One of the distinctive features of A-Level English Language is the analysis of spoken language data — transcripts of real conversations, interviews, radio programmes, or everyday interaction. Many students find this harder than analysing written texts, partly because spoken language looks “messier” on the page and partly because the vocabulary for describing it is different. Here’s a framework for approaching spoken data with confidence.
Start with the Context
Before looking at specific features, establish the context: Who is speaking to whom? What is the relationship between the participants (equal, unequal, familiar, professional)? What is the setting and purpose of the interaction? This context shapes every linguistic choice in the data, and your analysis should constantly return to it.
Key Features of Spoken Language to Analyse
Prosodic features: Intonation, stress, pace, pause, volume — these carry meaning in spoken language that punctuation carries in written language. Where transcripts include prosodic notation, engage with it: a rising intonation on a statement can signal uncertainty, seek confirmation, or function as a question.
Non-fluency features: False starts (“I was — she was already there when I”), filled pauses (“um”, “er”, “uh”), repetitions, self-corrections — these are normal features of spontaneous speech, not signs of unintelligence. Their presence tells you about the cognitive demands of the utterance; their absence may indicate rehearsed or planned speech.
Discourse markers: “Well…”, “So…”, “Right…”, “Okay…”, “I mean…”, “You know…” — these structure the conversation, signal topic shifts, and manage the relationship between speakers. They’re often dismissed as meaningless filler but they’re analytically significant.
Turn-taking: How do speakers take and yield turns? Are there overlaps, interruptions, or long pauses? Who speaks more? Who asks more questions? Turn-taking patterns reflect power dynamics and social relationships.
Deixis: Language that only makes sense in context — “here”, “this”, “that one over there”, “yesterday”, “him”. Spoken language is full of deixis because speakers share a physical context. It’s a marker of the situatedness of spoken interaction.
Avoid These Common Errors
- Treating non-fluency features as mistakes rather than as analytically significant features of spontaneous speech
- Applying written language categories uncritically to spoken data (spoken language has its own grammar)
- Describing what speakers say without analysing how they say it and what effect that creates
- Ignoring prosodic notation when it’s provided — this is evidence, and it should be used
Spoken language analysis — alongside all the written text frameworks — is taught throughout Thorogood English’s A-Level English Language programme. Join the waiting list for January 2027.
Leave a Reply