Summary: Argues that error correction (negative evidence) plays limited role in acquisition; comprehensible positive input is the central mechanism.]

Long, M. H. (1996). The role of the linguistic environment in second language acquisition. In W. C. Ritchie & T. K. Bhatia (Eds.), Handbook of Second Language Acquisition. Academic Press. [Summary: Interaction Hypothesis paper arguing that negative feedback via recasts and repair facilitates acquisition of specific features not available from positive input alone.]

Lyster, R., & Ranta, L. (1997). Corrective feedback and learner uptake: Negotiation of form in communicative classrooms. Studies in Second Language Acquisition, 19(1), 37–66. [Summary: Landmark study in French immersion classrooms; finds recasts produce less uptake than more explicit corrective moves; prompts major debate about implicit vs. explicit negative evidence.]

Mackey, A., & Goo, J. M. (2007). Interaction research in SLA: A meta-analysis and research synthesis. In A. Mackey (Ed.), Conversational Interaction in Second Language Acquisition. Oxford University Press. [Summary: Meta-analysis confirming that interaction and negative feedback studies show statistically significant effects on acquisition.]

Bitchener, J., & Knoch, U. (2010). Raising the linguistic accuracy level of advanced L2 writers with written corrective feedback. Journal of Second Language Writing, 19(4), 207–217. [Summary: Demonstrates that written corrective feedback for specific grammatical features (articles, prepositions) produces long-term accuracy improvements in advanced L2 writers.]