Error Analysis
SLAEA
Error Analysis (EA) is the systematic study of errors made by second language learners, with the aim of understanding the processes underlying L2 acquisition. The field was established by S. Pit Corder's seminal 1967 paper "The Significance of Learners' Errors" (published in International Review of Applied Linguistics, 5, 161–170), which argued that learner errors are not signs of failure but evidence of an active, rule-governed system — what Selinker (1972) would later call Interlanguage.
Historical Context
EA emerged as a reaction against Contrastive Analysis, which dominated applied linguistics in the 1950s–60s. Where CA predicted errors by comparing L1 and L2 systems before learners produced anything, EA worked after the fact — collecting actual learner output and analyzing it. Corder's insight was that errors reveal the learner's "built-in syllabus": the internal system they are constructing, which may follow a different sequence from the teacher's syllabus.
Corder's Key Claims (1967)
- Errors are systematic, not random. They reflect the learner's current state of knowledge — their transitional competence.
- Errors are significant in three ways:
- To the teacher: they show how far the learner has progressed and what remains to be learned.
- To the researcher: they provide evidence of how language is acquired.
- To the learner: they are a device for testing hypotheses about the target language.
- Errors ≠ mistakes. Corder distinguished systematic errors (reflecting gaps in competence) from unsystematic mistakes (slips caused by performance factors). See Error vs Mistake vs Slip.
Stages of Error Analysis
Corder (1974) and later Ellis (1994) formalized EA into a five-stage procedure:
| Stage | Description |
|---|---|
| 1. Collection | Gather a corpus of learner language (written or spoken) |
| 2. Identification | Identify which forms deviate from the target language norm |
| 3. Description | Describe the errors linguistically (omission, addition, misordering, misformation) |
| 4. Explanation | Determine the source — interlingual (L1 transfer), intralingual (Overgeneralisation), or other causes |
| 5. Evaluation | Judge the gravity of errors — which ones impede communication most? |
Error Categories
By linguistic level
- Phonological — mispronunciation of sounds, stress, intonation
- Morphological — incorrect inflections (goed, childs)
- Syntactic — word order, agreement, clause structure
- Lexical — wrong word choice, false cognates
- Discourse — inappropriate cohesion, register mismatch
- Pragmatic — inappropriate speech acts, politeness violations
By source (Richards 1974)
| Source | Description | Example |
|---|---|---|
| Interlingual | L1 transfer | Vietnamese learner drops articles |
| Intralingual: overgeneralisation | Extending a rule beyond its domain | He goed (regular past applied to irregular) |
| Intralingual: ignorance of rule restrictions | Applying a rule in contexts where it does not apply | He made me to go |
| Intralingual: incomplete application | Not fully applying a rule | He come yesterday (past tense not marked) |
| Intralingual: false concepts | Misunderstanding a distinction | Using was as a general past marker: He was go |
By gravity (Burt & Kiparsky 1972)
- Global errors — affect overall sentence structure and impede comprehension (The man who I saw him left)
- Local errors — affect a single element but the message remains clear (He have two book)
Criticisms of EA
- Avoidance is invisible. If learners avoid structures they find difficult, their errors never appear in the data — but the difficulty is real. Schachter (1974) demonstrated this with relative clauses: Japanese and Chinese learners produced few errors because they avoided the structure entirely.
- Ambiguity of classification. A single error may have multiple sources. Is He no like coffee (from a Spanish speaker) transfer from Spanish No me gusta or intralingual simplification?
- Focus on product, not process. EA describes what errors occur but struggles to explain the cognitive processes behind them.
- Ignores correct forms. EA only examines what learners get wrong, not what they get right — giving a skewed picture of competence.
Why It Matters for Teaching
- Diagnostic tool. Error patterns reveal what learners have and have not acquired, helping teachers prioritize Corrective Feedback and lesson focus.
- Informs syllabus design. Recurring errors across a class suggest gaps that need targeted instruction — connecting to Form-Focused Instruction.
- Reframes errors positively. Corder's legacy is that errors are evidence of learning, not failure. This attitude shift underpins communicative approaches that tolerate errors during fluency activities while addressing them during accuracy work.
- Complements CA. While EA replaced CA as the dominant paradigm, modern practice uses both: CA to predict likely areas of difficulty for a given L1 group, EA to confirm what actually proves difficult.
Key References
- Corder, S.P. (1967). The significance of learners' errors. IRAL, 5, 161–170.
- Corder, S.P. (1974). Error Analysis. In J. Allen & S.P. Corder (Eds.), The Edinburgh Course in Applied Linguistics, Vol. 3. Oxford University Press.
- Corder, S.P. (1981). Error Analysis and Interlanguage. Oxford University Press.
- Richards, J.C. (1974). Error Analysis: Perspectives on Second Language Acquisition. Longman.
- Burt, M. & Kiparsky, C. (1972). The Gooficon: A Repair Manual for English. Newbury House.
- Schachter, J. (1974). An error in error analysis. Language Learning, 24(2), 205–214.
- Ellis, R. (1994). The Study of Second Language Acquisition. Oxford University Press.