The Idiom Principle
Do we build sentences, or reach for ready-made ones?
The idiom principle is John Sinclair’s claim that speakers mostly choose ready-made, partly fixed phrases as single units, using word-by-word grammatical choice, the open-choice principle, as the fallback.
Why does a fluent speaker say I’m afraid I can’t make it and not I fear that I am unable to come? Both are grammatical, but only one sounds like something people actually say. Pawley and Syder (1983) called this the puzzle of nativelike selection: of all the grammatical ways to say something, fluent speakers pick a small, conventional set.
John Sinclair (1991) offered an explanation built on two principles. Under the open-choice principle, text is assembled slot by slot, and grammar is the only limit on which word comes next. Under the idiom principle, speakers draw on a large stock of ready-made or partly ready-made phrases, such as of course or set eyes on, that work as single choices even though they could be broken into parts. Sinclair argued that the idiom principle is the default mode, with open choice as the fallback.
Erman and Warren (2000) tested the idea by marking prefabs, word combinations that native speakers favour over equally possible alternatives. In their English samples, prefabs filled about 55 per cent of the text, a little more in speech than in writing. Wray (2002) gathered these units under the heading formulaic language: sequences that appear to be stored and retrieved whole. The psycholinguistic picture is less tidy. Siyanova-Chanturia and Martinez (2015) review evidence that familiar multiword expressions are processed faster than new phrases, but argue that this advantage does not prove they are stored as single wholes.
For teachers, the message is that fluency rests partly on a store of chunks and collocations, not only on rules plus words. That is the central bet of the Lexical Approach.
In actionOne sentence, mostly prefabs
/I’m afraid/ /I can’t make it/ to the meeting on Friday, but /I’ll let you know/ /as soon as/ I can.
An invented sentence, marked in the spirit of Erman and Warren’s analysis: slashes enclose prefabs, and the remaining words are chosen one by one.
In the classroom
- 1
Ask learners to underline sequences in a reading text that feel like one choice (on the other hand, it turns out that), then record them whole, each with a short example sentence.
- 2
Draw attention to useful chunks in the listening and reading texts you already use. In a small study by Boers and colleagues (2006), learners taught this way were judged more proficient in interviews.
- 3
Build speaking routines around frames such as The thing is, … or What I mean is …, and recycle them across several lessons so they come quickly under time pressure, as in task repetition.
Myth and reality
Formulaic language means idioms like *kick the bucket*, a decorative extra for advanced learners.
Most prefabs are plain and frequent: I don’t know, in order to, at the moment. Erman and Warren estimated that they make up about half of ordinary text, so they belong in teaching from the start.
Reflect
If about half of what your students hear is prefabricated, how much of your syllabus still treats English as single words plus grammar rules?
Worth a minute before your next lesson, or a conversation in the staffroom.
SourcesChecked against Crossref and library records
Sinclair set out the idiom and open-choice principles in Corpus, Concordance, Collocation (1991), drawing on his corpus work at Birmingham. Pawley and Syder’s 1983 puzzles of nativelike selection and fluency had pointed in the same direction.
Sinclair, J. (1991). Corpus, concordance, collocation. Oxford University Press.
Erman, B., & Warren, B. (2000). The idiom principle and the open choice principle. Text, 20(1), 29–62.
The study behind the famous estimate that prefabs make up about half of English text, with a clear method for spotting them.
Wray, A. (2002). Formulaic language and the lexicon. Cambridge University Press.
The standard book-length account of formulaic language: what it is, why we use it, and how learners acquire it.
Siyanova-Chanturia, A., & Martinez, R. (2015). The idiom principle revisited. Applied Linguistics, 36(5), 549–569.
Tests Sinclair’s idea against psycholinguistic evidence and asks what faster processing of chunks really shows.
Boers, F., Eyckmans, J., Kappel, J., Stengers, H., & Demecheleer, M. (2006). Formulaic sequences and perceived oral proficiency: Putting a Lexical Approach to the test. Language Teaching Research, 10(3), 245–261.
A small classroom experiment linking attention to chunks with how proficient learners sound.
Boers, F., & Lindstromberg, S. (2012). Experimental and intervention studies on formulaic sequences in a second language. Annual Review of Applied Linguistics, 32, 83–110.
A review of which teaching treatments for formulaic sequences have actually been tested, and how well they work.
ConnectionsSee it in the graph
- Semantic Prosodynot yet writtenSinclair’s units of meaning carry attitude as well as form.
- The Lexical Approachnot yet writtenA teaching approach built on the idea that language is largely chunks.
- Task-Based Language TeachingRepeating tasks gives learners chances to retrieve chunks quickly.
- Lexical Bundlesnot yet writtenFrequent word sequences identified automatically, one corpus route to formulaic language.
- Construction GrammarAlso treats stored form–meaning pairings, from fixed phrases to patterns, as basic.
Know a teacher who’d like this?
Send them the card. It’s made for Facebook, Zalo, LinkedIn and Instagram, and links back here.