Analysing spoken language: reading transcription conventions and the features of speech (fillers, false starts, turn-taking, prosody, deixis, spontaneity) and comparing speech with written texts.

What this dot point is asking

Unit 4 always includes a spoken text among the three, presented as a transcript. Speech has its own conventions and its own features, distinct from writing, and analysing it well means applying the right framework rather than treating it as flawed writing. The skill feeds both the close analysis (AO1, AO2) and the comparison with the written texts (AO4).

The answer

Read the transcript on its own terms

A transcript is not a failed essay. Judging it by written norms ("the grammar is wrong") misses the point and loses marks.

The features of speech

• Fillers and hesitations - signal real-time planning, hesitation or stance; the speaker buying thinking time.
• False starts and self-repair - show thought being revised mid-utterance, a marker of spontaneity.
• Turn-taking and overlaps - in dialogue, reveal how speakers co-construct meaning, who holds the floor, who interrupts.
• Deixis and context-dependence - speech leans on shared context ("over there", "this one"), so meaning is tied to the situation.
• Prosody - where marked, stress, pitch and pace carry meaning that punctuation carries in writing.

Use the spoken-written contrast

The richest comparative point is the contrast between spoken spontaneity and written craft. The written texts are planned, edited and structured; the spoken text is unfolding, interactive and context-bound. Naming this difference (and any surprising overlaps) gives you a strong AO4 connection rooted in the mode of production itself.

Examples in context

Reading a transcript feature for meaning. Suppose a transcript records: "it was er (.) it was absolutely terrifying I mean you you don't expect it do you." A weak answer marks this as ungrammatical and repetitive. The right framework reads it as meaningful spontaneous speech. The filler "er" and the micro-pause "(.)" capture the speaker planning in real time and reaching for the right word, which conveys the difficulty of articulating a frightening experience. The false start and repetition "it was... it was" enact hesitation, then the intensifier "absolutely terrifying" lands with more force for having been delayed. The discourse marker "I mean" signals the speaker clarifying or intensifying, and the tag question "do you" with the repeated "you you" invites the listener's agreement, showing speech co-constructing meaning interactively. None of this is error: every feature reveals spontaneity, stance or interaction. Comparing this with a written text's edited, complete syntax on the same subject, you can argue that the transcript's very non-fluency conveys an authenticity and immediacy the polished written text cannot, which is the comparative payoff AO4 rewards.

Try this

Q1. Why are fillers and false starts not errors in speech? [3 marks]

• Cue. Spontaneous speech is unplanned and unedited, so these features are normal and reveal real-time planning, hesitation or stance.

Q2. Name two features distinctive to spoken language. [2 marks]

• Cue. Any two of: fillers, false starts and self-repair, turn-taking and overlaps, deixis, back-channelling, discourse markers, marked prosody.

Q3. Analyse how a spoken transcript creates meaning and compare its methods with a written text. [20 marks]

• What the marker wants. Speech features named precisely and read as meaningful, the interaction analysed where present, and the spoken-written contrast used as a grounded comparative point.