Features of Unplanned Spoken Discourse
Unplanned spoken discourse is the language of real-time talk: conversation, interview, vox-pop, service encounter, casual narrative. Speakers compose under online cognitive load, monitor uptake, and revise mid-utterance. The resulting surface differs systematically from written prose and from the Scripted Listening Text that dominates ELT coursebooks. The features below are not errors or sloppiness; they are the structural fingerprint of speech, well documented in spoken corpora, chiefly CANCODE (the five-million-word Cambridge–Nottingham Corpus of Discourse in English assembled by Ronald Carter and Michael McCarthy) and the spoken sub-corpus of the Longman Grammar of Spoken and Written English.
The features
| Feature | Definition | Example |
|---|---|---|
| False start | Speaker begins, abandons, restarts | "I went — I mean, we went to the meeting." |
| Filler / hesitation | Voiced pause holding the turn | "It was, um, about an hour." |
| Self-repair | Mid-utterance correction | "She lives in Manchester, sorry, in Liverpool." |
| Ellipsis | Omission of recoverable material | "Coffee?" / "Going home now." |
| Repetition | Word or phrase repeated for fluency or emphasis | "I I really wanted to go." |
| Tail | Right-dislocated noun phrase clarifying a pronoun | "He's a nice bloke, that brother of yours." |
| Head | Left-dislocated noun phrase introducing a referent | "That brother of yours, he's a nice bloke." |
| Vague language | Imprecise reference forms | "stuff", "thing", "kind of", "or whatever" |
| Discourse marker | Interactional signal organising talk | "right", "well", "anyway", "so", "you know" |
| Backchannel | Listener token signalling attention | "mmhm", "yeah", "right" |
| Contraction / reduction | Reduced phonological forms | "gonna", "wanna", "she'd've", weak-form of /əv/ |
Corpus evidence
Ronald Carter (1998), drawing directly on CANCODE, demonstrates that ordinary conversation is structured around precisely these features rather than around the well-formed sentences of written grammar. Tails, heads, ellipsis, and vague language are not register-specific decorations but core organising resources of spoken English across speakers, regions, and registers within the corpus. McCarthy (1998) extends the argument to applied linguistics methodology: a grammar derived from written data systematically misdescribes the spoken language learners need to comprehend and produce.
Biber and colleagues (1999) deliver the comparable finding from the Longman Grammar of Spoken and Written English. Their conversation sub-corpus shows distinctive patterning in personal pronouns, contractions, ellipsis, and reduced subordinate clauses, with frequencies often differing from written registers by an order of magnitude. Carter and McCarthy's Cambridge Grammar of English (2006) builds the descriptive grammar that incorporates these findings.
Implications for listening materials
The pedagogic problem follows from the corpus findings. Coursebook dialogue, examined against spoken corpora, under-represents every feature listed above. Vague language is rare; backchannels are absent or token; tails and heads barely occur; reductions are softened by over-clear voice acting. Carter (1998), Gilmore (2007), and Field (2008) converge on the consequence: learners trained on scripted dialogue meet real spoken English as a foreign register on first contact, even at advanced levels. Spoken-grammar features are not stylistic add-ons that learners pick up later; they are the perceptual and structural environment in which content words appear, and decoding them is the task of real-time listening.
The corollary for materials development is twofold. Receptively, learners need exposure to texts that retain these features (Unscripted Listening Text or carefully briefed Semi-scripted Listening Text) because perception of vague language, ellipsis, and reductions does not transfer from cleaner input. Productively, the spoken-grammar tradition argues for explicit attention to high-frequency features (discourse markers, tails, ellipsis) at appropriate levels rather than treating them as advanced refinements.
References
- Biber, D., Johansson, S., Leech, G., Conrad, S., & Finegan, E. (1999). Longman Grammar of Spoken and Written English. Pearson Education.
- Carter, R. (1998). Orders of reality: CANCODE, communication, and culture. ELT Journal, 52(1), 43–56. https://doi.org/10.1093/elt/52.1.43
- Carter, R., & McCarthy, M. (2006). Cambridge Grammar of English. Cambridge University Press.
- Field, J. (2008). Listening in the Language Classroom. Cambridge University Press.
- Gilmore, A. (2007). Authentic materials and authenticity in foreign language learning. Language Teaching, 40(2), 97–118.
- McCarthy, M. (1998). Spoken Language and Applied Linguistics. Cambridge University Press.